Study On Land Use Change Detection Methods Base On High-resolution Remote Sensing Images

Change Detection of land use is not only important content of Global Change research, but also important basis of scientific decision-making and management in sustainable use of resources. Change Detection of land use by Remote Sensing is the most effective way to get information of land use change. High-resolution remote sensing images are applied in many fields with the fast-development of remote sensing technology and the improvement of satellite spatial sensors resolution. With more details, clearer geometry feature and more space information, the operation of high-resolution remote sensing image change fundamentally. Traditional remote sensing image processing will no longer apply to high-resolution remote sensing images. It needs comprehensive analysis of existing change detection technology and methods, and design Technical route of change detection suitable for high-resolution remote sensing image.An object-oriented remote sensing image processing system named ELU is designed and implemented by Remote Sensing Laboratory of Shanghai Jiao Tong University, which supports in part by the research project of Content-based Search in Image and Change Detection and Auto-updating of Special Target. It is administered by Shanghai Science and Technology Committee. Change detection of High-resolution remote sensing images is relatively deeply studied in this thesis supported by the project. Change detection module of ELU system is also designed and implemented according to the study.After study of the normal methods of changing detection, this paper mainly researches remote sensing changing detection method which is based on historical land use vector data, and explores change detection methods suitable for high-resolution remote sensing images. Specific studies as follows: First of all, this paper studies a work pattern combined vector data and raster data which based on raster coordinate system. researeh the way to use vector boundary to quickly extract the raster image data and a fast method of vectorization for remote sensing segmentation map. Secondly, this paper studies change detection methods based on vector data. Aiming to high-resolution remote sensing images, this paper uses the area of change pixel in adjacent to determine the span. Besides change of pixel, we must also consider change in the space between pixels, a single isolated or several adjacent change pixels can not determine the changes in span. Propose method Combined with many change detection algorithms integrated the advantages of the change detection algorithms. Finally, after the change detection, this paper studies updating of vector data through segmentation, classification and vectorization for the change span, product the new vector data.