[1]宋昱,等.基于图像块l0梯度最小化的边缘保持平滑算法[J].深圳大学学报理工版,2021,38(3):307-314.[doi:10.3724/SP.J.1249.2021.03307]
 SONG Yu,and SUN Wenyun,et al.Edge-preserving smoothing algorithm based on l0 gradient minimization of image-patch[J].Journal of Shenzhen University Science and Engineering,2021,38(3):307-314.[doi:10.3724/SP.J.1249.2021.03307]
点击复制

基于图像块l0梯度最小化的边缘保持平滑算法()
分享到:

《深圳大学学报理工版》[ISSN:1000-2618/CN:44-1401/N]

卷:
第38卷
期数:
2021年第3期
页码:
307-314
栏目:
电子与信息科学
出版日期:
2021-05-14

文章信息/Info

Title:
Edge-preserving smoothing algorithm based on l0 gradient minimization of image-patch
文章编号:
202103014
作者:
宋昱1 2 3孙文赟1 2 3
1)深圳大学电子与信息工程学院,广东深圳 518060
2)深圳大学深圳市媒体信息内容安全重点实验室,广东深圳 518060
3)深圳大学广东省智能信号处理重点实验室,广东深圳 518060
Author(s):
SONG Yu1 2 3 and SUN Wenyun1 2 3
1) College of Electronics and Information Engineering, Shenzhen University, Shenzhen 518060, Guangdong Province, P.R.China
2) Shenzhen Key Laboratory of Media Security, Shenzhen University, Shenzhen 518060, Guangdong Province, P.R.China
3) Guangdong Key Laboratory of Intelligent Information Processing, Shenzhen University, Shenzhen 518060, Guangdong Province, P.R.China
关键词:
信号与信息处理边缘保持图像平滑l0梯度最小化图像块局部统计特性
Keywords:
signal and information processing edge-preserving image smoothing l0 gradient minimization image patch local statistics characteristic
分类号:
TN911.73;TP391
DOI:
10.3724/SP.J.1249.2021.03307
文献标志码:
A
摘要:
l 0梯度最小化图像平滑算法可在保持边缘的同时滤除纹理和细节,但该算法使用图像梯度判决被平滑成分时会出现包含较小图像梯度(弱边缘)的区域会被平滑,而包含较大图像梯度(强纹理)的区域被保留的现象.为克服此缺陷,提出一种基于图像块l0梯度最小化算法(image-patch based l0 gradient minimization algorithm, 简称IP-l0算法)的图像平滑算法,通过对输入图像中的图像块而非整幅图像进行平滑,动态改变图像块目标函数中的权重参数,令主要包含强纹理的图像块以较大的力度进行平滑,而主要包含弱边缘的图像块以较小的力度进行平滑,再整合平滑后的图像块得到整个边缘保持平滑图像.对IP-l0算法、原始的l0梯度最小化算法、基于局部拉普拉斯滤波器的算法、基于相对全变差算法、基于树滤波的算法,以及2种基于深度学习的边缘保持算法进行仿真实验,结果表明,使用IP-l0算法滤波后的图像能在保持较弱的边缘的同时平滑强纹理.
Abstract:
Image smoothing algorithm based on l0 gradient minimization can smooth details and textures of image while preserving edges. Since the algorithm uses image gradient to determine the smoothed component, the region with smaller image gradients (weak edge) can be smoothed. However, the region with larger image gradients (strong edge) will be preserved. In order to overcome this drawback, we propose an image-patch based l0 gradient minimization image smoothing algorithm (IP-l0). Instead of globally smoothing the input image, our algorithm smoothed each image patch first and then combined the smoothed patches together to obtain the final smoothed image. The weight parameters in the objective function used to smooth each image patch are dynamically changed according to the local statistics of image patch, so that the patches containing strong textures will be smoothed with greater force, and vice versa. The experimental results show that compared to the original l0 gradient minimization algorithm and several other state-of-the-art edge-preserving image smoothing algorithms including the local Laplacian filter based algorithm, the relative total variation based algorithm, the tree filter based algorithm, and two kinds of deep-learning based smoothing algorithms, the proposed algorithm can effectively smooth strong textures and well preserve weak edges or structures, and the ability of edge preservation and texture smoothing is better than other algorithms.

参考文献/References:

[1] HE Jianzhong, ZHANG Shiliang, YANG Ming, et al. Bi-directional cascade network for perceptual edge detection[C]// IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, USA: IEEE, 2019: 3823-3832.
[2] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. DeepLab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis Machine Intelligence, 2018, 40(4): 834-848.
[3] YAN Zhicheng, ZHANG Hao, WANG Baoyuan, et al. Automatic photo adjustment using deep neural networks[J]. ACM Transcations on Graphics, 2016, 35(2): 11.
[4] BARRON J T, POOLE B. The fast bilateral solver[C]// Computer Vision: ECCV 2016. Amsterdam: Springer, Cham, 2016: 617-632.
[5] GASTAL E S L , OLIVEIRA M M. Domain transform for edge-aware image and video processing[J]. ACM Transcations on Graphics, 2011, 30(4): 69.
[6] HE Kaiming, SUN Jian, TANG Xiaoou.Guided image filtering[J]. IEEE Transcations on Pattern Analysis and Machine Intelligence, 2013, 35(6): 1397-1409.
[7] OCHOTORENA C N, YAMASHITA Y. Anisotropic guided filtering[J]. IEEE Transactions on Image Processing, 2019, 29: 1397-1412.
[8] MISHRA D, CHAUDHURY S, SARKAR M, et al. Edge probability and pixel relativity-based speckle reducing anisotropic diffusion[J]. IEEE Transactions on Image Processing, 2018, 27(2): 649-664.
[9] PARIS S, HASINOFF S W, KAUTZ J. Local Laplacian filters: edge-aware image processing with a Laplacian pyramid[J]. ACM Transactions on Graphics, 2011, 30(4): 68.
[10] FATTAL R. Edge-avoiding wavelets and their applications[J]. ACM Transactions on Graphics, 2009, 28(3): 22.
[11] KASS M, SOLOMON J. Smoothed local histogram filters[J]. ACM Transactions on Graphics, 2010, 29(4): 100.
[12] BAO Linchao, SONG Yibing, YANG Qingxiong, et al. Tree filtering: efficient structure-preserving smoothing with a minimum spanning tree[J]. IEEE Transactions on Image Processing, 2014, 23(3): 555-569.
[13] KARACANN L, ERDEM E, ERDEM A. Structure-preserving image smoothing via region covariances[J]. ACM Transactions on Graphics, 2013, 32(6): 176.
[14] YIN Hui, GONG Yuanhao, QIU Guoping.Side window filtering[EB/OL]. (2019-05-17)[2020-6-30]. http://arxiv.org/abs/1905.07177.
[15] FARBMAN Z, FATTAL R, LISCHINSKI D, et al. Edge-preserving decompositions for multi-scale tone and detail manipulation[J]. ACM Transactions on Graphics, 2008, 27(2): 67.
[16] BAI Sai, HAN Xiaoguang, YU Yizhou. An image transform for edge-preserving smoothing and scene-level intrinsic decomposition[J]. ACM Transactions on Graphics, 2015, 34(4): 78.
[17] XU Li, LU Cewu, XU Yi, et al. Image smoothing via gradient minimization[J]. ACM Transactions on Graphics, 2011, 30(6): 174.
[18] SUBR K, SOLER C, DURAND F. Edge-preserving multiscale image decomposition based on local extrema[J]. ACM Transactions on Graphics, 2009, 28(5): 147.
[19] XU Li, YAN Qiong, XIA Yang, et al. Structure extraction from texture via relative total variation[J]. ACM Transactions on Graphics, 2012, 31(6): 139.
[20] ZHOU Zhiqiang, WANG Bo, MA Jinlei. Scale-aware edge-preserving image filtering via iterative global optimization[J]. IEEE Transactions on Multimedia, 2018, 20(6): 1392-1405.
[21] XU Li, REN J S J, YAN Qiong, et al. Deep edge-aware filters[C]// The 32nd International Conference on Machine Learning. Lile, France: International Machine Learning Society, 2015: 1669-1678.
[22] FAN Qingnan, YANG Jiaolong, WIPF D, et al. Image smoothing via unsupervised learning[J]. ACM Transactions on Graphics, 2018, 37(6): 259.

相似文献/References:

[1]刘进忙,倪鹏,李超,等.复杂战场环境下的分坐标处理[J].深圳大学学报理工版,2014,31(2):138.[doi:10.3724/SP.J.1249.2014.02138]
 Liu Jinmang,Ni Peng,Li Chao,et al.The independent coordinate processing in complex battlefield[J].Journal of Shenzhen University Science and Engineering,2014,31(3):138.[doi:10.3724/SP.J.1249.2014.02138]
[2]贺成龙,秦洪,于永生.一种空中目标航迹的自适应跟踪算法[J].深圳大学学报理工版,2014,31(4):361.[doi:10.3724/SP.J.1249.2014.04361]
 He Chenglong,Qin Hong,and Yu Yongsheng.An adaptive tracking algorithm for aerial target[J].Journal of Shenzhen University Science and Engineering,2014,31(3):361.[doi:10.3724/SP.J.1249.2014.04361]

备注/Memo

备注/Memo:
Received:2020-05-09;Revised:2020-07-02;Accepted:2020-07-18
Foundation:China Postdoctoral Science Foundation Project (2019M663068); Natural Science Foundation of Guangdong Province (2020A1515010563); Basic Research and Applied Basic Research Foundation of Guangdong Province (2020A1515110425); Shenzhen Basic Research Foundation (JCYJ20180305124550725)
Corresponding author:Dr. SONG Yu.E-mail: 519559374@qq.com
Citation:SONG Yu, SUN Wenyun. Edge-preserving smoothing algorithm based on l0 gradient minimization of image-patch[J]. Journal of Shenzhen University Science and Engineering, 2021, 38(3): 307-314.(in Chinese)
基金项目:中国博士后科学基金资助项目(2019M663068);广东省自然科学基金资助项目(2020A1515010563);广东省基础与应用基础研究基金资助项目(2019A1515110425);深圳市基础研究计划资助项目(JCYJ201803051245 50725)
作者简介:宋昱(1988—),深圳大学博士后研究人员.研究方向:图像处理和机器学习.E-mail:songy@szu.edu.cn
引文:宋昱,孙文赟.基于图像块l0梯度最小化的边缘保持图像平滑算法[J]. 深圳大学学报理工版,2021,38(3):307-314.
更新日期/Last Update: 2021-05-30