Key-frame Selection in WCE Video Based on Shot Detection

Wireless Capsule Endoscopy (WCE) is an imaging technology that enables close examination of the interior of the entire small intestine. A major problem associated with this new technology is that a large number of images need to be manually examined by clinicians. It is therefore useful to automatically reduce the number of frames that need direct interpretation by a clinician. In this paper a technique based on shot detection method is presented for automatic key-frame selection in WCE videos. The frames in the small intestine zone containing relevant features are extracted from the video sequence as the key-frames. Experimental results show that the proposed key-frame selection techniques significantly reduce the number of frames that need to be directly viewed by clinicians, and speed up the diagnosis procedures.