From the Sector

Image processing system allows cost savings of more than 2,000 euros per year and weaving machine

MSc Marco Saggiomo from the Institut für Textiltechnik (ITA) and graduate engineer Marcin Kopaczka from the Institute of Imaging & Computer Vision of RWTH Aachen University (LfB) developed an image processing system for weaving machines that provides comprehensive benefits for woven fabric producers. The image processing system enables the weaving machine to detect faulty pickings autonomously. The application of the image processing system leads to cost savings of at least 2,210 euros per year for each weaving machine in comparison to manual faulty picking repairs.

The researchers won the “Best Student Paper Award” within the framework of the 7th International Conference on Pattern Recognition Applications and Methods (ICPRAM) for their joint paper on “Fully Automatic Faulty Weft Thread Detection using a Camera System and Feature-based Pattern Recognition”. The award confirms the success of the co-operation of both institutes, ITA and LfB, which will be continued in future research projects.

The image processing system is part of the doctoral thesis of Marco Saggiomo. Marcin Kopaczka supported ITA with the development of the image processing algorithm.
The joint paper is based on the successfully concluded project WeftAlert within the framework of the German Federation of Industrial Research Associations. Further results of the WeftAlert project include the extension of the state of the art of image-based process optimisation of air-jet weaving, excellent outcomes in image processing as well as the validation of image processing systems in laboratory and industrial environments.