基于计算机视觉测量的单针绗缝加工控制系统的研究
|
摘要
绗缝是利用缝线在物料上组合成无规则排列和独立的或者是有规则排列重复的各种图形的轨迹，这些图案不只是为了牢固地缝住三层物料，防止填充料移动，还可以增强缝制品的立体感，提高其装饰功能。因为物料的弹性和高速的绗缝加工，因此加工过程中物料有较大变形，线迹与预先设定的加工轨迹往往难以吻合。这种物料变形的数学模型难以建立，一般根据经验，采用某些补偿的方法，减少加工件与设计图案的误差，但加工复杂图案时却难以满足要求。
本文提出应用计算机视觉测量技术，对绗缝加工轨迹进行实时图像采集和处理、跟踪，实现对加工轨迹的动态修正，使加工图案满足设计要求。本文展开研究并取得一定成效：构建了基于PCI总线的微机实时图像采集系统；在采集的布料总图(鸟瞰图)的基础上，通过数字图像的数字滤波、图像增强、边缘检测等处理，提取布料图像的边缘，对轮廓的矢量化的象素点进行搜索，得到相应的图案矢量图，从而确定绗缝的加工轨迹，生成加工指令；在进给加工过程中，主计算机对动态局部图像与总图(鸟瞰图)的对应部分进行图像相关的匹配计算，应用数字图像理论，结合神经网络计算的改进最速下降法和模拟退火算法，提出独特而有效的相关迭代初始值赋值方法，形成稳定、高速和准确的相关运算，实现单针绗缝视觉测量和自动控制。
系统采用目前流行的控制系统发展的主要方向之一，将PC加专门数字控制器的开放式结构，作为单针绗缝控制系统，大大缩短开发周期，不仅可以利用PC技术的流行软件平台，开发出功能完善的人机界面，而且有效实现了图像采集、数字处理等复杂运算和实时控制，实用效果好。
在单针绗缝控制系统中把计算机视觉引入位置控制回路，实现对加工轨迹的实时测控，在国内未见有相关成果报道。这种方法推广到一般的柔性工件加工控制领域，将具有广阔的应用前景。
Quilting is the trace of suture which is woven on materials. By the means of ruleless arrangement, unique arrangement or inerratic reiteration, suture can form various patterns, which trace primarily intends to oversew three layers of materials hard, and prevent movements of fillings such as spray-glue cotton, wool, plume and so on. On the other hand, the trace can impress the sense of three-dimension and make it panoplied. Materials in quilting process are easy to distort, while the trace doesn't answer to the prearranged process track, and the distortion maths model is hard to establish, thus this article brings forward a view that we can use camera in real-time track of process trace.
In order to achieve the track of process trace, this article develops research in four aspects: establishing a graph collection system based on PCI bus. Quilting process track is from the edge of cloth figure(airscape) which is chosen as vector by users. In this article, we use usual digital image means to get the edge image of cloth, such as impressed image, digital filtering, edge detection and so on. In the vectorization of figure, we adopt strategies like this: regarding beeline the first and arc the second, so as to search for image element points. In the process after confirming the process track, the main controlling computer needs to run the image correlation procedure of match operating. The procedure functions in the compare between partial image of dynamic collection and corresponding image of the airscape. In chapter 5, basing on the analysis of correlative theory of digital image, we introduce the improved fasted-down algorithm and simulative anneal algorithm, which applies to NN calculation, an
d bring forward the unique and effective means, correlative original value evaluation. Basing on the combination of correlative arithmetic, a stable, high-speed and exact correlative arithmetic is formed, which makes it possible
to apply computer vision detection of single-needle quilting in industrial production. In this article, we adopt controlling system of PC+NC open-architecture to control the single-needle quilting processing. This structure can greatly shorten our development cycle, because our team had succeeded in applying some NC controller with similar structure in the control of multi-needle quilting process. Meanwhile, we can use PC tech and popular soft-ware tech to develop friendly human-machine interface with great function.
Adopted in computer vision detection, the single-needle quilting controlling system possesses brilliant future because of three points: on the one hand, it conducts real-time supervision on processing track, which equals to importing computer vision to location controlling loop, on the other hand, it is easy to generalize to processing and controlling field of common flexible workpiece.
本文提出应用计算机视觉测量技术，对绗缝加工轨迹进行实时图像采集和处理、跟踪，实现对加工轨迹的动态修正，使加工图案满足设计要求。本文展开研究并取得一定成效：构建了基于PCI总线的微机实时图像采集系统；在采集的布料总图(鸟瞰图)的基础上，通过数字图像的数字滤波、图像增强、边缘检测等处理，提取布料图像的边缘，对轮廓的矢量化的象素点进行搜索，得到相应的图案矢量图，从而确定绗缝的加工轨迹，生成加工指令；在进给加工过程中，主计算机对动态局部图像与总图(鸟瞰图)的对应部分进行图像相关的匹配计算，应用数字图像理论，结合神经网络计算的改进最速下降法和模拟退火算法，提出独特而有效的相关迭代初始值赋值方法，形成稳定、高速和准确的相关运算，实现单针绗缝视觉测量和自动控制。
系统采用目前流行的控制系统发展的主要方向之一，将PC加专门数字控制器的开放式结构，作为单针绗缝控制系统，大大缩短开发周期，不仅可以利用PC技术的流行软件平台，开发出功能完善的人机界面，而且有效实现了图像采集、数字处理等复杂运算和实时控制，实用效果好。
在单针绗缝控制系统中把计算机视觉引入位置控制回路，实现对加工轨迹的实时测控，在国内未见有相关成果报道。这种方法推广到一般的柔性工件加工控制领域，将具有广阔的应用前景。
Quilting is the trace of suture which is woven on materials. By the means of ruleless arrangement, unique arrangement or inerratic reiteration, suture can form various patterns, which trace primarily intends to oversew three layers of materials hard, and prevent movements of fillings such as spray-glue cotton, wool, plume and so on. On the other hand, the trace can impress the sense of three-dimension and make it panoplied. Materials in quilting process are easy to distort, while the trace doesn't answer to the prearranged process track, and the distortion maths model is hard to establish, thus this article brings forward a view that we can use camera in real-time track of process trace.
In order to achieve the track of process trace, this article develops research in four aspects: establishing a graph collection system based on PCI bus. Quilting process track is from the edge of cloth figure(airscape) which is chosen as vector by users. In this article, we use usual digital image means to get the edge image of cloth, such as impressed image, digital filtering, edge detection and so on. In the vectorization of figure, we adopt strategies like this: regarding beeline the first and arc the second, so as to search for image element points. In the process after confirming the process track, the main controlling computer needs to run the image correlation procedure of match operating. The procedure functions in the compare between partial image of dynamic collection and corresponding image of the airscape. In chapter 5, basing on the analysis of correlative theory of digital image, we introduce the improved fasted-down algorithm and simulative anneal algorithm, which applies to NN calculation, an
d bring forward the unique and effective means, correlative original value evaluation. Basing on the combination of correlative arithmetic, a stable, high-speed and exact correlative arithmetic is formed, which makes it possible
to apply computer vision detection of single-needle quilting in industrial production. In this article, we adopt controlling system of PC+NC open-architecture to control the single-needle quilting processing. This structure can greatly shorten our development cycle, because our team had succeeded in applying some NC controller with similar structure in the control of multi-needle quilting process. Meanwhile, we can use PC tech and popular soft-ware tech to develop friendly human-machine interface with great function.
Adopted in computer vision detection, the single-needle quilting controlling system possesses brilliant future because of three points: on the one hand, it conducts real-time supervision on processing track, which equals to importing computer vision to location controlling loop, on the other hand, it is easy to generalize to processing and controlling field of common flexible workpiece.
引文
[1] 郑南宁，计算机视觉与模式识别，北京：国防工业出版社，1998，1～80
[2] 徐太阳，张利锋，志成CS64型多针电脑绗缝机的功能与设计特点，中外缝制设备，1998，第5期：14-15
[3] 张书美，智能控制绗缝花型不变形方法的研究，青岛大学学报，2002，Vol．17，No．2：71-73
[4] 衣法臻，何平，机器人视觉动态目标跟踪的最优控制算法研究，哈尔滨工业大学学报，1999，Vol．31，No．3：122-125
[5] 沈胜宏，韩月秋，基于多DSP和FPGA的高速并行互相关图像跟踪技术，北京理工大学学报，2000，Vol．20，No．1：98-101
[6] 周锐，崔祜涛，多模自适应目标图像跟踪，宇航学报，1998，Vol．19， No．2：81-85
[7] 马颂德，张正友．计算机视觉-计算理论和算法基础，北京：科学出版社，1998．
[8] 刘成君，戴汝为．计算机视觉研究的进展．模式识别与人工智能，1995，12：59-70
[9] 龙甫荟，郑南宁．计算机视觉模型的研究与进展．信息与控制 1997，4(2)：112-116
[10] 崔屹，数字图象处理技术与应用，北京：电子工业出版社．1997
[11] 高文，陈熙霖．计算机视觉—算法与系统原理．北京：清华大学出版社，1999
[12] Tarr MJ, Black MJ, Dialogue, a computational and evolutionary perspective on the role of representation in vision. CVGIP: Image Understanding, 1994, 60(2):65-73
[13] Q.T.Luong & Faugeras, The Fundamental matrix: theory, algorithms and stability analysis, Int.j. of computer vision, 17(1): 43-75, 1996
[14] 张广军，李鑫，魏振忠．基于RBF神经网络的结构光三维视觉检测方法研究．计量学报．2002，23(4)．251-255
[15] Cesmeli E., and Wang.D., "Texture Segmentation Using Gaussian-Markov Random Fields and Neural Oscillator Networks", IEEE Transactions on Neural Networks, March 2001,Vol.12, No.2, pp.394-405
[16] 吴福朝，于洪川，韦穗．摄像机内参数自标定理论与实践．自动化学报，1999，11(6)：769 775
[17] R.C. Lue, M. H. Lin. Hierachical Robot multi-sensor data fusion system. NATO ASI series, vol. F58, 1990: 67-86
[18] 邹定海，叶声华，王春和，郭育．用于在线测量的视觉检测系统．仪器仪表学报，1995，Vol．16，NO．4：337-341
[19] 段发阶，刘风梅，叶声华．一种新型线结构光传感器结构参数标定方法．仪器仪表学报，2000，2(1)：108-110
[20] 唐露新、关建和、吴黎明等，多枪自动端焊的控制与实现，变压器2003年第12期
[21] 王佳，刘永东，梁晋文．工业动态跟踪测量的原则与技术特点．计量学报，2000，1(1)：34-39
[22] 吴福朝，于洪川，韦穗．与视点对应的视图插补．自动化学报，1999，11
[23] Igor Lackovic, Vedran Bilas etc., Measurement of gait parameters from free moving subjects Measurement. 2000, 27: 121-131
[24] 刘成君，戴汝为．计算机视觉研究的进展．模式识别与人工智能，1995，12：59-70
[25] 吴黎明，周曲，邓耀华，朱妙贤．基于小波分析的维氏硬度自动测试，上海：理化检验-物理分册．2004，Vol．40，No．1
[26] 吴黎明，周曲，邓耀华，朱妙贤等．维氏硬度试验压痕图的小波分析与自动计算硬度．中国机械工程．2004，Vol．5，No．6
[27] Comer.M.L., and Delp.E.J., "The EM/MPM Algorithm for
Segmentation of Textured Images: Analysis and Further Experimental Results", IEEE Transactions on Image processing. October 2000, Vol.9, No.10, pp.1731-1744.
[28] 史荣昌，矩阵分析，北京理工大学出版社，1995，104-126
[29] Canny J, "A computational approach to edge detection", T-PAMI. 1998, 8(6):679-698
[30] Kruizinga P., and Petkov. N., "Nonlinear Operator for Oriented Texture", IEEE Transactions on Image Processing, October 1999, Vol. 8, No.10. pp. 1395-1408.
[31] Zhang Y.J, "A Survey on evaluation methods for image segmentation", Pattern Recognition, 1996,29(8): 1335-1346
[32] 章毓晋．图像理解与计算机视觉[D]．北京，清华大学出版社，2000，8
[33] 清汉计算机工作室，VisuaI C++6.0多媒体开发实例，北京：机械工业出版社，2000
[34] 蒋平，沈世锦等．基于图像误差机器人运动目标自适应跟踪．机器人．1997，6
[35] 黄石生，毛鹏军，薛家祥，基于焊缝视觉跟踪的图像阈值化研究，电焊机，2001，Vol．31：20-22
[36] 周曲，吴黎明，唐露新，王桂棠．基于自动调焦的微机控制显微硬度计．广东自动化与信息工程．2001，Vol．22，No．4
[37] 祝世平，强锡富．基于计算机视觉的大型工件特征点三维坐标测最方法研究．仪器仪表学报，1997，12(6)：607-611
[38] 刘家豪，刘盛，傅建中．一种实用的计算机视觉三维微检测系统的研究．机床与液压．2003(3)：270-272，275
[39] Sahoo P, Wuukins, Yenger J, "Threshold Selection Using Renyi's Entropy", Pattern Recognition, 1997, 30(1):394-398
[40] 高立志，林志行，一种摄像机成像误差的模型化修正方法．机器人，1999，7(4)：266--271．
[41] H.A.Bruck, S.R. McNeill, et al., Digital Image Correlation Using Newton-Raphson Method of Partial Differential Correction, Exp. Mech, 1989, 29(3): 261-267
[42] 计宏伟，数字相关图像测量方法与应用的研究，天津大学博士论文，1997
[43] 王可定，计算机模拟及其应用，东南大学出版社，1997．
[44] B．T．科切特科夫，随动系统的统计分析和最优化，国防工业出版社，1984．
[45] 吴大方，刘文挺，陶涛，CCD图像处理技术应用于耐久性分析中疲劳断口的数据测量，现代力学测试技术，1998，352-355
[46] B.W.Smith, Error Assessment far Strain Mapping by Digital Image Correlation, Experimental Techniques, 1998, 19-21
[47] 陈卫福，杨建武，开放式数控系统及SERCOS接口应用技术，机械工业出版社，2003
[48] 薛均义，武自芳，微机控制系统及其应用，西安交通大学出版社，2003．2，374-380
[49] 王福瑞，单片机测控系统设计大全，北京航空航天大学出版社，1999
[50] Yusuf Altintas, Manufacturing Automation—Metal Cutting Mechanics, Machine Tool Vibrations, and CNC Design, 化学工业出版社, 2002
[51] 李秉操，张登举，付寿英等，单片机接口技术及其在工业控制中的应用，陕西电子编辑部 1991
[2] 徐太阳，张利锋，志成CS64型多针电脑绗缝机的功能与设计特点，中外缝制设备，1998，第5期：14-15
[3] 张书美，智能控制绗缝花型不变形方法的研究，青岛大学学报，2002，Vol．17，No．2：71-73
[4] 衣法臻，何平，机器人视觉动态目标跟踪的最优控制算法研究，哈尔滨工业大学学报，1999，Vol．31，No．3：122-125
[5] 沈胜宏，韩月秋，基于多DSP和FPGA的高速并行互相关图像跟踪技术，北京理工大学学报，2000，Vol．20，No．1：98-101
[6] 周锐，崔祜涛，多模自适应目标图像跟踪，宇航学报，1998，Vol．19， No．2：81-85
[7] 马颂德，张正友．计算机视觉-计算理论和算法基础，北京：科学出版社，1998．
[8] 刘成君，戴汝为．计算机视觉研究的进展．模式识别与人工智能，1995，12：59-70
[9] 龙甫荟，郑南宁．计算机视觉模型的研究与进展．信息与控制 1997，4(2)：112-116
[10] 崔屹，数字图象处理技术与应用，北京：电子工业出版社．1997
[11] 高文，陈熙霖．计算机视觉—算法与系统原理．北京：清华大学出版社，1999
[12] Tarr MJ, Black MJ, Dialogue, a computational and evolutionary perspective on the role of representation in vision. CVGIP: Image Understanding, 1994, 60(2):65-73
[13] Q.T.Luong & Faugeras, The Fundamental matrix: theory, algorithms and stability analysis, Int.j. of computer vision, 17(1): 43-75, 1996
[14] 张广军，李鑫，魏振忠．基于RBF神经网络的结构光三维视觉检测方法研究．计量学报．2002，23(4)．251-255
[15] Cesmeli E., and Wang.D., "Texture Segmentation Using Gaussian-Markov Random Fields and Neural Oscillator Networks", IEEE Transactions on Neural Networks, March 2001,Vol.12, No.2, pp.394-405
[16] 吴福朝，于洪川，韦穗．摄像机内参数自标定理论与实践．自动化学报，1999，11(6)：769 775
[17] R.C. Lue, M. H. Lin. Hierachical Robot multi-sensor data fusion system. NATO ASI series, vol. F58, 1990: 67-86
[18] 邹定海，叶声华，王春和，郭育．用于在线测量的视觉检测系统．仪器仪表学报，1995，Vol．16，NO．4：337-341
[19] 段发阶，刘风梅，叶声华．一种新型线结构光传感器结构参数标定方法．仪器仪表学报，2000，2(1)：108-110
[20] 唐露新、关建和、吴黎明等，多枪自动端焊的控制与实现，变压器2003年第12期
[21] 王佳，刘永东，梁晋文．工业动态跟踪测量的原则与技术特点．计量学报，2000，1(1)：34-39
[22] 吴福朝，于洪川，韦穗．与视点对应的视图插补．自动化学报，1999，11
[23] Igor Lackovic, Vedran Bilas etc., Measurement of gait parameters from free moving subjects Measurement. 2000, 27: 121-131
[24] 刘成君，戴汝为．计算机视觉研究的进展．模式识别与人工智能，1995，12：59-70
[25] 吴黎明，周曲，邓耀华，朱妙贤．基于小波分析的维氏硬度自动测试，上海：理化检验-物理分册．2004，Vol．40，No．1
[26] 吴黎明，周曲，邓耀华，朱妙贤等．维氏硬度试验压痕图的小波分析与自动计算硬度．中国机械工程．2004，Vol．5，No．6
[27] Comer.M.L., and Delp.E.J., "The EM/MPM Algorithm for
Segmentation of Textured Images: Analysis and Further Experimental Results", IEEE Transactions on Image processing. October 2000, Vol.9, No.10, pp.1731-1744.
[28] 史荣昌，矩阵分析，北京理工大学出版社，1995，104-126
[29] Canny J, "A computational approach to edge detection", T-PAMI. 1998, 8(6):679-698
[30] Kruizinga P., and Petkov. N., "Nonlinear Operator for Oriented Texture", IEEE Transactions on Image Processing, October 1999, Vol. 8, No.10. pp. 1395-1408.
[31] Zhang Y.J, "A Survey on evaluation methods for image segmentation", Pattern Recognition, 1996,29(8): 1335-1346
[32] 章毓晋．图像理解与计算机视觉[D]．北京，清华大学出版社，2000，8
[33] 清汉计算机工作室，VisuaI C++6.0多媒体开发实例，北京：机械工业出版社，2000
[34] 蒋平，沈世锦等．基于图像误差机器人运动目标自适应跟踪．机器人．1997，6
[35] 黄石生，毛鹏军，薛家祥，基于焊缝视觉跟踪的图像阈值化研究，电焊机，2001，Vol．31：20-22
[36] 周曲，吴黎明，唐露新，王桂棠．基于自动调焦的微机控制显微硬度计．广东自动化与信息工程．2001，Vol．22，No．4
[37] 祝世平，强锡富．基于计算机视觉的大型工件特征点三维坐标测最方法研究．仪器仪表学报，1997，12(6)：607-611
[38] 刘家豪，刘盛，傅建中．一种实用的计算机视觉三维微检测系统的研究．机床与液压．2003(3)：270-272，275
[39] Sahoo P, Wuukins, Yenger J, "Threshold Selection Using Renyi's Entropy", Pattern Recognition, 1997, 30(1):394-398
[40] 高立志，林志行，一种摄像机成像误差的模型化修正方法．机器人，1999，7(4)：266--271．
[41] H.A.Bruck, S.R. McNeill, et al., Digital Image Correlation Using Newton-Raphson Method of Partial Differential Correction, Exp. Mech, 1989, 29(3): 261-267
[42] 计宏伟，数字相关图像测量方法与应用的研究，天津大学博士论文，1997
[43] 王可定，计算机模拟及其应用，东南大学出版社，1997．
[44] B．T．科切特科夫，随动系统的统计分析和最优化，国防工业出版社，1984．
[45] 吴大方，刘文挺，陶涛，CCD图像处理技术应用于耐久性分析中疲劳断口的数据测量，现代力学测试技术，1998，352-355
[46] B.W.Smith, Error Assessment far Strain Mapping by Digital Image Correlation, Experimental Techniques, 1998, 19-21
[47] 陈卫福，杨建武，开放式数控系统及SERCOS接口应用技术，机械工业出版社，2003
[48] 薛均义，武自芳，微机控制系统及其应用，西安交通大学出版社，2003．2，374-380
[49] 王福瑞，单片机测控系统设计大全，北京航空航天大学出版社，1999
[50] Yusuf Altintas, Manufacturing Automation—Metal Cutting Mechanics, Machine Tool Vibrations, and CNC Design, 化学工业出版社, 2002
[51] 李秉操，张登举，付寿英等，单片机接口技术及其在工业控制中的应用，陕西电子编辑部 1991