纹理滤波器在分割肛管外病症图像中的应用

Application of image segmentation based on the texture filter in disease symptom outside the anal canal

采用均值滤波和中值滤波的方法对含有高斯噪声和椒盐噪声的肛管外外痔中的血栓性外痔图像进行滤波预处理,同时利用纹理滤波器分割图像将纹理图像、形态学操作、局部标准差滤波和Rangefilt滤波相结合,在保证分割图像整体性的同时,提高了分割图像的有效性,保留了图像的细节。通过血栓性外痔、息肉、混合痔、息肉血栓混合痔四种肛管外病症的分割图像结果的比较,表明纹理滤波器分割图像是一种性能良好的纹理图像分割方法,同时将其应用在肛管外病症图像的分割中,获得了很好的效果。

Firstly, we use the average value filter and the value filter＇s method to the Gauss noise and the spiced salt noise in the external hemorrhoid outside the anal canal anal, The images of the external hemorrhoid have been carried on the filter pre- treatment. What is more, combining the texture image, the morphology operation, the partial standard deviation filter ~th the range- filt filter, we use texture filter to achieve image segmentation. We guarantee the integrity of the image segmentation, enhaneing the division image effective and retaining the image detail at the same time. Experimental results show that by the comparison about segmentation image of thrombotic external hemorrhoids, polyps, mixed hemorrhoids and polyp of thrombosis combined mixed hemor- rhoids, the use of texture filter to split the image is a good texture segmentation methods, and application of image segmentation in disease symptom outside the anal canal got very good effects.