基于Retinex和HSV颜色空间的低照度图像增强算法

Low-Illumination Image Enhancement Algorithm Based on Retinex and HSV Color Space

针对低照度图像存在的对比度低、颜色失真、细节信息损失、噪声干扰等问题,提出一种基于Retinex的改进算法。首先,将原始图像转换到HSV颜色空间,采用改进算法和Gamma校正对V分量进行亮度调节;其次,基于对比度拉伸算法调整S分量,增强图像饱和度和对比度;最后,将融合后的图像转换回RGB颜色空间,输出图像。在MATLAB平台上选取不同场景低照度图像进行增强处理,采用图像信息熵(IE)、峰值信噪比(PSNR)和结构相似性指数(SSIM)三个客观指标进行均值对比,实验结果表明,新算法计算结果均优于传统的直方图均衡化算法(HE)和多尺度Retinex算法(MSR),有效保持图像边缘信息,提升图像质量。

An improved algorithm based on Retinex is proposed to solve the problems of low contrast,color distortion,loss of detail information and noise interference in low-illumination images.First,the original image was converted to HSV color space,and the V component was adjusted with improved algorithm and Gamma correction.Second,the s component was adjusted based on the contrast stretching algorithm,enhance image saturation and contrast.Finally,the fused image is converted back to RGB color space to output the image.The experimental results show that the new algorithm is better than the traditional Histogram equalization algorithm(HE)and the multi-scale Retinex algorithm(MSR),in the subjective evaluation and objective indicators have greater advantages,effectively maintain image edge information,improve image quality.