Machine vision can be defined as the application of computer vision to production issues. Its principle is to equip production machines with the ability to see in order to automate quality control or process control tasks.
This automation increases performance and production rates, makes production more reliable, improves product quality, ensures traceability, and guarantees safety.
Machine vision applications are numerous today and have been opened to all sectors of industry. Indeed, technical progress both in terms of cameras and lighting systems and computer systems have enabled a considerable widening of the field of application of industrial vision.
The advantage of an industrial vision system is that it is possible, with a single device, to systematically carry out, on all products, several different continuous checks which would require, without this device, different equipment:
the assembly conformity check makes it possible to verify the absence or the presence of elements constituting the product to be manufactured, as well as their positioning and their orientation;
the aspect control is intended to examine the surface conditions in order to detect defects in appearance such as scratches, scratches, holes, stains, defects in nuances of colors or textures …
dimensional control consists in measuring the dimensions of a part such as a length, a diameter, a depth, an angle or a particular geometry;
counting and sorting of parts can also be carried out by an industrial vision system;
the control of machines or robots is also carried out thanks to vision systems. The camera then becomes the robot’s eye, which allows the detection and localization of a part to manipulate, assemble or align it. The tracking of a trajectory for a removal operation can also be carried out;
the identification is intended to carry out a recognition or a verification of characters, as well as the reading of bar codes, matrix codes or color codes. This operation makes it possible to reference a product and ensure its traceability and statistical processing, as well as to monitor inventory management or the production flow;
on industrial sites, but also in other fields, vision allows surveillance and security operations through access control (analysis of fingerprints, faces, hands, eyes, license plates ) or following crowds or waves of people.
Machine vision systems are therefore installed in production plants, and more precisely at critical stages that justify a check:
upon receipt of the raw materials and parts necessary for the development of the product;
during manufacturing in order to control the transformation of raw materials and the assembly of parts or in order to control the manufacturing machines;
at the end of manufacturing for the control of the finished product;
to the packaging.