A color reproduction method with image enhancement for endoscopic images

Abstract

This paper presents an efficient color image enhancement method for endoscopic images. The enhancement is achieved in two parts: image enhancement at gray level followed by a color reproduction. The captured RGB endoscopic images are first converted into three 2-D spectral images using a well-known method called FICE (Fuji Intelligent Color Enhancement). Then the image with the maximum entropy is selected as the base image to be used for color reproduction in the next stage. In color reproduction, the chrominance map of a source color image is added to the base image. This chrominance map is found by matching luminance and texture information between these two images based on neighborhood statistic method. The distance between luminance components of base gray image and source color images is calculated using 2-norm Euclidean distance. The proposed method highlights some of the tissue characteristics in the base endoscopic image which will enable better diagnosis. The performance of the scheme is compared with other related algorithms in terms of simulation speed, image quality, efficiency of color reproduction and distortion.