Quality control for seams in the textile industry





    Born and raised in the cradle of the textile industry in southern Germany, these three young people Janine, Marie, Kristina and Sven had a high affinity for the textile industry from an early age. After studying the subject at university, the same passion for AI brought them together. Although Janine attended the same fashion school as Kristina and went to the same university as Marie in Hamburg, the trio came together by chance in the south of Germany. Sven joined through a mutual friend and is the perfect technical addition to this women team. They share a common goal of creating a more resource-efficient production within the textile industry with the help of faibrics' quality control system.

    With this bright vision of a sustainable and more environmentally friendly world for the textile industry, FAIBRICS was founded with the combined efforts of all four. Therefore they developed, due to a student research project, a quality control system on seams. While the sewing machine is running, the camera takes images from the textile with its sewed string. The image is sent to a quality control system, which should recognize seam defects. Since they have the multi-label classification, an image from the camera can show all attributes as true, false, or any other combination. Their goal was to validate the test images. Therefore they created heatmaps, which led to the correct classification decision. The attributes exceeded a prediction accuracy of around 97 -100 percent. Their mission is to establish an AI system for manufacturers and testing institutes in the textile business.

    The different business basics from the KI-Garage workshops will help them to expand this knowledge. To further optimize the artificial intelligence of faibrics, the individual coaching with the help of an AI expert and exchange with other founders about AI ideas will be a focus topic throughout the accelerator.

    Find out more by contacting