Moment-Based Features of Knitted Cotton Fabric Defect Classification by Artificial Neural Networks
The defect classification of knitted fabrics is a challenging area of research. Most of the defect detection works in India; Bangladesh is being carried by manually trained inspectors. The long working hours and the working environment at the company induces the fatigue, lack of concentration, and t...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2022-04-01
|
| Series: | Journal of Natural Fibers |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1080/15440478.2020.1779900 |
| Summary: | The defect classification of knitted fabrics is a challenging area of research. Most of the defect detection works in India; Bangladesh is being carried by manually trained inspectors. The long working hours and the working environment at the company induces the fatigue, lack of concentration, and triteness to the workers due to this they may not able to detect the defects on the clothes after it is manufactured. To overcome this problem, a computer-aided defect detection system is being developed using digital image processing and artificial neural Network methods. The two types of artificial neural networks were applied to compare the results obtained. The networks were: a back propagation based feed forward neural network and the other was Levenberg–Marquardt (LM) algorithm based back propagation network. Experimental results predicted detection of a high degree of variety of fabric defects. |
|---|---|
| ISSN: | 1544-0478 1544-046X |