Quality check of industrial production
Massive industrial production requires careful quality check before placing final products on the market.
Computer Vision techniques enable very precise automatic controls, allowing products’ exhaustive verification by intelligent agents able to identify imperfections or differences with respect to the design to be implemented
The problem
Quality check of large quantities of products is difficult and expensive, because it requires the use of specialized personnel and cannot be performed on the entire production, but only on a sample basis.
This problem becomes even more evident in the case of products that require the assembly of numerous components, which must be positioned correctly to ensure the assembled system works in the way it was designed.
The verification of the coherence between the design and the finished product can be carried out in an exhaustive way only through automatic agents able to visually analyze the objects with the same logic of a man.
The solution
Revelis developed Colibri, a platform that uses neural networks to implement Computer Vision techniques and enable quality analysis in the industrial sector.
The neural networks are trained on the basis of datasets of the customer’s products, thus guaranteeing complete customization of the solution.
Colibri main features are:
– Automatic analysis of industrial products images
– Identification of basic components and their position
– Automatic check for the absence of components necessary for the final product
– Compliance verification with respect to design
