The easy and ready access to Landsat datasets and the ever-lowering costs of computing make it feasible to monitor the Earth’s land cover at Landsat resolutions of 30 m. However, producing forest-cover products rapidly and on a large scale, such as intercontinental or global, is still a challenging task. By utilizing the huge catalog of satellite imagery as well as the high-performance computing capacity of Google Earth Engine, we proposed an automated pipeline for generating 30-m resolution global-scale forest map from time-series of Landsat images. We describe the methods to create products of forest cover at a global scale. First, we partitioned the landscapes into subregions of similar forest type and spatial continuity. Then, a multisource forest/nonforest sample set was established for machine algorithm learning training. Finally, a random forest classifier algorithm was used to obtain samples automatically, extract the characteristics of satellite images, and establish the forest/nonforest classifier models. Taking Landsat8 images in 2018 as a case, a novel 30-m resolution global forest cover (GFC30) map has been produced. The result shows that by the end of 2018, the total forest area in the world was 3.71 × 109 ha. The accuracy evaluation of GFC30 for 2018 was carried out using verification points via stratified random sampling of a MODIS land cover map (MCD12C1 product in 2012) and verified on high-resolution satellite imagery (e.g., Google Earth). According to the validation result, the overall accuracy of GFC30 for 2018 is 90.94%.
QuickBird satellite images are widely used in many fields, and applications have put forward high requirements for the
integration of the spatial information and spectral information of the imagery. A fusion method for high resolution
remote sensing images based on ISVR is identified in this study. The core principle of ISVS is taking the advantage of
radicalization targeting to remove the effect of different gain and error of satellites’ sensors. Transformed from DN to
radiance, the multi-spectral image’s energy is used to simulate the panchromatic band. The linear regression analysis is
carried through the simulation process to find a new synthetically panchromatic image, which is highly linearly
correlated to the original panchromatic image. In order to evaluate, test and compare the algorithm results, this paper
used ISVR and other two different fusion methods to give a comparative study of the spatial information and spectral
information, taking the average gradient and the correlation coefficient as an indicator. Experiments showed that this
method could significantly improve the quality of fused image, especially in preserving spectral information, to
maximize the spectral information of original multispectral images, while maintaining abundant spatial information.
This paper provides an analysis of main land types and image features of their change information by using
RADARSAT-2 satellite data of Chengdu, Yantai and Shantou. It also deals with the geometric rectification accuracy of
RADARSAT-2 data with a corner reflector. A method of fusing RADARSAT-2 data and optical image was used to
extract change in land use monitoring and to assess accuracy, in order to provide an effective supplement for the new
round investigation of the national resources.
Focusing on the fusion problem of the multispectral (Ms) and panchromatic (Pan) images from the same scene, a novel
image fusion method is proposed based on nonsubsampled contourlet transform (NSCT) and human visual system
(HVS). The most traditional fusion methods are IHS, PCA and Brovey transforms, which can bring the phenomenon of
spectral distortion. Avoiding this problem, the wavelet transform is usually used in image fusion in recent years, but it
only can capture limited directional information. Compared with the wavelet and other transforms, the contourlet
transform has the characteristics of multi-scale, time-frequency localization and multi-directions. However, due to the
lack of translation invariance of the contourlet transform, this paper uses the nonsubsampled contourlet transform. The
basic procedure consists of four steps. Firstly, the NSCT is performed on Pan image and the intensity component I of Ms
image with HIS transform, which can obtain the low frequency subband and highpass directional coefficients of each
image. Then a new fusion rule is presented based on HVS: corresponding low frequency and highpass components are
divided into several blocks, and contrast variance of every block is calculated, followed by a selection of an adaptive
threshold which can be used to construct the new low frequency and highpass components. The blocks with higher
contrast variance will be chosen. Thirdly, the new intensity component Inew with high spatial resolution is obtained by
performing the inverse NSCT on the attained coefficients. Finally, the inverse IHS using Inew component is performed
and the new fused multispectral image is obtained. According to the quantitative evaluation criteria, it is shown that the
proposed method can effectively preserve spectral information, improve spatial information of the fused image, and
outperform the traditional IHS, PCA, Brovey, wavelet and contourlet methods.
Position accuracy is the base of remote sensing image application. In this paper, the effect of the number, the
distribution and the accuracy of ground control points (GCPs) and DEM in different scales for the image rectification is
analyzed in detail. Quantitative evaluation of orthoimage is performed. The mathematical functions for calculating
the position accuracy of the orthoimage are given based on different georeference information. The relation between
the final accuracies of orthoimages and the accuracies of GCPs and DEM is analyzed based on the experiment results.
It shows that accuracies of the checked orthoimages coincide with the calculated accuracies. The final accuracy can be
estimated with the method described in this paper if the accuracy of control data is known. On the other hand, if the
final accuracy of the orthoimage were determined, the least requirements for the accuracies of GCPs and DEM could be
calculated by the mathematical functions.
With the development of remote sensing satellites, the data quantity of remote sensing image is increasing tremendously, which brings a huge workload to the image geometric rectification through manual ground control point (GCP) selections. GCP database is one of the effective methods to cut down manual operation. The GCP loaded from database is generally redundant, which may result in a rectification slowdown. How to automatically optimize these ground control points is a problem that should be resolved urgently. According to the basic theory of geometric rectification and the principle of GCP selection, this paper deeply comprehends some existing methods about automatic optimization of GCP, and puts forward a new method of automatic optimization of GCP based on voronoi diagram to filter ground
control points from the overfull ones without manual subjectivity for better accuracy. The paper is organized as follows: First, it clarifies the basic theory of remote sensing image multinomial geometric rectification and the arithmetic of how to get the GCP error. Second, it particularly introduces the voronoi diagram including its origin, development and characteristics, especially the creating process. Third, considering the deficiencies of existing methods about automatic optimization of GCP, the paper presents the idea of applying voronoi diagram to filter GCP in order to complete
automatic optimization. During this process, it advances the conception of single GCP's importance value based on voronoi diagram. Then by integrating the GCP error and GCP's importance value, the paper gives the theory and the flow of automatic optimization of GCPs as well. It also presents an example of the application of this method. In the conclusion, it points out the advantages of automatic optimization of GCP based on the voronoi diagram.