Fabric Defect Detection with Optimal Gabor Wavelet Based on Radon

Real-time is an important index of fabric defect detection. The optimal selection of Gabor wavelet is an effective method to solve the problems of data redundancy and low operation of multi-channel Gabor wavelet. In this research, several kinds of Gabor wavelet optimal selection detection algorithms commonly used in fabric defects are summarized. And these algorithm ideas are analyzed. According to the characteristics of fabric defects in actual production, a new optimal Gabor wavelet algorithm, Defect Direction Projection Algorithm(DDPA) is proposed. Last but not least, the operation speed and detection accuracy performance of six optimal selection algorithms are compared. The experimental results show that the proposed algorithm has certain advantages of performance and application.