基于最小二乘支持向量机的激光拼焊焊缝识别
|
摘要
激光拼焊焊缝质量结构光视觉检测中,对焊缝的准确识别是实现高精度检测的关键.针对检测图像中结构光光纹畸变特征不明显,无法准确识别焊缝的问题,依据焊缝纹理特征信息,提出了一种基于最小二乘支持向量机的焊缝识别方法.首先,分析并提取焊缝区和非焊缝区差异明显的纹理特征.其次,训练最小二乘支持向量机模型,对焊缝进行粗识别.最后,采用Laws纹理滤波提取焊缝区域,并通过阈值分割方法精确识别焊缝.针对不同工艺参数下的激光拼焊焊缝开展焊缝识别试验,结果表明,该方法能够有效地识别焊缝.
Accurate recognition of weld seam was the key for structural-light visual inspection of weld quality with high precision in tailored blank laser welding. Because of the problem that when the distortion of laser stripe was not obvious in the image and the welding seam cannot be recognized accurately, a recognition method according to the texture information of weld seam based on least squares support vector machine was proposed in this paper. Firstly, the textural features of the image were analysed and the textural features which had obvious difference between weld seam region and non-welded region were extracted. Secondly, the least square support vector machine model was trained and the coarse recognition of weld seam was accomplished. Finally, a fine recognition was achieved by Laws texture filter and threshold segmentation. The recognition experiments were carried out for weld seam in different welding parameters and the results showed that the weld seam can be recognized effectively by this method.
Accurate recognition of weld seam was the key for structural-light visual inspection of weld quality with high precision in tailored blank laser welding. Because of the problem that when the distortion of laser stripe was not obvious in the image and the welding seam cannot be recognized accurately, a recognition method according to the texture information of weld seam based on least squares support vector machine was proposed in this paper. Firstly, the textural features of the image were analysed and the textural features which had obvious difference between weld seam region and non-welded region were extracted. Secondly, the least square support vector machine model was trained and the coarse recognition of weld seam was accomplished. Finally, a fine recognition was achieved by Laws texture filter and threshold segmentation. The recognition experiments were carried out for weld seam in different welding parameters and the results showed that the weld seam can be recognized effectively by this method.
引文
[1]Irving B.Welding tailored blanks is hot issue for automakers[J].Welding Journal,1995,74(8):49-52.
[2]夏卫生,刘奋,韦春华,等.激光拼焊板HAZ力学性能的预测及有限元分析[J].焊接学报,2016,37(3):79-82.Xia Weisheng,Liu Fen,Wei Chunhua,et al.Prediction and finite element analysis of the mechanical properties of heat affected zone of laser welded blanks[J].Transactions of the China Welding Institution,2016,37(3):79-82.
[3]Chen Yanbin,Tao Wang,Li Liqun,et al.Visual sensing and morphological image processing of weld pool in laser spot welding[J].China Welding,2010,19(2):70-74.
[4]Nguyen H C,Lee B R.Laser-vision-based quality inspection sys-tem for small-bead laser welding[J].International Journal of Precision Engineering and Manufacturing,2014,15(3):415-423.
[5]伏喜斌,林三宝,杨春利,等.基于激光视觉传感的角焊缝外形尺寸检测[J].焊接学报,2008,29(7):47-50.Fu Xibin,Lin Sanbao,Yang Chunli,et al.Inspection of fillet weld shape dimension based on laser vision sensing[J].Transactions of the China Welding Institution,2008,29(7):47-50.
[6]Xu M,Zhao M Y,Zhang C N.Image processing method for weld quality inspection system of tailored blanks laser welding[C]//2010 International Conference on Measuring Technology and Mechatronics Automation.Changsha 2010:422-426.
[7]王平江,吴家勇,陈吉红,等.激光焊接中狭窄对接拼缝测量方法[J].中国机械工程,2010,21(17):2132-2137.Wang Pingjiang,Wu Jiayong,Chen Jihong,et al.A method for measuring seam topograph in tailored blank laser welding[J].China Mechanical Engineering,2010,21(17):2132-2137.
[8]Park S,Lee HS,Kim J.Seed growing for interactive image segmentation using SVM classification with geodesic distance[J].Electronics Letters,2017,53(1):22-23.
[9]宋召青,程子君,郑苏,等.基于平滑滤波与最小二乘支持向量机的指纹图像识别研究[J].海军航空工程学院学报,2010,25(2):172-176,180.Song Zhaoqing,Cheng Zijun,Zheng Su,et al.Research on the fingerprint image recognition based on smooth filter and least square support vector machine[J].Journal of Naval Aeronautical and Astronautical University,2010,25(2):172-176,180.
[10]王胜华,都东,曾凯,等.基于纹理特征的焊缝识别方法[J].焊接学报,2008,29(11):5-8.Wang Shenghua,Du Dong,Zeng Kai,et al.Weld recognition based on texture feature[J].Transactions of the China Welding Institution,2008,29(11):5-8.
[11]陈英,杨丰玉,符祥.基于支持向量机和灰度共生矩阵的纹理图像分割方法[J].传感器与微系统,2012,31(9):60-63.Chen Ying,Yang Fengyu,Fu Xiang.Method of texture image segmentation based on SVM and gray level co-occurrence matrix[J].Transducer and Microsystem Technologies,2012,31(9):60-63.
[12]Laws K I.Textured image segmentation[D].University of Southern California,1980.
[2]夏卫生,刘奋,韦春华,等.激光拼焊板HAZ力学性能的预测及有限元分析[J].焊接学报,2016,37(3):79-82.Xia Weisheng,Liu Fen,Wei Chunhua,et al.Prediction and finite element analysis of the mechanical properties of heat affected zone of laser welded blanks[J].Transactions of the China Welding Institution,2016,37(3):79-82.
[3]Chen Yanbin,Tao Wang,Li Liqun,et al.Visual sensing and morphological image processing of weld pool in laser spot welding[J].China Welding,2010,19(2):70-74.
[4]Nguyen H C,Lee B R.Laser-vision-based quality inspection sys-tem for small-bead laser welding[J].International Journal of Precision Engineering and Manufacturing,2014,15(3):415-423.
[5]伏喜斌,林三宝,杨春利,等.基于激光视觉传感的角焊缝外形尺寸检测[J].焊接学报,2008,29(7):47-50.Fu Xibin,Lin Sanbao,Yang Chunli,et al.Inspection of fillet weld shape dimension based on laser vision sensing[J].Transactions of the China Welding Institution,2008,29(7):47-50.
[6]Xu M,Zhao M Y,Zhang C N.Image processing method for weld quality inspection system of tailored blanks laser welding[C]//2010 International Conference on Measuring Technology and Mechatronics Automation.Changsha 2010:422-426.
[7]王平江,吴家勇,陈吉红,等.激光焊接中狭窄对接拼缝测量方法[J].中国机械工程,2010,21(17):2132-2137.Wang Pingjiang,Wu Jiayong,Chen Jihong,et al.A method for measuring seam topograph in tailored blank laser welding[J].China Mechanical Engineering,2010,21(17):2132-2137.
[8]Park S,Lee HS,Kim J.Seed growing for interactive image segmentation using SVM classification with geodesic distance[J].Electronics Letters,2017,53(1):22-23.
[9]宋召青,程子君,郑苏,等.基于平滑滤波与最小二乘支持向量机的指纹图像识别研究[J].海军航空工程学院学报,2010,25(2):172-176,180.Song Zhaoqing,Cheng Zijun,Zheng Su,et al.Research on the fingerprint image recognition based on smooth filter and least square support vector machine[J].Journal of Naval Aeronautical and Astronautical University,2010,25(2):172-176,180.
[10]王胜华,都东,曾凯,等.基于纹理特征的焊缝识别方法[J].焊接学报,2008,29(11):5-8.Wang Shenghua,Du Dong,Zeng Kai,et al.Weld recognition based on texture feature[J].Transactions of the China Welding Institution,2008,29(11):5-8.
[11]陈英,杨丰玉,符祥.基于支持向量机和灰度共生矩阵的纹理图像分割方法[J].传感器与微系统,2012,31(9):60-63.Chen Ying,Yang Fengyu,Fu Xiang.Method of texture image segmentation based on SVM and gray level co-occurrence matrix[J].Transducer and Microsystem Technologies,2012,31(9):60-63.
[12]Laws K I.Textured image segmentation[D].University of Southern California,1980.