计算机科学 ›› 2019, Vol. 46 ›› Issue (9): 277-283.doi: 10.11896/j.issn.1002-137X.2019.09.042

• 图形图像与模式识别 • 上一篇    下一篇

基于卷积网络的边缘保持滤波方法

石晓红1,2,3, 黄钦开4, 苗佳欣5, 苏卓4,5   

  1. (广州大学数学与信息科学学院 广州510006)1;
    (广州大学计算科技研究院 广州510006)2;
    (广东省数学教育软件工程技术研究中心 广州510006)3;
    (中山大学数据科学与计算机学院 广州510006)4;
    (中山大学国家数字家庭工程技术研究中心 广州510006)5
  • 收稿日期:2018-08-30 出版日期:2019-09-15 发布日期:2019-09-02
  • 通讯作者: 苏 卓(1985-),男,博士,CCF会员,主要研究方向为计算机视觉、图形图像处理,E-mail:suzhuo3@mail.sysu.edu.cn
  • 作者简介:石晓红(1978-),女,硕士,主要研究方向为图像处理;黄钦开(1994-),男,硕士生,主要研究方向为图像处理;苗佳欣(1994-),女,硕士生,主要研究方向为图像处理;
  • 基金资助:
    国家自然科学基金青年基金项目(61502541),2016年贵州省科技平台及人才团队专项资金项目(黔科合平台人才5609),2016年贵州省省级重点支持学科“计算机应用技术”(黔学位合字ZDXK20号),广州大学研究生创新能力培养资助计划(2018GDJC-D03)

Edge-preserving Filtering Method Based on Convolutional Neural Networks

SHI Xiao-hong1,2,3, HUANG Qin-kai4, MIAO Jia-xin5, SU Zhuo4,5   

  1. (School of Mathematics and Information Science,Guangzhou University,Guangzhou 510006,China)1;
    (Institute of Computing Science and Technology,Guangzhou University,Guangzhou 510006,China)2;
    (Guangdong Provincial Engineering and Technology Research Center for Mathematical Education Software,Guangzhou 510006,China)3;
    (School of Data and Computer Science,Sun Yat-sen University,Guangzhou 510006,China)4;
    (National Engineering Research Center of Digital Life,Sun Yat-sen University,Guangzhou 510006,China)5
  • Received:2018-08-30 Online:2019-09-15 Published:2019-09-02

摘要: 边缘保持滤波是计算机视觉、图像处理领域的重要基础理论研究,作为图像预处理操作对后续的处理结果有着重要影响。区别于传统滤波方法,边缘保持滤波方法不仅注重图像的平滑处理,还注重保持边缘细节。卷积神经网络在很多领域得到了应用,并取得显著的成果。本文将卷积神经网络引入边缘保持滤波,利用卷积神经网络的良好扩展性和灵活性来构建深度卷积神经网络模型(Deep Convolutional Neural Network,DCNN),通过3种类型的网络堆叠层,采用反向传播迭代更新网络参数,训练残差图像,实现基于DCNN的边缘保持滤波方法;还构建了基于梯度域的卷积神经网络模型(Gradient CNN,GCNN),对彩色图像的梯度信息进行学习,通过三层卷积对梯度图进行边缘保持平滑操作,得到边缘保持平滑梯度图,进而利用输入图像引导平滑梯度图进行彩色重建,得到彩色滤波图像。最后通过实验与常见的边缘保持滤波方法进行主观和客观评价对比。DCNN不仅在视觉上达到了其他滤波的效果,同时在处理时间上也存在较大优势,表明DCNN可以通过大量的数据训练有效地拟合出多种边缘保持滤波算法。与其他边缘保持滤波结果相比,GCNN在视觉上可以保持颜色风格与输入图像整体一致,而且图像相似度评价指标也更好,表明GCNN解决了部分滤波处理出现颜色偏差、梯度反转等问题,而且提高了处理效率。

关键词: 边缘保持, 卷积神经网络, 滤波, 平滑操作

Abstract: Edge-preserving filtering is a significant basic theory research in the fields of computer vision and image processing.As subsequent operation of pre-processing,edge-preserving filtering has great influence on final results of image processing.Different with traditional filtering,edge-preserving filtering focuses not only on smooth,but also on image edge details.Convolutional neural networks (CNNs) have been applied into a variety of research fields with great success.In this paper,CNN was introduced into edge-preserving filtering.Taking advantages of CNN’s excellent extensibility and flexibility,this paper constructed a deep convolutional neural network (DCNN).With three types of cascading network layers,DCNN iteratively updates its parameters by back propagation,produces a residual image and realizes a DCNN-based edge-preserving filtering.Besides,a gradient CNN model (GCNN) was constructed.The gradient of color images is learnt,edge-preserving smoothing operation is conducted for gradient images by three layers of convolution,and edge-preserving filtering gradient images are obtained.Subsequently,the input image is used to guide the filtering gradient image for reconstructing and obtaining color filtering image.Finally,experiments were made to evaluate the proposed methods and the proposed methods were compared with popular edge-preserving filtering methods subjectively and objectively.DCNN not only achieves the same visual effects as other methods,but also has big advantages in processing time,which demonstrates that DCNN can effectively and efficiently imitate various filtering methods through training on large amount of data.For GCNN,in terms of visual effects,its output conforms to the input in the color style globally.In terms of image similarity evaluation,it also outperforms other methods.This verifies that GCNN can address the problems of color shift and gradient inversion,as well as improve the filtering efficiency.

Key words: Convolutional neural networks, Edge-preserving, Filtering, Smoothing

中图分类号: 

  • TP391
[1]FARBMAN Z,FATTAL R,LISCHINSKI D.Edge-preservingdecompositions for multi-scale tone and detail manipulation [J].ACM Transactions on Graphics (TOG),2008,27(3):1-10.
[2]PARIS S,HASINOFF S W,KAUTZ J.Local Laplacian filters:edge-aware image processing with a Laplacian pyramid [J].ACM Transactions on Graphics (TOG),2015,58(3):81-91.
[3]GASTAL E S L,OLIVEIRA M M.Domain transform for edge-aware image and video processing[J].ACM Transactions on Graphics (TOG),2011,30(4):Article No.69.
[4]PARIS S,DURAND F.A fast approximation of the bilateral filter using a signal processing approach [J].International Journal of Computer Vision,2009,81(1):24-52.
[5]XU L,LU C,XU Y,et al.Image smoothing via L0 gradient mini-mization [J].ACM Transactions on Graphics (TOG),2011,30(6):174.
[6]TOMASI C,MANDUCHI R.Bilateral filtering for gray and co-lor images [C]//1998 IEEE Sixth International Conference on Computer Vision (ICCV’98).Piscataway,NJ:IEEE,1998:839-846.
[7]PORIKLI F.Constant time O (1) bilateral filtering [C]//2008 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’08).Piscataway,NJ:IEEE,2008:1-8.
[8]YANG Q,TAN K,AHUJA N.Real-time O (1) bilateral filtering [C]//2009 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’09).Piscataway,NJ:IEEE,2009:557-564.
[9]HE K,SUN J,TANG X.Guided image filtering [C]//European Conference on Computer Vision (ECCV’10).New York,NY:Springer,2010:1-14.
[10]PERONA P,MALIK J.Scale-space and edge detection using anisotropic diffusion [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1990,12(7):629-639.
[11]XU L,YAN Q,XIA Y,et al.Structure extraction from texture via relative total variation [J].ACM Transactions on Graphics,2012,31(6):Article No.139.
[12]DONG C,LOY C C,HE K et al.Image super-resolution using deep convolutional networks [J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2016,38(2):295-307.
[13]XIE J,XU L,CHEN E.Image denoising and inpainting withdeep neural networks [C]//25th International Conference on Neural Information Processing Systems.Red Hook,NY:Curran Associates,Inc.,2012:341-349.
[14]LIU S,PAN J,YANG M.Learning recursive filters for low-level vision via a hybrid neural network [C]//European Conference on Computer Vision (ECCV’16).New York,NY:Springer,2016:560-576.
[15]LI Y,HUANG J,AHUJA N,et al.Deep joint image filtering[C]//European Conference on Computer Vision (ECCV’16).New York,NY:Springer,2016:154-169.
[16]SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition [C]//International Conference on Learning Representations 2015 (ICLR2015).San Die-go,CA,2015.
[17]SZEGEDY C,LIU W,JIA Y,et al.Going deeper with convolutions [C]//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’15).Piscataway,NJ:IEEE,2015:1-9.
[18]XU L,REN J,YAN Q,et al.Deep edge-aware filters [C]//International Conference on Machine Learning.2015:1669-1678.
[19]ZHANG Q,SHEN X,XU L et al.Rolling guidance filter [C]//European Conference on Computer Vision (ECCV’14).New York,NY:Springer,2014:815-830.
[20]KARACAN L,ERDEM E,ERDEM A.Structure-preserving ima-ge smoothing via region covariances [J].ACM Transactions on Graphics,2013,32(6):Article No.176.
[21]HAM B,CHO M,PONCE J.Robust image filtering using joint static and dynamic guidance [C]//2015 IEEE Conference on Computer Vision and Pattern Recognition(CVPR’15).Piscata-way,NJ:IEEE,2015:4823-4831.
[1] 周乐员, 张剑华, 袁甜甜, 陈胜勇.
多层注意力机制融合的序列到序列中国连续手语识别和翻译
Sequence-to-Sequence Chinese Continuous Sign Language Recognition and Translation with Multi- layer Attention Mechanism Fusion
计算机科学, 2022, 49(9): 155-161. https://doi.org/10.11896/jsjkx.210800026
[2] 唐清华, 王玫, 唐超尘, 刘鑫, 梁雯.
基于M2M相遇区的PDR室内定位方法
PDR Indoor Positioning Method Based on M2M Encounter Region
计算机科学, 2022, 49(9): 283-287. https://doi.org/10.11896/jsjkx.210800270
[3] 李宗民, 张玉鹏, 刘玉杰, 李华.
基于可变形图卷积的点云表征学习
Deformable Graph Convolutional Networks Based Point Cloud Representation Learning
计算机科学, 2022, 49(8): 273-278. https://doi.org/10.11896/jsjkx.210900023
[4] 陈泳全, 姜瑛.
基于卷积神经网络的APP用户行为分析方法
Analysis Method of APP User Behavior Based on Convolutional Neural Network
计算机科学, 2022, 49(8): 78-85. https://doi.org/10.11896/jsjkx.210700121
[5] 朱承璋, 黄嘉儿, 肖亚龙, 王晗, 邹北骥.
基于注意力机制的医学影像深度哈希检索算法
Deep Hash Retrieval Algorithm for Medical Images Based on Attention Mechanism
计算机科学, 2022, 49(8): 113-119. https://doi.org/10.11896/jsjkx.210700153
[6] 沈祥培, 丁彦蕊.
多检测器融合的深度相关滤波视频多目标跟踪算法
Multi-detector Fusion-based Depth Correlation Filtering Video Multi-target Tracking Algorithm
计算机科学, 2022, 49(8): 184-190. https://doi.org/10.11896/jsjkx.210600004
[7] 檀莹莹, 王俊丽, 张超波.
基于图卷积神经网络的文本分类方法研究综述
Review of Text Classification Methods Based on Graph Convolutional Network
计算机科学, 2022, 49(8): 205-216. https://doi.org/10.11896/jsjkx.210800064
[8] 金方焱, 王秀利.
融合RACNN和BiLSTM的金融领域事件隐式因果关系抽取
Implicit Causality Extraction of Financial Events Integrating RACNN and BiLSTM
计算机科学, 2022, 49(7): 179-186. https://doi.org/10.11896/jsjkx.210500190
[9] 张颖涛, 张杰, 张睿, 张文强.
全局信息引导的真实图像风格迁移
Photorealistic Style Transfer Guided by Global Information
计算机科学, 2022, 49(7): 100-105. https://doi.org/10.11896/jsjkx.210600036
[10] 戴朝霞, 李锦欣, 张向东, 徐旭, 梅林, 张亮.
基于DNGAN的磁共振图像超分辨率重建算法
Super-resolution Reconstruction of MRI Based on DNGAN
计算机科学, 2022, 49(7): 113-119. https://doi.org/10.11896/jsjkx.210600105
[11] 刘月红, 牛少华, 神显豪.
基于卷积神经网络的虚拟现实视频帧内预测编码
Virtual Reality Video Intraframe Prediction Coding Based on Convolutional Neural Network
计算机科学, 2022, 49(7): 127-131. https://doi.org/10.11896/jsjkx.211100179
[12] 徐鸣珂, 张帆.
Head Fusion:一种提高语音情绪识别的准确性和鲁棒性的方法
Head Fusion:A Method to Improve Accuracy and Robustness of Speech Emotion Recognition
计算机科学, 2022, 49(7): 132-141. https://doi.org/10.11896/jsjkx.210100085
[13] 孙福权, 崔志清, 邹彭, 张琨.
基于多尺度特征的脑肿瘤分割算法
Brain Tumor Segmentation Algorithm Based on Multi-scale Features
计算机科学, 2022, 49(6A): 12-16. https://doi.org/10.11896/jsjkx.210700217
[14] 黄国兴, 杨泽铭, 卢为党, 彭宏, 王静文.
利用粒子滤波方法求解数据包络分析问题
Solve Data Envelopment Analysis Problems with Particle Filter
计算机科学, 2022, 49(6A): 159-164. https://doi.org/10.11896/jsjkx.210600110
[15] 吴子斌, 闫巧.
基于动量的映射式梯度下降算法
Projected Gradient Descent Algorithm with Momentum
计算机科学, 2022, 49(6A): 178-183. https://doi.org/10.11896/jsjkx.210500039
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!