YCbCr空间分治的双分支低照度图像增强网络  

Dual-branch low-light image enhancement network via YCbCr space divide-and-conquer

作　　者：闫晓阳 王华珂 侯兴松[1] 顿玉洁[1] Yan Xiaoyang;Wang Huake;Hou Xingsong;Dun Yujie(School of Information and Communications Engineering,Xi’an Jiaotong University,Xi’an 710049,China)

机构地区：[1]西安交通大学信息与通信工程学院,西安710049

出　　处：《中国图象图形学报》2023年第11期3415-3427,共13页Journal of Image and Graphics

基　　金：国家自然科学基金项目(62272376,61872286);陕西省重点研发项目(202DLGY04-05,S2021-YF-YBSF-0094)。

摘　　要：目的现有的低照度图像增强算法通常在RGB颜色空间采用先增强后去噪的方式提升对比度并抑制噪声,由于亮度失真和噪声在RGB颜色空间存在复杂的耦合关系,往往导致增强结果不理想。先增强后去噪的方式也放大了原本隐藏在黑暗中的噪声,使去噪变得困难。为有效处理亮度失真并抑制噪声,提出了一个基于YCbCr颜色空间的双分支低照度图像增强网络,以获得正常亮度和具有低噪声水平的增强图像。方法由于YCbCr颜色空间可以分离亮度信息与色度信息,实现亮度失真和噪声的解耦,首先将低照度图像由RGB颜色空间变换至YCbCr颜色空间,然后设计一个双分支增强网络,该网络包含亮度增强模块和噪声去除模块,分别对亮度信息和色度信息进行对比度增强和噪声去除,最后使用亮度监督模块和色度监督模块强化亮度增强模块和噪声去除模块的功能,确保有效地提升对比度和去除噪声。结果在多个公开可用的低照度图像增强数据集上测试本文方法的有效性,对比经典的低照度图像增强算法,本文方法生成的增强图像细节更加丰富、颜色更加真实,并且含有更少噪声,在LOL(low-light dataset)数据集上,相比经典的KinD++(kindling the darkness),峰值信噪比(peak signal-to-noise ratio,PSNR)提高了3.09 dB,相比URetinex(Retinex-based deep unfolding network),PSNR提高了2.74 dB。结论本文提出的空间解耦方法能够有效地分离亮度失真与噪声,设计的双分支网络分别用于增强亮度和去除噪声,能够有效地解决低照度图像中亮度与噪声的复杂耦合问题,获取低噪声水平的亮度增强图像。Objective The images acquired at night or backlight conditions always have poor visibility and have details that are hidden in the dark.Moreover,due to insufficient lighting and limited exposure time,the number of incident photons on these images decreases,thereby resulting in a large amount of non-negligible noise.Therefore,improving the contrast and removing noise from low-light images present a challenge.The existing low-light image enhancement algorithms usually enhance the contrast and suppress the noise in the RGB color space by way of enhancing and then denoising.However,due to the complex coupling relationship between brightness distortion and noise in the RGB space,enhancing-and-then denoising methods usually amplify the noise that is originally hidden in the dark,thus increasing the difficulty of the denois⁃ing task,affecting the aesthetic quality of images,and constraining subsequent image processing tasks on the aspects of image classification,object detection,and recognition.To effectively deal with brightness distortion and noise,a dualbranch low-light image enhancement network based on YCbCr space is proposed in this paper to yield enhanced images with minimal color distortion and less noise.Method The YCbCr space can separate luminance information from chromi⁃nance information.Experiments show that brightness distortion is mostly observed in luminance information and that chro⁃minance information is heavily polluted with noise.This paper designs a corresponding network structure based on the divide-and-conquer method to deal with different degradation modes,where different modules can effectively deal with a specific distortion and reduce the difficulty of network learning.A novel dual-branch network in YCbCr space is then devel⁃oped to realize the decoupling of luminance distortion and noise,which can enhance the luminance information and denoise the chrominance information.First,the low-light images are transformed from the RGB space to the YCbCr space.Sec⁃ond,the images in the YCbCr s

关 键 词：低照度增强 YCBCR颜色空间 双分支网络 噪声去除 分治策略 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]