Color Endoscopic Image Enhancement Technology Based on Nonlinear Unsharp Mask and CLAHE
Shanshan Fang, Chao Xu, Bo Feng, Yuzhe Zhu · 2021 IEEE 6th International Conference on Signal and Image Processing (ICSIP) · 2021
Since the existing endoscopic image enhancement algorithms cannot take into account the problems of detail enhancement, brightness preservation and color preservation, this paper proposes an endoscopic image enhancement based on an improved unsharp mask and CLAHE algorithm. First, perform local brightness adjustment and saturation correction on the input image, and secondly, it is also the key point to perform contrast enhancement, which is divided into low-frequency background enhancement and high-frequency detail enhancement. CLAHE is used to balance the background intensity of low frequency components, and then the high-frequency parts are nonlinearly sharpened. The improved nonlinear transfer function proposed is used, and the high-frequency texture part is emphasized in combination with canny edge detection. Finally, the objective and subjective quality assessment of the image is carried out. The results show that the proposed nonlinear method enriches the texture details such as blood vessels. The peak signal-to-noise ratio and Tenengrad gradient are better than those of the compared algorithm. And the algorithms in this paper are all enhanced in the brightness channel, which has the function of color preservation, so as to achieve the natural enhancement effect as much as possible.