基于改进判别性完整局部二值模式与格分割的织物瑕疵检测方法
|
摘要
为解决传统的完整局部二值模式在织物疵点检测时存在直方图维数过高和特征冗余并且在小区域图像变化幅度剧烈或变化幅度平缓时存在局限性的问题,提出一种改进判别性完整局部二值模式并结合自动格分割的织物瑕疵检测方法,该新算法可分为训练和测试2部分。通过实验将该算法、小波预处理的黄金图像相减方法、布林线指标方法、正则带方法进行对比,针对2种纹理3类瑕疵的织物图像数据集进行测试。结果表明,该方法对星形图案和箱形图案纺织品检测效果较好,一部分的查全率可达到0.99,大部分检测结果的查全率均在0.90以上。
The conventional central local binarization mode( CLBP) used in fabric defect inspection has the problems of high histogram dimension and feature redundancy,and limitation exists in conventional CLBP when the amplitude of the small part of the image varies greatly or the amplitude is flat. To solve the problems,a modified discriminant complete local binary pattern with lattice segmentation for fabric defect inspection was proposed. The proposed algorithm was divided into a training part and a testing part. The training stage was to calculate the feature value for each lattice after lattice segmentation in defect-free images and acquire the mean value of all feature values. The threshold was calculated by calculating the relative divergence between the feature value of every lattice and the mean of the feature values. The testing stage was to calculate the relative divergence and compare the result with the threshold. The lattice whose result was larger than the threshold was marked as a defect area. The proposed algorithm was compared with local binary patterns,boolean line indicator method,regular band method algorithms. Testing on fabric image datasets including 2 kinds of textures and 3 kinds of defects shows that the method has better inspection effect on star pattern and box pattern fabrics,one part of the true positive rate( TPR) value can reach 0. 99,and most of the inspection results of TPR are above 0. 90.
The conventional central local binarization mode( CLBP) used in fabric defect inspection has the problems of high histogram dimension and feature redundancy,and limitation exists in conventional CLBP when the amplitude of the small part of the image varies greatly or the amplitude is flat. To solve the problems,a modified discriminant complete local binary pattern with lattice segmentation for fabric defect inspection was proposed. The proposed algorithm was divided into a training part and a testing part. The training stage was to calculate the feature value for each lattice after lattice segmentation in defect-free images and acquire the mean value of all feature values. The threshold was calculated by calculating the relative divergence between the feature value of every lattice and the mean of the feature values. The testing stage was to calculate the relative divergence and compare the result with the threshold. The lattice whose result was larger than the threshold was marked as a defect area. The proposed algorithm was compared with local binary patterns,boolean line indicator method,regular band method algorithms. Testing on fabric image datasets including 2 kinds of textures and 3 kinds of defects shows that the method has better inspection effect on star pattern and box pattern fabrics,one part of the true positive rate( TPR) value can reach 0. 99,and most of the inspection results of TPR are above 0. 90.
引文
[1]CHO C S,CHUNG B M,PARK M J.Development of real-time vision-based fabric inspection system[J].IEEE Transactions on Industrial Electronics,2005,52(4):1073-1079.
[2]董蓉,李勃,徐晨,等.应用积分图的织物瑕疵检测快速算法[J].纺织学报,2016,37(11):141-147.DONG Rong,LI Bo,XU Chen,et al.Fast fabric defect detection algorithm based on integral image[J].Journal of Textile Research,2016,37(11):141-147.
[3]石美红,张正,郭仙草,等.基于显著纹理特征的织物疵点检测方法[J].纺织学报,2016,37(10):42-49.SHI Meihong,ZHANG Zheng,GUO Xiancao,et al.Fabric defect detection method based on texture saliency feature.[J].Journal of Textile Research,2016,37(10):42-49.
[4]NGAN H Y T,PANG G K H,YUNG N H C,et al.Wavelet based methods on patterned fabric defect detection[J].Pattern Recognition,2005,38(4):559-576.
[5]NGAN H Y T,PANG G K H.Novel method for patterned fabric inspection using bollinger bands[J].Optical Engineering,2006,45(8):112-115.
[6]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.
[7]NGAN H Y T,PANG G K H,YUNG N H C.Motifbased defect detection for patterned fabric[J].Pattern Recognition,2008,41(6):1878-1894.
[8]TSANG C S C,NGAN H Y T,PANG G K H.Fabric inspection based on the Elo rating method[J].Pattern Recognition,2016,51(4):378-394.
[9]JIA Liang,CHEN Chen,LIANG Jiuzhen,et al.Fabric defect inspection based on lattice segmentation and Gabor filtering[J].Neurocomputing,2017,238:84-102.
[10]GUO Z,ZHANG D.A completed modeling of local binary pattern operator for texture classification[J].IEEE Transactions on Image Processing,2010,19(6):1657-1663.
[11]GUO Y,ZHAO G,PIETIKINEN M,et al.Descriptor learning based on fisher separation criterion for texture classification[C]//Computer Vision-ACCV 2010.Berlin Heidel-berg:Springer,2011:185-198.
[12]OJALA T,PIETIKAINEN M,MAENPAA T.Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(7):971-987.
[13]WEN J,ZHAO J,WAN C.Improved morphological component analysis for interference hyperspectral image decomposition[J].Computer Electrical Engineering.2015,46(1):394-402.
[14]JIA Liang,LIANG Jiuzhen.Fabric defect inspection based on isotropic lattice segmentation[J].Journal of the Franklin Institute,2017,354:5694-5738.
[15]DO M N,VETTERLI M.The contourlet transform:an efficient directional multiresolution image representation[J].IEEE Transaltions on Image Processing,2005,14(12):2091-2106.
[16]TAJERIPOUR F,KABIR E.Fabric defect detection using modified local binary patterns[J].Eurasip Journal on Advances in Signal Processing,2007,2008(1):1-12.
[2]董蓉,李勃,徐晨,等.应用积分图的织物瑕疵检测快速算法[J].纺织学报,2016,37(11):141-147.DONG Rong,LI Bo,XU Chen,et al.Fast fabric defect detection algorithm based on integral image[J].Journal of Textile Research,2016,37(11):141-147.
[3]石美红,张正,郭仙草,等.基于显著纹理特征的织物疵点检测方法[J].纺织学报,2016,37(10):42-49.SHI Meihong,ZHANG Zheng,GUO Xiancao,et al.Fabric defect detection method based on texture saliency feature.[J].Journal of Textile Research,2016,37(10):42-49.
[4]NGAN H Y T,PANG G K H,YUNG N H C,et al.Wavelet based methods on patterned fabric defect detection[J].Pattern Recognition,2005,38(4):559-576.
[5]NGAN H Y T,PANG G K H.Novel method for patterned fabric inspection using bollinger bands[J].Optical Engineering,2006,45(8):112-115.
[6]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.
[7]NGAN H Y T,PANG G K H,YUNG N H C.Motifbased defect detection for patterned fabric[J].Pattern Recognition,2008,41(6):1878-1894.
[8]TSANG C S C,NGAN H Y T,PANG G K H.Fabric inspection based on the Elo rating method[J].Pattern Recognition,2016,51(4):378-394.
[9]JIA Liang,CHEN Chen,LIANG Jiuzhen,et al.Fabric defect inspection based on lattice segmentation and Gabor filtering[J].Neurocomputing,2017,238:84-102.
[10]GUO Z,ZHANG D.A completed modeling of local binary pattern operator for texture classification[J].IEEE Transactions on Image Processing,2010,19(6):1657-1663.
[11]GUO Y,ZHAO G,PIETIKINEN M,et al.Descriptor learning based on fisher separation criterion for texture classification[C]//Computer Vision-ACCV 2010.Berlin Heidel-berg:Springer,2011:185-198.
[12]OJALA T,PIETIKAINEN M,MAENPAA T.Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(7):971-987.
[13]WEN J,ZHAO J,WAN C.Improved morphological component analysis for interference hyperspectral image decomposition[J].Computer Electrical Engineering.2015,46(1):394-402.
[14]JIA Liang,LIANG Jiuzhen.Fabric defect inspection based on isotropic lattice segmentation[J].Journal of the Franklin Institute,2017,354:5694-5738.
[15]DO M N,VETTERLI M.The contourlet transform:an efficient directional multiresolution image representation[J].IEEE Transaltions on Image Processing,2005,14(12):2091-2106.
[16]TAJERIPOUR F,KABIR E.Fabric defect detection using modified local binary patterns[J].Eurasip Journal on Advances in Signal Processing,2007,2008(1):1-12.