Inflammatory Bowel Disease (IBD) is a pathologic condition that causes chronic inflammation. The most common type of IBD is Ulcerative Colitis (UC). This disease affects mucous membrane of the colon and the rectum and causes interconnected areas of inflammation and ulcer. So ulcer and abnormal red color of colon’s mucous are two important parameters which help experts to evaluate the severity of UC. Patients with UC have a significantly higher risk for the development of colorectal cancer. However, diagnostic accuracy for UC, is highly dependent on the experience and knowledge of the medical doctor, because there is considerable variety in the appearances of colonic mucosa within inflammations with UC and also there are many symptoms that common in IBD. This study consist of two steps. At the first we should separate colonoscopy images related to the healthy part of colon and the second is to evaluate the severity of UC according to ulcer and red color of colon’s mucous. Since visible vessel structure is the main sign of healty colon, in the first step a directional Hessian-based filter is applied to the image. This filter output expresses the percentage of belonging each pixel to the vessel structure. So by using Otsu’s binarization vessels and background are separated. in image of healthy people and the edge of ulcer in image of patient of UC so by using Otsu’s binarization vessel structure can be found. After post-processing which consists of morphological operator, we applied Uniform Local Binary Pattrtn method to image to estimate the uniformity; so we could separate UC images. Keywords Ulcerative colitis, Ulcer, Regions with abnormal red color Otsu’s binarization
