基于小波分析和BP神经网络的织物疵点识别
|
摘要
基于图像处理的织物疵点检测就是纹理分割和模式识别问题，本文提出了一种基于离散小波变换和BP神经网络的织物疵点识别方法，其主要包括织物疵点图像预处理、疵点图像小波变换、特征提取和疵点识别。
在预处理中，采用4种不同的小波域降噪方法，对织物疵点图像进行降噪处理，以峰值信噪比为标准，选择出贝叶斯阈值和软阈值收缩准则相结合的方法降噪效果最佳。根据织物疵点的特殊形状结构，本文创新性地提出了基于小波变换的自适应高斯平滑化和拉普拉斯算子卷积作差相结合的方法对织物疵点图像进行增强，以消除由重复的结构纹理单元形成的背景，突出疵点部分。
根据织物纹理和结构的特殊性质，本文提出了同时提取纹理特征和几何形状特征作为神经网络识别参数，以提高疵点正确识别率。在纹理特征提取部分，本文提出了小波域和空域分析相结合的方法：小波变换和灰度共生矩阵相结合的方法。详细介绍了小波基的选择、最佳分解层的确定方法、子带重构规则、灰度共生矩阵重要参数的选择方法和疵点纹理特征值的主成分分析。以给定的具有不同织物组织结构的无疵点织物图像经小波分解后输出的图像的高通子图像能量最小作为逼近条件从合适的小波库中优选出最佳分解小波基。采用相邻两层的高频子带能量之比值小于1作为最佳分解层数的选择依据。在最佳分解层，为了保证特征提取源的有效性，并减化运算的复杂程度，在图像子带重构时，提出子带重构规则。根据重构规则自动选择子带进行重构，对重构后的子带采用基于特征的小波融合方法进行融合。根据织物具有连续性和方向性的特征，本文提出了灰度共生矩阵差分距离和方向角自适应选择规则。对由灰度共生矩阵中提取的13个纹理特征参数进行主成分分析，以消除信息冗余。为综合表达疵点特征，对增强后的织物疵点图像采用最佳阈值边缘检测和形态学运算，从二值化后的织物疵点图像中提取长、短径之比作为几何形状特征，以描述疵点的形状特征。
在织物疵点识别网络设计时，提出了隐含层神经元个数不适宜规则和网络训练方法不适宜规则，以优化网络结构，提高训练速度。采用结构为：输入层7个神经元，隐含层16个神经元，输出层4个神经元，中间层神经元的激励函数为S型正切函数，输出层神经元的激励函数为S型对数函数的神经网络对织物疵点图像进行训练和识别。在本实验条件下，采用本文算法，对丝织物中常见疵点：断纬、断经、重纬、档疵、油污、破洞进行识别，疵点平均识别率为99.2％，无疵点图像识别率为100％。
The identification of fabric defects based on digital image processing techniques can beconsidered as texture segmentation and pattern recognition problems. In this paper, theidentification of fabric defects based on discrete wavelet transform and Back-Propagationneural network is proposed, the indispensable processes of which are the defect imagespre-processing, wavelet transform, feature extraction and defects identification.
During the pre-processing procedure, four different denoising methods based on wavelettransform are used to improve the peak-signal-to-noise ratio (PSNR) of the fabric defectimages. From the comparison of the 4 different denoising methods, the method basedBayesian thresholding and soft thresholding shrinkage is much better than the others.According to the special structure of the defects, a creative method is proposed to enhancethe defect images, which adopts self-adoptive Gaussian filter and Laplacian operatorconvolution and subtraction operation. With the enhancement procedure, all regular,repetitive texture patterns can be eliminated, and the defect areas become more significant.
According to the special properties of the fabric defects, both textural and geometricalfeatures are extracted to improve the identification accuracy. During the textural featureextraction procedure, wavelet transform domain combined with the spatial domain methodis proposed, that is, wavelet transform combines with spatial gray level co-occurrence(SGLC) matrices methodology. The selection of optimal wavelet base, automatic selectionof multiresolution levels, reconsturution rules for subimages, automatic selection ofimportant parameters of the SGLC and the principal component analysis (PCA) of theextracted feature parameters are presented in details. The approximation condition ofselection of optimal wavelet base choosed from the wavelet base bank is the energy of thehigh frequency subimages reaches the least value. The choice of the proper number ofmultiresolution levels is reached when the energy ratio of detail subimages in twoconsecutive levels is less than 1. Subimage reconsturution rules are proposed to ensure thevalidity of the extracted features and simplify the computation at the optimaldecomposition level. The image fusion is carried out based on wavelet transform when theproper reconsturution subimages are established. The automatic selection rules fordifference distance and the direction angle of the SGLC are proposed for the continuity anddirectionality of the fabrics. The principal component analysis is executed to reduce theinformation redundancy of the 13 textural features. To synthetically present the featuresof different fabric defects, geometrical features should be extracted as well during thefeatures extraction procedure. The ratio of the maximum length and the maximum width ofthe defect is extracted from the defect image executed with the optimal thresholdsegmentation algorithm and morphology operation.
The unsuitability rules of neuron number in hidden layer and training method are proposed to optimize the structure of the fabric defects identification neural network andimprove the training speed, during the identification neural network design. The 3-layerneural network is trained with the fabric defect images before identification with 7, 16and 4 neurons in the input layer, the hidden layer, the output layer respectively, and thetransfer functions for hidden layer and output layer are Logarithmic sigmoid andHyperbolic tangent sigmoid transfer function. The average recognition accuracy of defectand nondefect are 99.2% and 100% respectively, under the experimental condition, and thedefects include warp-lacking, weft-lacking, double weft, loom bar, oil stain, hole.
在预处理中，采用4种不同的小波域降噪方法，对织物疵点图像进行降噪处理，以峰值信噪比为标准，选择出贝叶斯阈值和软阈值收缩准则相结合的方法降噪效果最佳。根据织物疵点的特殊形状结构，本文创新性地提出了基于小波变换的自适应高斯平滑化和拉普拉斯算子卷积作差相结合的方法对织物疵点图像进行增强，以消除由重复的结构纹理单元形成的背景，突出疵点部分。
根据织物纹理和结构的特殊性质，本文提出了同时提取纹理特征和几何形状特征作为神经网络识别参数，以提高疵点正确识别率。在纹理特征提取部分，本文提出了小波域和空域分析相结合的方法：小波变换和灰度共生矩阵相结合的方法。详细介绍了小波基的选择、最佳分解层的确定方法、子带重构规则、灰度共生矩阵重要参数的选择方法和疵点纹理特征值的主成分分析。以给定的具有不同织物组织结构的无疵点织物图像经小波分解后输出的图像的高通子图像能量最小作为逼近条件从合适的小波库中优选出最佳分解小波基。采用相邻两层的高频子带能量之比值小于1作为最佳分解层数的选择依据。在最佳分解层，为了保证特征提取源的有效性，并减化运算的复杂程度，在图像子带重构时，提出子带重构规则。根据重构规则自动选择子带进行重构，对重构后的子带采用基于特征的小波融合方法进行融合。根据织物具有连续性和方向性的特征，本文提出了灰度共生矩阵差分距离和方向角自适应选择规则。对由灰度共生矩阵中提取的13个纹理特征参数进行主成分分析，以消除信息冗余。为综合表达疵点特征，对增强后的织物疵点图像采用最佳阈值边缘检测和形态学运算，从二值化后的织物疵点图像中提取长、短径之比作为几何形状特征，以描述疵点的形状特征。
在织物疵点识别网络设计时，提出了隐含层神经元个数不适宜规则和网络训练方法不适宜规则，以优化网络结构，提高训练速度。采用结构为：输入层7个神经元，隐含层16个神经元，输出层4个神经元，中间层神经元的激励函数为S型正切函数，输出层神经元的激励函数为S型对数函数的神经网络对织物疵点图像进行训练和识别。在本实验条件下，采用本文算法，对丝织物中常见疵点：断纬、断经、重纬、档疵、油污、破洞进行识别，疵点平均识别率为99.2％，无疵点图像识别率为100％。
The identification of fabric defects based on digital image processing techniques can beconsidered as texture segmentation and pattern recognition problems. In this paper, theidentification of fabric defects based on discrete wavelet transform and Back-Propagationneural network is proposed, the indispensable processes of which are the defect imagespre-processing, wavelet transform, feature extraction and defects identification.
During the pre-processing procedure, four different denoising methods based on wavelettransform are used to improve the peak-signal-to-noise ratio (PSNR) of the fabric defectimages. From the comparison of the 4 different denoising methods, the method basedBayesian thresholding and soft thresholding shrinkage is much better than the others.According to the special structure of the defects, a creative method is proposed to enhancethe defect images, which adopts self-adoptive Gaussian filter and Laplacian operatorconvolution and subtraction operation. With the enhancement procedure, all regular,repetitive texture patterns can be eliminated, and the defect areas become more significant.
According to the special properties of the fabric defects, both textural and geometricalfeatures are extracted to improve the identification accuracy. During the textural featureextraction procedure, wavelet transform domain combined with the spatial domain methodis proposed, that is, wavelet transform combines with spatial gray level co-occurrence(SGLC) matrices methodology. The selection of optimal wavelet base, automatic selectionof multiresolution levels, reconsturution rules for subimages, automatic selection ofimportant parameters of the SGLC and the principal component analysis (PCA) of theextracted feature parameters are presented in details. The approximation condition ofselection of optimal wavelet base choosed from the wavelet base bank is the energy of thehigh frequency subimages reaches the least value. The choice of the proper number ofmultiresolution levels is reached when the energy ratio of detail subimages in twoconsecutive levels is less than 1. Subimage reconsturution rules are proposed to ensure thevalidity of the extracted features and simplify the computation at the optimaldecomposition level. The image fusion is carried out based on wavelet transform when theproper reconsturution subimages are established. The automatic selection rules fordifference distance and the direction angle of the SGLC are proposed for the continuity anddirectionality of the fabrics. The principal component analysis is executed to reduce theinformation redundancy of the 13 textural features. To synthetically present the featuresof different fabric defects, geometrical features should be extracted as well during thefeatures extraction procedure. The ratio of the maximum length and the maximum width ofthe defect is extracted from the defect image executed with the optimal thresholdsegmentation algorithm and morphology operation.
The unsuitability rules of neuron number in hidden layer and training method are proposed to optimize the structure of the fabric defects identification neural network andimprove the training speed, during the identification neural network design. The 3-layerneural network is trained with the fabric defect images before identification with 7, 16and 4 neurons in the input layer, the hidden layer, the output layer respectively, and thetransfer functions for hidden layer and output layer are Logarithmic sigmoid andHyperbolic tangent sigmoid transfer function. The average recognition accuracy of defectand nondefect are 99.2% and 100% respectively, under the experimental condition, and thedefects include warp-lacking, weft-lacking, double weft, loom bar, oil stain, hole.
引文
1．李立轻，黄秀宝．图像处理用于织物疵点自动检测的研究进展[J]．东华大学学报：自然科学版，2002，28(4)：118-122．
2. Bodnarova A, Bennamoun M and Kubik K K. Suitability Analysis of Techniques for Flaw Detection in Textiles using Texture Analysis[J]. Pattern Analysis & Applications ,2000(3):254-266.
3. Anagnostopoulos C, Vergados D, Kayafas E, et al. A Computer Vision Approach for Textile Quality Control [J]. The Journal of Visualization and Computer Animation, 2001,12: 31-44.
4. Sung Hoon Jenog, Hyung Taek Choi, Sook Rae Kim, et al. Detecting Fabric with Computer Vision and Fuzzy Rule Generation. Part Ⅰ: Defect Classification by Image Processing [J]. Textile Research Journal, 2001, 71 (6): 518-526.
5. Radovan Stojanovic, Panagiotis Mitropulos, Christos Koulamas, et al. Real-Time Vision-Based System for Textile Fabric Inspection[J]. Real-Time Imaging,2001(7):507-518.
6. Tilocca A, Borzone P, Carosio S, et al. Detecting fabric defects with a neural network using two kinds of optical patterns[J]. Textile Research Journal, 2002,72(6): 545-550.
7. Anganostopoulos C, Anagnostopoulos I, Vergados D, et al. Sliding Windows: A Software Method Suitable for Real-Time Inspection of Textile Surfaces[J]. Textile Res. J., 2004,74(4): 646-651.
8. Ahmed Abouelela, Hazem M. Abbas, Hesham Eldeeb, et al. Automated Vision System for Localizing Structural Defects in Textile Fabrics[J]. Pattern Recognition Letters. 2005, 26: 1435-1443.
9. Atiqul Islam Md, Shamim Akhter, Tamnun E. Mursalin, et al. A Suitable Neural Network to Detect Textile Defects[J]. ICONIP 2006, Part Ⅱ ,LNCS 4233: 430-438.
10. Latif-Amet A, Ertuzun A, Ercil A. An efficient method texture defect detection: sub-band domain co-occurrence matrices[J]. Image and Vision Computing, 2000,18:543-553.
11. Karras D A, Mertzios B G. Improved Defect Detection Using Novel Wavelet Feature Extraction Involving Principal Component Analysis[J]. Neural Network Techniques, AI 2002, LNAI2557: 638-647.
12. Ceu L. Beirao, Mario A.T. Figueiredo. Defect Detection in Textile Images Using Gabor Filters[J]. ICIAR 2004,LNCS 3212: 841-848.
13. Sungshin Kim, Hyeon Bae, Seong-Pyo Cheon, et al. On-Line Fabric-Defects Detection Based on Wavelet Analysis[J]. ICCSA 2005, LNCS 3483: 1075-1084.
14. Ghazi Saeidi R, Latifi M, Shaikhzadeh Najar S. Computer Vision-Aided Fabirc System for On-Circular Knitting Machine[J]. Textile Res.J., 2005,75(6):492-497.
15. Jasper W, Joines J, Brenzovich J. Fabric defect detection using a genetic algorithm tuned wavelet filter[J]. Journal of Textile Institute, 2005,96(1): 43-54.
16. Arivazhagan S, Ganesan L, Bama S. Fault Segmentation in Fabric Images Using Gabor Wavelet Transform[J]. Machine Vision and Application,2006,16(6): 356-363.
17．贡玉南，华建兴，黄秀宝．基于高斯-马尔可夫随机场模型的织物纹理特征参数提取方法[J]．中国纺织大学学报，1999，25(2)：1-4．
18. Gong Yunan, Hua Jianxing, Huang Xiubao. Textile Texture Modeling Using Random Field Models[J]. Journal of China University (Eng. Ed.), 1999,16(4): 56-59.
19．贡玉南，华建兴，黄秀宝．基于匹配Gabor滤波器的规则纹理缺陷检测方法[J]．中国图象图形学报，2001，6(7)：624-628．
20. Gong Yunan, Hua Jianxing, Huang Xiubao. Fabric Defect Detection Using GMRF Model[J]. Journal of China Textile University, 1999,16(3): 10-13.
21．李立轻，黄秀宝．用于疵点检测的织物自适应正交小波的实现[J]．东华大学学报：自然科学版，2002，28(2)：77-81．
22．李立轻，黄秀宝．基于织物自适应正交小波的疵点检测[J]．东华大学学报：自然科学版，2001，27(4)：82-87．
23. Li Liqing, Huang Xiubao. Realization of Orthogonal Wavelets Adapted to Fabric Texture for Defect Detection [J]. Journal of Dong Hua University, 19(4):52-56.
24. Li Liqing, Huang Xiubao. Fabric Defect Detection Using Adaptive Wavelet Transform[J]. Journal of Dong Hua University, 2002,19(1): 35-39.
25．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)：10-14．
26．徐增波，贡玉南，黄秀宝．基于Wold纹理模型和分形理论的织物疵点检测[J]．中国纺织大学学报，2000，26(1)：7-11．
27．杨晓波．基于连续小波变换的非结构化畸变织物疵点自动检测[J]．江南大学：自然科学版，2005，4(6)：630-633．
28. Liu Jianli, Zuo Baoqi. The Recognition of Fabric Defects Based on Discrete Wavelet Transform(DWT) and BP Neural Network[C]. The 35th Textile Research Symposium at Hangzhou. 2006, August: 22-26.
29．张瑞林，徐轶峰．基于PCNN的织物疵点识别研究[J]．纺织学报，2004，25(6)：69-72．
30．卿湘运，段红，魏俊民．一种新的基于小波分析与神经网络的织物疵点检测与识别方法[J]．仪器仪表学报，2005，26(6)：618-622．
31．韩武鹏，陈文楷，刘正耀．纺织品检测中的模式识别应用[J]．控制理论与应用2003，20(3)：391-393．
32．韩武鹏，陈文楷，刘正耀．模糊小波算法在纺织品瑕点检测中的应用[J]．北京工业大学学报，2002，28(1)：8-10．
33．高晓丁，高滨，左贺．基于支持矢量机的织物疵点识别算法[J]．纺织学报，2006，27(5)：26-33．
34．石美红，张军英，段亚峰．基于改进型PCNN的织物疵点检测的研究[J]．丝绸，2002，6：14-17．
35．易丽华．织物疵点检测工业视觉系统的研究[D]．武汉：华中科技大学，2004．
36．高水平．织物疵点检测的图像处理技术[D]．青岛：青岛大学，2005．
37．姜平，周根荣．基于计算机的织物疵点自动检测[J]．传感器与维系统，2006，25(1)：73-75．
38．郑广，周万珍，马红霞．基于图像距离差的织物疵点检测算法研究[J]．河北科技大学学报，2006，2 7(3)：239-241．
39．陈俊杰，谢春萍．基于神经网络的织物疵点识别技术[J]．纺织学报，2006，27(4)：36-38．
40. Shu Yuan, Tan Zheng. Fabric defects automatic detection using Gabor filters[C]. WCICA 2004-Fifth World Congress on Intelligent Control and Automation, Conference Proceedings, 2004:3378-3380:
41. Lee Tin-Chi, Pang Grantham. Defect detection on fabrics by matched filters[J]. IEEE TENCON 2004-2004 IEEE Region 10 Conference Proceedings: Analog and Digital Techniques in Electrical Engineering, 2004: 89-92.
42. Yang Xuezhi, Pang Grantham, Yung Nelson. Discriminative Training Approaches to Fabric Defect Classification Based on Wavelet Transform, Pattern Recognition[J]. 2004,(37):889-899.
43. Pang Grantham, Yang Xuezhi. Minimum Classification Error-Based Neural Network For Inspection of Fabric Defects[C]. Proceeding of the 2na IASIED International Conference, Neural Networks and Computational Intelligence, February 23-25, 2004: 196-200.
44. Henry Y.T. Ngan, Grantham K.H. Pang, S.P. Yung, et al. Wavelet based methods on patterned fabric defect detection[J]. Pattern Recognition,2005, 38: 559-576.
45. Chung-Feng Jeffrey Kuo, Ching-Jeng Lee. A Back-Propagation Neural Network for Recognizing Fabric Defects [J]. Textile Res.J. 2003,73(2): 147-151.
46. Chung-Feng, Jeffrey Kuo, Ching-Jeng Lee, et al. Using Neural Network to Identify Fabric Defects in Dynamic Cloth Inspection [J]. Textile Research Journal, 2003, 73(3):238-244.
47. Chung-Feng, Jeffrey Kuo, Te-Li Su. Gray Relational Analysis for Recognizing Fabric Defects[J]. Textile Research Journal, 2003, 73(5):461-465.
48. Du-Ming Ysai, Cheng-Huei Chiang. Automatic band selection for wavelet reconstruction in the application of defect detect[J]. Image and vision computing. 2003,21: 413-431.
49．汪军，李立轻，夏冬胜．一种自动验布等级评定系统：中国，200510110244．6[P]．2006．04．19．
50．徐益荪．针织机织物疵点自动检测装置：中国，90212451．X[P]．1991．06．19．
51. Meier R, Uhlmann J, Leuenberger R. Uster, Fabriscan automatic quality inspection system for fabrics[J]. Melliangd English, 1999,5: 96-97.
52. Textile World. 2001,151(4): 56-60 (EVS).
53. Barco introduces new logistics software module for its textiles Manufacturing Execution System, http://www.barco.com.cn/bu/index.asp?indexkey=textiles.
54．Schmalfub H，schinner K L，俞雯译．“自动在线织物检验”[J]．国际纺织导报，1999，3：52-56．
55．孙延奎．小波分析及其应用[M]．北京：机械工业出版社，2005：1-2．
56．侯遵泽，杨文采．小波分析应用研究[J]．物探化探计算技术，1995，17(3)：1-10．
57．薛全会，程秀芳，姚桂艳．小波分析的应用现状与前景[J]．河北理工学院学报，2006，28(1)：46-49．
58．费佩燕，刘曙光．小波分析应用的进展与展望[J]．纺织高校基础科学学报，2001， 14(1)：72-78．
59．刘素美，李书光．小波分析的理论发展及应用[J]．河北理工学院学报，2005，5(2)：60-65．
60．杨福生．小波变换的工程分析与应用[M]．北京：科学出版社，1999：3-10．
61．彭玉华．小波变换与工程应用[M]．北京：科学出版社，1999：2-8．
62. Mallat SG. Multiresolution approximations and wavelet orthonormal bases of L_2(IR). Trans. Amer. Math. 1989, Soc-315(1): 69-87.
63. Lei Tian-hu, Udupa Janyarm K. Performance evaluation of finite normal mixture model-based image segmentation techniques[J]. IEEE Transactions on Image Processing, 2003, 12(10): 1153-1169.
64．李旭超．小波域图像降噪概述[J]．中国图象图形学报，2006，11(9)：1201-1209．
65．冯志林，尹建伟，刘洋等．Allen-Cahn水平集的提花织物图像去噪研究[J]．浙江大学学报，2005，39(2)：185-189．
66. Wang D, Haese-Coat V, Bruno A, et al. Some statistical properties of mathematical morphology [J]. IEEE Transaction on Signal Processing, 1995,43 (8): 1955-1965.
67．狄红卫，许瑶．数学形态学在图像滤波中的应用[J]，暨南大学学报(自然科学版)，2003，24(3)：42-45．
68．刘志敏，杨杰．基于数学形态学的图像形态滤波[J]．红外与激光工程，1999，28(4)：10-16．
69．薛继伟，张问银，陈东芳，王桂华．基于数学形态学的图像滤波方法[J]．大庆石油学院学报，2003，27(4)：48-51．
70．石宏理，蔡远利，邱祖廉．小波域隐Markov树模型降噪算法的改进[J]．工程数学学报，2005，22(3)：449-455．
71．庞丈俊，李会方．一种基于小波域HMT模型的图像去噪方法研究[J]．信息安全与通信保密，2005，9：108-109．
72．李会方，庞文俊，徐瑞萍．一种基于P2DHMT的小波域图像去噪算法[J]，计算机工程与应用，2006，4：45-48．
73．李会方，孙颖力，庞文俊．基于HMT模型的图像去噪方法研究[J]．2006，27(2)：309-311．
74. Chang S G, Yu B, Vetterli M. Adaptive wavelet thresholding for image denoising and compression [J]. IEEE Transactions on Image Processing, 2000,9(9): 1532-546.
75. Shapiro J M. Embedded image coding using zerotress of wavelet coefficients [J]. IEEE Transactions on Signal Processing, 1993,41(12): 3445-3462.
76. Mallat S, Hwang W L. Singularity detection and processing with wavelets[J]. IEEE Transactions on Information Theory, 1992, 38(2): 617-643.
77. Mallat S, Zhong S. Characterization of signals from multiscale edges [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1992, 14(7): 710-732.
78. Donoho D L, Johnstone Iain M, Adapting to unknown smoothness via wavelet shrinkage[J]. Journal of the America Statistical Association, 1995,90(432): 1200-1224.
79. Donoho D L. Denoising by soft-thresholding [J].IEEE Transactions on Information Theory,1995,41(3): 613-627.
80. Donoho D L, Johnstone Iain M. Ideal spatial adaptation by wavelet shrinkage [J]. Biometrika, 1994, 81(3): 425-455.
81. Abramovich F, Sapatinas T, Silverman Y B. Wavelet thresholding via Bayesian approach[J]. Journal of the Royal Statistical Society, 1998, 60: 725-749.
82．胡海平，莫玉龙．基于贝叶斯估计的小波阈值图像降噪方法[J]．红外与毫米波学报，2002，21(1)：74-76．
83. Grace Chang, Bin Yu, Martin Vetterli. Adaptive wavelet thresholding for image denoising and compression[J], IEEE Trans. on Image Proc., 2000,9:1532-1546.
84．冯象初，甘小冰，宋国乡．数值泛函与小波理论[M]．西安：西安电子科技大学出版社，2003：102-107．
85．王怀野，张科，李言俊．一种自适应各项异性高斯滤波方法[J]，计算机工程与应用，2004，40(10)：18-19．
86．陈书海，傅录祥．实用数字图像处理[M]．北京：科学出版社，2005：159-260．
87．贡玉南，华建兴，黄秀宝．纺织品表面纹理的图像分析方法[J]，中国纺织大学学报，1998，24(2)：111-114．
88. Anagnostopoulos C, Abbas, Hazem, El Deeb, et al. High performance computer application for textile quality control[C]. Proc. International Conference on Intelligent Information Processing, Beijing, August 21-25,2000: 344-347.
89. Anagnostopoulos C, Anagnostopoulos I, Vergados D,et al. Sliding windows: a software method suitable for real-time inspection of textile surfaces [J]. Textile Res. J.2004, 74(7):646-651.
90. Errol W. Applying Fourier and associatiated transforms to pattern characterization in textiles [J]. Textile Res. J. 1990, 60: 212-220.
91. Zhang Y F, Bresee R R. Fabric defect detection and classification using image analysis[J]. Textile Res. J., 1995, 65(1): 1-9.
92. Bugao X. Identifying fabric structures with Fast Fourier Transform techniques[J]. Textile Res. J., 1996, 66: 496-506.
93．张长胜，司伟，段亚峰．用图像处理技术快速检测机织物的经纬密度[J]．毛纺科技，2006，2：46-49．
94. I-Shou Tsai, Ming-Chuan Hu. Automatic inspection of fabric defects using an artificial neural technique [J]. Textile Res. J., 1996, 66(7): 472-482.
95. Gabor D. Theory of Communication [J]. IEEE Comcon.1964, 93:429-457.
96．童辉，张晓美，何国金．小波和傅立叶变换在图像变化检测上的应用比较[J]．计算机工程与应用，2005，32：87-93．
97．李立轻，黄秀宝．图像处理用于织物疵点自动检测的研究进展[J]．东华大学学报：自然科学版，2002，28(4)：118-122．
98. Yang Xuezhi, Pang Grantham, Yung Nelson. Discriminative training approaches to fabric defect classification based on wavelet transform[J]. Pattern Recognition, 2004,37: 889-899.
99. Ngan Henry Y T, Pang Grantham K H, Yung S P, et al. Wavelet based methods on patterned fabric defect detection[J]. Pattern Recognition 2004, 38:559-576.
100．卿湘运，段红，魏俊民，王璐娟．一种新的基于小波分析与神经网络的织物疵点检测与识别方法[J]．仪器仪表学报，2005，26(6)：618-622．
101. Jasper W, Joines J, Brenzovich J. Fabric detect detection using a algorithm tuned wavelet filter[J]. JOTI, 2005, 96(1): 43-54.
102. Tae Jin Kang, Soo Chang Kim, In Hwan Sul, et al. Fabric surface roughness evaluation using wavelet-fractal method[J]. Textile Res. J. 75(11): 751-770.
103. Haralick, R M, Shanmugam K, Dinstein I. Textural features for image classification[J]. IEEE Transactions on Systems, 19.73(6): 610-621.
104. I-Shou T. Applying an artificial neural network to pattern recognition in fabric defects[J]. Textile Res. J. 1995,65:21-32.
105. Chung-feng Jeffery Kuo, Te-li Su. Gray relational analysis for recognizing fabric defects[J]. Textile Res. J. 2003, 73(5):461-465.
106. Bodnarova A, Bennamoun M. Suitability analysis of techniques for flaw detection in textiles using texture analysis [J]. Pattern Analysis & Applications, 2000(3):254-266.
107．孙红光，李勇．基于双正交小波弱目标边缘提取方法[J]．长春理工大学学报，2004，27(2)：53-55．
108. Daubechies I, Ten Lectures on Wavelets[M]. CBMS, SIAM, 61, 1994: 198-261.
109. Du-Ming Tsai, Cheng-Huei Chiang. Automatic band selection for wavelet reconstruction in the application of defect detection[J].Image and Vision Computing, 2003(21): 413-431.
110．崔岩梅，倪国强，钟堰利．利用统计特征性进行图像融合效果分析及评价[J]．北京理工大学学报，2000，20(1)：102-106．
111．杨虎，刘琼荪，钟波．数理统计[M]．北京：高等教育出版社，2004：199-207．
112．胡小锋，叶庆泰，戴星．一种精确检测细胞在真实边缘的算法[J]．光学技术，2003，29(5)：595-598．
113．步红刚，黄秀宝．基于计算机视觉的织物疵点分类的近期进展[J]．东华大学学报：自然科学版，2006年32(4)：143-146．
114. Balakrishnan H, Venkataranman S, Jayaraman S. A vision-based system for the identification and classification of fabric defects[J]. Journal of Textile Institute, 1998,89(2): 365-380.
115．王士同．神经模糊系统及其应用[M]．北京：北京航空航天大学出版社，1998：1-2．
116. Hyung Taek Choi, Sung Hoon Jeong. Detecting fabric defects with computer vision and fuzzy rule generation. Part Ⅱ: Defect Identification by a fuzzy expert system[J]. Textile Res. J. 2001,71(7):563-573.
117．张立明．人工神经网络的模型及其应用[M]．上海：复旦大学出版社，1990：18-19．
118．蒋宗礼．人工神经网络导论[M]．北京：高等教育出版社，2001：3-10．
119．谭显胜，周铁军．BP算法改进方法的研究进展[J]．怀化学院学报，2006，25(2)：126-130．
120. Andrea Zanela, Sergio Taraglio. Improving a Real-Time Neural-Based Stereo Vision System[J]. Real-Time Imaging, 2001,7(1): 59-76.
121. Hirose Y. BP algorithm which varies the number of hidden units[J]. Neural Network, 1991,4(1): 61-66.
122．高大启．有教师的线性基本函数前向三层神经网络结构研究[J]．计算机学报，1998，21(1)：80-86．
123．伍春香，刘琳，王葆元．三层BP网络隐层节点数确定方法的研究[J]．武汉测 绘科技大学学报，1998，24(2)：177-179．
124．张清良，李先明．一种确定神经网络隐层节点数的新方法[J]．吉首大学学报：自然科学版，2002，23(1)：89-92．
125．周红晓，蔡俊，任德官．一种优化多层前馈神经网络中隐节点数的算法[J]．浙江师范大学学报，2002，25(2)：268-271．
126. Hagan M T, Menhaj M B. Training feed forward networks with the Marquardt algorithm[J]. IEEE Trans on Neural Networks, 1994,5(6): 989-993.
127. Bernard Widrow, Rodney Winter. Neural nets for adaptive filtering and adaptive pattern recognition[J]. IEEE Computer Society, 1988(3): 25-39.
128. Reidmiller. Proceedings of the IEEE Int. Conf. on NN (ICNN) (C). San Francisco, 1993:586-591.
129. Scales. Introduction to Non-Linear Optimization [M]. 1985:125-135.
130. Powell. Mathematical Programming[J]. 1977(12): 241-254.
131. Gill. Murray & Wright, Practical Optimization[M]. 1981:312-314.
132. Battiti. Neural Computation[J]. 1992(4): 141-166.
133. MacKay.Neural Computation[J]. 1992,4(3): 415-447.
134. Moller. Neural Networks[J]. 1993(6):525-533.
2. Bodnarova A, Bennamoun M and Kubik K K. Suitability Analysis of Techniques for Flaw Detection in Textiles using Texture Analysis[J]. Pattern Analysis & Applications ,2000(3):254-266.
3. Anagnostopoulos C, Vergados D, Kayafas E, et al. A Computer Vision Approach for Textile Quality Control [J]. The Journal of Visualization and Computer Animation, 2001,12: 31-44.
4. Sung Hoon Jenog, Hyung Taek Choi, Sook Rae Kim, et al. Detecting Fabric with Computer Vision and Fuzzy Rule Generation. Part Ⅰ: Defect Classification by Image Processing [J]. Textile Research Journal, 2001, 71 (6): 518-526.
5. Radovan Stojanovic, Panagiotis Mitropulos, Christos Koulamas, et al. Real-Time Vision-Based System for Textile Fabric Inspection[J]. Real-Time Imaging,2001(7):507-518.
6. Tilocca A, Borzone P, Carosio S, et al. Detecting fabric defects with a neural network using two kinds of optical patterns[J]. Textile Research Journal, 2002,72(6): 545-550.
7. Anganostopoulos C, Anagnostopoulos I, Vergados D, et al. Sliding Windows: A Software Method Suitable for Real-Time Inspection of Textile Surfaces[J]. Textile Res. J., 2004,74(4): 646-651.
8. Ahmed Abouelela, Hazem M. Abbas, Hesham Eldeeb, et al. Automated Vision System for Localizing Structural Defects in Textile Fabrics[J]. Pattern Recognition Letters. 2005, 26: 1435-1443.
9. Atiqul Islam Md, Shamim Akhter, Tamnun E. Mursalin, et al. A Suitable Neural Network to Detect Textile Defects[J]. ICONIP 2006, Part Ⅱ ,LNCS 4233: 430-438.
10. Latif-Amet A, Ertuzun A, Ercil A. An efficient method texture defect detection: sub-band domain co-occurrence matrices[J]. Image and Vision Computing, 2000,18:543-553.
11. Karras D A, Mertzios B G. Improved Defect Detection Using Novel Wavelet Feature Extraction Involving Principal Component Analysis[J]. Neural Network Techniques, AI 2002, LNAI2557: 638-647.
12. Ceu L. Beirao, Mario A.T. Figueiredo. Defect Detection in Textile Images Using Gabor Filters[J]. ICIAR 2004,LNCS 3212: 841-848.
13. Sungshin Kim, Hyeon Bae, Seong-Pyo Cheon, et al. On-Line Fabric-Defects Detection Based on Wavelet Analysis[J]. ICCSA 2005, LNCS 3483: 1075-1084.
14. Ghazi Saeidi R, Latifi M, Shaikhzadeh Najar S. Computer Vision-Aided Fabirc System for On-Circular Knitting Machine[J]. Textile Res.J., 2005,75(6):492-497.
15. Jasper W, Joines J, Brenzovich J. Fabric defect detection using a genetic algorithm tuned wavelet filter[J]. Journal of Textile Institute, 2005,96(1): 43-54.
16. Arivazhagan S, Ganesan L, Bama S. Fault Segmentation in Fabric Images Using Gabor Wavelet Transform[J]. Machine Vision and Application,2006,16(6): 356-363.
17．贡玉南，华建兴，黄秀宝．基于高斯-马尔可夫随机场模型的织物纹理特征参数提取方法[J]．中国纺织大学学报，1999，25(2)：1-4．
18. Gong Yunan, Hua Jianxing, Huang Xiubao. Textile Texture Modeling Using Random Field Models[J]. Journal of China University (Eng. Ed.), 1999,16(4): 56-59.
19．贡玉南，华建兴，黄秀宝．基于匹配Gabor滤波器的规则纹理缺陷检测方法[J]．中国图象图形学报，2001，6(7)：624-628．
20. Gong Yunan, Hua Jianxing, Huang Xiubao. Fabric Defect Detection Using GMRF Model[J]. Journal of China Textile University, 1999,16(3): 10-13.
21．李立轻，黄秀宝．用于疵点检测的织物自适应正交小波的实现[J]．东华大学学报：自然科学版，2002，28(2)：77-81．
22．李立轻，黄秀宝．基于织物自适应正交小波的疵点检测[J]．东华大学学报：自然科学版，2001，27(4)：82-87．
23. Li Liqing, Huang Xiubao. Realization of Orthogonal Wavelets Adapted to Fabric Texture for Defect Detection [J]. Journal of Dong Hua University, 19(4):52-56.
24. Li Liqing, Huang Xiubao. Fabric Defect Detection Using Adaptive Wavelet Transform[J]. Journal of Dong Hua University, 2002,19(1): 35-39.
25．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)：10-14．
26．徐增波，贡玉南，黄秀宝．基于Wold纹理模型和分形理论的织物疵点检测[J]．中国纺织大学学报，2000，26(1)：7-11．
27．杨晓波．基于连续小波变换的非结构化畸变织物疵点自动检测[J]．江南大学：自然科学版，2005，4(6)：630-633．
28. Liu Jianli, Zuo Baoqi. The Recognition of Fabric Defects Based on Discrete Wavelet Transform(DWT) and BP Neural Network[C]. The 35th Textile Research Symposium at Hangzhou. 2006, August: 22-26.
29．张瑞林，徐轶峰．基于PCNN的织物疵点识别研究[J]．纺织学报，2004，25(6)：69-72．
30．卿湘运，段红，魏俊民．一种新的基于小波分析与神经网络的织物疵点检测与识别方法[J]．仪器仪表学报，2005，26(6)：618-622．
31．韩武鹏，陈文楷，刘正耀．纺织品检测中的模式识别应用[J]．控制理论与应用2003，20(3)：391-393．
32．韩武鹏，陈文楷，刘正耀．模糊小波算法在纺织品瑕点检测中的应用[J]．北京工业大学学报，2002，28(1)：8-10．
33．高晓丁，高滨，左贺．基于支持矢量机的织物疵点识别算法[J]．纺织学报，2006，27(5)：26-33．
34．石美红，张军英，段亚峰．基于改进型PCNN的织物疵点检测的研究[J]．丝绸，2002，6：14-17．
35．易丽华．织物疵点检测工业视觉系统的研究[D]．武汉：华中科技大学，2004．
36．高水平．织物疵点检测的图像处理技术[D]．青岛：青岛大学，2005．
37．姜平，周根荣．基于计算机的织物疵点自动检测[J]．传感器与维系统，2006，25(1)：73-75．
38．郑广，周万珍，马红霞．基于图像距离差的织物疵点检测算法研究[J]．河北科技大学学报，2006，2 7(3)：239-241．
39．陈俊杰，谢春萍．基于神经网络的织物疵点识别技术[J]．纺织学报，2006，27(4)：36-38．
40. Shu Yuan, Tan Zheng. Fabric defects automatic detection using Gabor filters[C]. WCICA 2004-Fifth World Congress on Intelligent Control and Automation, Conference Proceedings, 2004:3378-3380:
41. Lee Tin-Chi, Pang Grantham. Defect detection on fabrics by matched filters[J]. IEEE TENCON 2004-2004 IEEE Region 10 Conference Proceedings: Analog and Digital Techniques in Electrical Engineering, 2004: 89-92.
42. Yang Xuezhi, Pang Grantham, Yung Nelson. Discriminative Training Approaches to Fabric Defect Classification Based on Wavelet Transform, Pattern Recognition[J]. 2004,(37):889-899.
43. Pang Grantham, Yang Xuezhi. Minimum Classification Error-Based Neural Network For Inspection of Fabric Defects[C]. Proceeding of the 2na IASIED International Conference, Neural Networks and Computational Intelligence, February 23-25, 2004: 196-200.
44. Henry Y.T. Ngan, Grantham K.H. Pang, S.P. Yung, et al. Wavelet based methods on patterned fabric defect detection[J]. Pattern Recognition,2005, 38: 559-576.
45. Chung-Feng Jeffrey Kuo, Ching-Jeng Lee. A Back-Propagation Neural Network for Recognizing Fabric Defects [J]. Textile Res.J. 2003,73(2): 147-151.
46. Chung-Feng, Jeffrey Kuo, Ching-Jeng Lee, et al. Using Neural Network to Identify Fabric Defects in Dynamic Cloth Inspection [J]. Textile Research Journal, 2003, 73(3):238-244.
47. Chung-Feng, Jeffrey Kuo, Te-Li Su. Gray Relational Analysis for Recognizing Fabric Defects[J]. Textile Research Journal, 2003, 73(5):461-465.
48. Du-Ming Ysai, Cheng-Huei Chiang. Automatic band selection for wavelet reconstruction in the application of defect detect[J]. Image and vision computing. 2003,21: 413-431.
49．汪军，李立轻，夏冬胜．一种自动验布等级评定系统：中国，200510110244．6[P]．2006．04．19．
50．徐益荪．针织机织物疵点自动检测装置：中国，90212451．X[P]．1991．06．19．
51. Meier R, Uhlmann J, Leuenberger R. Uster, Fabriscan automatic quality inspection system for fabrics[J]. Melliangd English, 1999,5: 96-97.
52. Textile World. 2001,151(4): 56-60 (EVS).
53. Barco introduces new logistics software module for its textiles Manufacturing Execution System, http://www.barco.com.cn/bu/index.asp?indexkey=textiles.
54．Schmalfub H，schinner K L，俞雯译．“自动在线织物检验”[J]．国际纺织导报，1999，3：52-56．
55．孙延奎．小波分析及其应用[M]．北京：机械工业出版社，2005：1-2．
56．侯遵泽，杨文采．小波分析应用研究[J]．物探化探计算技术，1995，17(3)：1-10．
57．薛全会，程秀芳，姚桂艳．小波分析的应用现状与前景[J]．河北理工学院学报，2006，28(1)：46-49．
58．费佩燕，刘曙光．小波分析应用的进展与展望[J]．纺织高校基础科学学报，2001， 14(1)：72-78．
59．刘素美，李书光．小波分析的理论发展及应用[J]．河北理工学院学报，2005，5(2)：60-65．
60．杨福生．小波变换的工程分析与应用[M]．北京：科学出版社，1999：3-10．
61．彭玉华．小波变换与工程应用[M]．北京：科学出版社，1999：2-8．
62. Mallat SG. Multiresolution approximations and wavelet orthonormal bases of L_2(IR). Trans. Amer. Math. 1989, Soc-315(1): 69-87.
63. Lei Tian-hu, Udupa Janyarm K. Performance evaluation of finite normal mixture model-based image segmentation techniques[J]. IEEE Transactions on Image Processing, 2003, 12(10): 1153-1169.
64．李旭超．小波域图像降噪概述[J]．中国图象图形学报，2006，11(9)：1201-1209．
65．冯志林，尹建伟，刘洋等．Allen-Cahn水平集的提花织物图像去噪研究[J]．浙江大学学报，2005，39(2)：185-189．
66. Wang D, Haese-Coat V, Bruno A, et al. Some statistical properties of mathematical morphology [J]. IEEE Transaction on Signal Processing, 1995,43 (8): 1955-1965.
67．狄红卫，许瑶．数学形态学在图像滤波中的应用[J]，暨南大学学报(自然科学版)，2003，24(3)：42-45．
68．刘志敏，杨杰．基于数学形态学的图像形态滤波[J]．红外与激光工程，1999，28(4)：10-16．
69．薛继伟，张问银，陈东芳，王桂华．基于数学形态学的图像滤波方法[J]．大庆石油学院学报，2003，27(4)：48-51．
70．石宏理，蔡远利，邱祖廉．小波域隐Markov树模型降噪算法的改进[J]．工程数学学报，2005，22(3)：449-455．
71．庞丈俊，李会方．一种基于小波域HMT模型的图像去噪方法研究[J]．信息安全与通信保密，2005，9：108-109．
72．李会方，庞文俊，徐瑞萍．一种基于P2DHMT的小波域图像去噪算法[J]，计算机工程与应用，2006，4：45-48．
73．李会方，孙颖力，庞文俊．基于HMT模型的图像去噪方法研究[J]．2006，27(2)：309-311．
74. Chang S G, Yu B, Vetterli M. Adaptive wavelet thresholding for image denoising and compression [J]. IEEE Transactions on Image Processing, 2000,9(9): 1532-546.
75. Shapiro J M. Embedded image coding using zerotress of wavelet coefficients [J]. IEEE Transactions on Signal Processing, 1993,41(12): 3445-3462.
76. Mallat S, Hwang W L. Singularity detection and processing with wavelets[J]. IEEE Transactions on Information Theory, 1992, 38(2): 617-643.
77. Mallat S, Zhong S. Characterization of signals from multiscale edges [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1992, 14(7): 710-732.
78. Donoho D L, Johnstone Iain M, Adapting to unknown smoothness via wavelet shrinkage[J]. Journal of the America Statistical Association, 1995,90(432): 1200-1224.
79. Donoho D L. Denoising by soft-thresholding [J].IEEE Transactions on Information Theory,1995,41(3): 613-627.
80. Donoho D L, Johnstone Iain M. Ideal spatial adaptation by wavelet shrinkage [J]. Biometrika, 1994, 81(3): 425-455.
81. Abramovich F, Sapatinas T, Silverman Y B. Wavelet thresholding via Bayesian approach[J]. Journal of the Royal Statistical Society, 1998, 60: 725-749.
82．胡海平，莫玉龙．基于贝叶斯估计的小波阈值图像降噪方法[J]．红外与毫米波学报，2002，21(1)：74-76．
83. Grace Chang, Bin Yu, Martin Vetterli. Adaptive wavelet thresholding for image denoising and compression[J], IEEE Trans. on Image Proc., 2000,9:1532-1546.
84．冯象初，甘小冰，宋国乡．数值泛函与小波理论[M]．西安：西安电子科技大学出版社，2003：102-107．
85．王怀野，张科，李言俊．一种自适应各项异性高斯滤波方法[J]，计算机工程与应用，2004，40(10)：18-19．
86．陈书海，傅录祥．实用数字图像处理[M]．北京：科学出版社，2005：159-260．
87．贡玉南，华建兴，黄秀宝．纺织品表面纹理的图像分析方法[J]，中国纺织大学学报，1998，24(2)：111-114．
88. Anagnostopoulos C, Abbas, Hazem, El Deeb, et al. High performance computer application for textile quality control[C]. Proc. International Conference on Intelligent Information Processing, Beijing, August 21-25,2000: 344-347.
89. Anagnostopoulos C, Anagnostopoulos I, Vergados D,et al. Sliding windows: a software method suitable for real-time inspection of textile surfaces [J]. Textile Res. J.2004, 74(7):646-651.
90. Errol W. Applying Fourier and associatiated transforms to pattern characterization in textiles [J]. Textile Res. J. 1990, 60: 212-220.
91. Zhang Y F, Bresee R R. Fabric defect detection and classification using image analysis[J]. Textile Res. J., 1995, 65(1): 1-9.
92. Bugao X. Identifying fabric structures with Fast Fourier Transform techniques[J]. Textile Res. J., 1996, 66: 496-506.
93．张长胜，司伟，段亚峰．用图像处理技术快速检测机织物的经纬密度[J]．毛纺科技，2006，2：46-49．
94. I-Shou Tsai, Ming-Chuan Hu. Automatic inspection of fabric defects using an artificial neural technique [J]. Textile Res. J., 1996, 66(7): 472-482.
95. Gabor D. Theory of Communication [J]. IEEE Comcon.1964, 93:429-457.
96．童辉，张晓美，何国金．小波和傅立叶变换在图像变化检测上的应用比较[J]．计算机工程与应用，2005，32：87-93．
97．李立轻，黄秀宝．图像处理用于织物疵点自动检测的研究进展[J]．东华大学学报：自然科学版，2002，28(4)：118-122．
98. Yang Xuezhi, Pang Grantham, Yung Nelson. Discriminative training approaches to fabric defect classification based on wavelet transform[J]. Pattern Recognition, 2004,37: 889-899.
99. Ngan Henry Y T, Pang Grantham K H, Yung S P, et al. Wavelet based methods on patterned fabric defect detection[J]. Pattern Recognition 2004, 38:559-576.
100．卿湘运，段红，魏俊民，王璐娟．一种新的基于小波分析与神经网络的织物疵点检测与识别方法[J]．仪器仪表学报，2005，26(6)：618-622．
101. Jasper W, Joines J, Brenzovich J. Fabric detect detection using a algorithm tuned wavelet filter[J]. JOTI, 2005, 96(1): 43-54.
102. Tae Jin Kang, Soo Chang Kim, In Hwan Sul, et al. Fabric surface roughness evaluation using wavelet-fractal method[J]. Textile Res. J. 75(11): 751-770.
103. Haralick, R M, Shanmugam K, Dinstein I. Textural features for image classification[J]. IEEE Transactions on Systems, 19.73(6): 610-621.
104. I-Shou T. Applying an artificial neural network to pattern recognition in fabric defects[J]. Textile Res. J. 1995,65:21-32.
105. Chung-feng Jeffery Kuo, Te-li Su. Gray relational analysis for recognizing fabric defects[J]. Textile Res. J. 2003, 73(5):461-465.
106. Bodnarova A, Bennamoun M. Suitability analysis of techniques for flaw detection in textiles using texture analysis [J]. Pattern Analysis & Applications, 2000(3):254-266.
107．孙红光，李勇．基于双正交小波弱目标边缘提取方法[J]．长春理工大学学报，2004，27(2)：53-55．
108. Daubechies I, Ten Lectures on Wavelets[M]. CBMS, SIAM, 61, 1994: 198-261.
109. Du-Ming Tsai, Cheng-Huei Chiang. Automatic band selection for wavelet reconstruction in the application of defect detection[J].Image and Vision Computing, 2003(21): 413-431.
110．崔岩梅，倪国强，钟堰利．利用统计特征性进行图像融合效果分析及评价[J]．北京理工大学学报，2000，20(1)：102-106．
111．杨虎，刘琼荪，钟波．数理统计[M]．北京：高等教育出版社，2004：199-207．
112．胡小锋，叶庆泰，戴星．一种精确检测细胞在真实边缘的算法[J]．光学技术，2003，29(5)：595-598．
113．步红刚，黄秀宝．基于计算机视觉的织物疵点分类的近期进展[J]．东华大学学报：自然科学版，2006年32(4)：143-146．
114. Balakrishnan H, Venkataranman S, Jayaraman S. A vision-based system for the identification and classification of fabric defects[J]. Journal of Textile Institute, 1998,89(2): 365-380.
115．王士同．神经模糊系统及其应用[M]．北京：北京航空航天大学出版社，1998：1-2．
116. Hyung Taek Choi, Sung Hoon Jeong. Detecting fabric defects with computer vision and fuzzy rule generation. Part Ⅱ: Defect Identification by a fuzzy expert system[J]. Textile Res. J. 2001,71(7):563-573.
117．张立明．人工神经网络的模型及其应用[M]．上海：复旦大学出版社，1990：18-19．
118．蒋宗礼．人工神经网络导论[M]．北京：高等教育出版社，2001：3-10．
119．谭显胜，周铁军．BP算法改进方法的研究进展[J]．怀化学院学报，2006，25(2)：126-130．
120. Andrea Zanela, Sergio Taraglio. Improving a Real-Time Neural-Based Stereo Vision System[J]. Real-Time Imaging, 2001,7(1): 59-76.
121. Hirose Y. BP algorithm which varies the number of hidden units[J]. Neural Network, 1991,4(1): 61-66.
122．高大启．有教师的线性基本函数前向三层神经网络结构研究[J]．计算机学报，1998，21(1)：80-86．
123．伍春香，刘琳，王葆元．三层BP网络隐层节点数确定方法的研究[J]．武汉测 绘科技大学学报，1998，24(2)：177-179．
124．张清良，李先明．一种确定神经网络隐层节点数的新方法[J]．吉首大学学报：自然科学版，2002，23(1)：89-92．
125．周红晓，蔡俊，任德官．一种优化多层前馈神经网络中隐节点数的算法[J]．浙江师范大学学报，2002，25(2)：268-271．
126. Hagan M T, Menhaj M B. Training feed forward networks with the Marquardt algorithm[J]. IEEE Trans on Neural Networks, 1994,5(6): 989-993.
127. Bernard Widrow, Rodney Winter. Neural nets for adaptive filtering and adaptive pattern recognition[J]. IEEE Computer Society, 1988(3): 25-39.
128. Reidmiller. Proceedings of the IEEE Int. Conf. on NN (ICNN) (C). San Francisco, 1993:586-591.
129. Scales. Introduction to Non-Linear Optimization [M]. 1985:125-135.
130. Powell. Mathematical Programming[J]. 1977(12): 241-254.
131. Gill. Murray & Wright, Practical Optimization[M]. 1981:312-314.
132. Battiti. Neural Computation[J]. 1992(4): 141-166.
133. MacKay.Neural Computation[J]. 1992,4(3): 415-447.
134. Moller. Neural Networks[J]. 1993(6):525-533.