Top 5 types of machine vision systems

For manufacturers who want to improve quality performance and automate production, machine vision systems are a key consideration. Constant improvement in vision system technology is contributing to the already rapid growth of Artificial Intelligence (AI).

In its simplest form, a vision system can be described as a computer with eyes. Smart technology enables it to not only identify something but also to inspect it and then communicate important data.

Here’s a summary of the top 5 types of machine vision systems:

1.       Identification

This consists of a range of applications which involve visually identifying a part or product. This can involve anything from identifying characters on a label, to confirming the correct part is in the right place on an assembly line before the next operation takes place. Some key benefits of this vision system include part traceability, part sorting and part feature verification. 

2.       Part Positioning

When it comes to high-speed line inspections or robotic pick and place tasks, machine vision is invaluable for part placement. Positioning tools can recognize and confirm the exact orientation and positioning of parts. Data is transferred to material handling devices that can better place the part in the correct position, if needed, or can trigger other automated operations to take place. 

3.       Flaw Detection

This is perhaps one of the most valuable tasks of machine vision systems; since quality control plays such a critical role in manufacturing. Machine vision can identify a range of quality issues, such as flaws, blemishes, cracks, pitting, or contamination for example. The main benefits are that issues can be picked up much faster, and far more accurately that if done by a human. In fact, the technology can pick up defects not even visible to the human eye. 

4.       Verification

The verification of parts, assemblies and packaged goods can be carried out by machine vision systems. Since there are such a vast range of verification options, they utilize the same tools used in other vision systems also. Some common examples are verification of a PCB assembly, blister packs, or the tolerances on an injection molded part.

5.       Tolerance Measurement

Vision system algorithms lend themselves to accurately checking tolerances on a part, to confirm it’s within spec and dimensionally accurate. This is a critical part in the quality process control stage of any manufacturing plant. Vision systems can analyse and verify the dimensions and tolerances of extremely complex parts and are extremely efficient.

The common benefit of each of these vision systems is the reduction of costly errors, improved productivity, and better overall quality; all of which ultimately equates to more profit. Because of this, Artificial Intelligence is set to keep developing steadily.

To learn more about the precedence of machine vision developments for autonomous vehicles, be sure to check out our article, Eyes on the Road: Machine Vision for Autonomous Driving.