Incremental Updating Of Remote Sensing Data Based On The Changed Image Blocks

With the rapid development of the earth observation technology, the amount of RS data is increasing quickly, and it brings a great deal of pressure to the memory storage capacity, the study of the remote sensing data incremental updating has been an urgent problem.Incremental updating of remote sensing data is an efficient image updating method. People detect the difference between old and current images in the same area, extract the changed information and make a separate store. When people check the historical image, just replace the corresponding data in current image with the changed image blocks to generate a snapshot of the old image.At present, the incremental updating of remote sensing data has just started, and the existing method has data redundancy. To solve this problem, author conducts an in-depth research of the remote sensing data incremental updating, and designs the updating method based on the changed image blocks. The main research works are as follows:1) The method of extracting the changed image blocks is designed in this paper. First, people convert the image into vector data, and then group the polygons by distance, and merger the same set of polygons into a convex polygon to cut the old image to generate the changed image blocks same as the changed region. Moreover, author designs a reasonable storage and management method of the image blocks.2) Author designs the generation method and process of historical image snapshot according to the changed image blocks and the current image. Based on the principle of image fusion, author fuses the changed blocks and the current image together after removing the background of the blocks, to avoid the data loss by the covered background pixel, and facilitate the users to browse the image snapshot.3) The image incremental updating system is developed based on Visual C# and ArcEngine in this paper, and the Landsat-TM images of ShanXi province are used to do some experiments. In addition, author compares her own method with the one based on the image pyramid to validate the superiority of her method.