基于MSRCR算法的夜间道路图像增强算法

Night road image enhancement algorithm based on MSRCR algorithm

文章提出了一种自适应伽马校正融合MSRCR算法的低照度图像增强算法,旨在解决现有MSRCR算法在处理低照度图像时存在的边缘信息模糊和白化现象等问题。该算法首先根据Retinex模型分别得到V通道图像中的反射图像和光照图像。在光照图像中,对小、中、大尺度图像分别采用MSRCR算法进行增强,然后根据像素的概率分布函数调节伽马校正的伽马值,以自适应校正图像亮度。在反射图像中,采用加权引导滤波进行去噪,有效保留了图像的细节信息和边缘信息。最后,反射图像和光照图像融合形成增强后的V通道图像,HSV色彩空间图融合形成增强后的低照度图像,并通过MSRCR算法的色彩恢复因子恢复丢失的色彩信息。实验结果表明,所提算法具有明显的优越性,相较于其他对比算法能更好地提高低照度图像的质量和视觉效果。

In this paper,a low illumination image enhancement algorithm based on adaptive gamma correction and MSRCR algo-rithm is proposed to solve the problems of edge information blurring and whitening in low illumination image processing by MSRCR algorithm.In this algorithm,the reflection image and light image in the V-channel image are obtained by Retinex model.In the light image,the MSRCR algorithm is used to enhance the small,medium and large scale images respectively,and then the gamma value of the gamma correction is adjusted according to the probability distribution function of the pixel to adjust the image brightness adaptively.In the reflected image,the weighted guided filter is used to denoise,which effectively preserves the details and edge information of the image.Finally,the reflection image and light image are fused to form the enhanced V-channel image,and the HSV color space image is fused to form the enhanced low-illumination image,and the lost color information is recovered by the color recovery factor of the MSRCR algorithm.The experimental results show that the proposed algorithm has obvious ad-vantages and can improve the quality and visual effect of low illumination image better than other comparison algorithms.