An expert in vision, he speaks from experience, having developed control systems in a wide variety of industries: “In agri-food, the product is alive, it is rarely 100% identical. When it is controlled by humans, the decision to comply is the result of consultation between different experts, on different types of defects. The challenge for the vision system is to take into account all these expertises and, at the same time, to smooth out human subjectivity.”

Baptiste Amato-Gagnon recommends a pragmatic and progressive approach to guarantee the performance of automated quality systems with vision because «there is no», he says, «magic in technology». For him, the only systems capable of meeting the needs of the agri-food industry are learner and supervised systems: "This will allow the machine to eliminate criteria that are not relevant to characterize a particular type of defect. As a result, the system will be much more accurate. I’ve seen unsupervised systems with huge ejection rates, not at all representative of actual quality.” A learning system is also a system that allows the company to capitalize on human expertise. “When an operator removes a product from the line, it adds value at a specific time. If he learns the machine why he did it, he creates a value that the company can exploit for years,” explains Baptiste Amato-Gagnon, who adds: “the machine will also return this value to the operator by making it easier for him to identify the root causes of his problems in production, to return more quickly to a compliant product.” For agri-food manufacturers who are still hesitant to get started, Baptiste Amato-Gagnon concludes with one last piece of advice: “Do not immediately seek to identify all defects. It is better to focus on the defects that make losing the most money. This results in very fast results, with returns on investment in a few months or even weeks.”