An appearance defect online inspection system of mesh fabric has been developed based on machine vision. The mean filter method is adopted to eliminate noise. An adaptive threshold method based on brightness is presented to eliminate the effects of uneven illumination and separate the foreground and background. By analyzing the texture characteristics and defect features of mesh fabric, the mesh fabric defect identification method based on texture features is proposed. The results have shown when the online inspection speed reaches 60m/min, the defect recognition rate can reach 95% or more and the online inspection system can meet the automation requirements of enterprises quite well.
GolnabiH.; AsadpourA.; Design and application of industrial machine vision systems;Robotics and Computer-Integrated Manufacturing2007, 23, 6, 630–637.
2.
ColetP.; Machine vision aids quality inspection;Quality2006, 45, 10, 22–23.
3.
LiR.; FPGA-based multi-sensor real time machine vision for banknote printing; InIS and T Electronic Imaging - Image Processing: Machine Vision Applications II; SPIE: USA, 2009; Vol. 7251.
4.
DarI.M.; Automated vision based quality control for electro-optical module manufacturing; InMachine Vision Applications, Architecture, and Systems Integration VI; SPIE: USA, 1997; Vol. 3205, pp. 236–245.
5.
KarathanassiV.; LossifidisChr.; RokosD.; Application of machine vision techniques in the quality control of pharmaceutical solutions;Computers in Industry1996, 32, 2, 169–179.
6.
AdelG.; FatenF.; RadhiaA.; Assessing cotton fiber maturity and fineness by image analysis;Journal of Engineered Fibers and Fabrics2011, 6, 2, 50–60.
7.
SunG.D.; ZhaoD.X.; LinQ.; Online defects inspection method for velcro based on image processing; In2010 2nd International Workshop on Intelligent Systems and Applications; IEEE: USA, 2010; pp. 1118–1121.
8.
WuL.L.; Identification of weed/corn using BP network based on wavelet features and fractal dimension;Scientific Research and Essay2009, 4, 11, 1194–1200.
9.
MedvedevaE.V.; KurbatovaE.E.; A two-stage image preprocessing algorithm;Pattern Recognition and Image Analysis2011, 21, 2, 304–308.
10.
WanY.; YaoL.; XuB.; Automatic segmentation of fiber cross sections by dual thresholding;Journal of Engineered Fibers and Fabrics2012, 7, 1, 114–120.
11.
MaJ.; TheilerJ.; PerkinsS.; Two realizations of a general feature extraction framework;Pattern Recognition2004, 37, 5, 875–887.
