带钢表面缺陷智能检测系统的设计与研究
|
摘要
针对目前自动化方法在带钢表面缺陷检测时准确度不高的问题,本文探讨了基于计算机视觉的智能检测系统总体设计方案及软、硬件构成,重点设计了其中的缺陷初检、分割和识别步骤解决方案。
针对实际生产中缺陷出现概率较低的现状,利用对图像分块差值图提取的灰度范围、方差等特征量,构造了缺陷初检判别函数,以提高检测算法的实时陛。对初检含缺陷的图像,因光照不均匀而引起传统的单一阈值分割结果十分不理想,提出了一种基于B样条拟合阈值曲面的分割算法,用于获得分割图像的阈值曲面。在分割出缺陷目标的基础上,选择了几何形状、不变矩等缺陷种类判别特征量,构建了相应的BP神经网络分类器完成对缺陷具体类别的识别。在上述带钢表面缺陷智能检测系统的设计思路下,本文对其中的缺陷初检、分割和识别等关键步骤进行了相应的算法实现与测试,取得良好效果。
This paper discussed the overall design of intelligent defects detection system, and focused on the primary detection, segmentation and recognition.
For the low probability of defects in actual production, the discriminant function which is used to detect whether the image contains defects or not has been constructed by extracting the features, such as gray range, variance and so on. The image that contains defects can not avoid the impact of uneven illumination, so the traditional segmentation also can not get the ideal division results. A segmentation method using threshold surface based on surface fitting is proposed to overcome such inadequacy. In the basis of the defect area was segmented, the features of the shape and the same moment are selected in order to construct the BP neural network classifiers for recognizing defects. The algorithms of primary detection, segmentation and recognition have been implemented and tested, achieved good results.
针对实际生产中缺陷出现概率较低的现状,利用对图像分块差值图提取的灰度范围、方差等特征量,构造了缺陷初检判别函数,以提高检测算法的实时陛。对初检含缺陷的图像,因光照不均匀而引起传统的单一阈值分割结果十分不理想,提出了一种基于B样条拟合阈值曲面的分割算法,用于获得分割图像的阈值曲面。在分割出缺陷目标的基础上,选择了几何形状、不变矩等缺陷种类判别特征量,构建了相应的BP神经网络分类器完成对缺陷具体类别的识别。在上述带钢表面缺陷智能检测系统的设计思路下,本文对其中的缺陷初检、分割和识别等关键步骤进行了相应的算法实现与测试,取得良好效果。
This paper discussed the overall design of intelligent defects detection system, and focused on the primary detection, segmentation and recognition.
For the low probability of defects in actual production, the discriminant function which is used to detect whether the image contains defects or not has been constructed by extracting the features, such as gray range, variance and so on. The image that contains defects can not avoid the impact of uneven illumination, so the traditional segmentation also can not get the ideal division results. A segmentation method using threshold surface based on surface fitting is proposed to overcome such inadequacy. In the basis of the defect area was segmented, the features of the shape and the same moment are selected in order to construct the BP neural network classifiers for recognizing defects. The algorithms of primary detection, segmentation and recognition have been implemented and tested, achieved good results.
引文
[1]董志洪,李立复.板带生产工艺现状及其发展趋势[J].钢铁,1999,34(11):68-71.
[2]罗成汉,解源,王飞.基于神经网络的钢材表面裂纹定量检测方法的研究[J].自动化技术与应用,2003,22(4):70-73.
[3]吴平川,路同浚,王炎.带钢表面自动检测系统研究状况与展望[J].钢铁,2000,35(6):70-75.
[4]罗志勇,王斌,刘栋王等.带钢表面缺陷检测系统的发展[J].钢铁1996,31(10):127-131
[5]吴平川,路同浚,王炎.钢板表面缺陷的无损检测技术与应用[J].无损检测,2000,22(7):312-315.
[6]苑惠娟,于晓洋,郭建英,李粉兰.数字式光处理技术及在三维测量中的应用[J].哈尔滨理工大学学报,2002,(3):35-37
[7]Rice S O.Mathematical Analysis of Random Noise[J].Bell Syst Tech,1945,24:46-156.
[8]郑健峰.带钢表面缺陷检测方法研究[D].西安建筑科技大学硕士论文,2006:14-17
[9]郎锐.数字图像处理学----Visual C++实现[M].北京希望电子出版社,2002:56-57.
[10]吕凤金.数字图像处理编程入门[M].北京:清华大学出版社,1999:201-205.
[11]容观澳.计算机图像处理[M].北京:清华大学出版社,2002:178-183.
[12]孙兆林.matlab6.x图像处理[M].北京:清华大学出版社,2002:226-227.
[13]许道峰.基于LVQ的带钢表面缺陷分类研究[J].中国仪器仪表,2004,2:5-7.
[14]崔屹.数字图像处理技术与应用[M].北京:电子工业出版社,1997:63-66.
[15]Sangwook L,Luh-Maan C,Miroslaw S.Automated recognition of surface defects using digital color image processing[J].Automation in Construction,2005,15(2006):540-449.
[16]Johnson R A,Wichem D W.Applied multivariate statistical analysis,5th ed[M].Prentice Hall:Upper Saddle River,NJ,2002:236-239
[17]章毓晋.图像工程(上册)----图像处理和分析[M].北京:清华大学出版社,2003,178-180.
[18]章毓晋.图像分割[M].北京:科学出版社,2001:23-30.
[19]Iwao S,Takio H,Notsu.A thresholding method using the mixture of normal density functions[A].Proc.of Inter.Symposiumon Speech,Image Processing,and Neural NetWorks[C].Hong Kong:IEEE,1994.304-307.
[20]张永平,郑南宁等.基于变分的图像分割算法[J].中国科学(E 辑),2002,32(1):133-144.
[21]郑毅,刘上乾等.利用变分的平面网格图像自适应分割算法[J].光电工程,2006,33(3):36-39.
[22]施法中.计算机辅助几何设计与非均匀有理B样条[M].北京:北京航空航天大学出版社,1994:207-208.
[23]陈元琰,张睿哲,吴东.计算机图形学实用技术(第二版)[M].北京:清华大学出版社,2007:173-175.
[24]刘鼎元,赵玉琦等.Bezier曲线和B样条曲线光顺拟合[J].计算数学,1984(4):360-365.
[25]高阳,李言俊.非均匀背景下的红外图像曲面拟合分割[J].光电工程,2006,33(7):83-88.
[26]曹利新,刘健等.面向截面测量数据B样条曲面光顺拟合及其精度计算[J].大连理工大学学报,2003,43(3):320-323.
[27]Woodward C D.Skinning techniques for interactive B-spline surface interpolation [J].CAD,1988,20(8):441-451.
[28]崔屹.图象处理与分析----数学形态学方法及应用[M].科学出版社,2000:1-59
[29]刘红冰.基于神经网络的冷轧带钢表面缺陷检测与分类技术研究[D].电子科技大学硕士论文,2005:35-36.
[30]高曙明.自动特征识别综述[J].计算机学报,1988,21(3):281-288.
[31]Dolye W.Operations useful for similarity-invariant pattern recognition[J].JACM,1998,9(2):259-267.
[32]Parke F I.Computer generated animation faces[J].Proceedings ACM ANN Conference,1972,4(1):451-457.
[33]Burks T E,Shearer S A,Gates R S.Back Propagation Neural Network Design and Evalution for Classifying Weed Species Using Color Image Texture[J].Transactions of the ASAE,2000,43(4):1029-1037.
[34]Hu M K.Visual pattern recognition by moment invariants[J].IEEE Trans on Information Theory,1962,8(2):179-187.
[35]Hecht Nielson R.Theory of the back propagation neural network[J].Int.J.Conf.On Neural Network,1989,27(1):593-605.
[2]罗成汉,解源,王飞.基于神经网络的钢材表面裂纹定量检测方法的研究[J].自动化技术与应用,2003,22(4):70-73.
[3]吴平川,路同浚,王炎.带钢表面自动检测系统研究状况与展望[J].钢铁,2000,35(6):70-75.
[4]罗志勇,王斌,刘栋王等.带钢表面缺陷检测系统的发展[J].钢铁1996,31(10):127-131
[5]吴平川,路同浚,王炎.钢板表面缺陷的无损检测技术与应用[J].无损检测,2000,22(7):312-315.
[6]苑惠娟,于晓洋,郭建英,李粉兰.数字式光处理技术及在三维测量中的应用[J].哈尔滨理工大学学报,2002,(3):35-37
[7]Rice S O.Mathematical Analysis of Random Noise[J].Bell Syst Tech,1945,24:46-156.
[8]郑健峰.带钢表面缺陷检测方法研究[D].西安建筑科技大学硕士论文,2006:14-17
[9]郎锐.数字图像处理学----Visual C++实现[M].北京希望电子出版社,2002:56-57.
[10]吕凤金.数字图像处理编程入门[M].北京:清华大学出版社,1999:201-205.
[11]容观澳.计算机图像处理[M].北京:清华大学出版社,2002:178-183.
[12]孙兆林.matlab6.x图像处理[M].北京:清华大学出版社,2002:226-227.
[13]许道峰.基于LVQ的带钢表面缺陷分类研究[J].中国仪器仪表,2004,2:5-7.
[14]崔屹.数字图像处理技术与应用[M].北京:电子工业出版社,1997:63-66.
[15]Sangwook L,Luh-Maan C,Miroslaw S.Automated recognition of surface defects using digital color image processing[J].Automation in Construction,2005,15(2006):540-449.
[16]Johnson R A,Wichem D W.Applied multivariate statistical analysis,5th ed[M].Prentice Hall:Upper Saddle River,NJ,2002:236-239
[17]章毓晋.图像工程(上册)----图像处理和分析[M].北京:清华大学出版社,2003,178-180.
[18]章毓晋.图像分割[M].北京:科学出版社,2001:23-30.
[19]Iwao S,Takio H,Notsu.A thresholding method using the mixture of normal density functions[A].Proc.of Inter.Symposiumon Speech,Image Processing,and Neural NetWorks[C].Hong Kong:IEEE,1994.304-307.
[20]张永平,郑南宁等.基于变分的图像分割算法[J].中国科学(E 辑),2002,32(1):133-144.
[21]郑毅,刘上乾等.利用变分的平面网格图像自适应分割算法[J].光电工程,2006,33(3):36-39.
[22]施法中.计算机辅助几何设计与非均匀有理B样条[M].北京:北京航空航天大学出版社,1994:207-208.
[23]陈元琰,张睿哲,吴东.计算机图形学实用技术(第二版)[M].北京:清华大学出版社,2007:173-175.
[24]刘鼎元,赵玉琦等.Bezier曲线和B样条曲线光顺拟合[J].计算数学,1984(4):360-365.
[25]高阳,李言俊.非均匀背景下的红外图像曲面拟合分割[J].光电工程,2006,33(7):83-88.
[26]曹利新,刘健等.面向截面测量数据B样条曲面光顺拟合及其精度计算[J].大连理工大学学报,2003,43(3):320-323.
[27]Woodward C D.Skinning techniques for interactive B-spline surface interpolation [J].CAD,1988,20(8):441-451.
[28]崔屹.图象处理与分析----数学形态学方法及应用[M].科学出版社,2000:1-59
[29]刘红冰.基于神经网络的冷轧带钢表面缺陷检测与分类技术研究[D].电子科技大学硕士论文,2005:35-36.
[30]高曙明.自动特征识别综述[J].计算机学报,1988,21(3):281-288.
[31]Dolye W.Operations useful for similarity-invariant pattern recognition[J].JACM,1998,9(2):259-267.
[32]Parke F I.Computer generated animation faces[J].Proceedings ACM ANN Conference,1972,4(1):451-457.
[33]Burks T E,Shearer S A,Gates R S.Back Propagation Neural Network Design and Evalution for Classifying Weed Species Using Color Image Texture[J].Transactions of the ASAE,2000,43(4):1029-1037.
[34]Hu M K.Visual pattern recognition by moment invariants[J].IEEE Trans on Information Theory,1962,8(2):179-187.
[35]Hecht Nielson R.Theory of the back propagation neural network[J].Int.J.Conf.On Neural Network,1989,27(1):593-605.