Crack detection in magnetic tile images using nonsubsampled shearlet transform and envelope gray level gradient. (1st May 2017)

• Record Type: Journal Article
• Authors: Yang, Chengli; Liu, Peiyong; Yin, Guofu; Wang, Ling
• Is Part Of: Optics & laser technology. Volume 90(2017)
• Page Start: 7
• Page End: 17
• Publication Date: 2017-05-01
• Subjects: Magnetic tile -- Crack detection -- Nonsubsampled shearlet transform -- Envelope gray level gradient; Optics -- Periodicals; Lasers -- Periodicals; Electronic journals; 621.366
• DOI: 10.1016/j.optlastec.2016.08.016
• Languages: English
• ISSNs: 0030-3992

Abstract

Abstract: A novel algorithm based on nonsubsampled shearlet transform (NSST) and envelope gray level gradient (EGLG) is proposed for detecting low-contrast cracks in variably illuminated magnetic tile images. The algorithm first adopted NSST to decompose the original images into multiple subbands at different directions and scales. Then a novel column filtering based on EGLG was employed to remove the uneven background in the approximation subband, and a shearlet coefficient variance discriminator (SCVD) was used to eliminate interferences of noise and textures in the detail subbands. Finally the cracks were extracted from the reconstructed images. To verify the validity of the proposed algorithm, extensive experiments were conducted in a novel machine vision based system and its comparison with traditional algorithms was given. Experimental results show that this method achieves an accuracy rate of 95.5% in detecting cracks longer than 0.9 mm with an average runtime of 0.576 s, and outperforms traditional methods in terms of accuracy and robustness. Highlights: Nonsubsampled shearlet transform is first used to detect cracks of magnetic tile. A filtering using envelope gray level gradient is proposed for background correction. A shearlet coefficient variance discriminator is used to detect cracks. A novel machine vision system has verified the validity of the proposed algorithm. The proposed method is effective in detecting cracks longer than 0.9 mm.