基于分数阶Retinex的低照度图像增强方法

Low illumination image enhancement method based on fractional Retinex

低照度彩色图像增强在生活中起着重要作用,传统的低照度彩色图像增强算法往往会引起图像的不同程度失真。为了增强低照度彩色图像而又不引起图像失真,本文提出了一种新的低照度图像自适应对比度增强算法。将分数阶微积分、传统Retinex变分法与分段对数变换饱和度增强法相结合,构造一种新的分数阶Retinex图像增强算法。实验结果表明,该方法具有增强图像对比度的同时又能保持边缘和纹理细节的能力。与传统低照度图像增强算法相比,能突出图像的细节纹理信息,同时图像色度和亮度也有明显改善。

Low illumination color images enhancement plays a significant role in our daily lives,and traditional algorithms for enhancing low illumination color images often cause varying degrees of image distortion.To improve the contrast of color images in low illumination without causing image distortion,a novel adaptive contrast enhancement algorithm of low illumination images is proposed in this article.A novel fractional Retinex image enhancement algorithm is constructed by combining fractional calculus,Retinex variational method,and the piecewise logarithmic transform saturation enhancement method.Experimental results show that this method can not only improve the contrast of color image,but also preserve the edge and texture details.Compared with existing low illumination image enhancement algorithms,it can highlight the detailed texture information of color image,and has a significant improvement in image chromaticity and brightness compared with existing state of the art enhanced algorithms.