Automatic colon polyp detection in endoscopic capsule images

Abstract

This paper presents a method of automatic detection of colon polyps in endoscopic capsule images. This approach uses color and geometrical analysis of the images and uses histogram of gradients features and multilayer perceptron neural networks for performing the detection at multiple scales. The detection performance is tested using a database of 30540 images, 540 of them containing at least one polyp.