利用各目标距离像之间的相关系数 ,可提高基于距离像建库的目标识别方法的性能。该文提出了利用相关系数构造特征矢量和用神经网络进行分类的雷达目标识别方法。研究了特征矢量的分类性能、最终判决量的确定、相关系数特征矢量法与最大相关系数法各自的特点等问题。对 6类飞机目标全方位角范围的识别结果表明 :最大相关系数法在高信噪比时的识别率较高 ,相关系数特征矢量法的抗噪性较强 ,两种识别方法有较强的互补性。若按某种方式将两者结合起来 ,将得到识别率和抗噪性俱佳的识别方法 By using the correlation coefficient between each target distance image, the performance of target recognition method based on distance image database can be improved. In this paper, a radar target recognition method using feature vectors and classification by neural networks is proposed. The classification performance of eigenvectors, the determination of the final decision, the respective characteristics of the correlation coefficient eigenvector method and the maximum correlation coefficient method are studied. The recognition results of the target range of 6 types of aircraft indicate that the maximum correlation coefficient method has a high recognition rate at high signal-to-noise ratio, and the correlation coefficient eigenvector method has strong anti-noise ability. The two recognition methods are strong Complementarity. If you combine the two in some way, you get a good recognition of noise and noise