基于Gabor小波变换的无监督纹理图像分割

本文提出了一种基于Ｇａｂｏｒ小波变换的无监督纹理分割方法，该方法首先对纹理图像进行Ｇａｂｏｒ小波变换，再经过一组Ｇａｕｓｓ波波器滤波，然后将得到的纹理特征送入由Ｈｏｐｆｉｅｌｄ网络构成的无监督分类器，最后得到纹理分割结果。实验和分析表明，该方法对于不同的纹理图像都具有良好的分割效果。

A Gabor Wavelet Transformation-Based Unsupervised Texture Images Segmentation

In this paper,we introduce a method called unsupervised texture segmentation based on Gabor wavelet transformation.First,we get the Gabor wavelet transformations of the texture images by a set of Gabor filters,then using a set of relative Gauss filters we can get the texture features of the images.After classifying the above texture features by an unsupervised classifier composed by a Hopfield neural network,we can get the segmentation of the texture images at last.After some experiments and analysis,it can be seen that our method can be used to segment different texture images and the results are quite better.