Research on Color Image Defogging Algorithm based on MSR and CLAHE

Abstract

To study the defogging of color images, an improved multi-scale Retinex (MSR) algorithm combined with contrast limited adaptive histogram equalization(CLAHE) algorithm is proposed in this paper. Two different defogging methods are proposed for the image without color deviation and the image with color deviation. For the image without color deviation, the RGB color space of the input image is converted to the lab color space. Then the image is converted to RGB color space after MSR processing of L channel. Finally, the image contrast is improved by CLAHE algorithm. Realize the image defogging while keeping the original color of the image. For the image with color deviation, the input image is processed by MSR, the stretch degree of contrast is determined according to the mean and variance of the separated reflection components. Finally, the image contrast is further improved by CLAHE processing to realize the defogging of the restored scene. The experimental results show that the two algorithms in this paper have better effects in image defogging, image primary color maintenance, image scene color recovery and so on.