Refiberd: AI based high-tech vision for textile waste transformation
In the heart of the global textile waste crisis, a formidable team of engineers has risen to the challenge. Refiberd, a women-led initiative, was founded with a vision to leverage cutting-edge AI research to revolutionize the fashion industry’s approach to sustainability. Spearheaded by Sarika Bajaj as the Co-Founder & Chief Executive Officer and Tushita Gupta as the Co-Founder & Chief Technology Officer, along with Julia Chatterjee, Research Associate, and Georg Menzl, PhD, AI Research Scientist, the team is on a mission to bring about a 100% circular economy.
Technology at the Core
Refiberd’s approach to addressing textile waste is grounded in a fusion of artificial intelligence and textile engineering. They firmly believe that the key to a sustainable future lies in innovative technological applications and cutting-edge processes. To realize this vision, they have developed an advanced material detection system using AI-based hyperspectral imaging.
Advanced Material Detection through Hyperspectral Imaging
At the heart of Refiberd’s technology is a state-of-the-art hyperspectral imaging system. This technology operates on the principle of analyzing how light interacts with different materials based on their chemical composition. This unique approach enables the identification of various materials by detecting their distinct light absorption and reflection patterns.
The hyperspectral imaging system is meticulously tuned for the most sensitive detection of textile fiber types. It can even discern trace amounts of materials and contamination, showcasing its efficacy in addressing the intricate challenges of textile waste management.
Process Explained
Hyperspectral Imaging: The textile is placed beneath the hyperspectral camera along with a line light source. The camera captures lines of hyperspectral data from the moving textile at a specified framerate.
Data Stitching: The collected data is then stitched together by a computer to create a hyperspectral cube. In this cube, each pixel of the image represents a spectrum.
Machine Learning: The hyperspectral cube is processed by a machine learning model, which has been trained on a proprietary dataset of thousands of custom textile samples. The model outputs a prediction of the material composition of the textile, providing a comprehensive understanding of the fibers present.
AI-Powered Precision
Refiberd’s technology goes beyond mere detection; it accurately identifies the composition of materials, including blended and layered fibers. The power of their artificial intelligence lies in its ability to process hyperspectral imaging data with unparalleled accuracy, ensuring a robust solution for textile waste sorting.
Refiberd’s commitment to a circular economy is evident in their innovative application of technology. By combining AI and hyperspectral imaging, they have created a solution that not only detects but also comprehensively identifies materials in textile waste. This marks a significant step forward in the journey towards sustainable fashion and a waste-free future.
Refiberd stands as a beacon of technological innovation in the fight against textile waste, showcasing the potential of AI to reshape industries for a more sustainable tomorrow.
