S变换在水下目标识别中的应用

摘要

研究了基于S变换模时频矩阵相似度的水下目标识别方法.根据测试样本对应的S变换模时频矩阵与标准样本对应的S变换模时频矩阵之间的相似度最大原则对测试样本进行识别.该方法不需要辅助分类器而直接实现目标识别,计算简单快速.仿真实验表明,该方法的识别率较高,且受噪声影响小,适合于水下目标识别.%Underwater target recognition based on S-transform with module time-frequency matrixes similarity was studied in the paper. Compute the similarity rate between S-transform with module time-frequency matrixes of test sample and which of normal sample. Test sample was recognized according to maximum similarity principle. The method can achieve target recognition directly without using auxiliary classifier. The computing process was simple and fast. Simulation and test show the method was suited to underwater target recognition because of its high recognition rate and insensitive to noise