Polyps Auto-Detection in Wireless Capsule Endoscopy Images Using Improved Method Based on Image Segmentation

Wireless Capsule Endoscopy (WCE) is a noninvasive instrument that widely used in screening the whole intestine and it has been utilized as a model especially for the examination of gastrointestinal (GI) diseases. However, it is numerous images of the detecting result produced by WCE that always burdens the physicians. To solve this problem, it is necessary to combine the manual diagnosis with the image segmentation technology. In this paper we proposed a feasible method by using K-means clustering and localizing region-based active contour segmentation for polyps auto-detection in WCE images. Experimental results shows the method is promising and efficient.