PCBs故障红外诊断系统中基于遗传算法的图像处理研究
|
摘要
随着红外热成像技术的发展，基于红外成像检测温度场和图像分析相结合的红外诊断技术，在电力、电子、石油化工、医学等领域，得到了广泛应用。目前，印制电路板（PCBs）故障的诊断已经成为一个具有重要意义的研究方向,传统的PCBs自动测试技术受到了前所未有的挑战。与常规方法相比较，红外诊断具有非接触性的特点，能快速诊断、隔离PCBs的热模式故障，大大节约了费用开支，减少了工作量。论文根据项目组课题“PCBs故障红外故障诊断系统研究”，针对系统中图像处理部分的图像特征提取，图像配准等关键性问题进行了深入研究。
论文在红外辐射理论的基础上引出运用红外热成像技术诊断PCBs故障的基本原理；研究了PCBs故障的热模式，分析了红外热成像技术对PCBs热模式故障的诊断能力;研究了数字图像处理技术在红外诊断中的应用，并对PCBs红外图像运用Roberts、Sobel、Prewitt、Laplacian、Canny等方法进行了边缘检测;分析、建立了图像配准的数学模型，将图像配准问题转化为基于遗传算法的最优配准参数的寻优问题，并运用遗传算法理论与技术对红外图像的配准问题进行了遗传算法设计。论文中解决问题的方法，具有一定的新意。
在遗传算法理论与技术基础上，深入研究了图像配准问题的编码、译码、适应度评价以及选择、交叉、变异算子的设计等。算法设计上体现了一定的新意，并对算法的优劣及其针对图像配准问题的可行性和有效性进行了分析与验证。最后论文设计了PCBs故障红外诊断系统中基于遗传算法的图像处理仿真系统，包括硬件组成、软件设计等。
仿真实验运用边缘检测方法提取了图像的主要特征，然后在图像边缘特征的基础上，对图像运用遗传算法进行了配准，获得了较为理想的配准参数，并根据配准参数，对图像进行了减影运算，图像减影结果准确定位了故障部位，仿真实现了对PCBs故障的诊断，验证了遗传算法用于图像配准问题的可行性和有效性。
论文对PCBs故障红外诊断系统中图像的处理进行了研究，获得了较为理想的仿真结果，并对算法设计提出了改进措施与方法，其中的研究方法具有重要
意义和一定的参考价值，为进一步的研究工作奠定了基础
With the development of Infrared Imaging Technology(IIT), the infrared diagnosing technology, based on infrared inspection of thermo-field and image analyzing, is applied more and more in field of electric power, electronics, petrochemistry, medicine, etc.. At present, the diagnosis of PCBs fault is one of diagnosis field with exploratory significance. Conventional automatic testing method come up against unprecedented challenge, compared with Automatic Test System(ATS), the advantages of infrared diagnosing technology are remote inspecting, capable of quickly diagnosing and isolating thermal fault, and laborsaving. In this thesis, according to project: the study of infrared diagnosing system of PCBs fault, we studied some key technologies of image processing deeply, such as edge inspection and image registration.
Based on fundamental Infrared Radiation(IR) theory, the infrared diagnosing technology is derived of. In this thesis, the thermal pattern of PCBs and the diagnosing ability of IIT are analyzed. The digital image processing technology and its application in infrared diagnosing of PCBs fault is studied, and the edge inspection of image is given by Roberts, Sobel, Prewitt, Laplacian, Canny methods. The model of image registration is analyzed and established, an infrared image registration method for diagnosing of PCBs fault based on genetic algorithm (GA) is proposed, and the algorithm is designed. Some new ideas are embodied in this thesis.
Based on GA theory and technology, for the image registration, the encoding, decoding, fitness evaluation, and the design of selection operator, crossover operator, mutation operator are studied deeply, some new ideas are embodied in the algorithm, the advantages, inferior position, feasibility and efficiency of the algorithm are analyzed particularly. At last, the simulating experimental system of image processing is built, include hardware, software, etc..
During the simulating experiments, first, the edge of image is inspected; next, the standard image and the test image are registrated by the algorithm procedure
based on GA, at last, the images subtraction is used to locate the fault elements, and some preferable results are obtained. The diagnosis of PCBs fault is realized by simulating system, and the feasibility and efficiency of the algorithm are validated.
The investigation on image processing of the infrared diagnosing system of PCBs fault is implemented theoretically and simulating experimentally, and some preferable results are obtained. The theoretical limits and the difficulties of algorithm are discussed, and the measures to improve the efficiency of algorithm are suggested. The works here are essential for further research on the study of infrared diagnosing system of PCBs fault
论文在红外辐射理论的基础上引出运用红外热成像技术诊断PCBs故障的基本原理；研究了PCBs故障的热模式，分析了红外热成像技术对PCBs热模式故障的诊断能力;研究了数字图像处理技术在红外诊断中的应用，并对PCBs红外图像运用Roberts、Sobel、Prewitt、Laplacian、Canny等方法进行了边缘检测;分析、建立了图像配准的数学模型，将图像配准问题转化为基于遗传算法的最优配准参数的寻优问题，并运用遗传算法理论与技术对红外图像的配准问题进行了遗传算法设计。论文中解决问题的方法，具有一定的新意。
在遗传算法理论与技术基础上，深入研究了图像配准问题的编码、译码、适应度评价以及选择、交叉、变异算子的设计等。算法设计上体现了一定的新意，并对算法的优劣及其针对图像配准问题的可行性和有效性进行了分析与验证。最后论文设计了PCBs故障红外诊断系统中基于遗传算法的图像处理仿真系统，包括硬件组成、软件设计等。
仿真实验运用边缘检测方法提取了图像的主要特征，然后在图像边缘特征的基础上，对图像运用遗传算法进行了配准，获得了较为理想的配准参数，并根据配准参数，对图像进行了减影运算，图像减影结果准确定位了故障部位，仿真实现了对PCBs故障的诊断，验证了遗传算法用于图像配准问题的可行性和有效性。
论文对PCBs故障红外诊断系统中图像的处理进行了研究，获得了较为理想的仿真结果，并对算法设计提出了改进措施与方法，其中的研究方法具有重要
意义和一定的参考价值，为进一步的研究工作奠定了基础
With the development of Infrared Imaging Technology(IIT), the infrared diagnosing technology, based on infrared inspection of thermo-field and image analyzing, is applied more and more in field of electric power, electronics, petrochemistry, medicine, etc.. At present, the diagnosis of PCBs fault is one of diagnosis field with exploratory significance. Conventional automatic testing method come up against unprecedented challenge, compared with Automatic Test System(ATS), the advantages of infrared diagnosing technology are remote inspecting, capable of quickly diagnosing and isolating thermal fault, and laborsaving. In this thesis, according to project: the study of infrared diagnosing system of PCBs fault, we studied some key technologies of image processing deeply, such as edge inspection and image registration.
Based on fundamental Infrared Radiation(IR) theory, the infrared diagnosing technology is derived of. In this thesis, the thermal pattern of PCBs and the diagnosing ability of IIT are analyzed. The digital image processing technology and its application in infrared diagnosing of PCBs fault is studied, and the edge inspection of image is given by Roberts, Sobel, Prewitt, Laplacian, Canny methods. The model of image registration is analyzed and established, an infrared image registration method for diagnosing of PCBs fault based on genetic algorithm (GA) is proposed, and the algorithm is designed. Some new ideas are embodied in this thesis.
Based on GA theory and technology, for the image registration, the encoding, decoding, fitness evaluation, and the design of selection operator, crossover operator, mutation operator are studied deeply, some new ideas are embodied in the algorithm, the advantages, inferior position, feasibility and efficiency of the algorithm are analyzed particularly. At last, the simulating experimental system of image processing is built, include hardware, software, etc..
During the simulating experiments, first, the edge of image is inspected; next, the standard image and the test image are registrated by the algorithm procedure
based on GA, at last, the images subtraction is used to locate the fault elements, and some preferable results are obtained. The diagnosis of PCBs fault is realized by simulating system, and the feasibility and efficiency of the algorithm are validated.
The investigation on image processing of the infrared diagnosing system of PCBs fault is implemented theoretically and simulating experimentally, and some preferable results are obtained. The theoretical limits and the difficulties of algorithm are discussed, and the measures to improve the efficiency of algorithm are suggested. The works here are essential for further research on the study of infrared diagnosing system of PCBs fault
引文
[1] 王德安, “热成像测试技术及其现状与发展趋势",红外技术,No.1,1996
[2] 彭焕良, “热成像技术发展综述",激光与红外,Vol.27,No.3, 1997
[3] 陈衡,“我国红外诊断技术的现状和展望", 激光与红外, Vol.28, No.5, Oct.1998
[4] 王格芳，张广喜，吴国庆,刘亚卿等，“印刷电路板红外测试系统”，红外技术，Vol.16
No.6, 1994
[5] 王立冬，赵飞宇，陈映平，“红外热成像技术在印制电路板故障诊断中的利用”，红外技术，Vol.21,No.5,1999
[6] 华顺芳,于胜学,王卫国，“红外热像技术诊断印刷电路板故障”，激光与红外 VoL.26 No.2 April, 1996
[7] 王格芳,陈国顺, “基于红外热成像的集成电路检测与诊断",激光与红外,Vol.29,No.5, Oct.1999
[8] 张维力,和丕训等著,红外诊断技术,水利电力出版社, 1991
[9] 徐淦卿，陈珏，陈东杰等著, 红外与物理技术，西安电子科技大学出版社，1989
[10] 吴宗凡等著,红外与微光技术, 国防工业出版社, 1998
[11] 吴国庆，“红外检测技术在电子领域的应用简介”，红外技术，Vol.17, No.1, 1995
[12] 陈衡，“红外扫描测温与热诊断学”，激光与红外，Vol.26, No.2, April 1996
[13] 陈水桥，陈洪山等，“红外热成像法研究电路板热分布特性”，物理实验，Vol.21，No.6
[14] 侯成刚，赵明涛，屈梁生，“红外热成像技术对印刷电路板故障的诊断能力”，红外技术，Vol.18, No.6, 1996
[15] 吴宗凡,“红外热像仪的原理和技术发展”，现代科学仪器，No.2,1997
[16] 周孝宽等著,实用微机图象处理,北京航空航天大学出版社，1994
[17] 阮秋琦著，数字图像处理学，北京：电子工业出版社，2001
[18] 王新成著，高级图像处理技术，北京：中国科学技术出版社，2000
[19] 吴晓波,“图像边缘特征分析”，光学精密工程，Vol.7,No.1, 1999
[20] 周凌翔，顾伟康，“最佳边缘检测的准则与算子”，模式识别与人工智能，Vol.8，No.1,Mar.1998
[21] 陈卉,姜莉等,“线性的医学图像边界检测方法中六种算子的比较", 山东生物医学工程,Vol.17, No.1, 1998
[22] 周道柄,朱卫刚, “几种边缘检测算子的评估",指挥技术学院学报,Vol.11,No.1, 2000
[23] Canny J. F，“A Computational Approach to Edge Detection”，IEEE Trans. On PAMI, 816, p.679-698, 1986
[24] 李国胜，“多源地理信息高精度纠正与配准的技术和方法",辽宁工程技术大学，硕士学位论文，Apr.2002
[25] P. E. Anuta,“Spatial registration of multispectral and multitemporal digital imagery using Fast Fourier Transformation techniques”，IEEE Tran.on GeoSci.Electron, Vol.GE.8,355-368,oct.1970
[26] A. Arcese,“Image Detest through Bipolar correlation”，IEEE Trans.Inform Theory, p.531-541,Sept,1970
[27] D. I. Barnea and H. F. Silverman,“A class of Algorithms for Fast Digital Image Registration”，IEEE Trans. On Computer, Vol.C-21, p.186-192, 1972
[28] M. Sredlow, C. D. MacGillen and Paul E. Anuta,“Image Registration: Similarity Measure and Preprocessing Method Comparison”，IEEE Trans. On Aerospace and Electronic Systems, Vol. AES-14, p.141-149, 1992
[29] 李智，张稚声，“基于轮廓特征的图像配准”，CCUA 图像分会，1997
[30] 王小睿，吴信才，“遥感多图象的自动配准方法”，中国图形图像学报，No.10，1997
[31] 郭海涛，刘智等，“基于遗传算法的快速影像匹配技术的研究”，测绘学院学报，No.9, 2001
[32] 熊兴华，钱曾波等，“遗传算法与最小二乘法相结合的遥感图像子像素匹配”，测绘学报，No.2, 2001
[33] 周明，孙树栋编著，遗传算法原理及应用，北京：国防工业出版社，1999
[34] 陈国良，王煦法等著，遗传算法及其应用，人民邮电出版社，2001
[35] 韩斌，刘以安等，“印刷缺陷检测中的图像对准”，华东船舶工业学院学报（自然科学版），Vol.16，No.2, Apr.2002
[36] 李俊山,沈绪榜, “图像匹配技术研究",微电子学与计算机,Feb.2000
[37] 鲁昌华，徐胜海，刘春, “数字图像处理技术在PCB板检测中的应用",仪器仪表学报,Vol.22,No.4增刊,2001
[38] 黄铜城，“图象几何变换与图象配准的关系研究”，邵阳师范高等专科学校学报，Vol.23,No.2,Apr.2001
[39] Inglada, J., Adragna, F., “Automatic multi-sensor image registration by edge matching using genetic algorithms”，Geoscience and Remote Sensing Symposium, 2001. IGARSS '01， IEEE 2001 International , Volume: 5 , 2001 Page(s): 2313 -2315 vol.5
[2] 彭焕良, “热成像技术发展综述",激光与红外,Vol.27,No.3, 1997
[3] 陈衡,“我国红外诊断技术的现状和展望", 激光与红外, Vol.28, No.5, Oct.1998
[4] 王格芳，张广喜，吴国庆,刘亚卿等，“印刷电路板红外测试系统”，红外技术，Vol.16
No.6, 1994
[5] 王立冬，赵飞宇，陈映平，“红外热成像技术在印制电路板故障诊断中的利用”，红外技术，Vol.21,No.5,1999
[6] 华顺芳,于胜学,王卫国，“红外热像技术诊断印刷电路板故障”，激光与红外 VoL.26 No.2 April, 1996
[7] 王格芳,陈国顺, “基于红外热成像的集成电路检测与诊断",激光与红外,Vol.29,No.5, Oct.1999
[8] 张维力,和丕训等著,红外诊断技术,水利电力出版社, 1991
[9] 徐淦卿，陈珏，陈东杰等著, 红外与物理技术，西安电子科技大学出版社，1989
[10] 吴宗凡等著,红外与微光技术, 国防工业出版社, 1998
[11] 吴国庆，“红外检测技术在电子领域的应用简介”，红外技术，Vol.17, No.1, 1995
[12] 陈衡，“红外扫描测温与热诊断学”，激光与红外，Vol.26, No.2, April 1996
[13] 陈水桥，陈洪山等，“红外热成像法研究电路板热分布特性”，物理实验，Vol.21，No.6
[14] 侯成刚，赵明涛，屈梁生，“红外热成像技术对印刷电路板故障的诊断能力”，红外技术，Vol.18, No.6, 1996
[15] 吴宗凡,“红外热像仪的原理和技术发展”，现代科学仪器，No.2,1997
[16] 周孝宽等著,实用微机图象处理,北京航空航天大学出版社，1994
[17] 阮秋琦著，数字图像处理学，北京：电子工业出版社，2001
[18] 王新成著，高级图像处理技术，北京：中国科学技术出版社，2000
[19] 吴晓波,“图像边缘特征分析”，光学精密工程，Vol.7,No.1, 1999
[20] 周凌翔，顾伟康，“最佳边缘检测的准则与算子”，模式识别与人工智能，Vol.8，No.1,Mar.1998
[21] 陈卉,姜莉等,“线性的医学图像边界检测方法中六种算子的比较", 山东生物医学工程,Vol.17, No.1, 1998
[22] 周道柄,朱卫刚, “几种边缘检测算子的评估",指挥技术学院学报,Vol.11,No.1, 2000
[23] Canny J. F，“A Computational Approach to Edge Detection”，IEEE Trans. On PAMI, 816, p.679-698, 1986
[24] 李国胜，“多源地理信息高精度纠正与配准的技术和方法",辽宁工程技术大学，硕士学位论文，Apr.2002
[25] P. E. Anuta,“Spatial registration of multispectral and multitemporal digital imagery using Fast Fourier Transformation techniques”，IEEE Tran.on GeoSci.Electron, Vol.GE.8,355-368,oct.1970
[26] A. Arcese,“Image Detest through Bipolar correlation”，IEEE Trans.Inform Theory, p.531-541,Sept,1970
[27] D. I. Barnea and H. F. Silverman,“A class of Algorithms for Fast Digital Image Registration”，IEEE Trans. On Computer, Vol.C-21, p.186-192, 1972
[28] M. Sredlow, C. D. MacGillen and Paul E. Anuta,“Image Registration: Similarity Measure and Preprocessing Method Comparison”，IEEE Trans. On Aerospace and Electronic Systems, Vol. AES-14, p.141-149, 1992
[29] 李智，张稚声，“基于轮廓特征的图像配准”，CCUA 图像分会，1997
[30] 王小睿，吴信才，“遥感多图象的自动配准方法”，中国图形图像学报，No.10，1997
[31] 郭海涛，刘智等，“基于遗传算法的快速影像匹配技术的研究”，测绘学院学报，No.9, 2001
[32] 熊兴华，钱曾波等，“遗传算法与最小二乘法相结合的遥感图像子像素匹配”，测绘学报，No.2, 2001
[33] 周明，孙树栋编著，遗传算法原理及应用，北京：国防工业出版社，1999
[34] 陈国良，王煦法等著，遗传算法及其应用，人民邮电出版社，2001
[35] 韩斌，刘以安等，“印刷缺陷检测中的图像对准”，华东船舶工业学院学报（自然科学版），Vol.16，No.2, Apr.2002
[36] 李俊山,沈绪榜, “图像匹配技术研究",微电子学与计算机,Feb.2000
[37] 鲁昌华，徐胜海，刘春, “数字图像处理技术在PCB板检测中的应用",仪器仪表学报,Vol.22,No.4增刊,2001
[38] 黄铜城，“图象几何变换与图象配准的关系研究”，邵阳师范高等专科学校学报，Vol.23,No.2,Apr.2001
[39] Inglada, J., Adragna, F., “Automatic multi-sensor image registration by edge matching using genetic algorithms”，Geoscience and Remote Sensing Symposium, 2001. IGARSS '01， IEEE 2001 International , Volume: 5 , 2001 Page(s): 2313 -2315 vol.5
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |