应用深度卷积神经网络的色织物缺陷检测

doi:10.13475/j.fzxb.20161001707

摘要/Abstract

摘要：

针对织物缺陷检测时传统人工的误检率、漏检率较高问题，提出一种应用深度卷积神经网络的色织物缺陷检测算法。因织物图像采集过程中含有较多噪声且信噪比较低，先对缺陷织物进行最优尺寸高斯滤波，有效滤除细节噪声；再根据织物图像特征建立深度卷积神经网络，利用径向基神经网络的非线性映射能力作用于卷积神经网络，并通过反向传播算法调整权值参数，获取无缺陷样本与训练样本之间的映射函数；最后，利用映射函数及特征字典重构图像并提取特征，根据Meanshift算法分割缺陷，确定缺陷位置。结果表明：应用深度卷积神经网络的缺陷检测算法对色织物图像库中的缺陷图像可实现提高检测效率、缩短检测时间，获取准确缺陷位置的目的。

关键词: 色织物, 图像库, 缺陷检测, 深度卷积神经网络, 映射函数

Abstract:

Focusing on the problems of high error detection and omission rate of traditional artificial fabric defect detection，this paper presents a yarn-dyed fabric defect detection method, which based on the deep-convolutional neural network. The fabric image contains much noise and has low signal noise ratio (SNR), and optimal dimension Gauss filter as preprocessing is conducted firstly for the sampled images to remove the detailed noise. Secondly, the deep-convolutional neural network is established based on the features of fabric samples, nonlinear mapping ability of radial basis function neural network acts upon convolutional neural network, weight parameters are adjusted via back propagation algorithm, and a mapping function between defect free samples and training samples can be obtained. Finally, the mapping function and features dictionary are used to reconstruct image and extract features, according to the Meanshift algorithm to segment the defects and determine the fabric defect position by two value. The experimental results demonstrate that the method based on the deep-convolutional neural network can achieve the purpose of improving efficiency, shortening the time of measurement, and obtaining an accurate defect image.

Key words: yarn-dyed fabric, image library, defect detection, deep-convolutional neural network, mapping funcition

景军锋范晓婷李鹏飞洪良. 应用深度卷积神经网络的色织物缺陷检测[J]. 纺织学报, 2017, 38(02): 68-74.

参考文献

[1] KUMAR A. Computer-vision-based fabric defect detection: a survey [J]. Industrial Electronics, IEEE Transactions on, 2008,55(1): 348-363. [2]ZHANG Y, JIANG G, YAO J, et al. Segmentation of jacquard warp-knitted fabric image based on hierarchical Markov random field model[J]. Journal of Textile Research, 2012, 33 (12): 102-106. [3] 潘如如, 高卫东, 钱欣欣等. 基于互相关的印花织物疵点检测[J]. 纺织学报, 2010, 31(12):134-138. [3]潘如如, 高卫东, 钱欣欣等. 基于互相关的印花织物疵点检测[J]. 纺织学报, 2010, 31(12):134-138. PAN Ruru, GAO Weidong, QIAN Xinxin et al. Defect detection of printed fabrics using normalized cross correlation[J]. Journal of Textile Research, 2010, 31(12):134-138. [4]NGAN H Y T, PANG G K H, YUNG S P, et al. Wavelet based methods on patterned fabric defect detection[J]. Pattern Recognition, 2005, 38(4): 559–576. [5]NGAN H Y T, YUNG S P, Automated fabric defect detection-A review, Image and Vision Computing, 2011, 29(7): 442-458. [6] NG M K, NGAN H Y T, YUAN X, et al. Patterned Fabric Inspection and Visualization by the Method of Image Decomposition[J]. IEEE Transactions on Automation Science & Engineering, 2014, 11(3):943 – 947. [7] 景军锋, 范晓婷, 李鹏飞等. 应用Gussian回代交替方向图像分解算法的色织物疵点检测[J]. 纺织学报, 2016,37(6):136-141. JING Junfeng, FAN Xiaoting, LI Pengfei et al. Yarn-dyed fabric defect detection based on Gaussian back substitution image decomposition[J]. Journal of Textile Research, 2016, 37(6):136-141. [8] NGAN H Y T, PANG G K H. Novel method for patterned fabric inspection using Bollinger bands[J]. Optical Engineering, 2006, 45(8): 087202-087217. [9] NGAN H Y T, PANG G K H. Regularity analysis for patterned texture inspection[J]. IEEE Transactions on Automation Science & Engineering, 2009, 6(1):131 - 144. [10]TAJERIPOUR F, KABIR E, SHEIKHI A. Defect Detection in Patterned Fabrics Using Modified Local Binary Patterns[C], IEEE International Conference on Computational Intelligence and Multimedia Applications, 2007,2:263-267. [11]NGAN H Y T, PANG G K H, YUNG N H C. Performance Evaluation for Motif-Based Patterned Texture Defect Detection[J]. IEEE Transactions on Automation Science & Engineering, 2010, 7(1):58-72. [12]朱丹丹. 基于图像分析的色织物疵点检测研究[D]. 江南大学, 2014. ZHU Dandan. Research of Defect Detection for Yarn-dyed Fabric Based on Image Analysis[D]. Jiangnan University, 2014. [13] 李文羽. 基于机器视觉和图像处理的色织物疵点自动检测研究[D]. 东华大学, 2014. LI Wenyu. Research on Automatic Detection for Yarn-dyed Fabirc Defect Based on Machine Vision and Image Processing[D]. Donghua University, 2014. [14]JING J, LIU S, LI P, et al. The fabric defect detection based on CIE L* a* b* color space using 2-D Gabor filter[J]. The Journal of The Textile Institute, 2015，107(10): 1-9. [15]王文远. 基于图像信噪比选择优化高斯滤波尺度[J]. 电子与信息学报, 2009, 31(10):2483-2487. WANG Wenyuan, Selecting the Optimal Gaussian Filtering Scale via the SNR of Image[J]. Journal of Electronics&Information Technology, 2009, 31(10): 2483-2487. [16] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet Classification with Deep Convolutional Neural Networks[J]. Advances in Neural Information Processing Systems, 2012, 25(2):1-9. [17] 余永维, 殷国富, 殷鹰等. 基于深度学习网络的射线图像缺陷识别方法[J]. 仪器仪表学报, 2014, 35(9): 2012- 2019. YU Yongwei, YIN Guofu, YIN Ying et al. Defect recognition for radiographic image based on deep learning network[J]. Chinese Journal of Scientific Instrument, 2014, 35(9):2012-2019. [18] WllAMPWSKI B M, CECATUI C, KOLBYSZ J, et al. A Novel RBF Training Algorithm for Short-term Electric Load Forecasting and Comparative Studies[J]. IEEE Transactions on Industrial Electronics, 2015, 62(10):1-1. [19] TAO W, JIN H, ZHANG Y. Color image segmentation based on mean shift and normalized cuts[J]. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2007, 37(5): 1382-1389.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed