We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
A Simplified pulse-coupled neural network for adaptive segmentation of fabric defects.
- Authors
Meihong Shi; Shoushan Jiang; Huiran Wang; Bugao Xu
- Abstract
This paper describes an adaptive image-segmentation method based on a simplified pulse-coupled neural network (PCNN) for detecting fabric defects. Defect segmentation has been a focal point in fabric inspection research, and it remains challenging because it detects delicate features of defects complicated by variations in weave textures and changes in environmental factors (e.g., illumination, noise). A new parameter called the deviation of the contrast (DOC) was introduced to describe the contrast difference in row and column between the analyzed image and a defect-free image of the same fabric. The DOC essentially weakens the influence of the weave texture and the illumination. The simplification of PCNN reduces the number of the network’s parameters by utilizing the local and global DOC information for the parameter selections. The validation tests on the developed algorithms were performed with fabric images captured by a line-scan camera on an inspection machine, and with images from TILDA’s Textile Texture Database () as well.
- Subjects
TEXTILES; WEAVING patterns; ARTIFICIAL neural networks; LIGHTING; ALGORITHMS
- Publication
Machine Vision & Applications, 2009, Vol 20, Issue 2, p131
- ISSN
0932-8092
- Publication type
Article
- DOI
10.1007/s00138-007-0113-z