Fabric Defect Detection in Textile Manufacturing: A Survey of the State of the Art

Table 2

Spectral algorithms for fabric defect detection.

Author	Proposed method	Dataset	Evaluation
Sulochan [25]	Multiscale wavelet features and fuzzy C-means clustering	Real and computer-simulated fabric images	Detection error rate
Vermaak et al. [26]	Dual-tree complex wavelet transform (DTCWT)	TILDA dataset	Detection success rate
Liu and Zheng [27]	The method based on information entropy and frequency domain saliency	Database created by the research associate of the industrial automation research laboratory	ACC, true positive rate(TPR), false positive rate(FPR), positive predictive value (PPV), negative predictive value (NPV), time, F-measure
Di et al. [28]	L0 gradient minimization method and two-dimensional fractional Fourier transform (2D-FRFT) for obtaining the saliency map of the quaternion image	Dataset from automation laboratory fabric database of Hong Kong University	True positive (TP), false positive (FP), true negative (TN), and false negative (FN)
Jing [29]	Gabor preprocessed golden image subtraction	Industrial automation laboratory at the University of Hong Kong and the TILD database	Detection success rate
Mohammed and Alhamdani [30]	Fuzzy back propagation neural network (FBPNN) with Gabor features	Collected dataset	Detection success rate
Yapi et al. [31]	Using learning-based local textural distributions in the contourlet domain	TILDA database	(TP, FP, TN, and FN) local precision (PL), local recall (RL), and local accuracy (ACCL)