In the context of sustainable development, the use of machine vision in the minerals industry aims directly at reducing the environmental impact by optimizing production facilities and improving quality control. The project I lead with my collegue Paolo Di Carlo takes place in the Marshal Plan. Indeed, the mineral industry occupies an important place in the Walloon economy and is facing important environmental challenges.
COGOLIN project is aimed at improving the control, by machine vision, of size and morphology of bulk rock fragments scrolling on a conveyor belt, in harsh industrial environments (quarries,mines, …). The project involves the design, validation and installation of a scanner based on laser triangulation providing surfometric images. Those images have then to be analysed with specific algorithms. Algorithms and the user interface application allowing for instant results reporting constitute a major part of the research.
Size and shape control is of major interest at many stages of the processes developed in the mineral industry. Such controls are usually performed in the industry by manual sampling and sieving. Unfortunately, these techniques are unable to provide regular measurements in « real time » and, thus, do not permit on-line control and optimization of the installations.