This PDF file contains the front matter associated with SPIE Proceedings Volume 12815, including the Title Page, Copyright information, Table of Contents, and Conference Committee list.

In the process of remote sensing image classification, remote sensing image itself has the characteristics of complex spatial relationship and diverse information in the image. This paper puts forward a remote sensing image classification method based on the residual network ResNet50, adding channel attention and spatial attention mechanism, replace the full connection layer. The experimental results prove its feasibility, and when the remote sensing image data does not meet the independent and identical distribution, it can still get better classification effect when it is input into the improved ResNet50, so as to further process the remote sensing image.

This paper proposes a shadow extraction method for GF-2 remote sensing images based on Stacking. The method uses Multiple Linear Regression Model as the second layer model to fuse the results of shadow extraction obtained by NDUI, NIS and C3 which are used as the first-layer base learners of the model. The comparison experiments demonstrate that the proposed method is superior to these traditional shadow operators such as NDUI, NSI and C3 in the accuracy of building shadow extraction. Finally, the height of the buildings is estimated using the shadow lengths and the imaging geometry method. The experimental results show that the average error of the building's height estimated is less than 1m.

Cities are primarily responsible for carbon dioxide peak and carbon neutrality. Understanding the carbon sequestration capacity within cities is a basic requirement for implementing the dual-carbon strategy. In this work, based on data from the Annual Land Use Change Investigation, we comprehensively used data from various sources to estimate carbon stock at patches scale in Wuhan, Hubei province, including Geographic National Conditions Monitoring, forest resource survey, wetland investigation, and soil survey. The spatial distribution characteristics of carbon stock were analyzed from two aspects: biomass carbon storage and soil carbon storage. And the relationships between different types of land use and carbon stock were also revealed in this work.

Traditional models for total phosphorus (TP) retrieval from remote-sensing images generally show low accuracy. Additionally, parameter adjustment and selection of parameter combinations of machine learning (ML) models exert significant influences on the regressive prediction effect and model performance. To solve these problems, this research proposed an extreme gradient boosting (XGBoost) model optimized by Bayesian optimization (BO), that is, BO-XGB. The optimal parameters of the model are sought automatically from a small sample size through BO, which shortens the training time. Taking Tiande Lake in Zhengzhou City (Henan Province, China) as the research region of interest, a BO-XGB model for TP retrieval is established based on GF1-WFV satellite data and TP water-quality data. Moreover, the accuracy of the model is compared with those of retrieval models established based on other four ML methods, namely, the XGBoost model, k-nearest neighbors (KNN), multilayer perceptron (MLP) and random forest (RF). Compared with the other four models, the BO-XGB TP retrieval model demonstrates the highest accuracy, with coefficients of determination R² , root mean square error (RMSE), and mean relative error (MRE) separately of 0.923, 2.15 × 10^-3 mg/L, and 1.81%. Finally, GF1-WFV satellite data are adopted to retrieve the spatial distribution of the TP concentration in Tiande Lake. The results show optimizing the XGBoost model using BO can significantly improve retrieval accuracy and the BO-XGB algorithm is more suitable for retrieving the TP mass concentration in Tiande Lake. The findings have implications for the inverse modelling of non-optical water quality parameters such as total nitrogen(TN).

The Three Rivers Source region is located in the high-altitude hinterland of the Qinghai-Tibet Plateau, and its ecological complexity and biodiversity render its ecological data invaluable. To address the issue of improving people's willingness to share data, an incentive system for ecological big data sharing based on contract theory and game theory is proposed in this paper. The combination of these two incentive mechanisms comprehensively considers various scenarios in actual situations and has a positive incentive effect on sharing ecological big data. The incentive mechanism based on contract theory formulates detailed contractual data sharing guidelines based on the quality, scarcity, and demand of shared data. This mechanism clarifies the accurate benefits and costs of data sharers and data demanders and effectively promotes the sharing of non-competitive data sets. Additionally, the incentive mechanism based on game theory designs the data sharing process as a game model. This approach maximizes the benefits between data sharers and data demanders, thus promoting the sharing of competitive data sets. Finally, the paper provides a specific implementation of the incentive system, and the results align with expectations.

As an interpolation model with strict mathematical form, the Kriging algorithm of geostatistics has the problem of negative weight when solving the weight value. As an important model of optimization probability, the minimum cross entropy algorithm is often used in insurance finance, power communication and other fields, but there is little research on weight optimization of spatial interpolation model. In this paper, the weight of negative weight points calculated by Kriging interpolation is used as a priori weight, and the optimization weight and corresponding optimization geological parameters (used to describe the values of elevation, mineral value, geothermal and other attributes) are obtained through the minimum cross entropy algorithm. By comparing the weight distribution and geological parameters before and after optimization based on the minimum cross entropy algorithm, the comparison results show that the Kriging model based on the minimum cross entropy algorithm eliminates the negative weight phenomenon and improves the estimation accuracy.

As the urbanization takes off, small and medium-sized cities play an increasingly vital role in the rapid expansion of cities. In order to settle the problems encountered in the built-up areas extraction in small and medium-sized cities, such as unclear boundaries of built-up area, confusion between built-up areas and villages or bare area, etc. This paper compares and analyzes different methods for built-up areas extraction in Xinzheng city, in view of NPP/VIIRS data and Sentinel-2B data, and proposes a method to extract built-up area using the fusion data of two types of data after overlaying. The results indicate that the extend of built-up areas may be entirely separated by using night light image, and Sentinel image can completely obtain the spatial distribution information of built-up areas in small and medium-sized cities; Compared with the single-source data extraction method, the extraction method combining two data sources has significantly improved the accuracy of the result. The overall accuracy is 94.60% and Kappa coefficient is 0.84, which is more appropriate for the built-up areas extraction in small and medium-sized cities.

Forest is one of the most important ecosystems on earth, and it plays an important role in maintaining ecological balance and protecting biodiversity. Therefore, the protection and construction of forests is one of the important contents of the implementation of ecological engineering in the Yellow River Basin (Gansu section), and the rapid and accurate extraction of forest resources has also become a research hotspot in this area. This study compared and analyzed the results of forest information extraction from multi-source remote sensing data based on feature fusion and decision fusion. and Kappa values were as high as 95.4% and 0.97, respectively. The distribution of forests extracted in mountainous areas, built-up areas, and farmland areas is basically consistent with visual interpretation. It shows that the forest extraction method of decision-level fusion of remote sensing data has high accuracy and reliability, and is suitable for high-precision forest acquisition in this study area.

In recent years, the ecological and environmental problems of the Yellow River Basin have gradually become our concern, and sedimentation at the mouth of the Yellow River can directly demonstrate the treating effect of the Yellow River. However, the research and monitoring of the Yellow River environment based on GIS technology is not in-depth enough, and only a few studies have a very effective interpretation of the future changes in the estuary of the Yellow River. Therefore, this study explores the inter-annual variation of sediment sedimentation in the estuary of the Yellow River by collecting Landsat satellite data and using QGIS software for data analysis, feature classification and inter-annual change. The linear fitting model of MATLAB software shows that the total sediment sedimentary area of the mouth of the Yellow River showed a downward trend in the decade from 2013 to 2022, which is closely related to natural factors such as ocean currents and waves and fortifications built by man. Therefore, the environment of the Yellow River has changed well in the past decade, and the governance will continue to strengthen on this basis.

With the aggravation of marine oil spill pollution, the safety of the marine ecosystem and the development of the marine industry are threatened. Synthetic aperture radar (SAR) has become an important tool for detecting oil spill pollution. In order to further improve the accuracy of oil spill detection, this paper adopts Mask R-CNN model for oil spill detection of SAR oil spill images and improves the non-maximum suppression algorithm. Through experimental validation, the SAR image oil spill detection method based on the improved Mask R-CNN model proposed in this paper successfully improved the S2AR oil spill detection precision rate, recall rate and F1 score, in which the oil spill detection accuracy reached 91.5%, an improvement of 9.1% compared with the traditional model. Therefore, the research in this paper has certain practical significance and value for improving the application and promotion of SAR images in marine oil spill detection.

Digital elevation model (DEM) is the core of geographic database and the basis of terrain analysis. Airborne LiDAR is widely applied in DEM research as it can quickly obtain the three-dimensional coordinates of the surface with high resolution and precision. Interpolation method is one of the important factors affecting the accuracy of DEM. This study explored the accuracy of different interpolation algorithms in DEM construction with consideration of different terrain complexity and laser point cloud data density. The results show that the density of point cloud data has lower impact on interpolation accuracy in low slope region than in rough region. The DEM accuracy in rough region is reduced seriously when the point cloud data extraction rate exceeds 10%. The accuracy of different interpolation methods is close in low slope area, while in rough area, different interpolation methods vary a bit large in DEM accuracy, and the Kriging method shows the best performance.

In the intelligent extraction of high-resolution remote sensing images, the scene classification of remote sensing images has great application prospects in many application fields. In past research, convolutional neural networks have shown great potential, and the introduction of the attention mechanism can further improve the feature representation ability in CNNs-based models. This letter proposes a new network architecture for remote sensing image scene classification. By introducing two attention mechanisms with different structures at different operation levels, the attention mechanism is merged at the level, and the performance of the basic network structure is effectively improved by using the new network architecture constructed. The overall accuracy show that the two attention structures can enhance the feature extraction effect overall and the details effectively. Experimental results on two public datasets, UC Merced and NWPU-RESISC45, show that the proposed model outperforms the current state-of-the-art methods in classification accuracy.

We propose a novel method, in this paper, to monitor the change of urban impervious surface area (ISA) using time series high-resolution remote sensing image. The distribution of urban ISA is an important indicator of urban ecological environment. The extraction of urban ISA by remote sensing technology is of great significance for understanding urbanization, hydrological cycle and urban heat island effect. Previous researches on ISA extraction mostly used medium and low-resolution images to study the change of ISA at the macro level. Because the resolution of data source is too low to detect the micro change of meter scale, it cannot meet the needs of sponge city and ecological city planning for fine modeling of urban underlying surface, and it is objectively necessary to carry out high-precision and fast ISA monitoring. For example, using high-resolution remote sensing images combined with machine learning algorithm, introducing time series features to construct a monitoring model. The difficulty of water surface monitoring based on time series is how to find the pseudo change, and how to extract the time-invariant spectral features is the difficulty. At the same time, time series images can solve the spectral confusion of surface features. The purpose of this study is to enhance the spectral difference between ISA and pervious surface by analyzing the time-variant spectral features extracted from time series high-resolution remote sensing satellite images. In this research, we took Hengqin New District, Zhuhai City, Guangdong Province of China as the study area. The area experienced rapid urbanization from 2009 to 2018. According to the ten-year time series of high-resolution remote sensing images, the ten-year ground features classification results of the study area acquired. The contribution of this paper is to use a method to extract time-series spectral features from time-series images to assist in the extraction of ISA. The average overall extraction accuracy of ISA in ten years increased from 87.8% to 91.1%.

The increase in marine development and transportation has led to a rise in oil spills. SAR can operate in all-weather and all-time, which allows for continuous monitoring of the Earth’s surface. However, the challenges of insufficient feature extraction from SAR images and the require large amounts of memory to store the parameters and weights, cannot be overlooked. We propose an improved lightweight model that utilizes SqueezeNet as the encoder within the U-Net framework to address the aforementioned issues. And three polarization features were extracted from dual-polarization Sentinel-1 image. The experimental results show that our improved model achieves higher accuracy and dice than the other two models while also having a smaller size. The method provides novel ideas for real-time and accurate oil spill detection.

Steep slope, plenty of rainfall and broken rocks led to frequent debris flow along Zha-Mo Highway in Tibet, which badly destroyed highway structure and restricted the local economic development. We discover the distribution and characteristics of debris flow along Zha-Mo Highway after investigating and sampling typical debris flow. Debris flow is densely distributed with 80 points along the road, and density is 0.69 per kilometer. Owing to great difference of elevation (reaching up to 3000m) along the highway, the distribution density and genetic type of debris flow appear obvious vertical zonal. In addition, the debris flow along the route has the characteristics of wide graduation, angular grain, heavy big unit weight, poor clay and low viscosity. Thus, a typical geological disaster data catalog database query system was developed to query and retrieve geological disaster data for road engineers. The proposed system makes the highway disaster treatment more closer to the actual engineering demand.

The traditional pixel-level extraction of building outlines is limited by the single feature, while the fitting of building outlines is limited by the problem of building corner extraction and complex shapes. This paper proposes a multi-level building contour extraction method that combines superpixel segmentation with LBP features. In this method, the SLIC method is used to segment the image, and the object-oriented method combined with LBP features is used to classify the building area for each segmented area, and then the building area is extracted using the pixel-level method combining LBP value and spectral feature. Building silhouettes. Experimental results show that this method is superior to traditional pixel-level extraction in terms of classification effect and accuracy.

Peanut ranks among the foremost cash crops in China. Accurate and timely understanding of peanut planting is very important for governments at all levels to manage production and formulate related policies. In this study, Pingtan was selected as the research area and Sentinel-2 and Landsat-8 data were acquired through the GEE cloud platform. The peanut planting areas in the study site were extracted using a random forest algorithm, and differences between the two datasets were compared. The experimental results show that: (1) The peanut planting area of Pingtan is all over the county, mainly concentrated in Haitan Island. There is a large peanut planting area in the northern part of Haitan Island, and other peanut planting areas are scattered in the southern part of Haitan Island and other islands; (2) The overall accuracy of Sentinel-2 data extraction was 94.44%, and the Kappa coefficient was 0.93; The overall accuracy of Landsat-8 data was 87.44%, and the Kappa coefficient was 0.84. (3) Remote sensing image data can be quickly used using GEE cloud platform, and crops can be quickly extracted based on the powerful processing capacity of GEE cloud platform.

Super-resolution is a burgeoning technique to enhance the spatial resolution of hyperspectral images (HSI) by fusing them with an auxiliary high-resolution multispectral image (MSI). However, most deep learning-based methods are based on supervised paradigm, which rely on large number of training triplets. Considering their deficiency, we propose a dual U-Nets autoencoders network inspired by the theory of spectral mixing, which purely operates on one pair of HSI-MSI for network optimization. Specifically, two coupled autoencoders are deployed as the backbone of our method, aiming to extract latent endmember and abundance of target image. To enrich the representation ability and enhance the interaction of two data, we design a novel dual U-Nets architecture for the encoder part by a parameter-shared strategy. Experiments in Chikusei dataset demonstrate the effectiveness of out method when compared with other state-of-the-art methods.

A good ecological environment is conducive to the pursuit of high quality of life and sustainable socio-economic development of human beings, and the rapid development of remote sensing technology provides great convenience for ecological environment monitoring. In this study, we construct a remote sensing ecological index (RSEI) model through principal component (PC) analysis method using four indicators of Normalized Difference Vegetation Index (NDVI), Soil humidity (WET), Land Surface Temperature (LST)and Normalized Difference Build and Soil Index (NDBSI) based on the Google Earth Engine (GEE) platform to evaluate the ecological environment quality of the nine cities of Shandong Province in the Lower Yellow River Region. The results show that (1) the overall trend of remote sensing ecological index in nine cities of the Lower Yellow River Region from 2001 to 2021 expresses an increasing trend, indicating that the ecological environment quality is improving. (2) The areas with higher ecological quality grades are distributed in the woodland area near Weishan Lake, and the area with higher vegetation cover near Mountain Tai, and the area at the mouth of the Yellow River, while the areas with lower grades are mainly distributed in areas such as urban construction land and unused land in the mountains. (3) The contribution rate of PC1 eigenvalues of RSEI in nine cities is around 60%, and NDVI and WET indexes are positively related to the ecological environment quality in nine cities in the Lower Yellow River Region, while LST and NDBSI are negatively related. (4) The overlay analysis of remote sensing ecological indices and land use types shows that forest land and grassland contribute to the improvement of ecological environment, and building land as well as unused land play a suppressive role in the improvement of ecological environment quality.

In order to reveal the characteristics of water and heat fluxes on different underlying surfaces of the old course of Yellow River under the background of climate warming, the M-K mutation test and Morlet wavelet analysis were used to analyze the annual average temperature in the past 71 years. The SEBS model was used to simulate the water and heat fluxes of different land use types in the old course of Yellow River. Finally, the energy balance characteristics of typical cultivated land in different seasons were analyzed by the eddy covariance system observation experiment. The results showed that the annual average temperature in the old course of Yellow River in the past 71 years showed an upward trend, and there were 6 years, 10 years, 15 years and 31 years of oscillation cycles. The area of dry land, forest land, grassland and unused land decreased from 1980 to 2020, and the area of water area, construction land and paddy field increased. Lakes, canals and other waters are the low-value areas of sensible heat flux in this area. The average value of latent heat flux in lakes, canals, and rivers is the highest, and the construction land is the lowest. The energy distribution is dominated by latent heat flux consumption in spring, summer and autumn, and by sensible heat flux consumption in winter. The results of this study will provide a scientific basis for coping with climate change and protecting the ecological environment of the basin.

This study provides a random forest (RF) based method to retrieve high wind speeds, which takes advantage of RF's powerful nonlinear fitting ability and prevents overfitting. The high wind speed data from the Stepping Frequency Microwave Radiometer (SFMR) are used as the reference data. Compared with two traditional GMF models, the C-band cross-polarized ocean model (C-2PO) and the C-band cross-polarized coupled parameter ocean model (C-3PO), the wind speed inversion results using the RF method have been greatly improved.

In practical applications of inverse synthetic aperture radar (ISAR), continuous long-time observation is not allowed due to transmission errors or noise interference. An effective way to reconstruct missing data would be to apply the sparse recovery (SR) method. However, the traditional SR methods need to discretization the parameter space, which inevitably leads to grid mismatch. The atomic norm minimization (ANM) method based on a continuous parameter estimation model can effectively eliminate the impact of grid mismatch and achieve high-resolution ISAR imaging. In this study, a sparse ISAR imaging method based on atomic norm minimization with Hankel-Toeplitz (ANM-HT) model was proposed to obtain better imaging performance. By reformulating the ANM-HT as semi-definite programming (SDP), the complete echo data can be recovered using SDP3 solvers. Real data results demonstrate the effectiveness and superiority of the proposed method.

Chang'E-3 landing camera takes high-resolution descent images with continuous time series for autonomous obstacle avoidance based on machine vision. The descent image sequence can provide rich surface topographic information. However, due to the dilemma of manual operation, high-precision slope maps are difficult to produce in real time based on descent image sequences for lander navigation and obstacle avoidance. In this paper, we use a blocking strategy for a single frame of Chang'E-3 descent images to estimate the slope of each sub-region independently without relying on the initial values of the terrain. Finally, we achieve a fast and accurate estimation of the distribution of hazard zones. The experiment proves the validity of the method, which only takes about 2 seconds to estimate more than 99% of the distribution of the hazard zones. This serves a reference for landing navigation and obstacle avoidance for future lunar exploration program.

Soil moisture, as the main characteristic parameter of the surface, plays an important role in climate change, energy exchange, and crop yield estimation. In order to remove the influence of vegetation coverage and roughness on soil moisture retrieval, a cooperative inversion method based on Water Cloud model and Oh model is proposed to retrieve soil moisture in winter wheat covered farmland areas using Sentinel-1/2 multi-source remote sensing data. The influence of winter wheat coverage on radar backscattering coefficient is removed by using the improved Water Cloud model with vegetation coverage and vegetation index fusion extracted from Sentinel-2 data. The attenuation factor in the Water Cloud model is solved by using the optimization theory. The soil moisture in the study area is retrieved based on the backscattering coefficient of VV and VH polarization Sentinel-1 data through the Look Up Table established by Oh model. The experimental results show that, based on the proposed method, compared with VH polarization, the soil moisture inversion from VV polarization data has a better effect, with a determination coefficient of 0.6577, a Root Mean Square Error of 0.0391 cm3 /cm3 , and a Mean Absolute Error of 0.0303 cm3 /cm3 , demonstrating the application potential of the proposed method in soil moisture retrieval.

Accurately and timely grasping agricultural information at the regional scale helps to solve food security issues and formulate agricultural policies. Remote sensing images have the advantages of wide monitoring range and the ability to eliminate human interference. In recent years, they have been increasingly valued in crop yield estimation. The CASA model is used to estimate the net primary productivity (NPP) of crops and then combine it with the harvest index (HI) to estimate crop yields. However, most studies use a fixed HI for yield estimation, which can lead to low accuracy. In this context, the HI is studied and improved for winter wheat yield prediction in this research. The main contents are as follows: Firstly, time-series Sentinel-2 optical data and meteorological data were used as inputs of the CASA model to calculate the NPP of winter wheat. Then, HI was calculated using the fieldwork data and a method was proposed to improve it due to its geographic location-related characteristics. Finally, the calculated NPP and improved HI were used to estimate the winter wheat yield. The experimental results prove that the proposed method has higher accuracy than original methods, with determination coefficient(R²) of 0.595, root mean square error(RMSE) of 793.4 kg/ha, and mean absolute error(MAE) of 659.53 kg/ha.

Semantic segmentation plays a crucial role in facilitating human understanding of remote sensing images. In recent years, significant advancements have been made in remote sensing semantic segmentation, largely attributed to the remarkable success of the fully convolutional network (FCN) in computer vision. The encode-decode architecture leverages the encoder to produce a continuous flow of semantic information, which are seamlessly integrated within the decoder. To fully exploit the potential of semantic features, several attention modules have been introduced to incorporate intermediate contextual information. However, these complex models often demand substantial computational resources. To tackle this challenge, this study presents a sophisticated model called the holistically guided fully convolution network (HGFCN). Unlike the conventional U-Net approach, which connects each relevant layer of the encoder and decoder, our model leverages holistically guided features derived from middle and high-level semantic features to recover spatial information. Empirical evaluations conducted on two high-resolution remote sensing datasets substantiate that the proposed method surpasses state-of-the-art semantic segmentation models.

Remote sensing image scene has the characteristics of complex background, large regional, seasonal and scale differences of similar scenes. This paper proposes a remote sensing image scene classification method based on migration learning and Swin Transformer model. The Swin-T model is pre-trained using ImageNet datasets to obtain the weight parameters of the pre-trained model, and then the migration learning method is used. Six remote sensing im-age datasets, including UCM, AID, NWPU, GID, NaSC-TG2 and SenseEarth Classify, are validated. The verification results showed that the overall classification accuracy was 99. 99%, 96.80%, 95.20%, 96.63%, 99.35% and 97. 09% respectively, and good classification results were obtained.

Knuckle arm CMM is a widely used standard in the field of industrial measurement. In this paper, the calibration of knuckle arm CMM and the industrial application of knuckle arm CMM are introduced. The calibration of articulated arm coordinate measuring machine can help us to know its measurement accuracy. This paper introduces the matters needing attention in the calibration process in detail. Knuckle-arm CMM is mainly used in automobile industry, heavy industry, surface inspection, blade measurement, casting blank inspection, reverse engineering and cultural relic digitization. This paper analyzes its application in automobile industry as a case.

The indoor large-size measuring standard device is an important standard device for high-precision geometric measurement instruments. In this paper, laser interferometer and standard steel tape measure are used as the comparison measurement object, and the measurement data are analyzed in detail. It shows that the indoor large-size measuring standard device has the characteristics of high accuracy and reliable data measurement results.

Studying the spatial and temporal changes of NPP in the Qinba area can provide scientific and technological data support for the realization of regional carbon neutrality goals. We use site parameters, geographical conditions monitoring data, MCD43A3 and a series of daily meteorological grid data to simulate daily NPP dataset at a spatial resolution of 500 m× 500 m during 2000-2020. The results show that the spatial distribution of NPP in the ecosystem of the Qinba area is significantly different. The NPP of the eastern part of the Qinba Mountains shows a decreasing trend, while the NPP in the western and southern parts shows an increasing trend. The significantly increased area accounts for 38.71% of the total area. Under the RCP4.5 scenario mode, it is predicted that the NPP in the Qinba Mountains will reach 337 gCm^-2a^-1 in the next 80 years. However, the positive effects of temperature, precipitation, and CO2 concentration on vegetation growth tend to flatten out in 2060. Climate change after 2060 may have a reverse inhibitory effect on vegetation growth.

With the strategy of Marine potestatem gradually promoting, satellite remote sensing technology has been widely used in chlorophyll-a concentration inversion and evaluation because of its advantages of large range and long time series. In this paper, using the data of GOCI with high temporal and spatial resolution and taking the fluorescence line height (FLH) as a sensitive factor, the chlorophyll-a concentration in Bohai Sea is inversed by CNN convolutional neural network, analyzing its temporal and spatial distribution characteristics and influencing factors. The results show that the spatial distribution of chlorophyll-a decreases from the nearshore to the interior of Bohai Sea, presenting annular distribution. The standard deviation of the overall change of chlorophyll-a concentration is less than 0.4mg/m3 during the five years from 2016 to 2020. However, the daily variation is large. In addition to the positive correlation between chlorophyll-a concentration and total suspended sediments concentration, sea surface temperature, tidal level, pollution discharge and ocean vortex are also factors that cannot be ignored.

The convolutional neural network (CNN) has been widely adopted in the processing of hyperspectral images (HSI) due to its remarkable capabilities in feature extraction. However, CNN-based methods for HSI classification still face the challenge of extracting spectral-spatial features that are more effective while considering all spectral bands. This paper reviews various spectral and spectral-spatial classifiers that have been developed to overcome the challenges of HSI classification, including the limitations of handcrafted features and the problems caused by spectral variability and nonlinearity. The paper proposes a novel method, the 3D ASPP Multi-Scale Feature Fusion Network (3A-MFFN), which simultaneously preserves high-resolution spatial detail information and high-level semantic information of HSI. The proposed network extracts and fuses multi-scale features for HSI classification using a multi-scale 3D ASPP convolution block to fuse different levels of feature output. The paper presents experiments and analysis to demonstrate the effectiveness of the proposed method. The results suggest that the 3A-MFFN algorithm can significantly enhance the classification accuracy of HSI with limited samples, making it a valuable contribution to the hyperspectral remote sensing community.

As a large structure, the truss track will inevitably wear and deform during long-term operation, which can negatively impact construction safety and quality. Therefore, it is necessary to safely, quickly and accurately detect the parallelism and flatness of the truss track. This paper proposes a new method for detecting parallelism and flatness of truss track based on point cloud data of 3D laser scanner. A corresponding data post-processing system is developed based on the method proposed above. Finally, the proposed method and system have been successfully applied to an actual detecting project of truss track in a large-scale iron and steel enterprise. The results show that the new method proposed in this paper is effective and feasible.

Based on the township economic statistics at the town scale, the change of the built-up area of Jiyuan City was calculated by remote sensing(RS) classification, and the relationship between agricultural mechanization and land urbanization and farmers' income in Jiyuan City from 2000 to 2019 was analyzed by using the type division method, spatial correlation analysis method and linear regression method, besides, the influencing factors were analyzed by constructing the GWR model. The results show that: (1) From the combination of the three types, the overall distribution pattern of “low-low-low” zone distributed in the west and “high-high-high” zone distributed in the east. (2) From the correlation analysis and linear regression, there was a significant positive correlation between land urbanization and farmers’ income in Jiyuan City during the study period, but there was no relationship between agricultural mechanization and farmers' income. From the perspective of the township, there was no correlation between land urbanization in Jiyuan City and farmers' income from 2013 to 2019. From 2013 to 2014, agricultural mechanization showed a positive correlation with farmers' income, and after 2014, the P value was greater than 0.05, which was not correlated with farmers' income. (3) From the perspective of the GWR model, the land urbanization coefficient showed an increasing trend, and the agricultural mechanization coefficient showed a decreasing trend, which confirmed the analysis that land urbanization has a greater impact on farmers' income.

Cross-border mining is an important topic of concern for surveying and mining technicians. This article takes the coal mine in Shenmu City as an example to determine the deformation aggregation area based on the average velocity map and time series analysis map obtained from InSAR monitoring of surface deformation. Using D-InSAR as a processing method to obtain the surface shape variables, surface cumulative shape variables, and maximum shape variables of subsidence map patches of SAR data in the study area, and conduct accuracy error analysis. Finally, the coal mining face and surface settlement range in the mining area are combined to accurately determine the boundary mining coal mine and export its geographical coordinates as the target area for suspected overlying boundary mining monitoring. InSAR technology can effectively monitor coal mining beyond the boundary, guide mining area production, overall planning and long-term development, and serve the sustainable development of mining areas.

This paper uses GIS and Fragstats4.2 software to quantitatively analyze the landscape space of Guanhaiwei Town, and visualize the data combined with image processing technology to enhance its spatial expression, analyze the hydrological characteristics of river corridors in Guanhaiwei Town, and the accessibility of traffic corridors, etc., and summarize the current problems. It is found that the construction land patches are concentrated on the traffic arteries and there are construction land disturbances within the woodland patches; the construction land is widely distributed, with scattered scattered points and irregular shapes. The woodland is mainly distributed in the south of the town, and the green patches in the town are less and the layout of green space is not reasonable enough, and the articulation with the surrounding landscape is weak. In the traffic land, the patches are highly fragmented and the landscape area is small. The patches of agricultural land are the landscape matrix, but their edges are highly fragmented, with small patches distributed sporadically and irregular in shape. The water patches were highly fragmented and no isolation zone was established. The study used GIS for image processing and spatial analysis, which fully reflects the importance and application of computer technology in geographic research.

The rapid development of China's power grid construction projects has led to the gradual extension of transmission line corridors to remote and mountainous areas, where engineering construction faces more complex terrain and difficult transportation environments. Line design requires survey data from transmission line corridors as support, while surveys also need line design data as their working basis. The traditional approach requires survey and design personnel to jointly go to the site for field work, and then process the office work indoors to finally determine the design or survey data, which is not intuitive enough and can easily lead to duplicate work. This study investigates the specific survey and design requirements for various professions. This study aims to achieve main functions such as rapid layout of power tower foundations, intelligent collection of land data, comprehensive survey methods for line intersection crossings, fast investigation methods for line house information, universal positioning and layout methods by comprehensively utilizing technologies such as 3D GIS, Beidou satellite navigation positioning, and intelligent speech recognition. These functions will assist field surveys in site selection and mapping, optimize work processes, reduce repeated work, and improve work efficiency

In early 2020, an outbreak of Corona Virus Disease 2019 (COVID-19) occurred in Wuhan. Due to Wuhan's special transportation location, COVID-19 spread to all provinces in mainland China in a short period of time. In China, the population and industry are highly concentrated in urban areas, and the mobility of people is strong, which makes the spread of the virus greatly increased. From a geographical perspective, analyzing the spatio-temporal dynamics and evolution characteristics of COVID-19 is conducive to clarifying the scope and extent of COVID-19 transmission. Therefore, statistical analysis and kernel density estimation methods are used to analyze the spatio-temporal characteristics, and the urban epidemic transmission network is constructed with the help of complex network theory to explore the degree of epidemic correlation between cities. The study found that the development process of COVID-19 in China has roughly experienced four stages: rapid growth, rapid decline, stable period, sporadic growth, with obvious spatial agglomeration characteristics. In the early stage of the epidemic, there was a strong connection between cities, and cutting off the transmission path of the epidemic could effectively control the spread of the epidemic.

A new method of improved double baseline steel wire method was used to monitor the shaft deformation in the auxiliary shaft of Niuerzhuang Mine in Fengfeng Mining Area. Based on the two steel wires point P1 and P2, the eight directions from Q1 to Q8 were determined by using the designed directional ruler coincide with the two points, and the distance from P1 to the shaft wall Q1 to Q8 was measured by laser rangefinder, so as to fit the cross section of the shaft. The deformation of the shaft is monitored by fitting the center point of the section and the change of radius. Compared with the traditional four-datum vertical line method, it is easier to measure, shorter in monitoring time and more efficient on the premise of ensuring accuracy.

The deterioration of natural ecological environment of Huai River Basin has been one of the most important factors of restricting the sustainable development of local social economy. However previous researches focused on the ecological environment are subjective and inaccuracy. RSEI is a newly developed aggregated index to fast detect ecological conditions solely using remote sensing data. However, it is unstable and its results are not unique, so it cannot be calculated in batches. As a result, the model can only use a single image to measure the comprehensive ecological environment index, and the results are accidental. This paper uses the improved RSEI model based on the full-time remote sensing big data and cloud computing technology to calculate the eco-environment quality and its changes of Huai River Basin. It is more objective and true to express the regional ecological environment status by using the average of the image measurement results of all phases of a year. Then this paper uses hot spot analysis, and geographic detector to analyze the temporal and spatial evolution characteristics and the influence factors of the eco-environment quality of Huai River Basin from 1990 to 2019. The results are as follows: (1) the ecological environment quality of Huai River Basin deteriorated gradually, and the ecological environment deteriorated most dramatically in 2015-2019, (2)the changes of the ecological environment index have obvious accumulation from 1990 to 2019 and the accumulation areas of cold and hot spots are decreasing gradually, (3) the interaction between human activities intensity, the diversity of land use, and the degree of topographic undulation have greater impact on the ecological environment than other factors. The number of factors who have a significant impact on the ecological environment is increasing gradually, and the ways these factors act are becoming more and more complex.

The Guangdong-Hong Kong-Macao Greater Bay Area (GBA) encompasses diverse and widespread wetlands, which have undergone significant changes in recent years due to various anthropogenic and natural factors. This paper investigated the dynamic changes of diverse wetlands in GBA from 2005-2020 using the Google Earth Engine (GEE). By integrating suitability atlas, the CA-Markov model was employed to predict wetland evolution in GBA. Findings revealed rapid urbanization between 2005-2015 led to swift degradation of natural wetlands. Conversely, from 2015- 2020, there was a recovery trend in natural wetlands, despite the emergence of new urban areas. By 2025, predictions suggest further recovery of natural wetlands, reduction in farmland and paddy fields, and continued urban expansion.

With the development of social economy, the limited land resources and the high concentration of population and industries bring great pressure on the rational allocation of national land space. In this paper, based on the classification system of Production–Living–Ecological Spaces(PLES), the land use data of nine cities along the Yellow River in Shandong Province for the four periods of 2005, 2010, 2015 and 2020 are divided into PLES, and the conflict intensity of PLES is diagnosed, and the conflict level is integrated with the appropriateness of the land space and the game model of PLES, and the resilient land space of the study area is divided into five categories: ecological protection zone, ecological restoration zone, ecological cultivation zone, ecological production zone and non-ecological zone. The results show that: (1) The overall conflict level of PLES shows a characteristic of "the proportion of high conflict level increases year by year, while the proportion of low conflict level decreases year by year". (2) The proportions of unsuitable, low suitable, general suitable and high suitable zones in the ecological suitability of land space are 42.21%, 30.13%, 22.39% and 5.27% of the total area of the study area, respectively. (3) The eco-restoration zone accounted for 15.53%, the eco-cultivation zone accounted for 0.86%, and the non-ecological zone accounted for 69.68% of the whole study area. The study delineates different ecological spaces and provides a new perspective for ecological space optimization.

The urban heat island theory plays a crucial role in the domains of urban climate and urban planning. However, conventional Urban Heat Island Functional Zoning (UHIFZ) methods often rely on empirical common sense, neglecting the spatial effects that can better explain the essence of UHIFZ. To address this gap, this paper proposes a novel method, named Geographically Weighted Kernel K-Means for UHIFZ (GWKKM-UHIFZ). By integrating spatial weights into the feature spatial distance metric process, this study demonstrates that the combination of feature spatial distance and machine learning classification algorithms can effectively achieve flexible urban heat island functional zoning and provide accurate data support for subsequent quantitative analysis. Additionally, the evaluation indexes can be adjusted according to specific requirements, thereby acquiring valuable thematic information for future urban planning and design upgrades.

With the rapid development of China's economy, cities are becoming more and more prosperous, and many cities are lit up at night. However, when urban lighting exceeds a certain limit and affects the life of residents, it becomes light pollution. In order to monitor and prevent light pollution, we first defined three kinds of light pollution accurately, and constructed a light pollution evaluation model for Shanghai based on ground-based satellite remote sensing technology and urban light sampling data, using the advantages of support vector machine model robustness, can effectively avoid "dimensional disaster", and simplify the multi-classification problem. The results showed that the distribution of nighttime illumination in Shanghai ranges from 0 to 60 lumens, with great variability from region to region, and the illumination level can reach 35 lumens or more in some important places such as large shopping malls and large squares, which require light pollution prevention. This study can realize the light pollution degree detection, provide data support and reference for reducing the impact of light pollution, and meet the demand of night lighting for scientific and sustainable development of the city.

Studying the spatial and temporal variation characteristics of land use and forest carbon sink effect in the Dabie Mountains is of great significance for optimizing the regional land use structure, forest ecosystem restoration and carbon balance. This paper is based on the eddy correlation system of the National Field Scientific Observation and Research Station of Forest Ecosystem in Dabie Mountains, Henan and spatial-temporal distribution model of geographic information system. The spatial-temporal distribution characteristics of land use/cover change and the carbon sink effect of the evergreen deciduous broad-leaved mixed forest ecosystem in Dabie Mountains were revealed by using the method of geo-information map. The results show that: from 1980 to 2020, the balance of land use/cover types showed unbalanced-balanced-balanced-unbalanced, and the fluctuation trend of the annual change ratio was small-big-big-small. Paddy fields, dry land, and construction land change most frequently. On monthly and seasonal scales, the forest ecosystem in the Dabie Mountains generally has a certain carbon sink effect.

The geometric decay factor, as an important part of the two-way reflection model, has an important modulating effect on the model. The integral geometric attenuation factor uses the normal distribution function of the microsurface elements for integration, which eliminates part of the unreasonable inflection point problem in the Blinn geometric attenuation factor in the curve. However, the integral geometric attenuation factor only considers the shading or shadowing effect under the same conditions, and the curve still has inflection points at the critical point between the shadowing and shading effects, especially at large incidence angles. To address these problems, under the same detection conditions, the detector detects a collection of reflected light from a large number of different microplane elements in the field of view, and the effects of masking and shadowing effects should be considered comprehensively. In the integral calculation of the reflected light from the microplane, it is necessary to assign the proportion of the incident light incident to each microplane element to the total incident light. The proposed modified integral geometric attenuation factor model, which is the product of the proportion of incident light striking the respective microplane element as a percentage of the incident light and the light intensity received by the detector from that surface as a percentage of the light intensity incident to this surface. The modified integral geometric attenuation factor model is compared with the integral geometric attenuation factor model, and the two-way reflection function curve of the modified integral geometric attenuation factor is compared with the measured two-way reflection function data by numerical simulation. The numerical simulation results show that the modified integral geometric attenuation factor model fits better with the measured two-way reflection function data, and can suppress the divergence trend of the integral model at large reflection angles.

Insect monitoring is vital for grassland management. In recent years, with the rise of the Internet of Things (IoT) technology, there are sufficient conditions for the automation of grassland insect population monitoring. This paper has designed and implemented an insect monitoring system for grassland management based on Python. The system allows 24-hour real-time monitoring of grassland insect populations and supports remote viewing of insect data by managers. The system uses embedded devices to monitor grassland insects in real time by taking pictures and uploading the insect picture information to the server through wireless network for picture processing, identifying and counting the number of insects, and then the insect information is displayed in real time on the information view screen in the server, so that managers can grasp the grassland insect situation remotely in real time. In this paper, through extensive testing, it is shown that the system achieves the insect monitoring function well and has the advantages of easy operation, high execution efficiency, strong stability and good scalability.

For the application scenarios of ocean surveying, the traditional motion reference unit (MRU) is limited in its ability to directly provide detailed wave characteristics such as the heave and attitude parameters of the vehicle. To address this limitation, a wave measurement algorithm based on the MRU is proposed. Among them, the filtering and integration algorithm in the frequency domain is proposed to process the vertical acceleration data obtained by the MRU, Moreover, the wave height and wave period measurement algorithms are further established according to the vertical displacement obtained by this algorithm. An algorithm for wave direction measurement based on the attitude angle of the vehicle motion obtained by the MRU is also designed. Ultimately, through experimental verification, the proposed algorithm demonstrates its effectiveness in accurately measuring wave height, wave period, and wave direction.

The coordination problem caused by economic development promoting urban expansion is concerned by various scholars. This paper is based on nighttime lighting data and socio-economic data from 2013 to 2020，constructed a quaternary system coupling coordination degree model to evaluate the coordination of urban expansion. To reveal the increasing and decreasing trend of urban coordination in Guangxi in recent years. The results show that: The total night-time light in all cities of Guangxi is on the rise, and the amplitude is different, which is consistent with the economic growth. Most cities in Guangxi are in an uncoordinated state, and the coordination of many cities in the study period shows an overall upward trend. Guangxi as a whole is developing in a coordinated and sustainable direction.

Synthetic aperture radar (SAR), as an active microwave remote sensor with high-resolution earth observation, is easily affected by radio frequency interference (RFI). Pulse radio frequency interference (PRFI), as a typical form of RFI, pollutes SAR raw data with bright stripes, which increases the difficulty of SAR image interpretation. However, the conventional eigen subspace projection (ESP) method decomposes the whole pulse into the interferences and the useful signals, resulting in the loss of useful signals in the non-interference positions with the pulse. To solve this problem, this paper proposes an improved SAR PRFI suppression method based on ESP. Firstly, the receiving pulses containing PRFI are detected and the specific position of PRFI in the pulse are determined by eigenvalue decomposition and energy accumulation in time-frequency domain, respectively. Secondly, the PRFI signal is separated from the useful signal by applying the ESP method only to the data where the PRFI is located. Finally, the PRFI signal projected by ESP is eliminated from the raw data. Simulation results shown that the proposed method can better protected useful signals and improve SAR image quality compared with conventional methods.

The main method of granite lithodemic mapping based on cogenetic magma evolution theory is identifying the evolutional sequence and contact relationship of the rock body. Huashan granite, the remelted complex pluton, generates different types of minerals in different evolutional cycles, so studying the metallogenic regularity by lithodemic classification is significant to delineate prospecting target areas. The magnetic susceptibility measurement of rocks in the area shows differences in each unit of the rock body and the exocontact zone. Four boundary enhancement methods, vertical derivative (VD), total horizontal derivative (THD), Tilt, and Theta map, are used to identify the lithodemic units by the aeromagnetic anomaly data of Huashan rock mass, and the Theta map can effectively identify the units inside the rock body with a higher resolution. This paper provides an effective geophysical method for lithodemic unit mapping in granite areas.

Rare earth resources are of great economic and strategic importance. The Bayan Obo, as the largest Nb-Fe-REE deposit in the world, have attracted much attention in mining field. This paper provides two index parameters of ore that can be used to quickly distinguish the ore types with exploitation value or industrial significance, which is helpful for geometallurgy and remote sensing prospecting. Seven ore types from the main and eastern orebodies of Bayan Obo deposits were collected for this study. The composition of mineral and element of each sample was acquired by TESCAN Integrated Mineral Analyzer (TIMA). Method of principal component analysis (PCA) was employed to analyze the compositional data. Metallogenic potential index of minerals and elements (MPIM and MPIE) were defined to make quantitative appraisal on the resource potentiality of Bayan Obo deposit. The ore types with a higher index are thought to be rich in REE. The results show that MPIM and MPIE are helpful to judge the metallogenic potential and mineral exploration in Bayan Obo deposit. The research is expected to be applied to the fields of geometallurgy and remote sensing exploration.

For the current millimeter wave radar imaging data volume than the mud level meter, rain gauge and other single-point monitoring means of data volume tens of times or hundreds of times geometrically increasing and other challenges, in order to meet the data acquisition computing needs, designed based on FPGA ground disaster system data acquisition computing design of colleges and universities. First of all, we analyzed the requirements of data acquisition of ground disaster system, designed the overall architecture of the system based on FPGA, developed XC7K325T as the main control chip, ADC analog-to-digital conversion and other related modules in one, and then conducted experimental verification of the data acquisition and operation system of ground disaster system, which proved that the system has good acquisition and processing performance, and effectively overcame the problems of large amount of data acquisition of ground disaster system and inaccurate acquisition and operation processing. The system can effectively overcome the problems of large volume of data collection and inaccuracy of data collection and processing, and is conducive to the subsequent data processing of the ground disaster system.

Tsunamis can cause significant loss of life, major property and environmental damage in coastal cities. The great destructive tsunamis have heightened the awareness of the study of potential risk. We adopt multiscale scenarios to test the tsunami risk, according to the analysis of historical seismic events and geologic structure. Local, regional and transoceanic tsunami sources might pose a significant potential threat to Fujian Province, which faces the open sea. The tsunami wave height distributions along coastline for hypothetical extreme rupture events are simulated. The tsunami risk of different level for coastal area was analyzed. The results are useful for understanding the potential risk of Fujian province. Furthermore, prevention measures for tsunami need to be prepared, and emergency plans must be made previously in coastal Provinces under the potential tsunami threat.

The phenomenon of hydrocarbon micro leakage in oil and gas reservoirs is common, and at least 85% of oil and gas fields in the world currently have hydrocarbon micro leakage. Remote sensing technology can quickly, efficiently and economically detect and extract surface mineral anomaly information caused by hydrocarbon leakage, so as to provide favorable exploration targets in the initial stage of oil and gas exploration, which has important social and economic significance for reducing exploration risks and accelerating exploration process. Based on the hydrocarbon microleakage theory, this study extracted clay mineralization, carbonation and red layer fading and alteration information in the study area using Landsat8 OLI image data, and used GIS tools for overlay analysis. Combined with the comprehensive analysis of geological structure and other factors, the oil and gas prospect areas in northeast Sichuan are predicted, designate 5 oil and gas prediction areas.

The western Sichuan plateau features a unique natural environment and profound humanistic connotations and boasts its booming tourism. However, Ganzi Prefecture has faced an arduous task in comprehensive prevention and control of geological disasters due to its unique natural environment, development of disaster-inducing factors, weak bearing capacity for disasters and various types of disasters. Due to frequent extreme weather, changeable climate, vulnerable primary environment and frequent natural disasters, e.g. geologic and meteorological disasters, floods, droughts and earthquakes on the Western Sichuan Plateau, the southwestern Sichuan Mountain areas and the mountainous areas around the Sichuan Basin, Sichuan is one of the provinces with the heaviest natural disasters in China. In addition, the extremely heavy rainfall process and earthquake disasters lead to frequent geological disasters such as collapse, hill-creep and debris flow, bringing a severe situation and great losses to Sichuan. Since the hidden dangers of geological hazards are distributed on very high mountains and in a large scale, and covered with vegetation, it is difficult to discover and prevent these hidden dangers. Frequent geological disasters pose a serious threat to the local residents’ life and property safety and seriously hinder the local social and economic development. Risk evaluation of geological disasters in Danba County is of great significance.

The combination of directional drilling and hydraulic fracturing for coalbed methane extraction has played an important role in gas management in low-permeability coal seams both domestically and internationally. Whether the drilling can meet the design requirements directly affects the extraction effect. This article introduces the system composition and construction method of a memory-type gamma logging tool with drilling trajectory measurement and lithology identification functions, and conducts on-site experiments on gas extraction holes in the Sangshuping No. 2 well. The experimental results show that the instrument works stably, the trajectory measurement and lithology identification are accurate, the working time can meet the needs of directional long drilling, the data obtained by the instrument can verify the drilling trajectory, and the detected lithology can provide guidance for directional long drilling gas extraction, improve extraction efficiency and reduce construction costs.

Using Google Earth Engine (GEE), Landsat 8 and Sentinel-2 data were analyzed to invert surface temperature and identify land cover types through a random forest algorithm. By analyzing data, investigated the impact of different types of land cover on the urban thermal environment. The results indicated that although the average temperature did not change significantly from 2018 to 2020, the range between the average maximum and minimum temperatures widened, leading to an increasingly evident urban heat environment. At the same time, lower temperature zones were mainly distributed in farmland, forest land, and other areas with high vegetation coverage and near water bodies. In contrast, moderate, high, and higher- temperature zones were predominantly found in built-up and unused lands with low vegetation coverage. In recent years, Xuzhou has transitioned from a mining-dependent city to a renewable-resource city, with most abandoned mines being reclaimed as ecological parks, farmland, etc. Therefore, the area of cultivated land increased by 21.9% from 2018 to 2020. Through our research, it is suggested that future efforts could be intensified to reclaim bare lands, such as abandoned mines, into vegetated lands, in order to alleviate the urban heat phenomenon in Xuzhou. These research findings can serve as a scientific basis for improving the urban thermal environment and provide a reference for Xuzhou to formulate eco-city construction strategies.

With the rapid economic development in the North China Plain, the significant emissions of black carbon (BC) aerosol have exerted notable impacts on climate, environment, and health, making it a hot topic in environmental studies in recent years. To compensate for the insufficient ground monitoring data, the PM2.5 satellite remote sensing inversion data and chemical model simulation data were utilized, applying the proportionality factor method to estimate the satellite concentration of BC. The estimation results were validated using BC monitoring data from seven stations in 2018, and based on this, the spatiotemporal variations of BC satellite concentration was obtained. The results showed that the estimated BC concentration using the proportionality factor method had good accuracy, with a correlation coefficient (R²) of 0.72. The spatial distribution of BC over the North China Plain in 2018 exhibited a decreasing trend from the inner region to the outer region, with high concentration areas mainly located in provinces with high straw burning and steel production. The seasonal variation of BC in the North China Plain in 2018 was pronounced, showing higher concentrations in autumn and winter and lower concentrations in spring and summer.

The distribution range of the dominant casualties of each intensity profile in different regions is basically the same, but the probability of casualties of the same magnitude in different regions is quite different.The Handan Fault is a hidden fault east of the Taihang Mountain Piedmont Fault. It is the boundary between the Handan Depression and the Cheng'an Low Uplift in the North China Plain Depression. It is located within the planning and construction range of the Eastern Handan New Area. Studying its activity is of great significance for earthquake risk assessment, and at the same time provides effective support for site selection and disaster prevention of major regional engineering projects. This paper uses the SLC data, IW mode, VV polarization, and de-orbit data of the Sentineil-1A satellite in 96 scenes from 2017 to 2021 to analyze the activity of the Handong fault based on SBAS-InSAR technology. The results show that in 2017-2021, the deformation characteristics of the two sides of the East Handan fault are significantly different. The west wall is uplifted and the east wall subsides. The average velocity of the west wall is 0.51mm/a, and the east wall velocity is -29.8mm/a. Combined with the data of regional tectonic movement, groundwater exploitation survey, and borehole detection, it is believed that the Handan Fault is a high-angle normal fault dipping westward, and its current activity is relatively low. The uplift of NW wall is mainly affected by the overall uplift of the Taihang Mountain, and the subsidence of the SE wall is mainly caused by the overexploitation of groundwater.

Real-time and dynamic monitoring of soil moisture is critically essential for farming activities and crop yield estimation. This paper focuses on the study of surface soil moisture (SSM) in agricultural fields with different vegetation sparsity. Based on the improved water cloud model (WCM) and convolutional neural network (CNN), the SSM inversion of agricultural fields covered with winter wheat at different growth stages was achieved. Firstly, the vegetation cover factor was introduced into the WCM to separate the scattering contribution under crop cover from the direct scattering contribution from the bare ground. Then, the remote sensing data was used to derive numerous characteristic parameters to the SSM. Finally, the CNN model was established for the purpose of SSM inversion. The better correlation between the backscattering coefficients obtained by the improved WCM simulations and the measured SSM was shown. The coefficients of determination were 0.46 and 0.39 under VV and VH polarization, respectively. Better inversion accuracy can be obtained by combining the two parameters with other characteristic parameters as input data into the CNN model for SSM inversion. The coefficient of determination was 0.75, root mean square error was 2.51 vol.% and mean absolute error was 2.12 vol.%. Therefore, the method can effectively separate the effects of agricultural crops and bare ground on radar signals, which contributes a novel research approach and guidance for handling SSM inversion in agricultural fields covered by vegetation.

In order to understand the Internet of Things technology, research on the application of Internet of Things technology in automatic monitoring of geological disasters is put forward. In this paper, firstly, the Internet of Things technology is used to improve the accuracy and precision of geological disaster monitoring and make the prevention and control of geological disasters more timely. When using the Internet of Things technology, it is necessary to analyze the business requirements in advance, find the key Internet of Things technology, and effectively infiltrate it with the geological disaster prevention system, so that the advantages of the Internet of Things technology can be reflected. Secondly, in view of the shortcomings of geological disaster monitoring and early warning means and methods at the present stage and the future development needs, combined with Internet of Things, communication and other technologies, deepen the Understanding of geological disasters, analyzing and studying the application of Internet of Things technology in monitoring and early warning of geological disasters (debris flow, collapse, landslide, etc.). Finally, combined with an application example of slope monitoring in a project, the automatic monitoring and real-time early warning functions of geological disasters are realized, which has high application value in geological disaster monitoring.

The Saint Elias Mountains is an important glacierized area in southeastern Alaska. Glacier velocity is a key parameter for understanding glacier variation. In the context of global warming, glacier velocity variation has been widely studied, but the details of their inter-annual, intra-annual and seasonal variation are unclear. Using SAR images with offset tracking and SBAS method, the three-dimensional velocity across the study area were obtained. The vertical velocity was further divided into surface parallel flow (SPF) and non-surface parallel flow (nSPF), where SPF is produced by the downhill glacier movement under its own gravity, while nSPF is caused by the deformation inside the glacier. The inter-annual, intra-annual and seasonal variations of glacier velocity in the study area during 2017-2020 and their influencing factors were analyzed. The results showed that: (1) The velocity of different glaciers varied significantly. In the vertical direction, the nSPF component was significantly greater than SPF regardless of the glacier velocity itself. (2) There was no striking inter-annual variation of velocity distribution and magnitude between 2017, 2018, 2019 and 2020. We also found different variation patterns in different seasons. (3) We used the flowline of Valerie glacier and analyzed the intra-annual velocity variation throughout 2018. The velocity was significantly higher during winter and spring. (4) The seasonal variation of velocity was not completely consistent with that of temperature and precipitation, and was mainly affected by subglacial hydrological conditions.

Most major earthquakes worldwide occur within subduction zones, but there is a death of studies on earthquake induced landslides in subduction zones. the Mw7.8 earthquake that jolted Nepal on April 25, 2015, the first strong earthquake in the Himalayan Subduction Zone in nearly 70 years, triggered a large landslide event, and the area served as a good venue for conducting research on earthquake-triggered landslides in Himalayan Subduction Zones. Based on remote sensing interpretation and field investigation, 2,072 sets of earthquake-triggered landslide information were obtained, revealing that the Nepalese earthquake-triggered landslides are distributed between 1,000-3,000 m above sea level, in the transition area between the High Himalaya and the Low Himalaya, basically along the Main Central Thrust, with a large topographic drop; The dominant distribution range of landslide slope values is 35-40°; the dominant distribution range of aspect values is 120-200°, which is closely related to the horizontal deformation field. The basic characteristics of earthquake-triggered landslides in the Himalayan Subduction Zone are initially discussed: The landslide points are clearly distributed in a faceted pattern and are closely associated with deep, low-dip fault rupture surfaces; the landslides are all located in the upper plate, and are easily induced by the inertia of seismic motion on slopes with slopes that are oriented in the same direction as the reverse thrust of the upper plate.

The types of Chinese place names in typhoon events are diverse and the place names are long, which increases the difficulty of place name recognition. At the same time, how to use two different granularity semantic units of Chinese characters and words to assist neural network models to recognize place names is also a problem. In order to improve the accuracy of Chinese place name recognition in typhoon events, a dictionary of place names in typhoon events was constructed. The dictionary of place names and HowNet dictionary are combined for learning features of place names in typhoon events. Based on the Lattice-BiLSTM-CRF model, the two granular semantics of Chinese are integrated to realize the recognition of Chinese place names. It is verified by an experiment that this method can improve the accuracy of Chinese place name recognition in typhoon events.

Oil spill accidents at sea are one of the major accidents that affect the environment and port safety, and can cause severe damage to the marine environment. Based on the hydrological and meteorological environment of the proposed crude oil terminal in Wenzhou, the diffusion and movement path of spilled oil and its impact range were calculated by constructing a mathematical model. Analyzed the impact of oil spills on some marine environmental protection zones, which has important reference value for engineering construction.

The key point and difficulty in realizing the rapid modeling of railway line 3D scenes is the automatic modeling technology of railway line entity model. Parametric modeling of linear structures is the committed step to realize automatic modeling of railway line structure. According to the characteristics of linear structures in line structure, this paper proposes an automatic modeling method based on lofting modeling for railway line 3D scenes. First, construct the section template library of the line structure. Call the template according to certain rules. Then, through the lofting procedures, the automatic modeling of the linear structure is realized under the parameter drive. Finally, by developing applications driven control parameter values, automated modeling of railway linear structures was completed. Compared with the traditional 3D modeling methods, the modeling method of lofting can realize the automatic modeling of linear structures in long strip railway line more quickly.

Plateau mountainous regions exhibit complex and diverse terrain characteristics, with varying altitudes and extensive land coverage. These areas are subject to a complex and diverse climate, particularly during the period from November to March, where the intricate microtopography in the plateau mountainous areas leads to highly complicated local weather conditions. Under adverse geographical and micro-meteorological environments, these regions are prone to transmission line icing disasters. The main factors influencing ice accretion on vulnerable microtopography differ according to different microtopographic types. This study focuses on the Guizhou region as the research object and divides it into different zones based on its geographical and climatic conditions. Within these zones, analysis and modeling are conducted for ridge lines, valleys, passes, and windward slopes. Finally, through cross-validation of the models and comprehensive analysis of the results, the key factors influencing ice accretion on various types of vulnerable microtopography in different zones are identified.

Accurate evapotranspiration (ET) simulation is of great significance for the study of surface hydrological cycle in alpine swamp meadow under the background of climate change. In this paper, the ET estimation model was constructed based on the meteorological and EDV data recorded at the Wayanshan Wetland observation Station in Qinghai Lake Basin, based on random forest (RF), extreme gradient lifting (XGB), support vector machine (SVR) and artificial neural network (ANN) learning algorithms, and the simulation accuracy of different machine learning models was evaluated. Cross-validation results show that the coefficient of determination of RF, XGB, SVR and ANN models is 0.97, 0.91, 0.85 and 0.75, and the root mean square error is 0.07, 0.04, 0.38 and 1.12, respectively. Among all models, RF has the best simulation effect, while XGB and SVR have better simulation effect. The ANN model has the worst simulation effect. According to the seasonal distribution trend of simulated ET and observed ET, all models can simulate the seasonal cycle of evapotranspiration well, and the average RMSE is 0.12-0.38 mm/d. The interpretability analysis of evapotranspiration and meteorological factors shows that air temperature, net radiation and soil water content are the key factors that dominate the evapotranspiration change of alpine marsh meadow in Qinghai Lake Basin.

Land use change is an important factor affecting the quality of the ecological environment, so the influence between land development and ecological environmental protection in the process of development and construction becomes an important proposition for building an ecologically civilized city in the new era. This paper takes Hangzhou Bay New Area as the research object and uses the remote sensing images of Hangzhou Bay New Area in four phases in 2013, 2015, 2017, and 2020 as the main data source to explore the relationship between land use and ecological environment quality in Hangzhou Bay New Area through the composition of different types of land use changes. The results show that the impact of land use changes on ecological environment quality is mainly focused on two aspects: in the change of land use structure, the ecological environment quality increases with the increase of forest land area, deteriorates with the expansion of construction land, and improves with the recovery of forest land area; In the change of land use mode, the ecological environment quality of other land converted to construction land becomes worse, the ecological environment quality of arable land converted to construction land does not change, and the ecological environment quality of unused land converted to forest land becomes better. In summary, the increase in the area of woodland in Hangzhou Bay New Area can strongly promote the improvement of ecological environment quality, and the protection of woodland and the selection of unused land for woodland development should be strengthened, while the continuous expansion of construction land should be restricted.

With the rapid development and widespread application of information technology, various information resources targeting complex application scenarios have been continuously enriched. Consequently, a large number of information management systems have emerged, and the capacity and processing requirements of information have far exceeded the capabilities of traditional processing methods. This presents significant challenges for comprehensive environmental perception that encompasses multiple sources of data from geographic, surveying, meteorological, oceanic, and human-related domains, involving the entire process from data generation to knowledge acquisition. In this paper, firstly, the concept and connotation of environmental comprehensive perception are expounded. Secondly, the key technology system of environmental comprehensive perception is explored, and the key links and cutting-edge research status of multi-source data acquisition, processing, management, analysis, mining and service of environmental comprehensive perception are sorted out, and a hierarchical environmental comprehensive perception system is constructed. Finally, the development trend of environmental comprehensive perception is prospected.

Classification of wetlands over large area is crucial for monitoring wetland ecological resources. In this paper, the Google Earth Engine (GEE) platform was integrated with deep learning technology to enhance the classification accuracy of wetlands in the Guangdong–Hong Kong–Macao Greater Bay Area (GBA). Firstly, the GEE was utilized for preprocessing the 2022 Landsat images of the GBA to determine the maximum wetland extent. Subsequently, the the Coatnet model with Auxiliary classifier(Aux-Coatnet) was employed to classify the annual wetland image. This model demonstrated an overall accuracy(OA) improvement between 6.13% and 24.52%, outperforming other models with balanced performance across diverse land use types. Particularly, it proved proficient in extracting wetland features like rivers, lakes or reservoirs, mangrove or swamp wetlands, and aquaculture pond. These findings highlight the potential of the Aux-Coatnet model in achieving high-accuracy extraction and classification of large-scale wetland features

NASA's Artemis III mission plans to land at the lunar south pole and has announced 13 pre-selected landing sites. The rugged terrain and complex illumination and communication conditions impose enormous challenges for landing and exploration at the lunar south pole. We simulated the illumination and Earth communication conditions for the Artemis III pre-selected landing sites between Jan. 1, 2024 and Jan. 1, 2026 at 5m/pixel and 1h intervals based on accurate LOLA terrain, and performed an engineering suitability analysis for 13 sites in combination with slope. The area of each site that meets the engineering constraints is calculated as an evaluation of the sites engineering suitability. Our analysis shows that Connecting Ridge has the best illumination conditions, the Malapert Massif and Leibnitz Beta Plateau areas have excellent Earth communication conditions, and Peak Near Shackleton and Nobile Rim 2 have excellent combined engineering suitability. We hope that this research will provide a useful reference for future lunar south pole exploration missions.

The spatial distribution of crop planting structure plays a crucial role in reflecting the layout and arrangement of farmland, making it highly significant for the advancement of precision agriculture and agricultural resource inventory. The extraction of crop planting structure through remote sensing satellite earth observation technology serves as a fundamental pillar for the development of precision agriculture informatization. This study focuses on the application of precision agricultural remote sensing in monitoring complex mountainous regions, employing the cognitive theory of remote sensing atlas to support the process. By integrating high-spatial resolution satellite images, medium-resolution time series satellite images, and ground observation data, a hierarchical zoning approach was developed, using farmland parcels as the basic spatial unit for crop type identification. A case study was conducted in Wushan County, Chongqing, serving as a demonstration zone. Vegetation indexes and phenological indexes were extracted by combining the high spatial resolution of GF satellites with the temporal resolution of middle-resolution satellite images. Subsequently, a multi-feature crop identification model at the parcel scale was constructed, facilitating the precise extraction of crop planting types in the southwest mountainous areas.

Calcite is usually the largest cement in hydrocarbon-bearing sandstone reservoirs, which has an important impact on reservoir densification and hydrocarbon accumulation. It is generally believed that calcite cement has multiple sources, phases and diversities, but the understanding is not completely consistent. In this paper, the occurrences and origins of calcite cement are studied in Upper Triassic Yanchang Formation sandstone reservoir in Ordos Basin. Due to the occurrences, the calcite cement could be divided into the disseminated calcite cement and the patchy calcite cement in Yanchang Formation sandstones. Through an integrated research of field investigations, petrographical, mineralogical and geochemical analyses, and fluid inclusion thermometrics, the conclusion is that the origins of the disseminated calcite cement and the patchy calcite cement are same-the released water from adjacent mudstones in the process of compaction, and their present different compositions and contact relations with the clastic particles are caused by their original occurrences. The disseminated calcite cement in the sandstone fills all the intergranular pores so that it could not contact with fluid, leaving no chance for water-rock interaction. While the patchy calcite cement in the sandstone is surrounded by pores so that it can get a chance to contact with acid fluid during hydrocarbon accumulations, which may lead to its in situ dissolution-reprecipitation reaction, the reset of its chemical compositions and the variation of its contact relation with the clastic particles. In view of this, the densification of the sandstones takes place before hydrocarbon accumulation. The conclusions reveal that detailed observations under optical and electron microscope are necessary for further research in calcite origin, which should be the basis for further mineralogical and geochemical analyses as well as fluid inclusion analyses.

The satellite thermal infrared thermal radiation anomaly caused by seismic activity has become a promising technical means in the field of seismic research. The surface temperature condition used to extract the pre-earthquake brightness temperature anomaly can be obtained from the thermal infrared remote sensing data through ground heating, which provides a good data basis for the research. Based on the data of the Fengyun-2 satellite, the background field of thermal infrared brightness temperature and the simultaneous offset index values of four earthquakes were constructed and analyzed in Xinjiang and Qinghai provinces, and the spatial and temporal distribution characteristics of abnormal brightness temperature were revealed. The results show that there is usually obvious warming before earthquakes and at least once. In space, the abnormal area in Xinjiang is roughly fine strip and the abnormal area in Qinghai Province is massive. The larger the earthquake magnitude, the more obvious the change in brightness temperature.

Using social media to mine disaster information (situational awareness) is of great significance for disaster management. With the development of technology, researchers began to attempt to use social media data for situational awareness. In this paper, a spatial and temporal evolution framework of disaster situation is proposed. Combining the LDA model, UIE model and kernel density estimation method, the spatial and temporal evolution of rainstorm disaster situation in the interested period is analyzed. The results indicate that the location in social media texts can more accurately reflect disaster information, help to quickly understand disaster dynamics, and provide scientific basis for decision-making by governments and rescue organizations.

Aiming at the problem that the low coherence point subsidence value cannot be accurately obtained in the monitoring process of small baseline subset interferometric synthetic aperture radar (SBAS-InSAR), a spatial subsidence prediction method in mining area combining random forest and SBAS-InSAR was proposed. Based on 17 Sentinel-1A images from 2020 to 2021, the land surface subsidence values of 825 high coherence points in Chaochuan Mine in Ruzhou City were obtained by SBAS-InSAR.785 points were randomly selected and input into the random forest model to predict the remaining 40 points. The prediction results were compared with the SBAS-InSAR monitoring results. The results showed that the average relative error and R² between them are 1.67% and 0.93, and the results are accurate. This method can be applied to the subsidence prediction of low coherence points.

In recent years, ports and port groups have fiercely competed for the source of goods, leading some to blindly expand their scale. This has resulted in excess capacity, leading to a great waste of resources. Therefore, it is vitally important to define the scope of the port hinterland scientifically and reasonably, since this is the foundation for improving the current situation of harmful competition in the hinterland. Based on analysis of pertinent research, this paper identifies various parameters for characterising the quality of ports and employs the entropy weight approach and the Newton model to devise a model for the division of port hinterlands. The proposed model is then applied to the hinterlands surrounding ports in Northeast China. The findings indicate that the distance separating the ports and the hinterland has the most significant effect on the partitioning of the port hinterland in smaller-scale areas. Conversely, in larger-scale areas, the size of the port is the most influential factor. This study offers valuable guidance to port clusters, enabling them to position themselves effectively in the market and prevent the wasteful expenditure of economic resources on competition.

This study undertakes a comprehensive assessment of the first order spectrum response from high frequency surface wave radar (HFSWR) during the severe tropical storm Lionrock, employing six consecutive days of current data collected via HFSWR within the southwestern region of the Taiwan Straits. The findings elucidate that the storm phase instigated marked modifications to ocean surface currents as observed by HFSWR. A substantial broadening of the first order spectrum is notable during the storm's duration, accompanied by an enhancement of the first order spectrum's SNR by roughly 2-10dB.

As the global climate warms, vegetation is more significantly affected by climate. The time and spatial resolution of the data is 1 month and 1 km, respectively, based on NDVI data from the Modis satellite collected in Inner Mongolia between 2001 and 2021. The Hurst index technique is used to conduct a thorough quantitative study on the vegetation change and projected development trend of Inner Mongolia using the maximum value synthesis methodology, trend analysis approach, Sen slope estimation analysis, and Mann-Kendall testing. The results show that: from 2001 to 2021, the proportions of slightly improved and significantly improved vegetation areas were 5.71% and 42.57%, respectively, and the proportions of slightly degraded and severely degraded areas were 12.03% and 4.93%, respectively. After 2021, the opposite trend of vegetation in the study period is more obvious than the same trend. The area of the opposite trend accounts for 76.45%, and the area of the same trend accounts for 23.54%. In Inner Mongolia, the areas where vegetation coverage will increase and continue to increase in the future account for 7.31% and 20.89% respectively, and the areas where vegetation coverage will decrease and continue to decrease account for 69.15% and 2.66% respectively. In the past 21 years, the overall vegetation has been in an improved state, but in the future, if the previous situation continues to develop, without any intervention measures, the vegetation will tend to degenerate.

Ultra-Wideband (UWB) Ground Penetrating Radar (GPR) is a short-range target detection based on UWB technology for the detection and identification of buried objects, e.g., for mine detection, earthquake rescue, etc. In this paper, an ultra-wideband MIMO radar system is investigated for the detection of underground targets. Different measurements of several stationary targets were carried out using the radar system and the experimental results agreed well with the simulation results. The reconstructed images show that the designed MIMO radar system is able to successfully detect and image subsurface targets.

When a basin-wide flood occurred in the flood season of 2020, there was the largest peak flow since the construction of the Three Gorges Reservoir in the upper reaches, while in the middle and lower reaches, the water level of the lower reaches of Jianli was in the forefront of history, and some stations continued to exceed the warning level for more than 40 days. In this paper, for the Yangtze River Basin in the summer of 2020, the rainfall characteristics are analyzed, and the causes of the abnormal flood-causing rainfalls are studied from the perspectives of the planetary-scale circulation characteristics (such as the subtropical high over western Pacific, the blocking high in the middle and high latitudes of Eurasia, and the summer monsoon in South Asia) and the classification of the low-level jet.

Spaceborne synthetic aperture radar (SAR) is an active radar system carried on a satellite, with the help of synthetic aperture technology to achieve all-day, all-weather, high-resolution imaging, it has been widely used in military reconnaissance, resource exploration, deep space exploration and other fields. As the electromagnetic environment becomes more and more complex, spaceborne SAR is more likely to receive radio frequency interference (RFI) from other radar equipment in the working frequency band, which will affect the interpretation of SAR images and lead to image quality degradation. Although there are endless methods for RFI detection and suppression, each method has its limitations and can only deal with specific interference scenarios. A unified, robust and high-precision SAR interference detection and suppression platform is urgently needed. Based on SAR signal processing technology, this paper uses multiple methods to jointly design a unified high-robust system for SAR data containing different types of RFI, and realizes efficient detection and suppression of spaceborne SAR RFI. The popularization and application of this system is of great significance to improve the anti-interference ability of SAR system, improve the processing efficiency of a large number of SAR data in complex scenes and ensure the accuracy of data.

Most existing depth networks that perform hyperspectral anomaly detection (HAD) using reconstruction errors tend to fit anomalous pixels, resulting in small reconstruction errors for anomalies, which are not favorable for separating targets from hyperspectral images (HSIs). To achieve a superior background reconstruction network for HAD purposes, a self-supervised blind-block reconstruction network (termed BockNet) with a guard window is proposed. BockNet creates a blind-block (guard window) at the center of the network's receptive field, making it unable to see the information inside the guard window when reconstructing the central pixel. This process seamlessly embeds a sliding dual-window model into our BockNet, where the inner window is the guard window, and the outer window is the receptive field outside the guard window. Naturally, BockNet uses only the outer window information to predict/reconstruct the central pixel of the perceptive field. During the reconstruction of pixels inside anomalous targets of different sizes, the targets typically fall into the guard window, weakening the contribution of anomalies to the reconstruction results and allowing these reconstructed pixels to converge to the background distribution of the outer window region. Accordingly, the reconstructed HSI can be regarded as a pure background HSI, and the reconstruction error of anomalous pixels will be further enlarged, thus improving the discrimination ability of the proposed network for anomalies. Extensive experiments on two datasets show the competitive performance of our BockNet compared to state-of-the-art detectors.

In the urban environment, multipath (MP) interference and non-line-of-sight (NLOS) reception are the main factors leading to the positioning deviation of the Beidou Navigation Satellite system. At present, double difference and sparse estimation are widely used to estimate the uncorrelated errors caused by MP interference and NLOS reception, but the identification of uncorrelated errors for general distributions is still limited. In this paper, to improve the recognition ability of MP interference and NLOS reception, we consider the comprehensiveness of the evaluation indicators and design a weighting matrix based on the dual-satellite carrier to noise ratio (C/N0) and single-satellite elevation and combine it with sparse estimation. In addition, to eliminate the estimated uncorrelated errors conforming to the general distribution, we employ the extended Kalman filter method to transform them into linear distributions and obtain the results of eliminating the errors. The experimental results show that the proposed method can improve the accuracy of position and speed by up to 28.12% and 93.47% compared with the existing double difference positioning and avoid the problem of local migration.

In order to solve the problem that traditional GIS methods are difficult to handle massive data, complex spatial relationships and accurate geographic phenomenon modeling. By fusing terrain and feature models, geographic phenomena can be efficiently and accurately represented and analyzed to provide a reliable basis for decision making.This paper proposes a fusion modeling technique for massive strip 3D terrain and feature models, aiming to solve the problem of effective fusion and modeling of 3D terrain and feature models for strip data in large-scale spatial data processing. Based on open-source geographic information technology, the traditional method is designed to unify and model 2D vector data, raster data, BIM data and 3D data from different sources, formats and scales to form a standardized spatial data set, in response to the problems of high computational complexity, large storage requirements and data redundancy in processing massive data. For the characteristics of strip data and 3D terrain and feature models, geometric matching and semantic matching methods are used to fuse and process vector data representing the same entity with different attribute information and spatial location information to solve the problem of data redundancy and duplication; it is used to describe the different levels and precision of spatial data, and the appropriate data precision can be flexibly selected in the fusion modeling process, so as to achieve a balance between data quality and calculation The purpose of balancing data quality and computational efficiency. Based on LOD hierarchical detail technology fusion processing of terrain-independent BIM models, terrain-dependent fusion of linear 3D feature models, to achieve the surface fitting of multi-source 3D model data in a specific scene, to improve the 3D scene display effect.In terms of data processing efficiency: traditional GIS methods may face bottlenecks in computation and storage when dealing with massive data. The efficiency of data processing can be improved by adopting fusion modeling technology for massive data. For spatial relationship accuracy: fusion modeling technology can capture the complex spatial relationships between terrain and features more accurately, providing more accurate simulation and analysis results.

With the completion of railroad line resurvey work, the measurement data are scattered in various units and lack unified storage and management, which brings many challenges and problems to railroad operation and maintenance. To solve this problem, this study proposes a multi-version, multi-temporal spatial data storage technology, as well as a distributed fast storage technology and a spatial service engine by studying the data management work, aiming to realize unified and fast storage, management and sharing of railroad survey data to improve the reliability, consistency and validity of data. First, this study focuses on the multi-version and multi-temporal spatial data storage technology. By adopting this technology, a large amount of spatial data generated by railroad line resurvey can be effectively handled, and the storage and management of different versions and temporal phases of data can be realized. Meanwhile, distributed fast storage technology is introduced to provide high-speed data access and processing capability to cope with the storage and processing demand of massive data. In addition, using the spatial service engine, fast query and analysis of measurement data can be realized. Second, this study also conducts in-depth research on the railroad line measurement data management system. First, the overall architecture of the system is designed to ensure the scalability and flexibility of the system. Then, the data structure of the system was optimized to accommodate the storage and query requirements of multi-version and multi-temporal data. In terms of technical architecture, distributed storage technology is adopted to realize fast storage and retrieval of remote sensing, 2D vector and 3D model data. In addition, the sharing and exchange of measurement data is realized through information sharing and information interface. The innovation of this study is to propose a multi-version and multi-temporal spatial data storage technology, and to realize the fast storage of different types of data by using distributed storage technology. By constructing a railroad line measurement data management system, the problems caused by decentralized data management can be effectively solved and the reliability, consistency and validity of data can be improved. This is of great significance for railroad operation and maintenance.

Visual navigation is an effective way to achieve high accuracy navigation and positioning of asteroids. Selecting effective navigation features and matching them accurately is critical to the visual navigation. The asteroid is small in size, rotates fast, and has a large change in viewpoint between images, thus the image features are significantly distorted. The visual navigation methods based on 2D feature extraction and matching cannot be adapted to such scenarios. Therefore, this paper proposes a visual navigation method based on 3D navigation features. First, the multi-view photometric method is used to produce high-resolution terrain based on low-resolution terrain and high-resolution images taken under different illumination conditions; then, based on a certain reflectance model, the high-resolution terrain is rendered using the current illumination and observation conditions to obtain navigation feature, which is consistent with the texture of the current captured image. Finally, considering the uncertainty of camera pose estimation, a motion-constrained discriminative correlation filter (DCF) is proposed to adaptively match features with distortion. Experiments show that the proposed method can be effectively employed in visual navigation.

Total suspended solids (TSS) represent a critical parameter for water quality assessment and an essential indicator for remote sensing water quality monitoring. The Gaofen (GF) series satellites are a crucial component of China's “Major Special Project of High Resolution Earth Observation System”, widely applied in fields like environmental protection and land resources monitoring. In our efforts to further explore the applicability of the satellite network composed of GF-1 and GF-1B for TSS change monitoring, we developed a TSS concentration model for the coastal waters south of Yantai, China, and discussed the impact of different aerosol models on inversion results. Our findings indicate that the optimal band combination for TSS inversion modeling in the coastal waters south of Yantai is "ln(B3)-ln(B2)". The model best suited for TSS concentration inversion in the study area is y = 9.9348x2 + 29.713x + 23.879, with a coefficient of determination (R^2) of 0.73. The choice of aerosol model when employing the FLAASH model for atmospheric correction influences the predictive accuracy of the inversion model for TSS concentration. GF-1 and GF-1B satellite data are promising for inverting TSS concentrations in water bodies, and high temporal resolution satellite constellation images are beneficial for enhancing regional water body environmental monitoring.

With the emergence of new geo-information industry, new products and new applications, the contradiction geoinformation security confidentiality requirements and the economic and social development of geo-information data sharing needs is increasingly acute, and through technical prevention and control is the necessary means to ensure the application of geo-information security at this stage, for different categories of geo-information, do not use the scene to take the corresponding technical measures is to achieve the whole process of geo-information security management The key to achieve the whole process of geo-information security management, and it is important to promote the application of geo-information security. Therefore, this paper reviews the latest progress and application of geo-information security technology in recent years, and on this basis puts forward several suggestions to meet the requirements of geo-information security to solve.

Serving as one of the important error sources in radio geodesy technology, the tropospheric delay correction aims to enhance the measurement accuracy of the earth observation technology in light of its accurate estimation. Given the difference of zenith tropospheric delay (ZTD) parameters estimated by the Global Pressure and Temperature 3 (GPT3) model in different regions, this paper used the ZTD calculated by the Global Navigation Satellite System (GNSS) to correct the residual of GPT3-derived ZTD, introduced Back Propagation (BP) neural network to establish the functional relationship between the ZTD residual and the geographic location of the station, and constructed the regional ZTD (RZTD) model that corporated the compensation of the ZTD residual. The RZTD model can estimate the ZTD parameters in real time by merely inputting the station position (longitude, latitude, and geodetic height) and time. The GNSS data of 19 continuously operating reference (CORS) stations in Hong Kong from 2018 to 2020 were selected for the experiment, and the accuracy of GNSS ZTD was verified using radiosonde (RS) data. It was found that the root mean square error (RMSE) and bias (Bias) of ZTD estimated by GNSS were 0.013 and 0.001 m, respectively. Taking the ZTD calculated by GNSS and RS as the true value, the accuracy of the RZTD model was verified. It was observed through the experiment that the average RMSE and Bias of the model were 0.013/0.001 m and 0.022/-0.003 m, respectively, exhibiting good stability in different seasons. The statistical results revealed that the RZTD model outperformed the GPT3 model with a 72% improvement in accuracy, which could meet the demand for real-time correction of regional tropospheric delay.

A five-channel imaging spectropolarimeter (AFIS) aimed for airborne remote sensing applications is demonstrated. Spectral information is obtained by grating diffraction in AFIS, and spatial image is captured by the moving airborne platform. The full-Stokes polarization information is acquired by a polarization modulator consisting of 2 thick retarders and one linear polarizer. The AFIS works in push sweep mode, and the spectral information and polarization information of one-dimensional spatial pixels can be recorded by one camera shot. The AFIS is less sensitive to the vibration compared to time modulated imaging spectrometer, and it is more suitable for airborne remote sensing applications. The principle and feasibility are proved by simulations.

In order to improve the efficiency of remote sensing satellite, with the limited performance of the existing optical remote sensing payload, a satellite constellation construction idea combining with the low and high resolution is proposed. The working mode of the narrow-area high-resolution satellite is guided by the pre-processed information on the wide-area low-resolution satellite. Full using of the onboard memory and the pre-installed base map, the satellites only transmit changes and self-update, effectively reduce the transmission pressure of remote sensing data. The on-board processing capacity is used to judge cloud images, generate guidance information, guiding narrow-domain high-resolution satellites to effectively photograph specific areas, and cooperate with SAR remote sensing satellites to achieve continuous and effective observation of specific regions. The proposed idea of cloud judgment guidance and regional coding can improve the observation efficiency, and provide value for future applications of on-board target recognition and communication between satellites.

In the smoothing velocity model, wave equation is often used to simulate the wave propagation. The one-way wave is faster than the two-way wave in calculation, higher in accuracy than the ray approximation calculation, and has a wide range of applications. Traditional one-way wave calculation only focuses on its travel time information, and its amplitude does not conform to spherical diffusion principle and energy conservation. Therefore, this paper proposes a one-way wave with amplitude preserving, and finds that the phase information of its one-way wave can be guaranteed through amplitude preserving processing, which is verified by simulation.

In this study, we conducted an applied analysis of radar echo characteristics during a severe hailstorm event in Pengshui County, Chongqing, on April 3, 2023, using data from the Wuchuan C-band dual-polarization weather radar ROSE2.0, as well as conventional meteorological observations from ground weather stations and sounding stations. The results indicate that (1) this event occurred under the dominance of low-level warm advection, with strong echo development at the leading edge of the low-level jet. The C-band dual-polarization weather radar effectively detected sidelobe echoes, bounded weak echo regions, and the typical hailstorm structure of supercell storms with widespread echo tilting above the bounded weak echo region. (2) In the mature hailfall phase, the area of the 50dBZ strong echo was greater than or equal to (10×10) km², showing a continuously thickening and deepening strong echo tilting, a strong echo wall behind it, and a well-organized structure within the mid-level cyclone, including a clearly visible bounded weak echo region (BWER) and a large-area of low-level radial wind convergence shear. The characteristic values BASE, TOP and HGT of mesocyclone increased significantly.(3) During the occurrence of hail, the ZDR (differential reflectivity) in the low-value region ranged from -3dB to 0dB. However, within the BWER, ZDR values are greater than 3dB. There was a significant difference in the lower-level CC , ranging from 0.5 to 0.92, which was similar to the indicators in the S-band.

Soil classification is a classic topic in the field of modern soil research. With the development of remote sensing technology, accurate soil classification can be predicted using different temporal satellite data. Therefore, it is crucial to compare the temporal response of satellite images. In this paper, the study area chosen was in Mingshui County of northeastern China, which is known as a black soil area, and a total of 34 years of Landsat satellite images were obtained, ranging from 1984 to 2018. We extracted six reflectance bands and three tasseled cap transformation indexes from each image as characteristics, and constructed a random forest (RF) model for classification. The results indicated that we could accurately predict the spatial distribution of the soil classes using hyper-temporal satellite images with an overall accuracy (OA) of 80.56% and a Kappa coefficient of 0.704. Compared with the mono-, bi-, and multi-temporal data, the overall accuracy using hyper-temporal data was increased by 17.78%, 12.23%, and 8.89%, and the Kappa coefficient using hyper-temporal data was increased by 0.265, 0.185, and 0.136, respectively. Our study provides a new perspective for updating the legacy soil data.

FMCW (frequency modulation continuous wave) radar can achieve range measurement without blind zones. Pulse compression is an indispensable processing technique for improving range resolution in radar systems. General processing procedures for pulse compression of the FMCW signal include data reading with software, fast Fourier transforming (FFT) in an embedded processor, and data saving in memory. However, it is hard to achieve real-time processing of FMCW signals. In this paper, a real-time pulse compression processing technology of FMCW signal based on field programmable gate array (FPGA) is proposed. First, the processing system (PS) in FPGA receives the data sent by the upper computer and transmits the data to the programmable logic (PL) through direct memory access (DMA) to realize data zero filling. Then a window function is added with the parallel pipeline method. Finally, the processing results are transmitted to the upper computer through a gigabit network based on the User Datagram Protocol (UDP) protocol on the PS end, and displayed in real-time. The experimental results prove that this method can achieve real-time processing of FMCW signals on FPGA.

In this paper, a method of environmental geometric modeling utilizing a digital elevation model is proposed to solve the problem that the environmental modeling based on spectral function cannot describe the actual terrain. At the same time, by combining the shooting and bouncing ray (SBR) method with the facet-based scattering model (FBSM), the SBR-FBSM algorithm is proposed. Finally, the simulation analysis is carried out for the actual environment, the target environment composite scene, and the complex scene respectively to verify the rationality, and effectiveness of the proposed electromagnetic scattering algorithm and its applicability in the SAR imaging field.

When employing Graph Neural Networks (GNN) for point cloud classification, global information is passed through the iteration of hidden states of graph nodes. However, this method is often associated with high memory and time costs. To address this issue, we propose a strategy that integrates prior OpenStreetMap (OSM) tags into LiDAR point cloud classification, which enhance iteration speeds by fixing the initial hidden states of certain graph nodes. First, we perform over-segmentation of the LiDAR point cloud to obtain super points and embedded features. Then, OSM tags are affixed to the super points and incorporated into the GNN to update other super point embedded features. Our results demonstrate that coupling OSM with GNN significantly improves classification accuracy for roof, tree, and impervious surface categories.

Detecting building changes based on high-resolution remote sensing images constitutes one of the most important tools for urban management, methods based on deep learning are substantially more efficient than traditional methods of detecting changes. Nevertheless, there are several problems with most models, such as incorrect edge detection results and missing detection results. Due to continuous down sampling, there is a loss of detailed information in the image as well as an insufficient ability to characterize features. Our paper proposes a multi-scale feature dependency network to improve the performance of change detection models. By combining the multi-scale embedding module with the spatiotemporal attention module, we can extract the original image features, preserve the rich context with multi-scale convolution, and establish long-short distance feature dependencies across different scale patches using long-short distance attention. Furthermore, we introduce a dynamic position deviation calculation method to accurately determine the relative position relationship between features during the attentional action phase. In this manner, the accuracy of model detection will be improved. Multi-level feature fusion helps improve the model's ability to perceive detail by fusing features at different scales, capturing semantic information from these features, and enhancing the feature representation capability. The publicly available building dataset LEVIR-CD was used to conduct comparative experiments with mainstream methods. In both visual detection and quantitative evaluation metrics, our network outperforms the comparison method.

Regional positioning can provide position information for objects in a certain area. In order to solve the positioning accuracy problem of regional positioning, a regional positioning system based on the integration of Inertial Navigation System (INS), Beidou System (BDS) and UWB System is proposed, so as to achieve accurate positioning in a certain area. Firstly, a multi-dimensional information fusion framework based on Extended Kalman filter (EKF), redundant information filter (RIF) and weighted fusion was proposed. In this framework, RIF_LP (Liner Prediction) algorithm and ILCTA (Information Lossless Coordinate Transformation Algorithm) are proposed. Finally, the experimental results show that the positioning fluctuation amplitude of the fusion positioning scheme is greatly reduced, and the positioning accuracy is higher than each single positioning scheme. The proposed fusion framework and algorithms have certain reference value for the construction of regional fusion positioning.

To address the challenges pertaining to the sluggish 3D reconstruction of the road surface, excessive workload, and limited adaptability, this study undertook the feature extraction of road point cloud data exhibiting diverse linear shapes, leveraging vehicle-mounted laser point cloud data. Additionally, a novel approach was devised to automatically construct the road triangulation network, thereby achieving a high-precision digital restoration of the road surface. Initially, the data attributes of the laser point cloud obtained from road vehicles are considered, and a data preprocessing technique is devised using the Alpha Shapes algorithm and the point cloud grid thinning algorithm. This method aims to retain crucial information pertaining to the elevation of the road surface, linear boundaries of the road surface, and skeletal features of the road surface. Subsequently, a technique is devised for constructing a Delaunay triangulation network using ordered point cloud data, based on the boundary points and feature points of the road surface. Additionally, an investigation is conducted on a method for constructing a road surface triangulation network with road boundary constraints, enabling the identification and removal of triangles located outside the road surface, thereby achieving precise restoration of the road surface. Ultimately, the assessment of the generated 3D pavement model's quality validates the method's feasibility and accuracy.

To solve the problem of insufficient information extraction in deep learning point cloud classification networks due to the rapid increase in dimension, as well as the issue of channel information balance leading to a lack of better discrimination ability for object features, we propose a point cloud classification network with steadily increasing feature dimensions and attention mechanism. By transforming multiple combinations of channel numbers and layers, the optimal combination of channel numbers and layers in the PointNet feature extraction module was determined, and the channel numbers were changed from (64, 64, 64, 128, 1024) to (32, 64, 128, 256, 512, 1024), resulting in a network with a feature dimension that increases steadily by a factor of 2.Incorporate the SENet channel attention mechanism into the evenly dimensionality-increased network, which weights the feature channels to enable the network to identify more accurate features and thus improve the discriminative power of object features. After training and testing on the ModelNet40 dataset, the improved classification network achieved an overall accuracy improvement of 2.58% compared to the PointNet classification network.