Novel detection strategy for abnormalities in WCE video clips

Wireless Capsule Endoscopy (WCE) is a revolutionary technology that allows physicians to examine the patients whole gastrointestinal tract, especially the small intestine. However, reviewing capsule endoscopic video is a labor intensive task and very time consuming. In this paper we propose a novel method to detect key frames with abnormalities. It is based on the adaptive non-parametric corner detection approach using both the color and texture features. Real world patient videos including abnormal findings are adopted to evaluate the performance of the proposed method. The experimental results demonstrate that the proposed approach leads to the reduction of the number of frames in the WCE video without losing critical information.