Identification of early decays in agricultural products using biospeckle laser screening and machine learning to reduce food waste
Research project
Food waste is a major issue, with 1.3 million tonnes lost annually in Sweden, often due to undetected early decay in fruits. Traditional methods are destructive, slow, and unsuitable for early spoilage. This project develops a fast, non-destructive technique using biospeckle imaging and AI to detect early decay. In collaboration with SLU and LTU, it aims to improve food quality control, reduce waste, and support sustainable agriculture through real-time monitoring
The project aims to develop a rapid, non-destructive method for detecting early decay in fruits and agricultural products. By combining advanced optical measurement techniques with AI, the system will reduce food waste, improve storage outcomes, and contribute to a more sustainable food system. The work is carried out in collaboration with SLU and LTU.
Food waste remains a significant global challenge, with more than one-third of food produced globally being lost or wasted. In Sweden alone, over 1.3 million tonnes of food were wasted in 2018, with substantial losses occurring in agricultural products such as tomatoes and potatoes—up to 50% and 40% respectively. A major contributor to this waste is the inability to detect early-stage decay in fruits and vegetables during storage and transport. Traditional methods for assessing food quality are destructive, time-consuming, and unable to identify subtle physiological changes before visible spoilage.
This project aims to address these limitations by developing a rapid, non-destructive optical inspection method for early decay detection in fruit and other agricultural products. The core of the system integrates biospeckle laser imaging with artificial intelligence to detect metabolic activity changes in produce, enabling early identification of decay processes before visible symptoms appear. By combining real-time imaging with machine learning classification, this approach will allow for high-throughput, automated quality control.
The project will be implemented in collaboration with the Swedish University of Agricultural Sciences (SLU) and Luleå University of Technology (LTU), ensuring a multidisciplinary approach. The outcomes will support the food industry by reducing waste, improving quality assurance, and contributing to sustainability and food security goals.