Patch based yarn defect detection using Gabor filters

This paper describes a simple and effective algorithm for texture defect detection in uniform and structured fabrics. The proposed approach is structured in two phases: feature extraction and defect identification. The texture features extraction phase relies on a complex symmetric Gabor filter bank and Principal Component Analysis for dimensionality reduction. Differently from most previous works, our analysis is performed on a patch basis, which has been more effective than simply considering raw pixels as features. The defect identification phase is fast as it is based on the evaluation of the Euclidean norm of the patch feature vectors, and on the comparison with fabric type specific parameters. A calibration procedure, performed offline, is adopted in order to estimate the optimal parameters. The performance of the algorithm has been extensively evaluated, via computer simulations, on the TILDA image database. The results show that our algorithm outperforms previous approaches in most of the considered cases, achieving a detection rate of 98.8% and a false alarm rate as low as 0.37%.