Quality control process of industrial production through Computer vision
The quality control process could be empowered by Computer Vision techniques that 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 in massive industrial production, where careful quality control is essential before placing final products on the market.
The problem
Quality control process of large quantities of products without Computer Vision techniques 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 is 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 human.
The solution
Among the different AI solutions, 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
