Automatic Detection and Localization of Surface Defects for Whole Piece of Ultrahigh-definition Leather Images

摘要

In the traditional detection system of leather defect based on CCD, the imaging view is small, the procedure is complex and the image mosaic algorithm is needed to form a whole piece of leather image. In order to address the above problems, a method of automatic detection and localization of surface defects for whole piece of leather images of ultra-high definition is proposed in this paper. Firstly, a whole piece of leather image is collected by an imaging system in one step with an ultra-high definition camera. Then the region of interest (ROI) of the whole piece of leather is obtained by a segmenting algorithm based on the image saturation channel according to the characteristics of the leather image. At last, the leather defects are detected and located by using the automatic detecting and locating method of leather defects based on the image gradients. Qualitative and quantitative experimental results show that the proposed method outperform the traditional methods, and is simple in imaging, can overcome the effects of uneven illumination, detect and locate the leather defects efficiently and precisely.