Machine vision is a topic that is ‘in fashion’. But we have to be very careful when incorporating technology into our company because on many occasions said technology is a fad or is far from being implementable. The case of machine vision is not that. Machine vision is here, it is a mature and reliable technology that can save us time, improve our quality and ultimately generate cost savings. Today we tell you the 7 applications that you have to know about machine vision:
1. Machine Vision for Defect Detection
This is perhaps the most common application of machine vision. Until now, defect detection has been carried out by trained people on selected batches, and full control of production is often not possible. With machine vision we can detect defects such as: cracks in metals, paint defects, bad impressions, etc. in sizes less than 0.05mm. Much better than the human eye! These vision cameras need an algorithm that is the ‘intelligent brain’ that is capable of differentiating between what is a defect and what is not. This algorithm is designed and trained specifically for each particular application through defect and non-defect images.
2. Machine Vision for Metrology
It is another of the queens of the applications. What until now had been done with complex laser metrology equipment or probes can now be measured using machine vision. In this application, the key is to make a good adjustment of the reference to be able to measure with the necessary precision, and above all, to use the appropriate lighting for each type of material and work environment. Using machine vision systems we can measure variable part sizes, straightness, parallelism…
3. Machine Vision for Intruder Detection
According to DZOptics, through hyperspectral cameras it is possible to differentiate between a fruit and a stone, which allows, especially in food, that the products are safer for the consumer. Hyperspectral cameras, which, after all, are a type of machine vision, are capable of differentiating the type of material through the measurement they make of the wavelength. In this way, we can differentiate a stone from a fruit, a plastic from a metal or other combinations as long as the material is different.
4. Machine Vision for Verification of assemblies
Every day more and more complex assemblies are made, with more parts or connections. Machine vision allows us to check, step by step, that each piece is in its place, or at the end of the process, that the final assembly is correct. This application is very useful for the assembly of machinery, equipment, electronic boards or highly complex pre-assemblies. These systems considerably reduce cycle times of very complex operations and reoperation times.
5. Machine Vision for Screen Reader
Sometimes it is not possible to extract data from a display screen either because it is a provider’s closed system or because said system is incompatible with ours. A solution to this problem is to install a vision camera to read the screen and extract the data that appears on it (temperatures, codes, voltages… any useful information that appears on the screen and you need). To do this, we look for the regions of interest in which the information is found, we use a character recognition algorithm (OCR) to extract it, and everything is perfect! Go for more info here.
6. Machine Vision for reading codes and characters (OCR)
Let’s be honest, designers are very nice people, but they tend to change from typography to more complex letters (and pretty, mind you!) quite quickly. We turn again to character recognition algorithms to have a vision recognition system trained artificial, that you have the typography that you have, be able to read them. It is a system so robust that it is capable of reading even handwritten letters. Proof of the best designers!
7. Machine Vision + robotics for bin picking
And finally, one of the most requested applications, combining collaborative robotics with machine vision to be able to carry out chaotic bin picking of parts. The pieces are out of order and therefore we need to optimize the trajectories and detect the grip coordinates. The robot needs help to tell it what a piece is and where it is, so that it can decide which is the best way to pick up the piece.