Construcciones Yamaro: AI system detects contamination in construction wood waste

Much of the contaminated wood from building sites is still destined for landfill. (Image: leszekglasner/stock.adobe.com)

Researchers from Monash University and Charles Darwin University (CDU) have developed an artificial intelligence (AI) system capable of identifying contaminated construction and demolition wood waste with 91 per cent accuracy.

Published in Resources, Conservation & Recycling, the study introduces a real-world image dataset of contaminated wood waste – a development that could help improve recycling rates and reduce landfill reliance in the construction sector.

The research team, led by Madini De Alwis and Dr Milad Bazli from CDU, and supervised by associate professor Mehrdad Arashpour, head of construction engineering at Monash, trained deep learning models to detect six common contamination types using standard RGB images.

“We curated the first real-world image dataset of contaminated construction and demolition wood waste,” said De Alwis, a PhD candidate at Monash’s Department of Civil and Environmental Engineering. “This new system could be deployed via camera-enabled sorting lines, drones or handheld tools to support on-site decision-making.”

Related stories:

• Monash files patent for construction AI technology
• Monash launches infrastructure testing facility to save lives
• Construction commences for $10 million CDU Trades Centre

While computer vision has been applied to general waste streams, its use in detecting contaminated wood waste has remained limited – until now.

“By fine-tuning state-of-the-art deep learning models, including CNNs and Transformers, we showed that these tools can automatically recognise contamination types in wood using everyday RGB images,” said Bazli.

Wood is among the largest contributors to global construction waste. Although much of it is recyclable, contamination makes recovery labour-intensive and expensive. By automating the identification process, the researchers aim to support scalable waste-sorting solutions that align with Australia’s circular economy strategy.

“This opens the door to scalable, AI-driven solutions that support wood waste reuse, recycling and reclamation,” said De Alwis. “This is a practical, scalable solution for a global waste problem.”

The post AI system detects contamination in construction wood waste appeared first on Inside Construction.

View Source