In Proceedings of The 10th Scandinavian Conference on Image Analysis, vol. I, pp. 317-322, Lappeenranta, Finland, June 9-11, 1997.

Abstract

An adaptive two-stage approach is proposed for the classification of surface defects. In the segmentation phase feature extraction is done and potential defect areas are marked. In the classification phase features describing the shape and internal structure of defects are extracted and defects are classified to different defect classes. The proposed segmentation scheme is taught with examples of faulty-free surface, and the classification scheme with examples of defects. The use of the Self-Organizing Map in segmentation phase makes the proposed method adaptable to different types of surfaces. The results of experiments with base paper samples are encouraging.