摘要
针对户外成像设备获取的沙尘图像存在色调偏黄、色彩不丰富及清晰度较低等问题,提出一种基于改进暗通道先验的沙尘图像清晰化算法.对图像偏色的问题,首先对高斯模型进行改进,采取自适应归一化方法调整图像暗部像素,并加权融合基于灰度世界的颜色校正方法,从而去除色偏影响;然后利用带色彩恢复的多尺度视网膜增强算法进行色彩恢复;对处理后存在的薄雾影响,基于暗通道先验方法重新选取大气光值,并对暗部像素进行亮度补偿;最后针对图像饱和度不足和对比度低的问题,将图像映射到HSI空间,分别利用自适应调整函数和改进双伽马校正算法进行提升.实验结果表明,该方法不仅能有效校正色偏,更好提升图像对比度及清晰度,且对图像色彩丰富度提升效果明显,可提高户外成像设备的图像质量.
Aiming at the problems of yellowish tones,lack of color richness and low clarity of sand dust images acquired by outdoor imaging devices,we proposesd a sand dust image clarity algorithm based on improved dark channel prior.For the problem of image color bias,firstly,we improved the Gaussian model by adopting the adaptive normalization method to adjust dark pixels of the image,and weighted fusion of a color correction method based on the gray world to remove the color bias effect.Secondly,the multi-scale retinal enhancement algorithm with color restoration was used for color restoration,for the fog effect existing after processing,the atmospheric light value was re-selected and the dark pixels were compensated for brightness using the dark channel-based prior method.Finally,for the problems of insufficient image saturation and low contrast,the images were mapped to HSI space and enhanced using the adaptive adjustment function and improved dual Gamma correction algorithm,respectively.The experimental results show that the method can not only effectively correct the color bias and better improve the image contrast and clarity,but also has a significant effect on the image color richness enhancement,which can improve the image quality of outdoor imaging equipment.
作者
牛宏侠
王春智
梁乐观
张瑞瑞
朱梦瑞
NIU Hongxia;WANG Chunzhi;LIANG Leguan;ZHANG Ruirui;ZHU Mengrui(School of Automation and Electrical Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China;Key Laboratory of Plateau Traffic Information Engineering and Control of Gansu Province,Lanzhou 730070,China;Key Laboratory of Opt-Electronic Technology and Intelligent Control of Ministry of Education,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处
《吉林大学学报（理学版）》
CAS
北大核心
2023年第6期1407-1418,共12页
Journal of Jilin University:Science Edition
基金
国家自然科学基金(批准号:61863024,62141303)
甘肃省自然科学基金(批准号:22JR5RA358,21JR7RA321)
甘肃省重点研发计划——工业类项目(批准号:23YFGA0049)。
关键词
沙尘图像
暗通道先验
偏色校正
色彩恢复算法
HSI颜色空间
sand dust image
dark channel prior
color bias correction
color restoration algorithm
HSI color space
