研究纹理图像的分类问题,纹理特证提取和分类器设计是决定分类正确率高低的关键.由于库存图像较多,且质量受到噪声影响,使图像特征提取比较困难.针对传统特征提取和分类算法分类正确率不高的难题,提出一种基于支持向量机的纹理图像分类算法.首先采用Gabor滤波器对纹理特征进行提取,采用主成分分析对提取后的特征进行选择,最后采用支持向量机进行纹理图像的分类.采用Brodatz纹理库进行测试实验,实验结果表明,支持向量机分类算法提高了纹理图像的分类正确率,降低了误分率和拒分率,且分类速度加快,适用于更为复杂的纹理分类,为图像提取提供了参考.%Study texture images classification. This paper put forward a texture images classification algorithm based on support vector machine. Firstly, texture features were extracted by Gahor filter, and then the principal com-ponent analysis was used to select the extracted features. Finally, support vector machine was used to build the tex-ture image classifier. The model performances were tested with Brodatz texture data. The experimental results show that the proposed algorithm improves the texture image classification accuracy, reduces the classification error rate and rejection rate, and increases classification speed. It can be used in complex texture classifications.
展开▼
