An automatic bleeding detection scheme in wireless capsule endoscopy based on statistical features in hue space

Wireless capsule endoscopy (WCE) is a recently developed video technology to detect small intestine diseases, such as bleeding. For analyzing WCE video frames, instead of using the most common RGB (red, green, blue) color scheme, in this paper, HSV (hue, saturation, intensity value) color scheme is used, which corresponds better to human perception system. The HSV color scheme exhibits less sensitivity to illumination changes, which helps in handling the problem of illumination variation in WCE videos due to the weakening of battery. Different statistical features computed from H, S, and V spaces of WCE images are investigated and it is found that hue provides a useful feature as it captures intrinsic information about the color of objects or surfaces in a scene. Hence in this paper, an automatic bleeding detection scheme from WCE video is proposed utilizing the hue space. Among different statistical measures, mean, standard deviation, variance and moment exhibit significantly distinguishable characteristics for bleeding and non-bleeding images. For the purpose of classification, K-nearest neighbor (KNN) classifier is employed. From extensive experimentation on several WCE videos collected from a publicly available database, it is observed that the bleeding detection performance of the proposed method in terms of accuracy, sensitivity and specificity is quite satisfactory in comparison to that obtained by some of the existing methods.