Complex Pattern Jacquard Fabrics Defect Detection Using Convolutional Neural Networks and Multispectral Imaging

The textile manufacturing industry faces considerable hurdles in identifying and mitigating fabric flaws, particularly due to the labor-intensive and financially expensive character of manual examination techniques used in the past. To address these problems, this research offers a novel approach which makes use of deep learning and multispectral imaging technologies to automatically detect faults in complex pattern jacquard fabrics. By combining the advantages of the hybrid InceptionV3 and ResNet50 algorithms, the suggested method offers a strong framework for precise and effective fabric fault detection. Through the incorporation of multispectral imaging, the system obtains a full perspective of the fabric, hence facilitating improved detection performance at different wavelengths. The combination of these modern technologies not only accelerates problem diagnosis but also greatly decreases associated expenses, making it a promising breakthrough in the textile sector. This automated flaw detection system has a lot of potential for increasing total fabric quality control, ensuring high-quality textile production while reducing manual inspection efforts.