Industrial Automation

    

Blog

Machine vision systems are an essential component of industrial automation solutions. They extract information from images taken during manufacturing processes and send that information over to equipment that will then act on it. In a sense, they are like computers that have eyes. Although there is a lot of variation among machine vision systems, they all include certain essential components to perform the basic functions of capturing and processing images, and then delivering an actionable result based upon features in the visual data.

CAPTURING THE DATA

The first step in putting together a machine vision system is setting up one or more cameras in key locations. Choosing the optimal lighting is essential, since proper lighting significantly reduces the likelihood of false information such as false positives in a presence/absence test.

The camera(s) will capture visual data with the goal of executing a series of processing steps to extract information that will be of use in guiding robots or ensuring product quality.

PROCESSING THE DATA AND GLEANING INFORMATION

In many machine vision systems, cameras are separated from the processing unit. They communicate with a computer via a specific protocol and send along the captured images so that the computer can process the visual data according to the needs of the manufacturer. Sometimes, however, the imager itself is combined with a processing apparatus in a single unit. These are known as smart cameras, and they have the advantage of being self-contained image capture and processing units. Cameras connected to computers tend to be more powerful and flexible, but smart cameras usually perform better and more reliably in harsh operating environments than their PC-based counterparts.

Whether captured by cameras hooked up to a computer or by a smart camera with processing technology included, the actual extraction of information looks very similar. Basic functions in a machine vision toolkit include pattern recognition, noting the presence/absence of a given item, counting items of a specific type, measuring the distance between two items, determining an item’s location, as well as code reading and code verification.

It’s important for manufacturers to select the appropriate software for their needs. Some companies have access to machine vision professionals who can take advantage of highly customizable software to perform a wide variety of tasks. Others do better by sticking with the basics, knowing that their fundamental requirements are being satisfied without giving anyone a headache. Because machine vision familiarity varies widely from manufacturer to manufacturer, Omron Microscan has made a point of incorporating two different user interfaces for its software – one for the average operator and another for the expert.

OUTPUTTING AND ACTING UPON THE DATA

Once the information has been gathered and processed, it’s time to put it to use. Logic incorporated into the software takes the information gleaned from machine vision technology and uses it to make pass/fail decisions or provide guidance for robotic systems. Hardware components of the system will react to the signals generated by the logic, perhaps by moving a robotic arm to a given location or booting a bottle with a botched cap off the production line.

Machine vision systems can gather useful information and process it with much greater speed and accuracy than human operators. They can also work around the clock with reliable performance. For these reasons, there is an enormous return on investment for manufacturers who implement a machine vision system. With proper setup and smart use of the image processing toolkit, manufacturers can cut costs, improve quality, fix bottlenecks and move closer to the factory of the future.

Learn here about the newest awe-inspiring addition to the family of Omron Microscan products - the new HAWK MV-4000 high-performance smart camera with near PC speeds! The golden middle you may be searching for.

Recent Posts


Leave a Comment

Comments

Written by on Reply