Germany’s textile-waste problem is no longer mainly about collection; it is about sorting mixed garments fast enough, accurately enough, and cheaply enough to keep them in circulation.
Every year, Germany generates about 1.4 million tonnes of used textiles, yet less than 1% enters a closed recycling loop, while only around 200,000 tonnes are actually checked and categorised. The bottleneck is not ambition. It is sorting. Manual classification is too slow and too dependent on expertise to cope with fast fashion’s volume and material complexity.
Cameras and AI replace the first human bottleneck
At the Recycling Atelier of the Augsburg Institute for Textile Technology (ITA), researchers have built DETEX, an AI-based textile-sorting system that uses two 13-megapixel IDS uEye XC cameras and trained neural networks to classify garments and identify material properties. One camera recognises the item type on the conveyor; the second examines fabric characteristics and flags features such as buttons or stains.
Better sorting determines whether recycling is worth doing
Mechanical recycling only works if garments are classified precisely enough to preserve fibre value during shredding and separation. DETEX is therefore not a gadget; it is an attempt to raise sorting accuracy so more material can flow into spinning, nonwovens or reuse rather than incineration or export.
From conveyor vision to a fuller recycling system
The next step is scale and sophistication. ITA plans to expand DETEX into a modular robotic system with 360-degree capture and robot-assisted handling, pushing textile sorting closer to an industrial, data-driven circular model.
