基于Gabor小波的无边缘活动围道纹理分割方法

该文提出了一种基于Gabor小波的活动围道纹理分割新方法。该方法先用Gabor小波提取图像的纹理特征，再用Chan-Vese模型进行分割。与其它基于Chan-Vese模型的纹理分割方法相比，基于Gabor小波的活动围道的纹理分割方法有两个优点:一是同时使用纹理特征和灰度信息演化围道，可分割纹理图像和非纹理图像，分割方法的灵活性好；二是在分割多类目标时，采用多级分层式曲线演化方法解决了初始围道难以选择的问题。对自然界真实图像和遥感图像的分割实验结果说明，该文提出的分割方法精度高。

Texture Image Segmentation Based on Gabor Wavelet Using Active Contours Without Edges

Texture is often divided into lots of isolate areas when a texture image is segmented. This paper develops a new method for texture image segmentation. It uses Gabor wavelet to extract texture features, and then Chan and Vese model is employed to segment a texture image. Compared to other texture segmentation methods using Chan and Vese models, the proposed method has two advantages. First, by combining the gray levels of pixels and texture information of an image, this method can be used for segmentation of a texture image or a none-texture image. Second, the hierarchical method proposed in this paper can avoid the problem due to the choice of initial conditions. The segmentation tests for remote sensing and natural texture images prove the proposed segmentation method is accurate and efficient.