A new Gabor filter combination and its application to fine defects detection is steel surfaces

Abstract

In this thesis, a new Gabor filter combination for detecting various sizes of defects, and its application to fine defects detection in steel surfaces. To detect various sizes of defects, two filters were effective than a single filter.

First, a new Gabor filter combination method was proposed. Response is obtained by linear combination of two Gabor-filtered images. Coefficient of linear combination is determined by energies of two Gabor-filtered images. The performance of proposed method was tested through the comparison with filter addition and filter selection.

Second, an algorithm for detecting seam cracks using Gabor filter combination was proposed. Seam cracks are narrow cracks generated in the edge regions of steel plates. Because the widths of seam cracks are various, we applied the Gabor filter combination method. To increase the performance, Gray value intensity transformation in pre-processing, and region of interest selection in post-processing, were proposed. New features were designed to increase the performance of classification. The proposed algorithm was tested on the images obtained in real production line.

Third, an algorithm for detecting pinholes was proposed. Pinholes are very tiny holes generated in the surfaces of scarfed slabs. To detect various sizes of pinholes, the Gabor filter combination was applied. A new image segmentation method was proposed in pre-processing. To reduce the number of pseudo-defects, dual thresholding method was used. We have increased the performance of classification by adding new morphological features to general texture based features. To evaluate the performance of proposed algorithm, an images obtained in real production line were used.