Multi-color Space Medical Endoscope Image Enhancement Method
Hongrui Liu, Chao Xu, Bo Feng, Kai Li · 2021 IEEE 6th International Conference on Signal and Image Processing (ICSIP) · 2021
Aiming at the problems of low contrast between the vascular part and the tissue part of the medical endoscope, and the vessels and lesions are not obvious enough, this paper proposes an improved method of medical endoscopic image enhancement based on the characteristics of medical images. First, we convert the image from RGB space to HSV space, split the V channel into a low frequency part and a high frequency part, use gamma correction for the low frequency part to adjust the brightness, denoise the high frequency part, and then do saturation correction on the S channel. Then, the image is converted from the HSV space to the YCrCb space, and the Y channel is weighted CLAHE to enhance the contrast. Next, we convert the image from YCrCb space to RGB color space. Finally, the Sobel operator edge detection is performed on the image, and the detected edge is enhanced with a mask to highlight the details. We use multiple sets of pictures to evaluate images from a subjective and objective perspective. The experimental results show that the algorithm proposed in this paper not only enhances the blood vessel, but also retains the authenticity of the image, and the effect is satisfactory.