[1]王建新１,朱 煜１,２,等.船体零件几何尺寸测量图像分割方法研究[J].机械与电子,2021,(05):68-73.
　WANG Jianxin,ZHU Yu,HU Xiaofeng,et al.Research on Image Segmentation for Geometric Dimension Measurement of Hull Parts[J].Machinery & Electronics,2021,(05):68-73.
点击复制

船体零件几何尺寸测量图像分割方法研究()
分享到：

机械与电子[ISSN:1001-2257/CN:52-1052/TH]

卷:
期数:
2021年05期
页码:
68-73
栏目:
智能工程
出版日期:
2021-05-24

文章信息/Info

Title:
Research on Image Segmentation for Geometric Dimension Measurement of Hull Parts
文章编号:
１００１-２２５７ ( ２０２１ ) ０５-００６８-０６
作者:
王建新１ ; 朱 煜１ ; ２ ; 胡小锋１ ; 张亚辉１
１． 上海交通大学机械与动力工程学院,上海 ２００２４０ ;?
２． 江南造船(集团)有限责任公司,上海 ２０１９１３
Author(s):
WANG Jianxin１ ; ZHU Yu１ ; ２ ; HU Xiaofeng１ ; ZHANG Yahui１
(１．School of Mechanical Engineering , Shanghai Jiao Tong University , Shanghai ２００２４０ , China ;
２．Jiangnan Shipyard ( Group ) Co．, Ltd． ,Shanghai ２０１９１３ , China)
关键词:
图像分割; 边缘检测; k-means 聚类; 轮廓提取; 边缘连接
Keywords:
image segmentation ;  edge detection ;  k-means clustering ;  contour extraction ;  edge linking
分类号:
TP３９１
文献标志码:
A
摘要:
针对船体零件表面划痕、锈蚀以及下表面成像等因素导致的伪边缘及边缘不连续问题,提出了基于边缘与 k-means 聚类的船体零件图像分割方法.对于表面划痕及锈蚀形成的伪边缘,在矩形边缘检测区域内利用轮廓法线方向与边缘梯度方向信息予以剔除;零件下表面成像导致的伪边缘,选取 k-means聚类区分零件的上下表面边缘;边缘不连续问题,基于测地距离和断点方向准确连接断点,形成封闭轮廓.实验结果表明,使用该方法可获得平滑、封闭的分割效果,船体零件的几何尺寸测量误差低于 ０.５mm ,且测得量值标准偏差的平均值为 ０.２６２mm ,测量重复性较高,有效保障了船体零件几何尺寸的测量精度.
Abstract:
Focused on the issue of pseudo edge and edge discontinuity caused by scratches , corrosion and imaging of undersurface of hullparts , an image segmentation method based on edge detection and edge linking is proposed．The normal direction and edge gradient direction information of the contour are used to eliminate the pseudo edges formed by scratches and corrosion in the rectangular edge detection area ; for pseudo edges caused by imaging of the lower surface of parts , the k-means clustering is selected to distinguish the upper and lower surface edges ; aiming at the problem of edge discontinuity , based on the geodesic distance and the direction of the break point , the break points areaccurately connected to form a closed contour．Experimental results show that the proposed method can obtain smooth and closed segmentation results , the measurement error of hull parts is less than ０.５mm , and the average value of the measured standard deviation is ０.２６２mm．This method effectively ensures the measurement precision of hull parts．

参考文献/References:

[１] 张辉,韦建光,吴卓琦,等 ．多尺度曲率角与灰度比例复合的角点检测方法[J ] ．东华大学学报(自然科学版),２０１５ , ４１ ( ４ ): ４１５-４２０ , ４２７．

[２] 钮圣虓,王盛,杨晶晶,等 ．完全基于边缘信息的快速图像分割算法[J ] ．计算机辅助设计与图形学学报, ２０１２ ,２４ ( １１ ): １４１０-１４１９．
[３] 赵渊博,赵慧洁 ．挠性接头薄筋厚度在线测量图像分割方法研究[J ] ．光学精密工程, ２００５ , １３ (增刊 １ ): １５３ -１５７．
[４] 徐红运,徐小力,左云波,等 ．复杂背景下的风力机组叶片图像 分 割 方 法 研 究 [ J ] ．电 子 测 量 技 术, ２０１８ , ４１(２２ ): ６０-６５．
[５] 张弘,郑培甜 ．灰度梯度映射函数改进的 Otsu 阈值法[J ] ．计算机工程与应用, ２０１７ , ５３ ( １２ ): １９０-１９５．
[６] HUANG Z P , YI F L , JIANG S Z , et al.The segmentation of liver and vessels in CT images using ３D hierarchical seeded region growing [ C ]// Proceedings of ２０１１International Conference on Computer Science and Automation Engineering．New York : IEEE Press , ２０１１ :２６４-２６９．
[７] 亢洁,丁菊敏,万永,等 ．基于分水岭修正与 U-Net 的肝脏图像分割算法[J ] ．计算机工程, ２０２０ , ４６ ( １ ): ２５５-２６１ , ２７０．
[８] 宗晓萍,田伟倩 ．采用 K- means 的脑肿瘤磁共振图像分割与特征提取[J ] ．计算机工程与应用, ２０２０ , ５６ ( ３ ):１８７-１９３．
[９] 王志彬,王开义,张水发,等 ．基于 K- means 聚类和椭圆拟合方 法的 白 粉虱 计数 算法 [ J ] ．农 业 工 程 学 报,２０１４ , ３０ ( １ ): １０５-１１２．
[１０] GONZALEZ R C , WOODS R E． 数字图像处理[ M ] ．阮秋 琦,阮 宇 智,译 ．３ 版 ．北 京:电 子 工 业 出 版社,２０１１．
[１１] 王艺璇 ．机器视觉测量中的工件外轮廓提取方法[ D ] ．武汉:华中科技大学,２０１９．
[１２ ] 许宏科,秦严严,陈会茹 ．一种基于改进 Canny 的边缘检测算法[J ] ．红外与激光工程, ２０１６ , ４５ ( ６ ):２０４-２１０．
[１４] WANG Z , ZHANG H．Edge linking using geodesic distance and neighborhood information [ C ]// International Conference on Advanced Intelligent Mechatronics , ２００８ : １５１-１５５．
[１５] 全国法制计量管理计量技术委员会 ．测量不确定度评定与表示:JJF１０５９．１ — ２０１２ [ S ] ．北京:中国质检出版社,２０１２．

相似文献/References:

[1]唐德谦,李钧,姚保良,等.智能巡检机器人的电缆缺陷检测关键技术研究[J].机械与电子,2019,(03):76.
　TANG Deqian,LI Jun,YAO Baoliang,et al.Research on Key Technologies of Cable Inspection for Intelligent Inspection Robot[J].Machinery & Electronics,2019,(05):76.
[2]王鹏宇,游有鹏,杨雪峰.结合密度峰聚类的 Ｋ 均值图像分割算法[J].机械与电子,2019,(02):40.
　,K-means Image Segmentation Algorithm Combined with Density Peak Clustering[J].Machinery & Electronics,2019,(05):40.
[3]黄亚伟,陈 悦,黄晓华.基于遗传算法的二维最大类间方差法的优化[J].机械与电子,2018,(04):20.
　HUANG Yawei,CHEN Yue,HUANG Xiaohua.Optimization of Two-Dimensional Otsu Method Based on Genetic Algorithm[J].Machinery & Electronics,2018,(05):20.
[4]林丽红,马铁军,徐培.Gabor变换在轮胎X光图像处理的应用[J].机械与电子,2016,(04):59.
　LIN Lihong,MA Tiejun,XU Pei.Application of Gabor Transform in X-ray Tire Image Processing[J].Machinery & Electronics,2016,(05):59.
[5]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,(05):62.
[6]魏 哲１,张 阳１,张明明２,等.视觉引导划片机轨迹优化算法研究[J].机械与电子,2019,(12):38.
　WEI Zhe,ZHANG Yang,ZHANG Mingming,et al.Research on Trajectory Optimization Algorithm of Visual Guide Grinding Wheel Dicing Machine[J].Machinery & Electronics,2019,(05):38.
[7]于晓海,张 阳,须 颖.一种改进自适应阈值的 Ｃａｎｎｙ算法[J].机械与电子,2020,(01):6.
　,An Improved Adaptive Threshold Canny Algorithm[J].Machinery & Electronics,2020,(05):6.
[8]高英姿,姚爱琴,孙运强,等.基于二进小波和抗噪形态学的烟雾边缘检测算法[J].机械与电子,2021,(08):18.
　GAO Yingzi,YAO Aiqin,SUN Yunqiang,et al.Smoke Edge Detection Algorithm Based on Dyadic Wavelet and Anti Noise Morphology[J].Machinery & Electronics,2021,(05):18.
[9]田 婷,陈 平,刘 宾.基于立体匹配技术的 X 射线分层成像方法研究[J].机械与电子,2021,(09):3.
　TIAN Ting,CHEN Ping,LIU Bin.Research on X-ray Layered Imaging Method Based on Stereo Matching Technology[J].Machinery & Electronics,2021,(05):3.
[10]包建国,邱鹏程,蒋兴福,等.基于图像分割的蒸发器二次侧管板清洗喷嘴自适应定位[J].机械与电子,2021,(11):3.
　BAO Jianguo,QIU Pengcheng,JIANG Xingfu,et al.Adaptive Positioning of Tubeplate Cleaning Nozzle on Secondary Side of Evaporator Based on Image Segmentation[J].Machinery & Electronics,2021,(05):3.

备注/Memo

备注/Memo:
收稿日期: ２０２０-１２-１６
基金项目:国家重点研发计划资助项目( ２０１８YFB１７００５００ )
作者简介:王建新 (１９９６－),男,河北唐山人,硕士研究生,研究方向为图像处理及视觉测量;朱 煜 ( １９７４－),男,上海人,博士研究生,研究员,研究方向为智能制造;胡小锋 (１９７７－),男,浙江兰溪人,博士,副教授,博士研究生导师,研究方向为信息物理融合装配系统;张亚辉 (１９８７－),女,河北石家庄人,博士,助理研究员,研究方向为智能制造以及装配规划.
更新日期/Last Update: 2021-06-04