總結而言，本研究旨在解決胚布瑕疵檢測中的挑戰，提出一種輕量且可解釋的AI方法，適用於少量標記資料和現場限制。 這將有助於提高瑕疵檢測的效率和準確性，同時增加在工業場域中的可行性和可信度。;Defect detection in fabrics is crucial for ensuring product quality in industrial settings. However, it is challenging due to the diverse patterns of different fabric types. Deep learning methods have shown promise in defect detection, but they require a large amount of labeled data, which is often difficult to obtain in practical scenarios. Additionally, the need for lightweight models arises from on-site hardware limitations and real-time processing requirements.
To overcome these challenges, we propose an AI approach that is effective with limited labeled data and offers real-time performance. We also prioritize interpretability, recognizing its importance in industrial settings. Our approach not only focuses on model accuracy but also emphasizes interpretability to gain trust from domain experts.
In summary, our research aims to tackle the challenges of defect detection in fabrics by introducing a lightweight and interpretable AI method suitable for limited labeled data and on-site constraints. This approach will enhance the efficiency and accuracy of defect detection while ensuring feasibility and credibility in industrial environments.
