Intestinal polyp recognition in capsule endoscopy images using color and shape features

Authors:
Baopu Li;Yichen Fan;Max Q.-H. Meng;Lin Qi
Affiliations:
Department of Electronic Engineering, The Chinese University of Hong Kong, Hong Kong, SAR, China;Department of Electronic Engineering, The Chinese University of Hong Kong, Hong Kong, SAR, China;Department of Electronic Engineering, The Chinese University of Hong Kong, Hong Kong, SAR, China;Department of Electronic Engineering, The Chinese University of Hong Kong, Hong Kong, SAR, China
Venue:
ROBIO'09 Proceedings of the 2009 international conference on Robotics and biomimetics
Year:
2009

Citing 5
Cited 1

Color indexing

International Journal of Computer Vision
Robust Histogram Construction from Color Invariants for Object Recognition

IEEE Transactions on Pattern Analysis and Machine Intelligence
Evaluation of MPEG-7 shape descriptors against other shape descriptors

Multimedia Systems
Texture analysis for ulcer detection in capsule endoscopy images

Image and Vision Computing
Computer-aided tumor detection in endoscopic video using color wavelet features

IEEE Transactions on Information Technology in Biomedicine

Small bowel image classification based on Fourier-Zernike moment features and canonical discriminant analysis

Pattern Recognition and Image Analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

Capsule endoscopy (CE) has gradually seen its wide application since it can directly view the small bowel in human body for the first time. However, a challenging problem is that too many images produced in each examination pose a tough workload to physicians. In this paper, we propose a new scheme aiming for intestinal polyp detection for CE images in order to partially solve this problem. This new scheme utilizes color and shape information, also powerful clues employed by physicians, to recognize polyp images. An illumination invariant color feature based on chromaticity histogram is first proposed. Integrating it with Zernike moments which are scale, translation and rotation invariant, we exploit the combined information as color and shape features to discriminate polyp from normal regions. Experiments on our present image data verify that it is promising to employ the proposed features to recognize polyp images.