Smart industrial spectral cameras via artificial intelligence (SISCAI)

This project works on spectral imaging. Spectral imaging allows to perceive subtle optical signals or invisible wavelengths. The technology has many industrial applications in waste or food sorting and quality control. Powerful spectral image processing algorithms based on AI and deep learning exist, but are often limited to lab environments with controlled conditions and high-quality spectral cameras. The SISCAI project will bring these applications from the lab to practice. For this, it will develop smart industrial spectral cameras that will be optimized for their task. The cameras will be able to run AI-based processing algorithms on the device. The optimization of the camera design is based on the modeling of the design space, camera simulation and advanced optimization methods. These conceptual developments are demonstrated in two industrial contexts: the determination of fruit quality parameters in a greenhouse, and the quality inspection of food products in a free-fall sorting machine.