Fu Hsun Fiber Industries has officially introduced a prototype of its FabGoalie AI Fabric Defect Detection System at TITAS 2025, developed in collaboration with National Yang Ming Chiao Tung University (NYCU). The innovation is being showcased at booth L1109, marking a significant milestone in the integration of artificial intelligence within textile quality control.
At the heart of FabGoalie AI lies an advanced algorithm capable of detecting objects and classifying them into various defect categories. This comprehensive system integrates AI-powered software, specialized inspection hardware, and an intuitive graphical user interface (GUI), offering a complete solution for textile manufacturers seeking accuracy, consistency, and efficiency.
The system can identify over ten different types of fabric defects, including oil stains, dye stains, holes, broken yarns, snags, and creases. Beyond defect detection, FabGoalie AI can also analyze color consistency, evaluating perceptual color variations across the left, center, and right areas of the fabric roll.
A Milestone in Automated Textile Quality Control
“The prototype is an important milestone for us. With FabGoalie AI, fabric producers can overcome the limitations of traditional manual inspection. I am glad that we have been able to push the boundaries once again,” said Tony Chen, Chairman and Board Member of Fu Hsun Fiber Industries. “Of course, there is even more potential ahead, and we look forward to feedback from the textile industry at TITAS.”
The FabGoalie AI system supports roll-to-roll fabric inspection, allowing material to move directly from the cart to the winding station. Equipped with seven cameras and an optical marking system, it can inspect fabric at speeds of 25–30 meters per minute. Despite its powerful configuration, the system’s compact hardware is designed for space-efficient inline integration, making it ideal for large-scale production environments.
Real-Time Monitoring and Data-Driven Insights
To meet the evolving needs of modern manufacturing, FabGoalie AI provides an interactive and multilingual graphical interface optimized for real-time decision-making. Its GUI features preconfigured data tables and charts, along with a dedicated display monitor attached to the machine and a workstation for production floor management.
The system also supports QR code access for quick dashboard viewing and a web-based platform for remote monitoring and long-term analytics. This ensures that production teams can access critical inspection data from multiple locations and maintain consistent quality standards.
Collaboration and Future Outlook
Fu Hsun Fiber Industries and NYCU are currently testing FabGoalie AI in real production environments. The partners are now inviting fabric producers worldwide to collaborate in expanding the system’s deployment across multiple manufacturing sites.
By exposing the AI model to a wider range of materials, structures, and defect types, the collaboration aims to enhance its ability to identify rare and complex fabric flaws, paving the way for the next generation of intelligent textile inspection systems.
