According to the features of sonar image, a methodology of side scan sonar image classification using support vector machine and gray level co-occurrence matrix is presented.The texture features extracted with gray level co-occurrence matrix and principal component analysis method, and then we give out the most suitable the texture features for side scan sonar image.Sea classification is carried out with support vector machine using the backscatter combined with sonar image.It is concluded that the result of classifica-tion based on the backscatter combined with the texture features is better than that based on solely the backscatter.%根据侧扫声纳影像的特征,提出一种基于SVM和GLCM的侧扫声纳影像分类方法,利用灰度共生矩阵提取其纹理特征,采用主成分分析法对纹理特征进行筛选,选择适合侧扫声纳影像的最佳纹理特征,结合侧扫声纳影像的回波强度,应用支持向量机对侧扫声纳影像进行分类。研究结果表明,纹理特征结合回波强度的支持向量机分类精度高于只依靠回波强度的支持向量机分类精度。
展开▼