水面船舶图像去雾方法研究  

Research on Dehazing Method of Surface Ship Image

作　　者：郑晓 白前看 余坤 Zheng Xiao;Bai Qiankan;Yu Kun(Wuhan Institute of Marine Electric Propulsion,Wuhan 430064,China;Naval University of Engineering,National Key Laboratory of Science and Technology on Vessel Integrated Power System,Wuhan 430033,China)

机构地区：[1]武汉船用电力推进装置研究所,武汉430064 [2]海军工程大学舰船综合电力技术国防科技重点实验室,武汉430033

出　　处：《船电技术》2023年第8期49-53,共5页Marine Electric & Electronic Engineering

摘　　要：传统去雾方法不能有效适应水面船舶图像环境,存在去雾效果对比度较低、颜色偏暗等不足,直接影响水面船舶对环境自主感知能力。本文提出基于改进四叉树搜索策略和子窗口引导滤波的方式分别估计大气光值A和透射率参数,有效去除水面船舶图像雾气影响。定性和定量实验结果表明,相比较于对比方法,所提出方法能有效去除水面船舶图像中雾气影响,去雾效果更明显,纹理细节更丰富,主观感官上更自然,贴近无雾水面环境。实验结果表明所提方法的图像质量性能和计算效率指标均优于对比方法。The traditional dehazing methods can not effectively adapt to the image environment of the surface ship,there are some problems such as low contrast and darker color of the dehazing image,which directly affect the ability of the surface ship to perceive the environment autonomously.In this paper,an improved quadtree search strategy and a sub-window guided filtering method are proposed to estimate the atmospheric light value A and the transmittance parameters respectively,to effectively remove the influence of fog on the images of ships on the water surface.The experimental results of qualitative and quantitative evaluation show that,compared with the comparison methods,the proposed method can effectively remove the influence of fog in the images of surface ships,the dehazing effect is more obvious,the texture details are more abundant,the subjective sense is more natural,and it is close to the fog-free surface environment.The experiment results show that the image quality and computational efficiency of proposed method are better than the comparison methods.

关 键 词：水面船舶有雾图像 图像去雾 四叉树搜索 子窗口引导滤波 

分 类 号：U661.7[交通运输工程—船舶及航道工程]