Remote Sensing Image Recongnition Based On Neural Networks

Remote sensing(RS)is a technology by adjudging,measuring and analyzing character of targets at long bowls.It realizes the collection,process,recognition and classification of image information on the earth.RS technology has good advantages on dynamic cycle,data,abundant information and easily acquisition as a tool in age of information, so it is the most effective technology and means of obtaining space-time information. RS image is mostly used to map relief maps,make orthograph and thematic maps by professional interpretation,which can be saved in the databases of geographical information system( GIS)or up dated the GIS databases.Patern classification is a key technique in remotely sensed image processing. Although the research history of patern classification techniques is quite long,users require for more accurate classification result and smaller computing load now. So there is an urgent need for modern patern classification methods to solve the modern remote sensing applicationsIn recent years,with the development of the theory about artificial neural network system,the neural network technology is becoming increasingly an efective means of classification processing of remote sensing images.Compared with classification of the traditional Bayesian statistics, the results show it has not only the highest accuracy but also the fastest speed of classification.Based on the conclusion of existing research fruit,the thesis discusses some artificial neural network methods, such as BP, Kohonen, FKCN and AFKCN neural network, which is applied to RS image classification.In this thesis,the following aspects have been researched:1) At first,this thesis reviews some principled problems about the practical application of the methods to remote sensing data classification.2) After analyzing the methods of traditional supervised and unsupervised classification, the thesis presents classified thought and algorithmic flow to six kinds of traditional classification algorithms.Pointing out the short coming of traditional classification method based on statistics pattern recognition and combining with the method of fuzzy mathematics,fuzzy pattern recognition is inducted into the remote sensing image classification.3) BP neural network is widely used for classification of remote sensing image data nowadays.And then the thesis practices supervised classification with BP algorithm on the base of the clustering image supported by ERDAS software.4) Fuzzy Kohonen neural network(FKCN) is a network which integrate FCM algorithm and Kohonen network,and it can adjust the parameters of the network automaticly.The experiment show that remote sensing image clustering by fuzzy neural network is an efective method.5) Athough Fuzzy Kohonen clustering network(FKCN) shows great superiority in processing the ambiguity and uncertainty of image , but there are many defects such as the number of network nodes can't be determined automatically ,the speed of network convergence is very slow ,and the computation cost is too large. To overcome these defects ,an adaptive FKCN model(AFKCN) is presented in this paper ,which can determine the network structure automatically according to the gray level distribution character of the image. By using the new fuzzy intensification operator and implementing a sample space transition in the network learning procedure ,the network convergence speed is greatly improved and the clustering result is also improved.