基于稀疏约束的离焦图像多尺度去模糊方法  

Multi-scale Deblurring Method for Defocused Image Based on Sparse Constraint

作　　者：蔡小爱 张海民 CAI Xiaoai;ZHANG Haimin(School of Information and Finance,Xuancheng Vocational and Technical College,Xuancheng 242000;School of Computer and Software Engineering,Anhui Institute of Information Engineering,Wuhu 241000,China)

机构地区：[1]宣城职业技术学院信息与财经学院,安徽宣城242000 [2]安徽信息工程学院计算机与软件工程学院,安徽芜湖241000

出　　处：《常熟理工学院学报》2021年第2期65-69,共5页Journal of Changshu Institute of Technology

基　　金：安徽省教育厅2019年高校自然科学重点研究项目“面向智慧社区的视频监控异常行为识别方法的研究”(KJ2019A1295);安徽省教育厅2020年高校优秀青年人才支持计划项目(gxyq2020107)。

摘　　要：由于传统方法没有充分考虑加性噪声对图像处理效果的影响,导致去模糊处理后图像易丢失细节信息,峰值信噪比较小以及图像清晰度较差的问题.为此提出基于稀疏约束的离焦图像多尺度去模糊方法.通过建立观测模型,判断图像斑点位置,消除离焦图像中因噪声引起的序列斑点,降低加性噪声的影响程度.进而分离离焦图像中的前景聚焦和离焦背景区域,提取出离焦图像的深度图.最后通过稀疏约束获取模糊核,并运用该模糊核实现对离焦图像的去模糊处理.实验结果表明,所提方法在图像细节保留、清晰度提升和峰值信噪比方面均具有优越性,有效提升了图像的感官效果.Because the traditional methods do not fully consider the effect of additive noise on image processing,it is easy for the image to lose detailed information,the peak signal-to-noise ratio is small,and the image clarity is poor.Therefore,a multi-scale defocus image deblurring method based on the sparse constraint is proposed.By establishing the observation model,the position of image spots can be judged,the sequence speckle caused by noise can be eliminated in the defocused image,and the influence of additive noise can also be reduced.The depth map of the defocused image is extracted by separating the foreground focus area and the defocused background area in the defocused image.Finally,the fuzzy kernel is obtained by the sparse constraint,and the defocused image is deblurred by using the fuzzy kernel.The experimental results show that the proposed method has advantages in image detail retention,clarity improvement and PSNR,which effectively improves the sensory effect of the image,and fully verifies the application value of the method.

关 键 词：稀疏约束 离焦图像 去模糊 图像去噪 观测模型 加性噪声 模糊核 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]