Automatic detection and removal of bubble frames from wireless capsule endoscopy video sequences

Abstract

Wireless Capsule Endoscopy (WCE) is a new challenging clinical technique to obtain a detailed video of complete gastrointestinal tract (GIT). The major disadvantage of this technique is that it generates a huge number of frames which is difficult to diagnose for clinicians. In this particular paper, we propose a method to reduce the time of visualisation by removing frames which have redundant information of targeted disease. There are several types of redundant frames which can be categorised as informative and non-informative frames. The key method of the proposed strategy is to define dynamic threshold for intensity and hue in HSV or HSI colour space using the chosen ratio of training and testing data sets. Canny Edge Detector is used to find edges and Water segmentation method is used for segmentation. We acquire training and testing dataset from various WCE video segments to do this pilot research. An experimental result shows that the proposed method achieves very promising performance for detection.