基于FPGA的高速轨距测量技术的研究
摘要
轨距是最基本的轨道参数之一。自有轨检车以来，轨距一直是检测的一项内容。如果轨距不能精确的检测，将会造成重大的铁路安全隐患。为了适应快速、重载的需要，确保列车安全、平稳、无间断的运行，实现建立“数字”化的铁路的目标，迫使轨检车的检测能力需要提高，工作性能需要优化。如何精确而高效地测量出钢轨轨距一直是国内外铁道部门的一个重要的研究课题。
本文首先简要介绍了钢轨轨距检测的开发背景、研究现状及FPGA的有关知识。然后针对轨距自动检测系统，论述了系统的组成及检测原理，对摄像机标定、坐标变换、图像细化都作了详细介绍。最后利用VHDL语言编程实现轨距检测中的坐标变换，并仿真。比较了查找表和乘法器两种实现方案。FPGA是视频图像实时处理技术中最佳的折中方案，能够满足轨距自动检测系统高速、实时的检测要求。
Gauge is one of the basic rail parameters and it has always been one of the necessary parameters to measure ever since the gauge measurement vehicle developed. If the gauge can not be measured precisely, it will lead to weighty rail hidden trouble. In order to adapt to the need of speediness and over loading, to guarantee the train run safely, smoothly and uninterruptedly, to realize the object of digital railway, we must improve the measurable capacity and optimize the performance .So, it is very important for the rail gauge to be measured precisely and high efficiently and it is still an important research of the train department in every country to be in face.
This paper first introduces the developing background and the research status quo of the automatic measurement system for railway gauge and also some knowledge about FPGA. Then it illustrates the structure and the measurement principle of the gauge automatic measurement system. And also it carefully introduces the camera calibration the coordinate transform and the image thinning .At last I use VHDL to realize the coordinate transform in the gauge automatic measurement system and achieve simulation. And then it compares the LUT method with the multiplier method. FPGA is the best eclectic project in video image real time processing technology and it can satisfy the requirements of high speed and real time in gauge automatic measurement system.
本文首先简要介绍了钢轨轨距检测的开发背景、研究现状及FPGA的有关知识。然后针对轨距自动检测系统，论述了系统的组成及检测原理，对摄像机标定、坐标变换、图像细化都作了详细介绍。最后利用VHDL语言编程实现轨距检测中的坐标变换，并仿真。比较了查找表和乘法器两种实现方案。FPGA是视频图像实时处理技术中最佳的折中方案，能够满足轨距自动检测系统高速、实时的检测要求。
Gauge is one of the basic rail parameters and it has always been one of the necessary parameters to measure ever since the gauge measurement vehicle developed. If the gauge can not be measured precisely, it will lead to weighty rail hidden trouble. In order to adapt to the need of speediness and over loading, to guarantee the train run safely, smoothly and uninterruptedly, to realize the object of digital railway, we must improve the measurable capacity and optimize the performance .So, it is very important for the rail gauge to be measured precisely and high efficiently and it is still an important research of the train department in every country to be in face.
This paper first introduces the developing background and the research status quo of the automatic measurement system for railway gauge and also some knowledge about FPGA. Then it illustrates the structure and the measurement principle of the gauge automatic measurement system. And also it carefully introduces the camera calibration the coordinate transform and the image thinning .At last I use VHDL to realize the coordinate transform in the gauge automatic measurement system and achieve simulation. And then it compares the LUT method with the multiplier method. FPGA is the best eclectic project in video image real time processing technology and it can satisfy the requirements of high speed and real time in gauge automatic measurement system.
引文
[1]褚振勇，翁木云．FPGA设计及应用[M]．西安：西安电子科技大学出版社，2002．
[2]谷口庆治．数字图像处理．基础篇[M]．北京：科学出版社，2002．
[3]谷口庆治．数字图像处理．应用篇[M]．北京：科学出版社，2002．
[4]阮秋琦．数字图像处理学[M]．北京：电子工业出版社．2001
[5]夏良正．数字图像处理[M]．南京：东南大学出版社，1999
[6]赵曙光，郭万有，扬颂华．可编程逻辑器件原理、开发与应用[M]．西安：西安电子科技大学出版社，2000．
[7]曾繁泰，侯亚宁，崔元明．可编程器件应用导论[M]．北京：清华大学出版社，2001
[8]容观澳．计算机图像处理[M]．北京：清华大学出版社，2000．
[9]侯伯亭，顾新．VHDL硬件描述语言与数字逻辑电路设计[M]．西安：西安电子科技大学出版社，1997．
[10]Kenneth．R．Castleman著，朱志刚，林学，石定机等译．数字图像处理[M]．北京：电子工业出版社，1998
[11]Stefan Sjoholm，Lennart Lindh著，边计年，薛宏熙译．用VHDL设计电子线路[M]．北京：清华大学出版社，2000
[12]赵雅兴．FPGA原理、设计与应用[M]．天津：天津大学出版社，1999．
[13]吴立德，计算机视觉．复旦大学出版社，1993
[14]李介谷．计算机视觉的理论和实践．上海交通大学出版社，1991
[15]贾云得．机器视觉．科学出版社，2000
[16]Electronic Rail Eye. RT&S, 48-50, September, 1986
[17]Kunio Takeshita, Kinai Takagi, Yoshie Narita. Method of Rail Profile Measurement with Light section Method. RTRI Report, Vol. 7, No. 11, 1993
[18] S. Soeleiman. The Measurement of Rail Head Wear. Rail Track, 1990
[19] Ir. Coenraad Esveld, MINPROF-Wheel and Rail Profile Measuring System. Rail Engineering International Edition, No. 4, 1992
[20] G.W. Small, Z.S. Hegedus. Measurement of rail cross-section with solid-state sensor arrays. Optics and Laser Technology, February, 1986: 43-47
[21] RAC rapidly records rail condition. RT&S, April, 1987
[22] Tsai. RY, A Versatile Camera Calibration Technique for High-Accuracy 3D Machine Vision Metrology Using Off-the-Shelf TV Cameras and Lenses. IEEE Journal of Robotics and Automation, Vol. RA-3, No. 4, August 1987, pp:323-344.
[23] E. Guillou, D. Meneveaux, E. Maisel, K. bouatouch. Using vanishing points for camera calibration and coarse 3Dreconstruction from a single image. Visual Computer. 2000(16):396-410
[24] Peter M. Will, Keith S. Pennington. Grid Coding: Preprocessing Technique for Robot and Machine Vision. Artificial Intelligence, 2(3):319-329,1971
[25] Peter M. will, Keith S. Penningtion, Grid Coding: A Novel Technique For Image Processing. Proceedings of the IEEE. Vol. 60, No. 6:669-680, June 1972
[26] 罗飞路，傅恩锡．CCD摄像机内外部参数快速准确的标定方法．湖南大学学报，Vol．24，No．2 April 1997
[27] 杨长江，汪威，胡占义．一种基于主动视觉的摄像机内参数自标定方法．计算机学报，第21卷，第5期，1998年5月：428-435
[28] 雷成，吴福朝，胡占义．一种新的基于主动视觉系统的摄像机自标定方法．计算机学报，第23卷，第11期，2000年11月：1130-1139
[29] 张昌凡，龙永红，彭涛。可编程逻辑器件及VHDL设计技术[M]。华南理工大
学出版社，2001
[30]Shinichi Hirai, Masakazu Zakouji, Tatsuhiko Tsuboi. Implementing Image Processing Algorithms on FPGA-based Real Time Vision System Department of Robotics, Ritsumeikan University Kusatsu, Shiga 525-8577 Japan
[31]Anthony Edward Nelson. Implementation of Image Processing Algorithms on FPGA Hardware。Thesis submitted to theFaculty of the Graduate School of Vanderbit University. May, 2000
[32]Steffen Klupsch, Sorin A. Huss, M. Rumpfo Real Time Image Processing based on Reconfigurable Hardware Acceleration。Workshop Heterogeneous Reconfigur(?)ble System on Chip(SoC),April, 2002
[33]来晓岚，赵佳明，卢焕章。DSP+FPGA实时信号处理系统。电子设计技术
[34]www.fpga.com.cn
[2]谷口庆治．数字图像处理．基础篇[M]．北京：科学出版社，2002．
[3]谷口庆治．数字图像处理．应用篇[M]．北京：科学出版社，2002．
[4]阮秋琦．数字图像处理学[M]．北京：电子工业出版社．2001
[5]夏良正．数字图像处理[M]．南京：东南大学出版社，1999
[6]赵曙光，郭万有，扬颂华．可编程逻辑器件原理、开发与应用[M]．西安：西安电子科技大学出版社，2000．
[7]曾繁泰，侯亚宁，崔元明．可编程器件应用导论[M]．北京：清华大学出版社，2001
[8]容观澳．计算机图像处理[M]．北京：清华大学出版社，2000．
[9]侯伯亭，顾新．VHDL硬件描述语言与数字逻辑电路设计[M]．西安：西安电子科技大学出版社，1997．
[10]Kenneth．R．Castleman著，朱志刚，林学，石定机等译．数字图像处理[M]．北京：电子工业出版社，1998
[11]Stefan Sjoholm，Lennart Lindh著，边计年，薛宏熙译．用VHDL设计电子线路[M]．北京：清华大学出版社，2000
[12]赵雅兴．FPGA原理、设计与应用[M]．天津：天津大学出版社，1999．
[13]吴立德，计算机视觉．复旦大学出版社，1993
[14]李介谷．计算机视觉的理论和实践．上海交通大学出版社，1991
[15]贾云得．机器视觉．科学出版社，2000
[16]Electronic Rail Eye. RT&S, 48-50, September, 1986
[17]Kunio Takeshita, Kinai Takagi, Yoshie Narita. Method of Rail Profile Measurement with Light section Method. RTRI Report, Vol. 7, No. 11, 1993
[18] S. Soeleiman. The Measurement of Rail Head Wear. Rail Track, 1990
[19] Ir. Coenraad Esveld, MINPROF-Wheel and Rail Profile Measuring System. Rail Engineering International Edition, No. 4, 1992
[20] G.W. Small, Z.S. Hegedus. Measurement of rail cross-section with solid-state sensor arrays. Optics and Laser Technology, February, 1986: 43-47
[21] RAC rapidly records rail condition. RT&S, April, 1987
[22] Tsai. RY, A Versatile Camera Calibration Technique for High-Accuracy 3D Machine Vision Metrology Using Off-the-Shelf TV Cameras and Lenses. IEEE Journal of Robotics and Automation, Vol. RA-3, No. 4, August 1987, pp:323-344.
[23] E. Guillou, D. Meneveaux, E. Maisel, K. bouatouch. Using vanishing points for camera calibration and coarse 3Dreconstruction from a single image. Visual Computer. 2000(16):396-410
[24] Peter M. Will, Keith S. Pennington. Grid Coding: Preprocessing Technique for Robot and Machine Vision. Artificial Intelligence, 2(3):319-329,1971
[25] Peter M. will, Keith S. Penningtion, Grid Coding: A Novel Technique For Image Processing. Proceedings of the IEEE. Vol. 60, No. 6:669-680, June 1972
[26] 罗飞路，傅恩锡．CCD摄像机内外部参数快速准确的标定方法．湖南大学学报，Vol．24，No．2 April 1997
[27] 杨长江，汪威，胡占义．一种基于主动视觉的摄像机内参数自标定方法．计算机学报，第21卷，第5期，1998年5月：428-435
[28] 雷成，吴福朝，胡占义．一种新的基于主动视觉系统的摄像机自标定方法．计算机学报，第23卷，第11期，2000年11月：1130-1139
[29] 张昌凡，龙永红，彭涛。可编程逻辑器件及VHDL设计技术[M]。华南理工大
学出版社，2001
[30]Shinichi Hirai, Masakazu Zakouji, Tatsuhiko Tsuboi. Implementing Image Processing Algorithms on FPGA-based Real Time Vision System Department of Robotics, Ritsumeikan University Kusatsu, Shiga 525-8577 Japan
[31]Anthony Edward Nelson. Implementation of Image Processing Algorithms on FPGA Hardware。Thesis submitted to theFaculty of the Graduate School of Vanderbit University. May, 2000
[32]Steffen Klupsch, Sorin A. Huss, M. Rumpfo Real Time Image Processing based on Reconfigurable Hardware Acceleration。Workshop Heterogeneous Reconfigur(?)ble System on Chip(SoC),April, 2002
[33]来晓岚，赵佳明，卢焕章。DSP+FPGA实时信号处理系统。电子设计技术
[34]www.fpga.com.cn