שמור ב:
| Main Authors: | , |
|---|---|
| פורמט: | Recurso digital |
| שפה: | אנגלית |
| יצא לאור: |
Zenodo
2025
|
| נושאים: | |
| גישה מקוונת: | https://doi.org/10.5281/zenodo.17608799 |
| תגים: |
הוספת תג
אין תגיות, היה/י הראשונ/ה לתייג את הרשומה!
|
תוכן הענינים:
- <p>Fabric defect detection is a crucial task in the textile industry, where early identification of faults ensures product quality<br>and minimizes production losses. Conventional inspection methods, often manual and subjective, are insufficient for realtime and large-scale textile monitoring. While deep learning has significantly advanced defect detection performance, its<br>effectiveness is hindered by the scarcity of labeled textile datasets and the domain gap between generic image features and<br>fabric-specific textures. To overcome these limitations, this paper proposes a Two-Way Transfer Learning (TWTL)Approach<br>for fabric fault detection, which leverages bidirectional knowledge transfer between a source domain (e.g., ImageNet) and a<br>target domain (TILDA datasets). Our method integrates both forward and backward transfer mechanisms to enhance<br>feature adaptability and defect classification accuracy. Experimental results on benchmark datasets such as TILDA<br>demonstrate that the proposed approach outperforms traditional CNNs and one-way transfer learning models with notable<br>improvements in performance metrics such as accuracy, precision etc. The optimized lightweight architecture of the model<br>facilitates low-latency inference on edge devices, ensuring its applicability for real-time automated textile inspection in<br>resource-constrained environments.</p>