Image inspecting method and device for detecting faults
The present invention relates to the automatic detection of faults in fabric material. Linear limited pulse response filter with optimal energy separating capacity is used. The detection of industrial product includes the following steps: automatic design of optimal filters, filtering the collected product image, calculate the energy of each pixel in local area, and finel threshold treatment of the pixels to separate faults. The present invention also relates to the method of detecting unknown faults in fabric by means of data synthesis.
Inventor: PENG K H G | COOMA A
Priority Number: CN100401043C
IPC Current: G01N002189 | D06H000308 | G01N0021898
Assignee Applicant: The University of Hong Kong
Title: Image inspecting method for detecting faults | A kind of method detection contain defect fabric
Usefulness: Image inspecting method for detecting faults | A kind of method detection contain defect fabric
Summary: For detecting defects such as mixed filling, mispicks, kinky filling, misread, wrong-draw, colored yarn, broken end, dirty end, double weft, big not, broken yarn, triple-warp, etc., of plain fabric material gathered from looms, in textile industry. Also for detecting defects of paper, plastic, wood, and steel in production lines.
Novelty: Web material inspection e.g. for textile paper, involves computing energy of each pixel in acquired image of web material based on output of optimal filter to attenuate pixels in image based on defect free region
纺织/服装
织品工程
香港
