基于Gabor小波和主元分析的纹理图像分割

Texture Image Segmentation Based on Gabor Wavelet and Principle Component Analysis

提出了基于Gabor小波和主元分析相结合的纹理图像分割算法。首先对纹理图像进行多通道滤波,获得了一系列滤波后的纹理图像。其次,借助于“能量测度”的概念,求解出各象素有效的纹理特征。为了进一步减少特征之间的信息冗余,降低聚类分析的计算负荷,采用主元分析(PCA)对所得的纹理特征进行降维。然后利用K M ean算法实现纹理图像的分类。最后针对所提算法,进行了仿真试验。

This paper puts forward a texture segmentation method which is based on the combination of Gabor Wavelet and Principal Components Analysis. First, a set of Gabor Wavelet filters is applied to the texture images to obtain textures features described with the “energy measure”. Second,in order to decrease the dimension of features greatly and reduce the complexity of computation,the Principal Components Analysis （ PCA ）method is used. Third, a K- Mean clustering method is used to classify features. Finally,simulations are performed on the segmentation algorithm which are presented in this paper.