[1]沈钧贤,梅劲松,王 干,等.基于 CycleGan 的动车零件图像高光消除方法[J].机械与电子,2024,42(07):10-15.
　SHEN Junxian,MEI Jinsong,WANG Gan,et al.A Highlight Elimination Method for Train Parts Image Based on CycleGan[J].Machinery & Electronics,2024,42(07):10-15.
点击复制

基于 CycleGan 的动车零件图像高光消除方法()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
42
期数:
2024年07期
页码:
10-15
栏目:
研究与设计
出版日期:
2024-07-26

文章信息/Info

Title:
A Highlight Elimination Method for Train Parts Image Based on CycleGan
文章编号:
1001-2257 ( 2024 ) 07-0010-06
作者:
沈钧贤 1 ; 梅劲松 1 ; 王 干 2 ; 陈苏扬 3
1. 南京航空航天大学自动化学院,江苏 南京 211106 ;?
2. 南京拓控信息科技股份有限公司,江苏 南京 210004 ;?
3. 南京地铁运营有限责任公司,江苏 南京 210000
Author(s):
SHEN Junxian1 ;  MEI Jinsong1 ;  WANG Gan2 ;  CHEN Suyang3
( 1.College of Automation Engineering , Nanjing University of Aeronautics and Astronautics , Nanjing 211106 , China ;
2.Nanjing Tuokong Information Technology Co. , Ltd. , Nanjing 210004 , China ;
3.Nanjing Metro Operation Co. , Ltd. , Nanjing 210000 , China )
关键词:
高光消除; 图像处理; 光照补偿; 循环对抗生成网络
Keywords:
highlight removal ;  image processing ;  light compensation ;  CycleGan
分类号:
TP391.41 ; U266
文献标志码:
A
摘要:
应用光学无接触方法检测动车零件损伤时,采集图像的照射光易在零件上造成镜面反射,使得采集图像上存在高光,不利于对损伤部位的辨识。提出一种基于循环对抗生成网络( CycleGan )和光照补偿算法结合的方法来修复动车轮部图像的高光部分。首先通过生成手段初步修复高光区域,然后改进 Retinex 光照补偿算法,实现对生成图像的进一步增强,达到高光消除的目的。实验结果表明,所提方法大幅改善了高光区域的像素灰度,使得该部分像素灰度接近图像整体像素灰度,同时强化了暗部细节和图像信息,提高了损伤部分的辨识度。
Abstract:
When using optical non contact methods to detect damage to motor vehicle parts , the illumination light of the collected images can easily cause specular reflection on the parts , resulting in high light on the collected images , which is not conducive to the identification of damaged parts.This paper proposes a method based on a combination of CycleGan and illumination compensation algorithm ( Retinex ) to repair the highlight part of the moving wheel image.First , the highlight area is initially repaired through generation means , then the Retinex illumination compensation algorithm is improved to further enhance the generated image.Experimental results show that the method used greatly improves the pixel grayscale of the highlight area , making the pixel grayscale of this part close to the overall pixel grayscale of the image , which achieve the goal of highlight removal.Meanwhile , it strengthens the details and image information of the dark parts , and improves the recognition of the damaged parts.

参考文献/References:

[ 1 ] SU T , ZHOU Y , YU Y , et al.Highlight removal of multi-view facial images [ J ] .Sensors , 2022 , 22( 17 ): 6656.

[ 2 ] SOUZA A C S , MACEDO M C F , NASCIMENTO V P , et al.Real-time high-quality specular highlight removal using efficient pixel clustering [ C ] ∥2018 31st SIBGRAPI Conference on Graphics , Patterns and Images ( SIBGRAPI ) .New York : IEEE , 2018 : 56-63.
[ 3 ] 阮子行,黄勇,王梦,等 . 基于图像处理的番茄表面高光去除方法[ J ] . 中国瓜菜, 2023 , 36 ( 4 ): 64-70.
[ 4 ] SHAFER S A.Using color to separate reflection components [ J ] .Color research and application , 1985 , 10 ( 4 ):210-218.
[ 5 ] ZHANG Z , REN W H , LU Y , et al.Highlightremoval with orthogonal decomposition [ C ]∥2022 4th International Conference on Data-driven Optimization of Complex Systems ( DOCS ) .New York : IEEE , 2022 : 1-6.
[ 6 ] 李明悦 . 基于稀疏表示的人脸图像高光检测和消除[ J ] . 现代计算机, 2019 ( 19 ): 41-45.
[ 7 ] HOYER P O.Non-negative matrix factorization with sparseness constraints [ J ] .The journal of machine learning research , 2004 , 5 ( 9 ): 1457-1469.
[ 8 ] LI L , CHEN J C , CAO W J , et al.New framework for highlights removal based on cycleGAN [ C ] ∥2022 34th Chinese Control an Decision Conference ( CCDC ) . New York : IEEE , 2022 : 5082-5086.
[ 9 ] HU G W , ZHENG Y F , YAN H R , et al.Maskguided cycle-GAN for specular highlight removal [ J ] .Pattern recognition letters , 2022 , 161 : 108-114.
[ 10 ] KIM S Y , ABERMAN K , KANAZAWA N , et al. Zoom-to-inpaint : image inpainting with high-frequency details [ C ] ∥2022 IEEE / CVF Conference on Computer Vision and Pattern Recognition ( CVPR ), 2022 : 477-487.
[ 11 ] 宋文净,金晅宏 . 基于生成对抗网络的面部遮挡还原算法研究[ J ] . 电子测量技术, 2020 , 43 ( 23 ): 109-114.
[ 12 ] CRIMINISI A , PEREZ P , TOYAMA K.Object removal by exemplar-based inpainting [ C ] ∥2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition ( CVPR ) .New York : IEEE , 2003 : II.
[ 13 ] ZHU J Y , PARK T , ISOLA P , et al.Unpaired image-to-image translation using cycle consistent adversarial networks [ C ] ∥2017 IEEE International Conference on Computer Vision ( ICCV ) .New York : IEEE , 2017 : 2223-2232.
[ 14 ] LAND E H , MCCANN J J.Lightness and retinex theory [ J ] .Journal of the optical society of America , 1971 , 61( 1 ): 1-11.
[ 15 ] 刘雪峰,刘佳明,付民 . 生成对抗网络扩充样本用于高光谱图像分类 [ J ] . 电子测量技术, 2022 ,45 ( 3 ):146-152.
[ 16 ] ZHANG C , KUPPANNAGARI S R , KANNAN R , et al.Generative adversarial network for synthetic time series data generation in smart grids [ C ] ∥2018 IEEE International Conference on Communications , Control , and Computing Technologies for Smart Grids ( SmartGridComm ) .New York : IEEE , 2018 : 1-6.
[ 17 ] HOU Q B , ZHOU D Q , FENG J S.Coordinate attention for efficient mobile network design [ C ] ∥2021 IEEE / CVF Conference on Computer Vision and Pattern Recognition ( CVPR ) .New York : IEEE , 2021 : 13713-13722.
[ 18 ] 曹建芳,靳梦燕,李朝霞,等 . 基于循环生成对抗网络的壁画色彩修复算法[ J ] . 山东科技大学学报(自然科学版),2023 , 42 ( 4 ): 101-112.
[ 19 ] 牛群峰,刘江鹏,王莉,等 . 基于残差网络与迁移学习的柑橘类型识别模型研究[ J ] . 现代农业科技, 2023( 8 ): 73-76 , 79.
[ 20 ] FU G , ZHANG Q , ZHU L , et al.A multi-task network for joint specular highlight detection and removal [ C ] ∥2021 IEEE / CVF Conference on Computer Vision and Pattern Recognition ( CVPR ) .New York : IEEE , 2021 : 7752-7761.

相似文献/References:

[1]宋孟华１,曹金龙２,鲍成伟１. 基于图像处理的嵌入式火焰检测系统[J].机械与电子,2019,(01):58.
　, Embedded Flame Detection System Based on Image Processing[J].Machinery & Electronics,2019,(07):58.
[2]闫莉莉,张云鹤,方静,等.碳纤维增强铝基复合材料裂纹图像检测[J].机械与电子,2017,(08):36.[doi:1001-2257(2017)08-0036-05]
　YAN Lili,ZHANG Yunhe,FANG Jing,et al.The Interlaminar Crack Image Detection of the Carbon Fiber Reforced Aluminum Matrix Composites[J].Machinery & Electronics,2017,(07):36.[doi:1001-2257(2017)08-0036-05]
[3]刘朝涛,于彪.基于Arduino的桥梁检测设备的研发[J].机械与电子,2016,(08):76.
　LIU Chaotao,YU Biao.Research and Development of a Bridge Detection Equipment based on Arduino[J].Machinery & Electronics,2016,(07):76.
[4]张小洁,段峻.改进的Hough变换在零件圆心定位中的应用[J].机械与电子,2016,(10):62.
　ZHANG Xiaojie,DUAN Jun.Application of Improved Hough Transforming in Parts Center Positioning[J].Machinery & Electronics,2016,(07):62.
[5]陈 红,胡恩才,白静静.交叉路口信号灯智能管理系统设计[J].机械与电子,2019,(11):63.
　,Design of Intelligent Management System for Intersection Signal Lights[J].Machinery & Electronics,2019,(07):63.
[6]李 勋１,张伟军１,徐智卿２,等.狭小空间内孔高测量机器人设计及视觉定位方法研究[J].机械与电子,2019,(12):60.
　,,et al.Research on Design and Visual Positioning Method of Hole Height Measurement Robot in Narrow Space[J].Machinery & Electronics,2019,(07):60.
[7]潘泽锴,杨浩然.基于嵌入式的剑麻绳图像采集与检测系统[J].机械与电子,2020,(06):37.
　PAN Zekai,YANG Haoran. The Embedded Sisal Rope Image Acquisition and Detection System[J].Machinery & Electronics,2020,(07):37.
[8]孙 佩,胡小敏.基于图像特征提取的地铁客流量自动统计系统[J].机械与电子,2020,(10):17.
　SUN Pei,HU Xiaomin.Metro Passenger Flow Statistics System Based on Image Feature Extraction[J].Machinery & Electronics,2020,(07):17.
[9]李晓飞,张周强,张守京,等.基于LabVIEW的迈克尔逊干涉微位移测量方法研究[J].机械与电子,2021,(01):8.
　LI Xiaofei,ZHANG Zhouqiang,ZHANG Shoujing,et al.Research on Michelson Interference Micro-displacement Measurement Method Based on LabVIEW[J].Machinery & Electronics,2021,(07):8.
[10]杨玥旻,闫维新.基于图像处理的爬壁机器人焊缝识别与跟踪[J].机械与电子,2021,(03):65.
　YANG Yuemin,YAN Weixin.Welding Seam Recognition and Tracking Based on Image Processing for Wall-climbing Robot[J].Machinery & Electronics,2021,(07):65.

备注/Memo

备注/Memo:
收稿日期: 2023-10-18
作者简介:沈钧贤 ( 1998- ),男,江苏苏州人,硕士研究生,研究方向为图像处理、无损检测;梅劲松 ( 1967- ),男,江苏东台人,副研究员,研究方向为无人飞行器控制、嵌入式系统、无损检测系统、检测技术与自动化装置;王 干 ( 1992- ),男,江苏南京人,硕士研究生,图像工程师,研究方向为目标检测、无损检测;陈苏扬 ( 1989- ),男,江苏扬州人,中级工程师,研究方向为地铁车辆检修。
更新日期/Last Update: 2024-08-28