基于双三次插值算法的集成电路在线检测快速模板匹配
|
摘要
为了解决目前电路板缺陷检测视觉识别系统对多目标识别检测过程实时性差、识别效率低,甚至出现无法识别等问题,首先对集成电路板图像按一定比例进行双三次插值算法处理,并在该基础上创建模板图像,然后进行归一化积相关匹配快速地识别检测出目标的大概位置,最后以双三次插值算法缩放比例值还原目标在原图像中的位置,从而精确识别检测出的目标。实验结果表明:该算法能达到100%识别,运算时间比传统算法快2～11倍,可见该技术提高了检测效率,具有很好的实用性。
In order to solve problems in the current circuit board defect detection visual recognition system,such as poor real-time performance in multi-target recognition detection,low recognition efficiency,and even cases of being unrecognizable,Firstly,the algorithm performed bicubic interpolation algorithm on the integrated circuit board image by a certain ratio,and created a template image on the basis of this,and then performed normalized product correlation matching to quickly identify the approximate position of the detected target,finally,the bicubic interpolation algorithm is used to scale the value to restore the position of the target in the original image,thereby accurately identifying the detected target. The experimental results show that the algorithm can achieve 100% recognition and the operation time is 11 times faster than the traditional algorithm. It can be seen that the technology improves the detection efficiency and has good practicability.
In order to solve problems in the current circuit board defect detection visual recognition system,such as poor real-time performance in multi-target recognition detection,low recognition efficiency,and even cases of being unrecognizable,Firstly,the algorithm performed bicubic interpolation algorithm on the integrated circuit board image by a certain ratio,and created a template image on the basis of this,and then performed normalized product correlation matching to quickly identify the approximate position of the detected target,finally,the bicubic interpolation algorithm is used to scale the value to restore the position of the target in the original image,thereby accurately identifying the detected target. The experimental results show that the algorithm can achieve 100% recognition and the operation time is 11 times faster than the traditional algorithm. It can be seen that the technology improves the detection efficiency and has good practicability.
引文
1胡艺,杨帆,潘国峰.一种改进的印刷电路板缺陷检测分割算法[J].科学技术与工程,2017,17(9):221-228Hu Yi,Yang Fan,Pan Goufeng. An improved segmentation algorithm for defect detection of PCB[J]. Science Technology and Engineering,2017,17(9):221-228
2 梁杰民,郭宝平,杨欧,等.干涉滤光片在PCB机器视觉检测中的应用研究[J].科学技术与工程,2007,7(6):1048-1051Liang Jiemin,Guo Baoping,Yang Ou,et al. Application of machine vision technique in the identification for welding spots of printed circuit board(PCB)[J]. Science Technology and Engineering,2007,7(6):1048-1051
3 张驰,廖华丽,周军.基于单目视觉的工业机器人智能抓取系统设计[J].机电工程,2018,35(3):283-287Zhang Chi,Liao Huali,Zhou Jun. Intelligent grasping system for industrial robot based on monocular vision[J]. Journal of Mechanical&Electrical Engineering,2018,35(3):283-287
4 白冰峰,温秀兰,张中辉.基于Haar小波和形状模板的图像快速匹配算法[J].组合机床与自动化加工技术,2017(2):37-40Bai Bingfeng,Wen Xiulan,Zhang Zhonghui. Fast image matching algorithm based on wavelet and shape template[J]. Modular Machine Tool&Automatic Manufacturing Technique,2017(2):37-40
5 丁筱玲,赵强,李贻斌,等.基于模板匹配的改进型目标识别算法[J].山东大学学报(工学版),2018,48(2):1-7Ding Xiaoling,Zhao Qiang,Li Yibin,et al. Modified target recognition algorithm based on template matching[J]. Journal of Shandong University(Eengineering Science),2018,48(2):1-7
6 秦海报.基于模板匹配的快速少纹理目标识别算法[J].现代计算机,2019,32(1):47-51Qin Haibao. Texture-less object detection algorithm with speed-up based on template matching[J]. Electronic Science and Technology,2019,32(1):47-51
7 段亮弟,宋平,陈众,等.基于均值漂移的快速模板匹配算法[J].西北工业大学学报,2018,36(4):792-799Duan Liangdi,Song Ping,Chen Zhong,et al. A target tracking algorithm based on mean shift and fast template matching[J]. Journal of Northwestern Polytechnical University,2018,36(4):792-799
8 黄超灵,谢光汉,孙涛.一种新型智能跳跃式模板匹配算法[J].科学技术与工程,2008,8(7):1821-1824Huang Chaoling, Xie Guanghan, Sun Tao. New intelligent skip method of template matching[J]. Science Technology and Engineering,2008,8(7):1821-1824
9 陈志坚,李哲,石磊.一种基于图像特征块匹配的电子稳像算法[J].科学技术与工程,2008,8(11):2512-2515Chen Zhijian,Li Zhe,Shi Lei. Electronic image stabilization method based on feature block matching[J]. Science Technology and Engineering,2008,8(11):2512-2515
10 陈洵凛,杨煜俊.基于小生境遗传算法的激光切割快速模板匹配[J].激光技术,2019,43(1):125-130Chen Xunlin,Yang Yujun. Fast template matching for laser cutting based on niche genetic algorithm[J]. Laser Technology,2019,43(1):125-130
11 焦圣喜,张利辉,江绛.图像检测技术在工件在线分选中的应用[J].机床与液压,2010,38(5):71-72Jiao Shengxi,Zhang Lihui,Jiang Jiang. The application of image inspection technique in workpiece online selection[J]. Machine Tool&Hydraulics,2010,38(5):71-72
12 唐琎,李青.一种快速的模板匹配算法[J].计算机应用,2010,30(6):1559-1561Tang Jin,Li Qing. Fast template matching algorithm[J]. Journal of Computer Applications,2010,30(6):1559-1561
13 刘锦峰.图像模板匹配快速算法研究[D].长沙:中南大学,2007Liu Jinfeng. Research on fast algorithm for image template matching[D]. Changsha:Central South University,2007
14 Stefano L D. Mattoccia S. Fast template matching using bounded partial correlation[J]. Machine Vison and Applications 2003,13(4):213-221
15 Gary Bradski,Adrian Kaehler.学习OpenCV[M].于仕琪,刘瑞祯,译.北京:清华大学出版社,2009:149-158Gary Bradski,Adrian Kaehler. Learning OpenCV[M]. Yu Shiqi,Liu Ruizhen,translated. Beijing:Tsinghua University Press,2009:149-158
16 何东健.数字图像处理[M].西安:西安电子科技大学出版社,2015:207-209He Dongjian. Digital image processing[M]:Xi'an:Xidian University Press,2015:207-209
17 毛星云,冷雪飞. OpenCV3编程入门[M].北京:电子工业出版社,2015:223-237Mao Xingyun,Leng Xuefei. Getting started with OpenCV3 programming[M]. Beijing:Publishing House of Electronics Industry,2015:223-237
18 高成敏,陈良,林永和.双三次卷积模板算法[J].计算机工程与应用,2009,45(17):151-154Gao Chengmin,Chen Liang,Lin Yonghe. Bicubic convolution template algorithm[J]. Computer Engineering and Applications,2009,45(17):151-154
19 章毓晋.计算机视觉教程[M]. 2版.北京:人民邮电出版社,2017:260-262Zhang Yujin. A course of computer vision[M]. 2nd ed. Beijing:Posts&Telecom Press,2017:260-262
20 郑咸义,姚仰新,雷秀仁,等.应用数值分析[M].广州:华南理工大学出版社,2008:74-82Zhen Xianyi,Yao Yangxin,Lei Xiuren,et al. Applied numerical analysis[M]. Guangzhou:South China University of Technology Press,2008:74-82
2 梁杰民,郭宝平,杨欧,等.干涉滤光片在PCB机器视觉检测中的应用研究[J].科学技术与工程,2007,7(6):1048-1051Liang Jiemin,Guo Baoping,Yang Ou,et al. Application of machine vision technique in the identification for welding spots of printed circuit board(PCB)[J]. Science Technology and Engineering,2007,7(6):1048-1051
3 张驰,廖华丽,周军.基于单目视觉的工业机器人智能抓取系统设计[J].机电工程,2018,35(3):283-287Zhang Chi,Liao Huali,Zhou Jun. Intelligent grasping system for industrial robot based on monocular vision[J]. Journal of Mechanical&Electrical Engineering,2018,35(3):283-287
4 白冰峰,温秀兰,张中辉.基于Haar小波和形状模板的图像快速匹配算法[J].组合机床与自动化加工技术,2017(2):37-40Bai Bingfeng,Wen Xiulan,Zhang Zhonghui. Fast image matching algorithm based on wavelet and shape template[J]. Modular Machine Tool&Automatic Manufacturing Technique,2017(2):37-40
5 丁筱玲,赵强,李贻斌,等.基于模板匹配的改进型目标识别算法[J].山东大学学报(工学版),2018,48(2):1-7Ding Xiaoling,Zhao Qiang,Li Yibin,et al. Modified target recognition algorithm based on template matching[J]. Journal of Shandong University(Eengineering Science),2018,48(2):1-7
6 秦海报.基于模板匹配的快速少纹理目标识别算法[J].现代计算机,2019,32(1):47-51Qin Haibao. Texture-less object detection algorithm with speed-up based on template matching[J]. Electronic Science and Technology,2019,32(1):47-51
7 段亮弟,宋平,陈众,等.基于均值漂移的快速模板匹配算法[J].西北工业大学学报,2018,36(4):792-799Duan Liangdi,Song Ping,Chen Zhong,et al. A target tracking algorithm based on mean shift and fast template matching[J]. Journal of Northwestern Polytechnical University,2018,36(4):792-799
8 黄超灵,谢光汉,孙涛.一种新型智能跳跃式模板匹配算法[J].科学技术与工程,2008,8(7):1821-1824Huang Chaoling, Xie Guanghan, Sun Tao. New intelligent skip method of template matching[J]. Science Technology and Engineering,2008,8(7):1821-1824
9 陈志坚,李哲,石磊.一种基于图像特征块匹配的电子稳像算法[J].科学技术与工程,2008,8(11):2512-2515Chen Zhijian,Li Zhe,Shi Lei. Electronic image stabilization method based on feature block matching[J]. Science Technology and Engineering,2008,8(11):2512-2515
10 陈洵凛,杨煜俊.基于小生境遗传算法的激光切割快速模板匹配[J].激光技术,2019,43(1):125-130Chen Xunlin,Yang Yujun. Fast template matching for laser cutting based on niche genetic algorithm[J]. Laser Technology,2019,43(1):125-130
11 焦圣喜,张利辉,江绛.图像检测技术在工件在线分选中的应用[J].机床与液压,2010,38(5):71-72Jiao Shengxi,Zhang Lihui,Jiang Jiang. The application of image inspection technique in workpiece online selection[J]. Machine Tool&Hydraulics,2010,38(5):71-72
12 唐琎,李青.一种快速的模板匹配算法[J].计算机应用,2010,30(6):1559-1561Tang Jin,Li Qing. Fast template matching algorithm[J]. Journal of Computer Applications,2010,30(6):1559-1561
13 刘锦峰.图像模板匹配快速算法研究[D].长沙:中南大学,2007Liu Jinfeng. Research on fast algorithm for image template matching[D]. Changsha:Central South University,2007
14 Stefano L D. Mattoccia S. Fast template matching using bounded partial correlation[J]. Machine Vison and Applications 2003,13(4):213-221
15 Gary Bradski,Adrian Kaehler.学习OpenCV[M].于仕琪,刘瑞祯,译.北京:清华大学出版社,2009:149-158Gary Bradski,Adrian Kaehler. Learning OpenCV[M]. Yu Shiqi,Liu Ruizhen,translated. Beijing:Tsinghua University Press,2009:149-158
16 何东健.数字图像处理[M].西安:西安电子科技大学出版社,2015:207-209He Dongjian. Digital image processing[M]:Xi'an:Xidian University Press,2015:207-209
17 毛星云,冷雪飞. OpenCV3编程入门[M].北京:电子工业出版社,2015:223-237Mao Xingyun,Leng Xuefei. Getting started with OpenCV3 programming[M]. Beijing:Publishing House of Electronics Industry,2015:223-237
18 高成敏,陈良,林永和.双三次卷积模板算法[J].计算机工程与应用,2009,45(17):151-154Gao Chengmin,Chen Liang,Lin Yonghe. Bicubic convolution template algorithm[J]. Computer Engineering and Applications,2009,45(17):151-154
19 章毓晋.计算机视觉教程[M]. 2版.北京:人民邮电出版社,2017:260-262Zhang Yujin. A course of computer vision[M]. 2nd ed. Beijing:Posts&Telecom Press,2017:260-262
20 郑咸义,姚仰新,雷秀仁,等.应用数值分析[M].广州:华南理工大学出版社,2008:74-82Zhen Xianyi,Yao Yangxin,Lei Xiuren,et al. Applied numerical analysis[M]. Guangzhou:South China University of Technology Press,2008:74-82