基于灰度共生矩阵和BP神经网络的织物组织结构识别
|
摘要
目前织物组织结构的分析还是借助于放大镜、分析针等简单工具,完全依赖人工目测完成。由于人工目测不可避免会受到个人的视力、情绪、疲劳、光线等因素的影响,带有很大的个人主观性,往往难以保证识别质量,造成测试结果具有一定的主观性和不可靠性,而且耗费操作人员大量的时间重复做相同的工作。加上效率低下已无法适应目前纺织生产小批量、多品种、高效率的需要。这成为了提高纺织企业信息化、生产自动化程度的一个瓶颈。如果能设计出一套织物组织结构参数的自动识别系统,将会大大推动纺织生产效率,实现全面自动化。图像处理技术和人工智能近年来取得了很大的发展,这使得利用图像处理技术实现织物组织结构的自动分析成为可能。
本文运用灰度共生矩阵对织物图像进行特征提取,然后设计了一个三层BP神经网络对提取出来的特征值进行识别分类,经试验验证对平纹、斜纹、缎纹三种组织织物图像的正确识别率可达93.45%。在织物分类成功的基础上,运用小波分解理论,提取织物图像经纬向亮度信息,对织物组织点进行定位,利用组织点图像表面纹理信息,提取相关性特征,对其经纬属性进行判断,从而分析出织物具体组织结构。
At present, the work to analyze fabric structure still depends on artificial visual measurement, which is easily influenced by personal sight, mood, mental state as well as optical line. Therefore this method of testing is of great individual subjectivity and unreliability. The recognition quality is difficult to guarantee as a result of that. Meanwhile, it takes much more time for operators to do the same testing work repeatedly with poor efficiency, which cannot meet the needs of current textile production for small quantities, multi-item and high efficiency. With so many drawbacks, artificial visual measurement has become a great bottleneck of textile industry to improve information technology and production automation. An automatic identification system on fabric structural parameters will realize automation fully and enhance textiles’production efficiency powerfully. With the development of image processing technology and artificial intelligence, automatic analysis on fabric structure as a replacement of manual labor is of great possibility.
In this thesis, features of fabric-image have been extracted by GLCM (Gray Level Co-occurrence Matrix). These features were employed to a three layer BP neural network for analyzing. It was verified by experiments on three kinds of fabric structures, such as plain weave, cross grain and satin, that all the structures can be correctly identified up to the rate of 93.45%. Furthermore, wavelet decomposition theory was applied to extract fabric image brightness information to locate point area. Related features could be drawn from the surface texture information of interlacing images, hence the properties of warp and weft can be judged to analyze the specific textual structure of fabric.
本文运用灰度共生矩阵对织物图像进行特征提取,然后设计了一个三层BP神经网络对提取出来的特征值进行识别分类,经试验验证对平纹、斜纹、缎纹三种组织织物图像的正确识别率可达93.45%。在织物分类成功的基础上,运用小波分解理论,提取织物图像经纬向亮度信息,对织物组织点进行定位,利用组织点图像表面纹理信息,提取相关性特征,对其经纬属性进行判断,从而分析出织物具体组织结构。
At present, the work to analyze fabric structure still depends on artificial visual measurement, which is easily influenced by personal sight, mood, mental state as well as optical line. Therefore this method of testing is of great individual subjectivity and unreliability. The recognition quality is difficult to guarantee as a result of that. Meanwhile, it takes much more time for operators to do the same testing work repeatedly with poor efficiency, which cannot meet the needs of current textile production for small quantities, multi-item and high efficiency. With so many drawbacks, artificial visual measurement has become a great bottleneck of textile industry to improve information technology and production automation. An automatic identification system on fabric structural parameters will realize automation fully and enhance textiles’production efficiency powerfully. With the development of image processing technology and artificial intelligence, automatic analysis on fabric structure as a replacement of manual labor is of great possibility.
In this thesis, features of fabric-image have been extracted by GLCM (Gray Level Co-occurrence Matrix). These features were employed to a three layer BP neural network for analyzing. It was verified by experiments on three kinds of fabric structures, such as plain weave, cross grain and satin, that all the structures can be correctly identified up to the rate of 93.45%. Furthermore, wavelet decomposition theory was applied to extract fabric image brightness information to locate point area. Related features could be drawn from the surface texture information of interlacing images, hence the properties of warp and weft can be judged to analyze the specific textual structure of fabric.
引文
[1]张翔.基于计算机视觉的织物组织结构参数自动识别[D].东南大学,2006.
[2]杨帆等编著.数字图像处理与分析[M].北京航空航天大学出版社,2007:275-276.
[3] Sobus J,Pourde yhimi,B.Gerde J. Assessing Changes in Texture Periodicity Due to Appearance Loss in Carpets: Gray Level Co-occurrence Analysis. Textile Res. J.1991, 61(10):557-567.
[4] Lin J. Applying a Co-occurrence Matrix to Automatic Inspection of Weave Density for Woven Fabrics. Textile Res. J., 2002, 72(6):486-490.
[5] Huang C,Liu S,and Yu W. Woven Fabric Analysis by Image Processing Part 1:Identification of Weave Patterns. Textile Res. J, 2000, 70(6):481-485.
[6] Kang T,Kim C. Automatic Recognition of Fabric Weave Patterns by Digital Image Analysis. Textile Res. J.,1999, 69(2):77-83.
[7]庄国瑜,刘凤英,万振凯.织物组织的二维谱分析与相关分析.纺织学报,1998, 19(4):241-244.
[8]张翔.基于计算机视觉的织物组织结构自动识别系统——图像拼接算法研究与实现[D].东南大学,2006.
[9] Ravandi S.,Toriumi.. Fourier Transform Analysis of Plain Weave Fabric Appearance. Textile Res. J.,1995,65(11):676-683.
[10] Sakaguchi A.,Kim H.,Matsumoto. Woven Fabric Quality Evaluation Using Image Analysis. Textile Res. J., 2000, 70(11):950-956.
[11] Xu B.. Identifying Fabric Structures with Fast Fourier Transform Techniques Texti1e Res. J., 1996,66(8):496-506.
[12]高卫东,刘基宏,徐伯俊,薛卫,狄炜.织物中纬纱排列参数的自动识别棉纺织技术,2002,30(1):28-31.
[13]高卫东,刘基宏,徐伯俊,薛卫,狄炜.织物中经纱排列参数的自动识别.棉纺织技术,2002,30(3):31-34.
[14]高卫东,刘基宏,徐伯俊,薛卫,狄炜.织物组织结构的自动识别.棉纺织技术,2002, 30(4):26-28.
[15]辛斌杰,余序芬,吴兆平.机织物经纬密测量的图像处理技术[J].中国纺织大学学报,1999.3.
[16]李艳梅,徐伯俊,高卫东.织物组织结构参数自动识别系统的研究.北京纺织,2002, 23(2):54-57.
[17]彭玉华.小波变换与工程应用.科学出版社.1999. 9.
[18] Hu M.,Tsai.Fabric Inspection Based on Best Wavelet Packet Bases.Textile Res. J, 2000,70 (8):662-670.
[19]冯毅力,李汝勤.用小波变换法自动测量机织物经纬密度.纺织学报,2001,22 (2):94-95.
[20]庄国瑜,孙悦.织物纹理计算机图象处理的多频道分解方法.天津工业大学学报,2001,20(3):26-28.
[21]孙悦,庄国瑜.小波基的选择与织物图像处理.天津纺织工学院学报,1999,18(5):6-9.
[22]马云芳,王小华.基于小波理论的织物纹理特征提取杭州电子工业学院学报,2002,22(6):85-88.
[23]胡觉亮.基于贝叶斯方法的织物分类研究[J].纺织学报2004,25(l):48-49.
[24] Jeon B Bae J Moon W. Automatic Recognition of Woven Fabric Patterns by Artificial Neural Network. Textile Res.J.2003,73(7):645-650.
[25]张瑞林,蒋静坪.人工神经网络识别丝织物的研究.2002,23(2).
[26] Kuo C.,Shih C.,Lee J.. Automatic Recognition of Fabric Weave Pattern by a Fuzzy C-Means Clustering Method. Textile Res.J., 2004, 74 (2):107-111.
[27]张一,耿兆丰,基于基元特征匹配的织物组织结构分析与识别[J].微计算机信息,2006, 22(1):268-271.
[28]西松丰典.Automatic Recognition of the Pattern by using Image Information Analysis[J]. Textile Mach.Soc.Jpn.} 1991,44(6):51-57.
[29]太田建一,野中敬人,石井富久等. Automatic Analyzing of A Fabric Design with requency Compobebts, Part 1: Automatic Analyzing of A Wearing Design. J. Textile Mach.Soc.Jpn.,1998, 49(9): 61-68.
[30]李凯华,方晓初,吴世林,李元科.织物组织计算机图像分析系统的开发[J].武汉科技学院学报,2003,16(2):85-88.
[31]杨帆等编著.数字图像处理与分析[M].北京航空航天大学出版社,2007:275-276.
[32] B. Julesz, E. N. Gilbert, L. A. Sheep, and H. L. Frisch,“Inablity of humans to discriminate between visual textures that agree in second-order statistics-revisited,”Perception, 2,1973,391-405.
[33] R. M. Haralick, K. Shanmugam, and I. Dinstein,“Textural features for image classification”, IEEE Trans. Syst. Man Cybern, SMC-3, 1973,610-621.
[34]苑丽红,付丽,杨勇,苗静.灰度共生矩阵提取纹理特征的实验结果分析[J].计算机应用,1998,29(4):1018-1021.
[35]何东建,耿楠,张义宽.数字图像处理[M].西安:西安电子科技大学出版社.2003,7:213～218。
[36]杨帆等编著.数字图像处理与分析[M].北京航空航天大学出版社,2007:279-281.
[37]刘娅静,基于计算机视觉的麦田杂草识别技术研究[D],河北工业大学,2006.
[38] M. C. Hu,I, S. Tsai. The Inspection of Fabric Defects by Using Wavelet Transform [J]. The Journal of the Textile Institute 2000. 91(3):420-433.
[39]托德元,何小海,李舒平,吴小强.小波变换及其在图像处理中的应用[J].四川大学学报(自然科学版),1994,31(4):481-488.
[40] Takato.M.Takagi,and MoriT.Automated fabric inspection using image processing techniques, SPIE, Automated inspection and high speed architectures, 1988:151-158.
[41]常利利,马俊,邓中民,蒋蕾.基于灰度共生矩阵的织物组织结构差异分析[J].纺织学报,2008 29(4) :43-46.
[42]董长虹编著.Matlab神经网络与应用[M].国防工业出版社,2007:8-10.
[43]焦李成.神经网络系统理论[M].西安电子科技大学出版社,1995.
[44] Simon Haykin.神经网络原理[M].叶世伟,史忠植,译.机械工业出版社,2004.
[45]从爽.面向MATLAB工具箱的神经网络理论与应用[M].中国科学技术大学出版社,1998.
[46]田景文,高美娟.人工神经网络算法研究与及应用[M].北京理工大学出版社,2006.
[47]杨行峻,郑君里.人工神经网络[M].高等教育出版社,1992.
[48]沈清,胡德文,时春.神经网络应用技术[M].国防科技大学出版社,1993.
[49]张立明.人工神经网络模型及其应用[M].复旦大学出版社,1993.
[50]张铃,张钹.人工神经网络理论及应用[M].浙江科学技术出版社,1997.
[51]张德丰.MATLAB神经网络仿真与应用[M].电子工业出版社,2009:11-12.
[52]韩力群.人工神经网络理论设计与应用[M].化学工业出版社,1997.
[53]阎平凡,黄端旭.人工神经网络——模型·分析与应用[M].安徽教育出版社,1993.
[54]周继成.人工神经网络——第六代计算机的实现[M].科学普及出版社,1993.
[55]闻新,周露,李翔等.MATLAB神经网络仿真与应用[M].北京科学出版社, 2003.
[56]陈明.神经网络模型[M].大连理工大学出版社,1993.
[57]楼顺天等.基于Matlab的系统分析与设计——神经网络[M].西安电子科技大学出版社,2000.
[58]飞思科技产品研发中心.MATLAB6.5辅助神经网络分析与设计[M].电子工业出版社,2003.
[59]张良均,曹晶,蒋世忠.神经网络的实用教程[M].机械工业出版社,2007.
[60]焦李成.神经网络的应用与实现[M].西安电子科技大学出版社,1996.
[61]魏海坤.神经网络结构设计的原理与方法[M].中国铁道出版社,2000.
[62]罗四维.人工神经网络建造.中国铁道出版社,1997.
[63]钟小勇,基于Curvelet变换和BP神经网络的织物疵点检测[D],苏州大学,2006.
[64] Matlab user’s Guide.The mathworks.Inc, 2003.
[65] Neural Network Toolbox User’s Guide, The math Works, Inc,2003.
[66] MATLAB中文论坛编著.MATLAB神经网络30个案例分析[M].北京航空航天大学出版社,2010.
[67]李建华,李万社.小波理论发展极其应用[J].河西学院学报,1999,22(2):27-30.
[68]秦前清,杨中凯.实用小波分析[M].西安:西安电子科技出版社,1998,14-29.
[69]夏德深,傅德胜.现代图像处理技术与应用[M].南京:东南大学出版社,1997.189-199.
[70]徐增波,贡玉南,黄秀宝.基于二维连续小波变换的织物疵点检测[J].中国纺织大学学报,2000,26(2):10-13.
[71]马云芳,王小华.基于小波理论的织物纹理特征提取[J].杭州电子工业学院学报,2002,22(6):85-88.
[72]孙悦,庄国瑜.小波基的选择与织物图像处理[J].天津纺织工学院学报,1999,18(5):6-9.
[73]李兵兵,常义林,胡征.小波基的构造方法和算法设计[J].电子科学学刊,1994,16(6):646-650.
[74]李立轻,黄秀宝.基于织物自适应小波的疵点检测东华大学学报(自然科学版),2001,27(4):82-87.
[75]刘泓,莫玉龙.基于二维自适应异小波基小波变换的规则纹理缺陷检测[J].中国图像图形学报,1999,4(4):349-352.
[76]徐造坤,基于小波分析的机织物识别研究[D],武汉理工大学,2006.
[77]贾婷婷,李立轻.基于图像分解的织物组织自动识别[J].东华大学学报,1999,34(6):668-671.
[78]薛文良,魏孟媛,陈革,程隆棣.纱线结构中纤维形态的数学表征[J].纺织学报,2010.5,31(5):30-33。
[79]余伟东.纺织材料学[M].中国纺织出版社,2006:197.
[2]杨帆等编著.数字图像处理与分析[M].北京航空航天大学出版社,2007:275-276.
[3] Sobus J,Pourde yhimi,B.Gerde J. Assessing Changes in Texture Periodicity Due to Appearance Loss in Carpets: Gray Level Co-occurrence Analysis. Textile Res. J.1991, 61(10):557-567.
[4] Lin J. Applying a Co-occurrence Matrix to Automatic Inspection of Weave Density for Woven Fabrics. Textile Res. J., 2002, 72(6):486-490.
[5] Huang C,Liu S,and Yu W. Woven Fabric Analysis by Image Processing Part 1:Identification of Weave Patterns. Textile Res. J, 2000, 70(6):481-485.
[6] Kang T,Kim C. Automatic Recognition of Fabric Weave Patterns by Digital Image Analysis. Textile Res. J.,1999, 69(2):77-83.
[7]庄国瑜,刘凤英,万振凯.织物组织的二维谱分析与相关分析.纺织学报,1998, 19(4):241-244.
[8]张翔.基于计算机视觉的织物组织结构自动识别系统——图像拼接算法研究与实现[D].东南大学,2006.
[9] Ravandi S.,Toriumi.. Fourier Transform Analysis of Plain Weave Fabric Appearance. Textile Res. J.,1995,65(11):676-683.
[10] Sakaguchi A.,Kim H.,Matsumoto. Woven Fabric Quality Evaluation Using Image Analysis. Textile Res. J., 2000, 70(11):950-956.
[11] Xu B.. Identifying Fabric Structures with Fast Fourier Transform Techniques Texti1e Res. J., 1996,66(8):496-506.
[12]高卫东,刘基宏,徐伯俊,薛卫,狄炜.织物中纬纱排列参数的自动识别棉纺织技术,2002,30(1):28-31.
[13]高卫东,刘基宏,徐伯俊,薛卫,狄炜.织物中经纱排列参数的自动识别.棉纺织技术,2002,30(3):31-34.
[14]高卫东,刘基宏,徐伯俊,薛卫,狄炜.织物组织结构的自动识别.棉纺织技术,2002, 30(4):26-28.
[15]辛斌杰,余序芬,吴兆平.机织物经纬密测量的图像处理技术[J].中国纺织大学学报,1999.3.
[16]李艳梅,徐伯俊,高卫东.织物组织结构参数自动识别系统的研究.北京纺织,2002, 23(2):54-57.
[17]彭玉华.小波变换与工程应用.科学出版社.1999. 9.
[18] Hu M.,Tsai.Fabric Inspection Based on Best Wavelet Packet Bases.Textile Res. J, 2000,70 (8):662-670.
[19]冯毅力,李汝勤.用小波变换法自动测量机织物经纬密度.纺织学报,2001,22 (2):94-95.
[20]庄国瑜,孙悦.织物纹理计算机图象处理的多频道分解方法.天津工业大学学报,2001,20(3):26-28.
[21]孙悦,庄国瑜.小波基的选择与织物图像处理.天津纺织工学院学报,1999,18(5):6-9.
[22]马云芳,王小华.基于小波理论的织物纹理特征提取杭州电子工业学院学报,2002,22(6):85-88.
[23]胡觉亮.基于贝叶斯方法的织物分类研究[J].纺织学报2004,25(l):48-49.
[24] Jeon B Bae J Moon W. Automatic Recognition of Woven Fabric Patterns by Artificial Neural Network. Textile Res.J.2003,73(7):645-650.
[25]张瑞林,蒋静坪.人工神经网络识别丝织物的研究.2002,23(2).
[26] Kuo C.,Shih C.,Lee J.. Automatic Recognition of Fabric Weave Pattern by a Fuzzy C-Means Clustering Method. Textile Res.J., 2004, 74 (2):107-111.
[27]张一,耿兆丰,基于基元特征匹配的织物组织结构分析与识别[J].微计算机信息,2006, 22(1):268-271.
[28]西松丰典.Automatic Recognition of the Pattern by using Image Information Analysis[J]. Textile Mach.Soc.Jpn.} 1991,44(6):51-57.
[29]太田建一,野中敬人,石井富久等. Automatic Analyzing of A Fabric Design with requency Compobebts, Part 1: Automatic Analyzing of A Wearing Design. J. Textile Mach.Soc.Jpn.,1998, 49(9): 61-68.
[30]李凯华,方晓初,吴世林,李元科.织物组织计算机图像分析系统的开发[J].武汉科技学院学报,2003,16(2):85-88.
[31]杨帆等编著.数字图像处理与分析[M].北京航空航天大学出版社,2007:275-276.
[32] B. Julesz, E. N. Gilbert, L. A. Sheep, and H. L. Frisch,“Inablity of humans to discriminate between visual textures that agree in second-order statistics-revisited,”Perception, 2,1973,391-405.
[33] R. M. Haralick, K. Shanmugam, and I. Dinstein,“Textural features for image classification”, IEEE Trans. Syst. Man Cybern, SMC-3, 1973,610-621.
[34]苑丽红,付丽,杨勇,苗静.灰度共生矩阵提取纹理特征的实验结果分析[J].计算机应用,1998,29(4):1018-1021.
[35]何东建,耿楠,张义宽.数字图像处理[M].西安:西安电子科技大学出版社.2003,7:213～218。
[36]杨帆等编著.数字图像处理与分析[M].北京航空航天大学出版社,2007:279-281.
[37]刘娅静,基于计算机视觉的麦田杂草识别技术研究[D],河北工业大学,2006.
[38] M. C. Hu,I, S. Tsai. The Inspection of Fabric Defects by Using Wavelet Transform [J]. The Journal of the Textile Institute 2000. 91(3):420-433.
[39]托德元,何小海,李舒平,吴小强.小波变换及其在图像处理中的应用[J].四川大学学报(自然科学版),1994,31(4):481-488.
[40] Takato.M.Takagi,and MoriT.Automated fabric inspection using image processing techniques, SPIE, Automated inspection and high speed architectures, 1988:151-158.
[41]常利利,马俊,邓中民,蒋蕾.基于灰度共生矩阵的织物组织结构差异分析[J].纺织学报,2008 29(4) :43-46.
[42]董长虹编著.Matlab神经网络与应用[M].国防工业出版社,2007:8-10.
[43]焦李成.神经网络系统理论[M].西安电子科技大学出版社,1995.
[44] Simon Haykin.神经网络原理[M].叶世伟,史忠植,译.机械工业出版社,2004.
[45]从爽.面向MATLAB工具箱的神经网络理论与应用[M].中国科学技术大学出版社,1998.
[46]田景文,高美娟.人工神经网络算法研究与及应用[M].北京理工大学出版社,2006.
[47]杨行峻,郑君里.人工神经网络[M].高等教育出版社,1992.
[48]沈清,胡德文,时春.神经网络应用技术[M].国防科技大学出版社,1993.
[49]张立明.人工神经网络模型及其应用[M].复旦大学出版社,1993.
[50]张铃,张钹.人工神经网络理论及应用[M].浙江科学技术出版社,1997.
[51]张德丰.MATLAB神经网络仿真与应用[M].电子工业出版社,2009:11-12.
[52]韩力群.人工神经网络理论设计与应用[M].化学工业出版社,1997.
[53]阎平凡,黄端旭.人工神经网络——模型·分析与应用[M].安徽教育出版社,1993.
[54]周继成.人工神经网络——第六代计算机的实现[M].科学普及出版社,1993.
[55]闻新,周露,李翔等.MATLAB神经网络仿真与应用[M].北京科学出版社, 2003.
[56]陈明.神经网络模型[M].大连理工大学出版社,1993.
[57]楼顺天等.基于Matlab的系统分析与设计——神经网络[M].西安电子科技大学出版社,2000.
[58]飞思科技产品研发中心.MATLAB6.5辅助神经网络分析与设计[M].电子工业出版社,2003.
[59]张良均,曹晶,蒋世忠.神经网络的实用教程[M].机械工业出版社,2007.
[60]焦李成.神经网络的应用与实现[M].西安电子科技大学出版社,1996.
[61]魏海坤.神经网络结构设计的原理与方法[M].中国铁道出版社,2000.
[62]罗四维.人工神经网络建造.中国铁道出版社,1997.
[63]钟小勇,基于Curvelet变换和BP神经网络的织物疵点检测[D],苏州大学,2006.
[64] Matlab user’s Guide.The mathworks.Inc, 2003.
[65] Neural Network Toolbox User’s Guide, The math Works, Inc,2003.
[66] MATLAB中文论坛编著.MATLAB神经网络30个案例分析[M].北京航空航天大学出版社,2010.
[67]李建华,李万社.小波理论发展极其应用[J].河西学院学报,1999,22(2):27-30.
[68]秦前清,杨中凯.实用小波分析[M].西安:西安电子科技出版社,1998,14-29.
[69]夏德深,傅德胜.现代图像处理技术与应用[M].南京:东南大学出版社,1997.189-199.
[70]徐增波,贡玉南,黄秀宝.基于二维连续小波变换的织物疵点检测[J].中国纺织大学学报,2000,26(2):10-13.
[71]马云芳,王小华.基于小波理论的织物纹理特征提取[J].杭州电子工业学院学报,2002,22(6):85-88.
[72]孙悦,庄国瑜.小波基的选择与织物图像处理[J].天津纺织工学院学报,1999,18(5):6-9.
[73]李兵兵,常义林,胡征.小波基的构造方法和算法设计[J].电子科学学刊,1994,16(6):646-650.
[74]李立轻,黄秀宝.基于织物自适应小波的疵点检测东华大学学报(自然科学版),2001,27(4):82-87.
[75]刘泓,莫玉龙.基于二维自适应异小波基小波变换的规则纹理缺陷检测[J].中国图像图形学报,1999,4(4):349-352.
[76]徐造坤,基于小波分析的机织物识别研究[D],武汉理工大学,2006.
[77]贾婷婷,李立轻.基于图像分解的织物组织自动识别[J].东华大学学报,1999,34(6):668-671.
[78]薛文良,魏孟媛,陈革,程隆棣.纱线结构中纤维形态的数学表征[J].纺织学报,2010.5,31(5):30-33。
[79]余伟东.纺织材料学[M].中国纺织出版社,2006:197.
© 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 |