一种新型小波图像融合的水下目标增强算法  

A Novel Algorithm for Underwater Image Enhancement Based on Wavelet Image Fusion

作　　者：刘志强[1] 朱大奇[1] LIU Zhi-qiang;ZHU Da-qi(Shanghai Engineering Research Center for Intelligent Maritime Search and Rescue and Underwater Robots,Shanghai Maritime University,Shanghai 201306,China)

机构地区：[1]上海海事大学智能海事搜救与水下机器人上海工程技术研究中心,上海201306

出　　处：《控制工程》2022年第12期2235-2243,共9页Control Engineering of China

基　　金：国家自然科学基金资助项目(62033009);上海市科委创新行动计划项目(20510712300,21DZ2293500)。

摘　　要：针对水下图像存在色偏、清晰度低和边缘细节不足的问题,提出了一种新的水下图像增强算法。首先,以颜色校正的方式改进了一种基于暗通道先验(DCP)的水上图像增强方法,使其可以更好地用于水下图像,并将增强后的图像作为图像融合的一个输入。然后,利用非锐化掩蔽(USM)增强图像的纹理特征,并将再次增强的图像作为图像融合的另一个输入。最后,采用小波融合的方式对以上两个输入图像进行融合,进一步提升图像质量。与其他方法的比较结果表明,利用所提方法增强的图像具有更自然的外观、更高的清晰度、更多的细节和边缘信息,所提方法对水下图像有良好的增强效果。In order to solve the problems of color cast,low sharpness and insufficient edge detail in underwater images,a novel algorithm for underwater image enhancement is proposed in this paper.Firstly,a water image enhancement method based on dark channel zrior(DCP)is improved by color correction,so that it can be used to enhance underwater images.The enhanced image is used as an input for image fusion.Then,unsharp masking(USM)is used to enhance the texture features of the enhanced image,and the re-enhanced image is used as another input for image fusion.Finally,the two input images are fused by wavelet fusion to further improve the image quality.Comparison with other methods shows that the image enhanced by the proposed method has more natural appearance,higher sharpness,more details and edge information.The method presented in this paper has a good enhancement effect on underwater images.

关 键 词：水下图像增强 颜色恒常性 图像融合 暗通道先验 小波变换 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]