Machine vision sorts cotton fibers

Textile manufacturing is a leading industry in India and producers are increasingly implementing machine-vision-controlled automation as a consequence of higher quality standards. Premier Evolvics (Coimbatore, India; www.premierevolvics.com), a manufacturer of quality testing and online monitoring equipment for the textile industry, recently upgraded its Automated Rapid Tester (ART) fiber testing machines by switching to digital cameras.

Cotton samples are placed in a box with a transparent bottom and inspected by a built-in camera combined with lighting. The images are processed by software modules evaluating the fiber length, color quality, and the percentage of dark particles (trash). The upgrade was developed by system integrator Lucid Imaging (Bangalore, India; www.lucidimaging.in), and included the migration from analog CCD cameras with interlaced scan to Guppy FireWire-based cameras with interlaced sensors from Allied Vision Technologies (Stadtroda, Germany; www.alliedvisiontec.com).

Lucid replaced the licensed image-processing software with an open-source-based application. This offered both quality and cost benefits: The full digital data transmission increased the reliability of the system while cutting the frame grabber and the open-source software helped save license costs. “Because of the similar interlaced sensor technology, Premier Evolvics could take the benefits of a migration to a digital interface without having the drawbacks,” explains Anand Chinnaswamy, managing director of Lucid Imaging. “No significant changes had to be done to the system architecture such as lighting or software algorithms since the image data delivered by the digital camera have the same specifications as those delivered by the analog camera.”