A sar image recognition method

Abstract

The invention belongs to the technical field of aperture radar (Synthetic Aperture Radar, SAR) automatic target recognition, in particular to SAR image target recognition using a two-dimensional linear discriminant analysis method for feature extraction. A SAR image recognition method, given two-dimensional SAR images of class F, calculating the inter-class distance between all class F images, and then calculating a weight coefficient matrix according to the calculated inter-class distance construction, using this weight coefficient matrix Calculate a projection subspace, and judge whether the calculated projection subspace converges. If not, repeat the previous operation to recalculate a new weight coefficient matrix to obtain a new feature space until the feature space converges. The present invention can greatly improve the convergence speed of the feature matrix, thereby ensuring the recognition rate of images with higher misrecognition rates is improved, and at the same time ensuring the recognition rates of other images to the greatest extent.

Description

technical field [0001] The invention belongs to the technical field of aperture radar (Synthetic Aperture Radar, SAR) automatic target recognition, in particular to SAR image target recognition using a two-dimensional linear discriminant analysis method for feature extraction. Background technique [0002] The principle of SAR image target recognition is to establish a feature library based on the known training sample target category information, perform feature extraction on the test sample, and select the type of training sample corresponding to the highest similarity in the library as the classification result of the test sample. [0003] The rapid development of SAR technology has greatly improved the resolution of the resulting image, and the target information in the SAR image has also shown explosive growth, which has brought about a substantial increase in the amount of corresponding data. Facing the huge amount of data, Key technologies in object detection and reco...

AI Technical Summary

Problems solved by technology

However, when using 2DLDA for classification and recognition, there will be a large misrecognition rate due to the small inter-class distance of the image in the projection space.

In the literature (Chong Lu, SenJian An, Wanquan Liu, Xiaodong Liu, "An innovative Weighted 2DLDA Approach for Face Recognition", JSign Process Syst (2011) 65:81-87), the author Chong Lu et al. proposed a weighted 2DLDA Algorithm, this algorithm calculates the inter-class distance between various types, and redefines the inter-class discrete matrix according to these calculation results, this algorithm can make full use of the distance relationship between various types, and iteratively find the optimal projection matrix, but This algorithm is not very effective in SAR image recognition, and the matrix convergence speed is too slow. Although it can improve the recognition rate of images with low recognition rates, it also greatly reduces the recognition rate of images with high recognition rates.

Images