International Journal of Science and Research (IJSR)

International Journal of Science and Research (IJSR)
Call for Papers | Fully Refereed | Open Access | Double Blind Peer Reviewed

ISSN: 2319-7064

Downloads: 2 | Views: 32 | Weekly Hits: ⮙1 | Monthly Hits: ⮙2

Research Paper | Computer Science and Information Technology | India | Volume 13 Issue 5, May 2024 | Rating: 5.3 / 10

Boosting Fabric Defect Recognition and Classification using Grey Wolf Optimizer with Deep Learning Model

N. Sajitha | Dr. S. Prasanna Priya

Abstract: Fabric defect detection has been instrumental in the textile production method, however, still, there are certain problems in accurately and rapidly identifying defects. Fabric defect recognition and categorization were examined for their significance in industrialization and urbanization. Meanwhile, the fabric industries are majority industrialists of textile, so the possibility of defect will added in the small - scale industries. Therefore, classification based on certain configurations can also be performed by the computer vision - based automated scheme. The complicated information attained from the real - world fabric model has a high complication owing to the multi - dimensional data set, higher order variable and high - variety data made from the images. In recent years, the usage of the deep learning (DL) approach in the textile industry for defect recognition has become a growing tendency. This article introduces Fabric Defect Recognition and Classification using Grey Wolf Optimizer with Deep Learning (FDRC - GWODL) model. The FDRC - GWODL technique applies bilateral filtering (BF), an innovative image processing method, to effectively mitigate noise interference and optimize image clarity. Leveraging the Residual Network (ResNet) model as a feature extractor, the technique extracts discriminatory features critical for precise defect demonstration. Hyperparameter tuning is performed by the Grey Wolf Optimizer (GWO), enhancing model parameters to increase recognition accuracy. Consequently, defect classification is performed by the Random Forest (RF) classifier, recognized for its efficiency and robustness in managing complicated classifier tasks. Simulation outcomes illustrate the betterment of the presented FDRC - GWODL technique in fabric defect recognition, illustrating its possibility to enhance quality control measures in textile manufacturing process.

Keywords: Fabric Defect Detection; Grey Wolf Optimizer; Deep Learning; Image Processing; Defect Classification

Edition: Volume 13 Issue 5, May 2024,

Pages: 1164 - 1172

How to Download this Article?

Type Your Valid Email Address below to Receive the Article PDF Link

Verification Code will appear in 2 Seconds ... Wait