海军航空大学学报2025，Vol.40Issue(4)：600-606,640,8.DOI:10.7682/j.issn.2097-1427.2025.04.010
基于Box-Cox变换的海洋物种图像增强方法
Marine Species Image Enhancement Method Based on Box-Cox Transform
摘要
Abstract
Due to the absorption and scattering of underwater media,capturing images of marine species usually suffer from color distortion,low contrast and blurred details.To solve these problems,a marine species image enhancement method based on Box-Cox transform is proposed.Specifically,the zero-point symmetry criterion is first applied in the CIELAB color space to compress the L channel of the underwater distorted image,while pixel equalization is adjusted to the A and B channels to generate a reference image for the color transfer image.Then,the source image is generated by minimizing the loss function to compensate for the attenuated color channel of the marine species image.On this basis,a color transfer strategy is adopted to migrate the color features of the reference image to the source image to obtain a color-corrected image.Finally,the Box-Cox transform strategy further enhances the color-corrected image contrast.The pro-posed method is experimentally validated,and the experimental results show that the method is superior to the existing classical enhancement methods in improving image quality.关键词
图像增强/颜色转移/Box-Cox变换/直方图均衡化/对比度增强Key words
image enhancement/color transfer/Box-Cox transform/histogram equalization/contrast enhancement分类
航空航天引用本文复制引用
张卫东,王中,郝志斌,高兴运..基于Box-Cox变换的海洋物种图像增强方法[J].海军航空大学学报,2025,40(4):600-606,640,8.基金项目
中国博士后科学基金(2024M750747) （2024M750747）
河南省科技研发计划联合基金青年科学家项目(235200810066) （235200810066）
河南省重点研发与推广专项(242102210075) （242102210075）
河南省教师教育课程改革研究项目(2024-JSJYYB-099) （2024-JSJYYB-099）
国家级大学生创新训练计划项目(202410467001、202410467017) （202410467001、202410467017）
