January 2008

Machine vision: A matter of pixels and software

Machine vision (MV) is the application of computer vision to industry and manufacturing. 

It most often requires digital input/output devices and computer networks to control other manufacturing equipment such as robotic arms. 

One of the most common applications of machine vision is the inspection of manufactured goods such as semiconductor chips, automobiles, food, and pharmaceuticals. 

Much as human inspectors working on assembly lines visually inspect parts to judge the quality of workmanship, so machine vision systems use digital cameras, smart cameras, and image processing software to perform similar inspections. MV systems perform defined tasks such as counting objects on a conveyor, reading serial numbers, and searching for surface defects. Manufacturers favor machine vision systems for visual inspections that require high-speed, high-magnification, 24-hour operation, and/or repeatability of measurements. 

Imagine a part moving on a conveyor. It moves into position for inspection, and a sensor triggers the camera to take a picture of the part as it passes beneath. At the same time, a lighting pulse freezes and sharpens the image. The lighting highlights features of interest and minimizes the appearance of features that are not of interest (like shadows or reflections). The image then reverts to a framegrabber, which is a device that converts the output of the camera to digital format. 

This format is a two dimensional array of numbers, corresponding to the luminous intensity level of the corresponding point in the field of view. It is a pixel. 

The image goes to computer memory so the software can make the decision as to the suitability of the manufactured piece. This is the machine vision software.

The software goes through any number of steps including manipulation to reduce noise or to convert many shades of gray to a simple combination of black and white. In addition, it counts, measures, and/or identifies objects, dimensions, defects, or other features in the image. As a final step, the software passes or fails the part according to programmed criteria. 

If a part fails, the software may signal a mechanical device to reject the part; alternately, the system may stop the production line and warn a human worker to fix the problem that caused the failure.

Though most machine-vision systems rely on black-and-white cameras, the use of color cameras is becoming more common. 

It is also becoming popular for MV systems to include digital camera equipment for direct connection rather than a camera and separate framegrabber, thus reducing signal degradation.

Nicholas Sheble (nsheble@isa.org) writes and edits the Automation Basics department. Sources this month are the Banner Engineering and Wikipedia.


Beaming the points

A photoelectric sensor produces one beam of light. Therefore, there is only one point for inspection. 

For multiple points of inspection, you may have to use multiple sensors. 

A vision sensor "sees" an area and senses many thousands of distinct points (or pixels) in this area. 

Vision inspects multiple points on an object. By changing the lens of the sensor, one can make the field of view smaller or larger. 

A vision sensor can also sit further away from the part.


Source: Banner Engineering