电子测量与仪器学报

周巍巍,高银,吴仪芳,李俊.局部高斯均差变分保边图像平滑算法[J].电子测量与仪器学报,2024,38(4):94-107

局部高斯均差变分保边图像平滑算法

Edge-preserving image smoothing with local Gaussian mean-difference variation

DOI：

英文关键词:image smoothing local Gaussian mean-difference variation detail-preserving isolated noise

基金项目:国家自然科学基金(62001452)、中国福建光电信息科学与技术创新实验室（闽都创新实验室）(2021ZZ116)、福州市科技计划项目(2022-ZD-001)资助

作者	单位
周巍巍	1.兰州交通大学数理学院兰州730070；2.中国科学院海西研究院泉州装备制造研究中心泉州362216
高银	2.中国科学院海西研究院泉州装备制造研究中心泉州362216； 3.中国科学院福建物质结构研究所福州350000
吴仪芳	中国科学院海西研究院泉州装备制造研究中心泉州362216
李俊	1.兰州交通大学数理学院兰州730070；2.中国科学院海西研究院泉州装备制造研究中心泉州362216； 3.中国科学院福建物质结构研究所福州350000

Author	Institution
Zhou Weiwei	1.School of Mathematics and Physics, Lanzhou Jiaotong University, Lanzhou 730070, China; 2.Quanzhou Institute of Equipment Manufacturing, Haixi Research Institute, CAS, Quanzhou 362216, China
Gao Yin	2.Quanzhou Institute of Equipment Manufacturing, Haixi Research Institute, CAS, Quanzhou 362216, China; 3.Fujian Institute of Research on the Structure of Matter, CAS, Fuzhou 350000, China
Wu Yifang	Quanzhou Institute of Equipment Manufacturing, Haixi Research Institute, CAS, Quanzhou 362216, China
Li Jun	1.School of Mathematics and Physics, Lanzhou Jiaotong University, Lanzhou 730070, China; 2.Quanzhou Institute of Equipment Manufacturing, Haixi Research Institute, CAS, Quanzhou 362216, China; 3.Fujian Institute of Research on the Structure of Matter, CAS, Fuzhou 350000, China

摘要点击次数: 303

全文下载次数: 13016

中文摘要:

针对图像平滑过程中无法保留细节的问题，提出了基于局部高斯均差变分的保边图像平滑算法。首先，通过统计学分析建立局部高斯均差变分算子。其可以衡量局部梯度与高斯滤波处理后的梯度差异，区分结构和纹理。其次，构建局部高斯均差变分平滑模型，由稀疏求解得到初始平滑图像。最后，针对复杂纹理图像存在纹理残留的问题，提出孤立噪声去除模型。模型通过自适应窗口设定像素值，在不影响结构的前提下去除初始平滑图像中的纹理残留。通过主观、客观实验，与经典的算法对比，证明该算法有更高质量的平滑结果。评价指标整体提升了0.7%。通过压缩伪影去除、HDR色调映射、图像去雾和拉普拉斯金字塔加速的扩展实验，验证该算法在不同视觉任务上的适用性和效率可提升性。

英文摘要:

An edge-preserving smoothing algorithm based on local Gaussian mean-difference variation is proposed to address the issue of detail not being preserved during the process of image smoothing. Firstly, a local Gaussian mean-difference variational operator is established by statistical analysis. To differentiate between structure and texture, the operator is employed to quantify the difference between the local gradient and the gradient after Gaussian filtering. Secondly, a local Gaussian mean-difference variational smoothing model is developed, and a sparse solution is used to produce the initial smooth image. Finally, an isolated noise removal model is suggested to address the issue of texture residue in images with complex texture. The model adjusts pixel values using an adaptive window and eliminates texture residue from the initial smooth image without changing the structure. It has been demonstrated through subjective and objective experiments that this algorithm produces smoothing results of superior quality than traditional algorithms. Evaluation indicators improved by 0.7% overall. Extended experiments verify the algorithm's applicability and efficiency enhancement potential across various visual tasks, including compression artifact removal, HDR tone mapping, image dehazing, and accelerated Laplacian pyramid.

查看全文查看/发表评论下载PDF阅读器