Image Processing Using Color and Object Information for Wireless Capsule Endoscopy

Wireless capsule endoscopy provides real-time images in the digestive tract. Capsule images are usually low resolution and are diverse images due to travel through various regions of human body. Color information has been a primary reference in predicting abnormalities such as bleeding. Often color is not sufficient for this purpose. In this study, we took morphological shapes into account as additional, but important criterion. First, we processed gastric images in order to indentify various objects in the image. Then, we analyzed color information in the object. In this way, we could remove unnecessary information and increase the accuracy. Compared to our previous investigations, we could handle images of various degrees of brightness and improve our diagnostic algorithm.
{"references":["","A. Mata, J. Llach, and JM. Bordas, \"Wireless Capsule Endoscopy,\"\nWorld J Gastroenterol, April 7, 2008, Vol. 14, no.13, pp. 1969 -1971.","A. Glukhovsky, \"Wireless capsule endoscopy\", Sensor Review, vol. 23,\nno. 2, 2003, pp. 128–133.","D. G. Adler and C. J. Gostout, \"Wireless Capsule Endoscopy\", Hospital\nPhysician, 2003, pp.14-22.","Y.-G. Lee, and G. Yoon, \"Bleeding Detection Algorithm for Capsule\nEndoscopy,\" World academy of science engineering and technology, vol.\n81, pp. 672-677, Sem., 2011.","Y.-G. Lee, and G. Yoon, \"Real-Time Image Analysis of Capsule\nEndoscopy for Bleeding Discrimination in Embedded System Platform,\"\nWorld academy of science engineering and technology, vol. 59, pp.\n2526-2530, Dec., 2011.","Y.-G. Lee, and G. Yoon, \"Improvement of Blood Detection Accuracy\nusing Image Processing Techniques suitable for Capsule Endoscopy,\"\nWorld academy of science engineering and technology, vol. 65, pp.\n1096-1099, May, 2012.","Tony Lindeberg, \"Edge Detection and Ridge Detection with Automatic\nScale Selection,\" International J of Computer Vision, Nov, 1998, vol. 30,\nno. 2, pp. 117-156.","Heath, M. D., Sarkar, S., Sanocki, T., Bowyer, K. W., \"A robust visual\nmethod for assessing the relative performance of edge-detection\nalgorithms,\" IEEE Trans on Pattern Analysis and Machine Intelligence,\nvol. 19, no. 12, pp. 1338-1359, Dec, 1997.","F Catté, PL Lions, JM Morel, T Coll, \"Image Selective Smoothing and\nEdge Detection by Nonlinear Diffusion,\" SIAM Journal on Numerical\nAnalysis, vol. 29, no.1, pp. 182-193, Feb, 1992\n[10] JH Elder, SW Zucker, \"Local scale control for edge detection and blur\nestimation,\" IEEE Trans on Pattern Analysis and Machine Intelligence,\nvol. 20, no. 7, pp. 699-715, Jul, 1998.\n[11] J. F. Canny, \"A Computational Approach to Edge Detection,\" IEEE\nTrans. Pattern Analysis and Machine intelligence, vol. PAMI-8, no. 6,\npp. 679–698, Nov. 1986."]}