Design of fabric defect detection software correspond to ultrawide image scanner

LUO Ranran, BAO Min, SHEN Weiping, FEI Xin

(School of Machinery & Automation, Zhejiang SciTech University, Hangzhou 310018, China)

Abstract: A bundled software for ultrawide image scanner was provided to solve the problems of image geometric distortion when images were captured with a ultra wide scanner, and the problems of spot defects when the textile was weaved. The system procedure was programmed with the visual programming tool Delphi. The correct image was achieved by the use of coordinate transformation and gray interpolation method, and the theory of sequence images registration was used to realize the purpose of detect inspection. The results indicate that the software can detect the fabric defects quickly and accurately, and it also has important practical significance to improve the detection technology level and to enhance the quality control of products.

Key words: ultrawide scanner; digital image; geometric distortion; coordinate transformation; image registration; detect inspection