The QUALISPECTRA project, in partnership with CETIC and the Walloon Agricultural Research Center (CRA-W), uses advanced machine learning algorithms and hyperspectral imaging to improve food safety. By integrating these technologies, QUALISPECTRA aims to provide rapid and reliable control of agri-food products, focusing on product authentication and contaminant detection. This initiative is part of innovations aimed at agile and safe design and production methods, as well as the agri-food chains of the future.
Expertises:
Data Science ⊕
Domains
Innovation theme: Artificial Intelligences ⊕
Factsheet:
The primary goal of the QUALISPECTRA project is to enhance food safety by using hyperspectral imaging methods for quality control of agri-food products. Faced with the challenges of reliable product authentication and effective contaminant management, QUALISPECTRA leverages a combination of artificial intelligence algorithms and hyperspectral imaging, enabling rapid and objective observation of samples. The CETIC, through its Data Science department, plays a key role in identifying the latest hyperspectral image analysis techniques while optimizing image acquisition and processing protocols to ensure precise, real-time diagnostics. The project also focuses on integrating AI into image analysis to develop future agri-food chains, with specific applications such as animal feed and spices.
QUALISPECTRA focuses on integrating AI into image analysis and developing future agri-food chains. In collaboration with CETIC, the project is developing innovative tools to identify contaminants and ensure product consistency in the food industry. CETIC, in partnership with CRA-W, is also working on optimizing image annotation protocols to guarantee the quality of data used in image processing algorithms. Major advances include the analysis of spectral variations over time and the extraction of sub-pixel level information, notably in the detection of Fusarium head blight in wheat. The project also plans to implement these algorithms in real time on CRA-W cameras, enabling rapid exploitation of the information contained in hyperspectral images and the direct detection of anomalies in products.
Thanks to its expertise in data science, CETIC significantly contributes to innovation in the field of hyperspectral imaging. Its role is essential in integrating cutting-edge AI algorithms into real-time analysis systems, enabling the efficient processing of large volumes of data. This approach strengthens the ability of agri-food companies to reliably and quickly ensure the quality and safety of their products. With the support of CETIC, QUALISPECTRA enables better management of food quality while providing innovative and scalable solutions for the industry. The project also aims to raise awareness among Walloon companies about the use and potential of hyperspectral image acquisition, whose industrial and agronomic applications remain largely underappreciated.
Finally, the QUALISPECTRA project has sparked great interest among several industrial players, particularly in the agri-food sector, who see this initiative as an opportunity to improve their quality control processes and optimize food safety.
