Surface and Corner Defect Detection on Tile Images Using Gabor Features, Level Set Segmentation and Dot Product

Abstract

Quality inspection is one of the most important problems in industry. Digital image processing techniques are used, in order to be able to detect defects by using advanced systems and to prevent human faults during product categorization process based on quality. In this paper, an automatic surface and corner defect detection algorithm using Gabor feature, level set segmentation and dot product are proposed in tile images. This algorithm detects two groups of defects, surface defects such as pinhole, bulb, spot, crack, and chip corner defect, produced during the tile production process. In this algorithm, after pre-processing stage on images by using selective binary and Gaussian filtering regularized level set (SBGFRLS) method, the tile's image are segmented from its background with the lowest error rate and time and then filtering is used for analyzing random texture surface by a novel way of Gabor filters. At last the desired defect is detected by using a sensitive threshold to changes. The proposed algorithm is evaluated on a perfect set of tile images. The obtained results show effectiveness of our proposed algorithm in real and operational conditions.