The patent-pending system has been integrated into the company’s WebSpector platform and validated through factory trials on a purpose-built full scale in-house demonstration system with sophisticated fabric transport capabilities.

Traditional methods for defect detection rely on human inspection which is ineffective, with detection rates under 65%, while the Shelton WebSpector machine vision system offers a sophisticated platform for automated defect detection of over 97%, but until now has been restricted to plain textiles.

“What our new system basically does is essentially make the ‘good’ pattern invisible to the detection software,” explains Shelton Vision Managing Director and CEO Mark Shelton. “Building on our market leading vision system for plain textiles, we have developed template matching techniques for the resolution of complex pattern deformations in order for the system to pick up defects in the pattern as well as underlying defects.”

While pattern matching and neural network approaches have previously been tried for patterned textiles they have failed to provide a practical solution due to the extreme complexity associated with pattern matching on deformable substrates like textiles, as well as the time required to train a neural network for each pattern type.

“The challenge is that fabrics are not rigid and can be creased or stretched and are also subject to local distortion,” says Mark. “As a result, inspection without the technique we have developed, would lead to thousands of false positives. Our sophisticated pattern inspection software techniques ensure a clean image, allowing the detection of faults on fabrics running at speeds of up to a hundred metres a minute.”