Research On Change Information Extraction Of Remote Sensing Image Based On ANN

The change detection technology occurred not long ago in remote sensing image processing technology, it is used to recognize the change of the object on land with time pass. Due to the facility of obtaining and multi-source of remote sensing image, and the technology of image processing develops rapidly, the change detection technology becomes a important method in detecting the change of land use and land cover, the change of forest cover, the change of environment, the range of the disaster.The artifical neural network(ANN) is a hot question for discussion in the application of remote sensing, it is a self-adaptive dynamical system which is widely connected by large amount of neural units, it bases on distributing store and parallel processing, it is much the same as the manner of human processing information. ANN has the parallel architecture and the capability of parallel realizing, it study by exercise, every neural unit can compute and process the input information by itself. It has the capacity of settling down the trouble, integrating the information, synthesis reasoning, and rapid overall processing capacity, it can solve the regular problem arise from remote sensing image processing, due to its capacity of powerful processing capacity, it is widely used in the application of remote sensing.First the significance of change detection and the development overview of remote sensing was discussed, and the development overview of change detection technology was summarized, the superiority , shortcoming and its use in practice of every change detection method were analysed. The ANN methods used in remote sensing application were summarized. Two type of ANN (the self-organizing competitive neural network and the self-organizing feature map neural network (SOFM)) were used in change detection using remote sensing data in this paper, the arithmetic was researched and improved. The detection result was contrasted to conventional methods, the experiment verified that the methods used in this paper is efficient and is feasible in practice.