Classifying land cover using high-resolution remote-sensing images is challenging. The emergence of deep learning provides improved possibilities, but owing to the limitations of network structures, traditional convolutional neural network methods lose essential information about boundaries and small ground objects. We propose a superpixel-optimized convolutional neural network (SOCNN) framework to overcome this weakness. The SOCNN includes three modules: a semantic segmentation module, a superpixel optimization module, and a fusion module. The performance of the first module was evaluated using several common networks. PSPNet outperformed other networks, obtaining a pixel accuracy of 83.25%, a Kappa coefficient of 0.7862, and a mean intersection over union of 64.19%. Weighted loss was introduced to alleviate the effect of category imbalance, and the class pixel accuracy of category 11 improved by 19.77% with a weight of 20. The subpixel model was evaluated, and the pixel accuracy reached 83.43% with the superpixel-FCN method. Our superpixel optimized module improved the pixel accuracy of the object boundary by 1.37% when the fusion factor was 0.65. These results show that the SOCNN method is effective for recovering boundary information.
The difficulty of accurate and large-scale measurement for surface parameters limits the regional surface soil moisture (SSM) estimation using synthetic aperture radar (SAR). Moreover, the coarse resolution of soil moisture products generated by existing methods, which fuse SAR and passive microwave products, cannot fully satisfy the requirement of specific regional applications. To solve this problem, an SAR-optical data fusion method for soil moisture estimation (SOFSME) based on a cascade neural network is proposed in this study. SOFSME obtains surface parameters from historical soil moisture images and related environmental images to estimate a SSM image with high resolution at large scale from Sentinel-1A C-band SAR data. Validation experiments in single and multiple land-use type areas showed that the SOFSME performed best on bare soil areas with a median root mean square error of 0.0203. The median universal image quality index of estimated soil moisture image was 0.1454, which was better for single cropland areas than multi-land-use type areas. The Pearson correlation coefficient showed a median value of 0.7645 in both experiments. These results showed that the SOFSME had high accuracy, availability, and stability in regional soil moisture estimation. Compared with existing methods, the SOFSME can provide high-quality soil moisture images and does not directly depend on field measurement data. Thus, the proposed SOFSME method is of great value for high-resolution soil moisture estimation in more regional applications.
Currently, many satellite data are used to invert soil moisture. However, there is no study about quantitative evaluation of observation capability of GF-1 wide field of view (WFV) sensors for soil moisture inversion. Therefore, we proposed a method to evaluate it. We used WFV, Landsat8 Operational Land Imager (OLI), and Moderate-resolution Imaging Spectroradiometer (MODIS) data to invert soil moistures in Wuhan from September 2013 to September 2014 based on the Perpendicular Drought Index (PDI) and modified PDI (MPDI). From the estimated results, the R2 values, and standard error, we found that both the PDI and MPDI had a significantly negative linear correlation with soil moisture (P<0.01). Through the values of R, mean absolute error, mean relative error, and root mean square error, we found that a strong relativity existed between the estimated and observed soil moistures. It was evident from the results for the WFV, OLI, and MODIS that the performances of WFV and OLI were consistent and that WFV performed better than MODIS. All the results indicated that WFV sensors had a high observation capability for soil moisture inversion in Wuhan. The comprehensive evaluation results for the performance of the PDI and MPDI proved that the MPDI performed better for soil moisture inversion than did the PDI.
KEYWORDS: Web services, Databases, Human-machine interfaces, Data modeling, Data centers, Taxonomy, Data processing, Information fusion, Standards development, Data integration
Antarctica plays a key role in many scientific questions, of which those related to global climate change are probably the
most prominent examples. There are many researches on Antarctic are carried out at present, and some special institutes
sponsored by public and private communities are responsible for antarctica data management and maintenance.
Antarctic Spatial Data Infrastructure (AntSDI) [1]sponsored by SCAR's Standing Committee on Antarctic Geographic
Information (SC-AGI) is the one responsible for Antarctica spatial data maintenance and sharing by means of OGC
standard and specification. Antarctica Spatial Data Infrastructure (AntSDI) has already collected huge volumes of
geospatial data and offer an opening geospatial information service. In order to management and use Geospatial data
efficiently, and enable most of the users can access to Geospatical data and service at will, we firstly must registry data
and service into one or more registry center, then we should construct a building system which can supply users a
uniform interface to access data and service in registry center and user also can add their own data and service to system
and become part of system's capability. in this paper we present GeoAnt, a prototype interoperable AntSDI building
system. GeoAnt is a three-tier standard-based open geospatial web service system which fully automates data discovery,
access, and integration steps of the geospatial information discovery process under the interoperable service framework.
The paper discusses the system architecture, the individual components of the system and the use of the system in the
international project- Grove Mountains GIService Portal (GMGP).
KEYWORDS: Sensors, Data modeling, Process modeling, Surface plasmons, Web services, Performance modeling, Data processing, Environmental sensing, Flame detectors, Computer security
A common Sensor Web data service framework for Geo-Processing Workflow (GPW) is presented
as part of the NASA Sensor Web project. This framework consists of a data service node, a data
processing node, a data presentation node, a Catalogue Service node and BPEL engine. An abstract
model designer is used to design the top level GPW model, model instantiation service is used to
generate the concrete BPEL, and the BPEL execution engine is adopted. The framework is used to
generate several kinds of data: raw data from live sensors, coverage or feature data, geospatial
products, or sensor maps. A scenario for an EO-1 Sensor Web data service for fire classification is
used to test the feasibility of the proposed framework. The execution time and influences of the
service framework are evaluated. The experiments show that this framework can improve the quality
of services for sensor data retrieval and processing.
KEYWORDS: Databases, Web services, Image retrieval, Sensors, Data processing, Java, Internet, Remote sensing, Standards development, Geographic information systems
Recent advances in open geospatial web service, such as Web Coverage Service as well as corresponding web ready data
processing service, have led to the generation of large amounts of OGC enabled links on Internet. Recently a few search
engines that are specialised with respect to geographic space have appeared. However, users do not always get the
effective OGC WCS link information they expect when searching the Web. How to quickly find the correct spatial aware
web service in a heterogeneous distributed environment has become a "bottleneck" of geospatial web-based applications.
In order to improve the retrieval efficiency of OGC Web Coverage Service (WCS) on WWW, a new methodology for
retrieving WCS based on clustering capability aware spatial search service middleware is put forward in this paper.
KEYWORDS: Raster graphics, Web services, Databases, 3D visualizations, Whole body imaging, Internet, Systems modeling, Data integration, Data modeling, Satellite imaging
With the development of earth observation technology provides more geo-information, a large amount of spatial data
with different format can be obtained through the Internet. So in the construction of spatial data infrastructure and digital
earth, how to effectively utilize and integrate the spatial information to fulfill the need of people from different
departments and fields in the distributed environment is an important issue. Based on the web coverage service
specification provided by OGC, a new coverage data service architecture including multi raster data source(MRDS),
raster data engine(RDE), raster transfer object (RTO), raster business object(RBO), WCS business object (WBO), WCS
business interface (WBI) and WCS Facade (WF)is put forward, which can effectively publish many kinds of imagery
and DEM data, which cab be deployed in distributed network environment, published through a unified map definition
file and implemented GetCapabilities, DescribeCoverage and GetCoverage interface. Some key technologies including
GML coverage profile process and multi WCS combination are described in detail. Finally, a client which could get
imagery and DEM data from Web coverage Service for 3D visualization is designed and implemented. It is proven a
feasibility way that Web 3D rendering by means of integration of multi Web Coverage service.
KEYWORDS: Geographic information systems, Internet, Databases, Web services, Data acquisition, Connectors, Data processing, Logic, Associative arrays, Roads
It is the GIS developing direction to integrate GIS with Internet. Internet GIS is not only used to publish distributed geographic information on line, but also to provide on line tools for distributed geographic information real-time analysis and processing. Internet GIS should make use of existing multi-sources and functions for geographic information analysis and processing. The distributed feature is inherent to geographic information, which needs distributed organization and processing approach. Internet GIS (Geographic Information System) is the integration of Internet and GIS, being used to implement publishing on line and distributed real-time analysis of geographic information. The problem of geographic information distributed management and processing should be solved by Internet GIS. This paper provides three approaches to the problem, especially the distributed component and Geo-Connector approach to Internet GIS. The distributed component method includes homogeneous, heterogeneous and conjoint methods, for the problem of distributed geographic information publishing and real - time analysis. With instance, the feasibility and practicability of distributed component method are interpreted in detail. The concept of Geo-connectors in GIS application server side for organization and management of distributed heterogeneous database is presented and a case of heterogeneous geo-data access and interoperability is studied
With the increasing maturity of distributed object technology, CORBA, .NET and EJB are universally used in traditional IT field. However, theories and practices of distributed spatial database need farther improvement in virtue of contradictions between large scale spatial data and limited network bandwidth or between transitory session and long transaction processing. Differences and trends among of CORBA, .NET and EJB are discussed in details, afterwards the concept, architecture and characteristic of distributed large-scale seamless spatial database system based on J2EE is provided, which contains GIS client application, web server, GIS application server and spatial data server. Moreover the design and implementation of components of GIS client application based on JavaBeans, the GIS engine based on servlet, the GIS Application server based on GIS enterprise JavaBeans(contains session bean and entity bean) are explained.Besides, the experiments of relation of spatial data and response time under different conditions are conducted, which proves that distributed spatial database system based on J2EE can be used to manage, distribute and share large scale spatial data on Internet. Lastly, a distributed large-scale seamless image database based on Internet is presented.
KEYWORDS: Geographic information systems, Java, Data communications, Control systems, Databases, Human-machine interfaces, Data processing, Visualization, Logic, Internet
A systematic plan of building Web Geographic Information System (WebGIS) using ActiveX technology is proposed in this paper. In the proposed plan, ActiveX control technology is adopted in building client-side application, and two different schemas are introduced to implement communication between controls in users¡¯ browser and middle application server. One is based on Distribute Component Object Model (DCOM), the other is based on socket. In the former schema, middle service application is developed as a DCOM object that communicates with ActiveX control through Object Remote Procedure Call (ORPC) and accesses data in GIS Data Server through Open Database Connectivity (ODBC). In the latter, middle service application is developed using Java language. It communicates with ActiveX control through socket based on TCP/IP and accesses data in GIS Data Server through Java Database Connectivity (JDBC). The first one is usually developed using C/C++, and it is difficult to develop and deploy. The second one is relatively easy to develop, but its performance of data transfer relies on Web bandwidth. A sample application is developed using the latter schema. It is proved that the performance of the sample application is better than that of some other WebGIS applications in some degree.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.