The Research On Equipment Image Recognition Technology Based

With the wide application of information technology in reconnaissance intelligence equipment,the application of image intelligence has become more and more extensive,and image processing technology has become an important part of comprehensive processing of battlefield intelligence.At present,the artificial mode is mainly used for image interpretation,which requires a lot of manpower and time to comp lete,and is not suitable for fast battlefield information against rhythm.Therefore,the research on equipment image intelligent recognition technology and system development based has great application value in the military.This paper first introduces the basic concepts of intelligent image recognition,as well as image recognition techniques such as clustering,support vector machine and neural network related to the research of this paper.Then,the weapon image clustering,aircraft image recognition and vehicle equipment image recognition are respectively carried out.Research and system function implementation,the main contents are as follows:The AP clustering algorithm and K-means clustering algorithm are used to cluster the weapon equipment images.Using Matlab as the platform,a cluster-based weapon image annotation system is designed and implemented to solve the problem of laborious and inefficient low-volume data marking.The intelligent recognition system of aircraft based on support vector machine is studied.The acquisition of image information,image feature extraction and selection,and intelligent recognition of classifier training and decision process are realized.The system is implemented based on Matlab.The military vehicle equipment picture recognition system based on neural network is studied,which realizes the functions of candidate frame search,CNN feature extraction,neural network construction and training.The military vehicle equipment image data with a small number of annota tions was constructed,and a CNN model for military vehicle equipment image recognition with considerable precision and ready to use was trained.