Guide image based enhancement method for wireless capsule endoscopy

Good image quality of the wireless capsule endoscopy is key for doctors to diagnose gastrointestinal (GI) tract diseases. However, the limited illumination and complex environment in the GI tract usually result in low quality endoscopic images. Existing image enhancement methods only use the information of the image itself or multiple images of the same scene to accomplish the enhancement. In this paper, we propose a brand-new image enhancement method called guide image based enhancement (GIE). GIE enhances low quality images by using the information of a good quality image of the similar scene. Firstly, the low quality image is preprocessed by the proposed logarithm based contrast limited pointwise histogram equalization (LCLPHE) algorithm. After that it is corrected by a selected good quality image. Experimental results show that GIE improves the average intensity of endoscopic images by 24.67%, the entropy by 16.37% and the average local entropy (MLE) by 56.68%, which outperforms the state-of-art methods.