Artificial intelligence could hold the key to sorting through vast volumes of construction and demolition waste, with new and emerging technologies deployed to pinpoint timbers that can be recycled for future projects. Wood Central understands that this technology could not only shake up the construction waste industry, responsible for 44% of the waste produced in Australia, but also drive the pivot toward a fully circular economy.
That is according to a group of Australian researchers who, in research published last week, trained and tested deep-learning models to detect different types of wood contamination from high-resolution images with 91.67% accuracy.
As it stands, contaminated wood is one of the most challenging products to recycle due to the difficulties in identifying and sorting through materials contaminated by paint, treatments, metals, and other materials. However, thanks to AI-trained modelling, researchers observed strong precision and recall across six types of contaminated wood, including 1) Asbestos-Contaminated Wood, 2) Creosote-Contaminated Wood, 3) Fungi-Contaminated Wood, 4) Metal-Contaminated Wood, 5) Mould and Mildew-Contaminated Wood, and 6) Painted Wood.
“We curated the first real-world image dataset of contaminated construction and demolition wood waste,” said Madini De Alwis, PhD candidate at Monash’s Civil and Environmental Engineering Department, who worked with Dr Milad Bazli (from Charles Darwin University) on the project. “The new system could be deployed via camera-enabled sorting lines, drones or handheld tools to support on-site decision-making.”
Wood Central understands that while computer vision has been used extensively in general waste streams, its use for sorting through contaminated wood waste has been limited—until now.
“By fine-tuning state-of-the-art deep learning models, including CNNs (Convolutional Neural Networks) and transformers, we showed that these tools can automatically recognise contamination types in wood using everyday RGB images,” Dr Bazli said, adding that “this opens the door to scalable, AI-driven solutions that support wood waste reuse, recycling and reclamation.” De Alwis said the result is a practical and scalable solution for a global waste problem: “By enabling automated sorting, we’re giving recyclers and contractors a powerful tool to recover valuable resources and reduce landfill dependency.”
Next-gen skip bins – Robotics is the key to sorting timber from rubbish
The research comes after Wood Central reported last year that Monash researcher Diani Sirimewan is, in separate research, using AI and deep learning to develop a ‘skip master’ for construction waste and was now trawling Melbourne’s construction sites to find materials that could be reused in construction projects.
According to Sirimewan, also a PhD candidate at Monash’s Civil and Environmental Engineering Department, the problem is that waste is often laid down on the floor and manually sorted by labourers looking for valuable materials like timber: “Because labourers are manually sorting contaminated waste, there are health and safety concerns, as well as the need to move heavy and bulky materials.”
That’s when Sirimewan decided to use AI to trace materials “in the wild” across Melbourne’s construction sites. “If you feed a new image to the model, the model recognises whether it’s concrete, timber or metal,” she said, adding that waste management giants are already using robotics to sort through domestic waste.
Moving forward, Sirimewan hopes the new research will lead to investment in robotics and automation, improving Australia’s waste processing and recycling approach – a key pillar in the country’s pledge to become fully circular by 2030: “We are trying to ensure safety and health, not take away job opportunities,” she said. “In fact, automation will likely create more skilled roles. We still need workers to monitor the process and handle the machines and technology.”
- For more information about the circularity of timber and timber-based products, please review Wood Central’s special report.
