基于颜色校正与改进的CLAHE多尺度融合水下图像增强  

Underwater Image Enhancement Based on Color Correction and Improved CLAHE Multi-scale Fusion

作　　者：弭永发 迟明善 张强[1] 刘鹏杰 王天佑 MI Yongfa;CHI Mingshan;ZHANG Qiang;LIU Pengjie;WANG Tianyou(School of Navigation and Shipping,Shandong Jiaotong University,Weihai 264200,China)

机构地区：[1]山东交通学院航运学院,山东威海264200

出　　处：《无线电工程》2024年第6期1470-1480,共11页Radio Engineering

基　　金：山东省自然科学基金面上项目(ZR2022ME087);山东交通学院2023年校级研究生科技创新项目立项(2023YK097);山东交通学院博士科研启动基金(BS2024009)。

摘　　要：由于水体对光线的吸收和散射作用,水下图像易出现色偏、对比度低等问题。针对上述问题,提出一种基于颜色校正与改进的限制对比度自适应直方图均衡(Contrast Limited Adaptive Histogram Equalization,CLAHE)多尺度融合水下图像增强方法。提出自动对比度调整的改进版白平衡方法对图像颜色进行校正;采用结合自适应伽马校正的CLAHE提高图像的对比度、亮度、轮廓与细节;采用高斯与拉普拉斯金字塔提取图像不同特征图进行多尺度融合,从而实现水下图像的增强处理。此外,还对提出的算法与其他算法进行主客观评价与对比分析。对比实验结果表明,采用所提方法增强处理后的水下图像质量明显提升,水下图像颜色质量评价(Underwater Color Image Quality Evaluation,UCIQE)、信息熵、水下图像质量评价(Underwater Image Quality Measure,UIQM)的均值分别为0.6390、7.7149、0.7183,均证明了所提方法的有效性。另外,消融实验结果也证明所提算法的合理性,能够有效解决水下图像色偏问题,均衡提高水下图像的对比度与亮度,增强图像的边缘和细节特征。Due to the absorption and scattering of light by water bodies,the underwater images are prone to color cast and low contrast.In order to address the above problems,a multi-scale fusion underwater image enhancement method based on color correction and improved Contrast Limited Adaptive Histogram Equalization(CLAHE)is proposed.Firstly,an improved version of white balance method with automatic contrast adjustment is proposed to correct the color of the image.Secondly,the CLAHE with adaptive gamma correction is used to improve the contrast,brightness,contour features and details of the image.Finally,a multi-scale fusion of different feature maps extracted using Gaussian and Laplacian pyramids is performed to achieve enhancement of underwater images.Additionally,the subjective and objective evaluations and comparative analysis of the proposed algorithm with other algorithms are conducted.The comparative experimental results show that the quality of underwater images enhanced by the proposed method is significantly improved.The average values of Underwater Color Image Quality Evaluation(UCIQE),information entropy and Underwater Image Quality Measure(UIQM)evaluation indicators are 0.6390,7.7149,and 0.7183,respectively,which all prove the effectiveness of the proposed method.In addition,the ablation experimental results also prove the rationality of the proposed algorithm.The proposed algorithm can effectively solve the problem of color cast in underwater images,improve the contrast and brightness of underwater images in a balanced manner,and enhance the edge and detail features of the image.

关 键 词：水下图像增强 多尺度融合 白平衡 自适应伽马校正 限值对比度自适应直方图均衡 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]