Study on Classification of Matching Area of Underwater Topography Based on BP Neural Network

In this paper,a new method for recognition and classification of matching background area and unmatching area,base on BP neural network,is proposed,aimed on unmatching phenomenon problem in terrain matching navigation.Firstly,statistics of the background area is researched and the statistical parameters with higher significance are given;secondly,BP neural network is studied,and the statistical parameters as mentioned above are used as input to train the BP network;finally,base on the relationship between statistical parameters and the precision of matching,the self-recognition and classification of matching are implemented with a good result.