Wireless Capsule endoscopy (WCE) allows a physician to examine the entire small intestine. Capsule endoscopy helps physicians to view small intestine's internal structure- an area that isn't easily reached with conventional endoscopy in a more distinct way. The most crucial problem with this technology is that too many images produced by WCE causes a tough task to physicians, so it is very significant if we can help physicians do diagnosis using computerized methods. In this paper, a new method aimed for small bowel tumor detection of WCE images is proposed. This new approach mainly focuses on texture feature, a powerful clue used by physicians, to detect tumor images with Supervisedclassification. Analysis is also done using Principle Component Analysis (PCA) based Local Binary Pattern (LBP) and Unsupervised classification to discriminate tumor regions from normal regions. Simulation results show an efficiency of around 99% using Supervised Classification.
展开▼
