Sustainability in textile industry is dependent on the transition from linear to a circular economy, where textile items are continually reused, refurbished and recycled. One of the key challenges in the textile sustainability is proper and accurate sorting of materials to enable recycling. Currently, only 1-2% of the textile waste globally is recycled into new clothes or products. Conventional textiles products involve a diverse range of materials such as cotton, wool, polyester, lycra etc. and blended textiles that combine different materials to improve quality and performance characteristics. The recycling of textile lack a reliable and scalable method, for accurate identification of textile composition, makes it difficult to recycle; leading to huge landfill accumulation and wastage of resources after its use. This is due to the fact that different recycling mechanisms are required for diverse textile materials. For example, mechanical recycling involves breaking down of textile items into reusable fibres or raw materials, but it requires homogenous material for input. On the other hand, chemical recycling ensures separation of complex fibre blends into monomers or reusable fibres. Both processes require accurate composition identification to optimize recycling yields and avoid contamination. Therefore, a proper technology to identify single or blended textile items and their composition is required, to enhance the recycling process and reduce the textile waste.Traditionally, manual sorting and burn tests have been used to identify the textile composition but these are not accurate and are prone to errors, labour-intensive and inefficient for handling large volumes of textile. Advanced technologies are needed for effective and non-destructive fibre identification, which can enable textile identification in non-destructive, real-time in production mode. We have developed a method which is capable of rapid, precise and automated analysis of textile composition, to enable efficient material recovery and recycling. The mechanism uses a unique combination of spectroscopic sensor module, a data process unit and an artificial intelligence-based algorithm for proper identification of textile item’s composition. The method is scalable and an automated system for textile identification may be developed in future using the proposed mechanism.

optical sensing,textile identification,Recycling，,Non-Destructive Test