作者: K.L. Mak , P. Peng , H.Y.K. Lau
DOI: 10.1109/ICIT.2005.1600684
关键词: Robustness (computer science) 、 Image segmentation 、 False alarm 、 Artificial intelligence 、 Machine vision 、 Engineering 、 Computer vision 、 Gabor filter 、 Software 、 Image texture 、 Filter (signal processing)
摘要: This paper proposes a real-time computer vision system for detecting defects in textile fabrics. The developments of both the hardware and software platforms are presented. design prototyped defect detection ensures that fabric moves smoothly evenly so high quality images can be captured. also new filter selection method to detect defects, which automatically tune Gabor functions match with texture information. is further developed into segmentation algorithm. scheme tested on-line off-line by using variety homogeneous different defects. results exhibit accurate low false alarm, thus confirming robustness effectiveness proposed