The widespread use of automated fabric inspection systems is one of the first ways that artificial intelligence (AI) and machine learning are expected to have a positive and dynamic impact on the textile industry. According to Calvin Wong, CEO of AiDLab, the Laboratory for Artificial Intelligence in Design, located in Hong Kong, most of the world’s top mills still rely on human inspection in their operations. This is a hot topic in the business right now.
“At least half of these manufacturers will be using AI-based systems in the next ten years,” he stated at the recent ITMAconnect Innovative Technologies webinar, which can be watched in its entirety at goto.itmaconnect.com. The most recent camera-based vision systems for inspecting textile materials can precisely check color hues, eradicate flaws, and save a significant amount of money. “Quality will significantly reduce returns, which are currently extremely costly to the industry and the environment,” the speaker continued.
WiseEye is an AI-based textile material inspection system created by AiDLab that can identify color shading and fabric flaws quickly and precisely in real-time examination settings. The device combines deep learning methods with machine vision algorithms to achieve over 90% accuracy and high-speed inspection at up to 60 meters per minute. Additionally, the technology provides automatic flaw labeling in accordance with industry standards. It can examine different kinds of fabric and detect more than 40 typical fabric flaws.
During the ITMAconnect webinar, the need of high-quality data was emphasized by both Florian Pohlmeyer, head of digitalization at RWTH Aachen University’s Institute for Textile Technology (ITA), and Sverker Evefall, senior application manager for Swedish vision systems company ACG Eyetech.
In 2017, ITA established the Innovation and Learning Center (ILC) in Aachen to help businesses get ready for the digital era. Digital solutions for production are being created in its Model Factory 4.0, and enterprises are receiving fundamental knowledge on Industry 4.0, condition monitoring, sensor technology, and automation through workshops and seminars.
ACG EyeTech uses robotic guiding, measurement, and inspection using 2D, 3D, and deep learning algorithms. ACG Kinna and ACG Nowo, sibling firms, showcased a robotic pillow filling process at ITMA 2023 in Milan. The method can fill and finish around 3,840 pillows in an eight-hour shift. Machine vision systems are being added by ACG EyeTech to these systems in order to provide flawless and error-free fiber handling, filling, sewing, and packing.
Evolutionary process
The most sophisticated machines “will not achieve what they are fully capable of without qualitative data,” according to Pohlmeyer the textile business has numerous processes and a convoluted value chain. At every stage, data is gathered, yet it’s frequently dispersed across various software programs, databases, and equipment. Every one of them has a unique format, thus it’s critical that they all adhere to the same standards so that systems can interact.
Evefall continued, saying that just like training a child, every visual system needs to be shown what to perform.The algorithm needs to be taught what to search for; it is not an intelligent entity that can just detect flaws. Don’t underestimate the amount of time required Many manufacturers believe artificial intelligence (AI) can handle everything but putting it into practice is an evolutionary process that requires time. Large volumes of data must be displayed to the system.”
People power
The panel chair, Dr. Andre West, associate professor and director of the Zeis Textiles Extension (ZTE) at the Wilson College of Textiles at North Carolina State University, added that working with AI experts was essential to bringing knowledge into a company. Both agreed that having good people was essential to success.
According to Pohlmeyer, “training employees who use AI in production is crucial because this goes beyond technology and involves skill development.” To integrate these new tools with domain expertise in the textile industry, businesses require personnel in a variety of sectors, including cyber security and data analytics.
