Home » Greentech: TU Berlin gives clothing a second life with AI

Greentech: TU Berlin gives clothing a second life with AI

by admin
Greentech: TU Berlin gives clothing a second life with AI

A second life for clothing: Researchers are developing new processes for the automated sorting of old clothing and textile waste

The conveyor belt accelerates rapidly. A wide strip of light shines on a white blouse before it falls into a basket. A yellow sweater rolls up and is scanned, then blue jeans and a red coat. Photos of the scanned garments appear on a screen as well as information about the product type, color and target group.

Greentech Circular Recycling Kleider mit KI TU Berlin FOTO Artificial Photography on Unsplash

The black textile scanner, developed by a team of researchers from TU Berlin, FU Berlin and the company circular.fashion, has the potential to revolutionize the sorting of used clothes and thus make the second-hand business significantly more attractive. An urgently needed step if you look at the climate footprint of clothing.

Triggered by the fast fashion trend, the fashion industry causes an estimated eight percent of global CO2 emissions.

TrendingGreentech.LIVE Conference 25.-27.04.2024 (Spring Edition)

Greentech Science helps with clothing recycling

“So far, old clothes have been sorted by hand. “It’s a time-consuming process; each piece of clothing has to be handled two to three times until the product type and quality are determined,” says Karsten Pufahl from the TU Department of Nonlinear Optics, who leads the research in the project.

If the yellow sweater is in good condition, comes from a high-quality brand or corresponds to current fashion trends, it is suitable for the domestic or Western European second-hand market.

According to Pufahl, this product is known in the industry as cream products and is where you can earn the most money. The demand for second-hand clothing is increasing, flea markets, online sales platforms and even large fashion chains offer previously worn clothing. Anything that is no longer wearable is made into wiping rags, painter’s fleece or car seat fillings.

See also  Test: Ubiquiti G5 Bullet, great image quality and smart detections, at a fair price!

The rest of German and European used textiles are mainly exported to Africa, often including unusable clothing. That will probably change in the near future. The European Union wants to ensure that clothing that is no longer usable is no longer exported to third countries and causes waste there. In the future, more sorting will have to be done in Europe, for which there are currently not enough skilled workers.

AI-supported image analysis

This is exactly where the CRTX project comes into play. The black scanner is equipped with AI-supported image analysis, which recognizes the type of product in seconds and whether the item of clothing is suitable for the second-hand market or should be recycled.

Another advantage: Karsten Pufahl’s team observes the current fashion market to recognize trends, trains the AI ​​and thus helps second-hand retailers to respond directly to customers’ wishes. This means that items of clothing will increasingly find a second life.

In the future, the scanner will recognize additional product features and thus enable the best possible utilization. This high-quality sorting is being worked on in the follow-up project, which has been running since the middle of the year and is funded by the Federal Ministry for Economic Affairs and Climate with around 2.3 million euros, of which 1.3 million euros are for the TU Berlin.

The scanner is currently at a large recycling company in Hamburg, where the AI ​​is supposed to learn different quality levels based on thousands of items of clothing. There are already people interested in the final product and a patent application is also underway.

See also  The Italian startup that fights digital touts: "This is how we map ticket prices online"

Fiber-to-fiber recycling

In a second part of the project, the team dedicated itself to sorting for fiber-to-fiber recycling, in which yarn of the same quality is obtained from old textiles and can then be processed into a new textile product. This high-quality recycling method has not yet been used much because fiber detection with the near-infrared spectroscopy previously used is too imprecise.

For a few textiles, such as white bed sheets or jeans, it can be said with some certainty what material they are made of and what harmful substances they contain, so that new textiles can be created from them. This is more difficult with clothing, which often consists of a mixture of different fibers. As part of CRTX, Karsten Pufahl and his colleagues have developed a process that can precisely detect any type of textile using Raman spectroscopy.

Raman spectroscopy, which is often used in laboratory analysis, has previously had a problem with fluorescence. Because organic materials, which textiles are made of, fluoresce, they interfere with the measurement. The research team has further developed the process so that this problem no longer exists.

Greentech for pollutant detection

Even certain pollutants can now be measured in textiles. “It’s not quite as trivial because it’s complicated both to get the pollutants and to find contaminated clothing that you need to train the software to recognize pollutants,” says Karsten Pufahl.

It is also not yet clear which specific process will prevail for fiber-to-fiber recycling. There is currently a gaping gap between the sorting requirements of recyclers and what is technically feasible in material recognition.

See also  Check out our first test with the new Chevrolet Spin 2025

According to Pufahl, what is needed here is more openness and further exchange on both sides in order to find out exactly how the gap can be closed. It would be an important step to recover fibers of the same quality and free from harmful substances and to be able to use them to produce new, non-toxic textile products – in keeping with the circular economy.

Sources and links:

Website of the crtx.ai project

Website of the fashionsort.ai project

You may also like

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More

Privacy & Cookies Policy