Case Study: PICVISA Harnesses Hyperspectral Imaging to Revolutionize Textile Sorting

Textile waste is a growing global concern with far-reaching environmental, social, and economic implications. The global textile industry is one of the largest and fastest-growing industries, producing an enormous quantity of textile products each year. Estimates suggest that less than 1% of textiles are currently recycled. A significant portion of textile waste ends up in landfills where natural fibres, such as cotton, take years to decompose, while synthetic fibres, like polyester, persist for much longer, polluting the environment.

The EU requires all its member countries to implement systems to manage 100 per cent of their textile waste, more than 16 million tons per year, to boost textile recycling by January 1st, 2025. Despite efforts to regulate and promote recycling, textile recycling faces challenges due to the complexity of sorting textile materials, including blends of different fibres, presenting a pressing need for efficient textile sorting technologies.

PICVISA is on a mission to automate textile sorting

PICVISA, a Spanish company specializing in optical sorting, robotics, AI, and deep learning solutions, had a growing number of clients with a significant interest in advanced textile separation solutions, which led to their decision to invest in textile sorting.

In pursuit of innovative textile sorting technology, PICVISA successfully developed a fully automated machine that can classify and automatically separate several types of textile waste by composition (cotton, polyester, viscose, and other fibres), colour, and shape. Thanks to the technological solution implemented by PICVISA, Coleo Recycling in A Coruña, Galicia, classifies and traces some 5,000 tons of textile waste annually.

Hyperspectral Imaging Enables Accurate Textile Identification and Sorting

The key component of PICVISA’s fully automated textile sorting machine is the Specim FX17 hyperspectral camera. The Specim FX17 operates in a line-scan mode, collecting hyperspectral data in the near-infrared (NIR) region spanning from 900 to 1700 nm. NIR hyperspectral imaging allows the identification of the composition of textile products since different textile fibres (natural, artificial, and synthetic) have unique spectral characteristics that can be used for classification.

Using the Specim FX17 hyperspectral camera, PICVISA can capture images and analyze the spectral responses of different materials with exceptional precision.

Daniel Carrero, Technical Director of PICVISA, explains the impact of the Specim FX17 camera: “This technology is a game-changer. It enables us to identify materials and compositions of garments for pre-selection before recycling, providing a complete analysis with excellent spectral resolution.”

Market Intelligence News & Insights: 

How Will Apple’s $500 Billion Investment for US Manufacturing, Education, and AI Play Out?

On February 24^th, Apple announced a “commitment… to spend and invest more than $500 billion in the U.S. over the next four years”, with plans for new manufacturing plants, enhanced support for manufacturing partners via their Advanced Manufacturing Fund, and a myriad of other initiatives. As part of this plan, Apple projects 20,000 new hires, with roles focused on R&D, silicon engineering, software development, and AI/ML engineering.

Read More

Sensor-fusion for Enhanced Sorting Accuracy

In addition to the Specim FX17 camera, PICVISA integrates various complementary technologies into their textile sorting machine to ensure unparalleled accuracy. These technologies encompass colour separation, defect identification, and contaminant extraction. RGB systems are employed for colour classification, while inductive sensors identify metals in garments.

Artificial intelligence (AI) is also critical in PICVISA’s sorting process. By utilizing AI algorithms, the company achieves the classification of materials that may be identical in colour and composition but differ in appearance or shape, guaranteeing precise sorting outcomes.

Responding to Customers’ Needs Is the Path to Success

PICVISA’s diligent efforts have resulted in an extensive classification library comprising over 20 compositions. However, customer demand for an even more comprehensive range of classifications remains steadfastly increasing. Notably, tackling the identification of elastane has proven challenging for PICVISA, as its recognition is contingent upon particular compositions and proportions. To overcome the obstacle and further develop their solution, PICVISA is actively exploring regression-based classification besides class-based as a potential solution.

Regulations Drive Innovation and Growth of Textile Recycling

PICVISA envisions significant growth and development in the textile sorting industry. With the implementation of the EU regulations, there is an accelerated need for effective processes to handle textile waste, making the technologies and solutions provided by PICVISA increasingly vital.

Lluís Seguí, Managing Director of PICVISA, states, “For PICVISA, the textile sorting market represents one of the fastest-growing future sectors. Developing chemical recycling for mixed-fiber garments and finding a solution for already-sorted pure garments will be crucial.”

Specim is a partner with exceptional products and service

PICVISA already had experience working with Specim, as the company has previously implemented Specim’s hyperspectral cameras in their automated sorting machines for other segments, such as plastic recycling.

PICVISA’s collaboration with Specim has been highly satisfactory, setting the benchmark for successful partnerships. Daniel Carrero emphasises their decision to work with Specim, “We chose Specim for their exceptional product and service. It’s a safe bet, as they offer an industrial-grade solution with fantastic capabilities.”