Break down industrial machine vision into 5 steps

Machine vision companies continue to spend time resources in person, print, and online, educating people about the basics of industrial machine vision. Control Engineering has written more than a few pieces on the topic, also. The following explanation is based on a presentation from Kyle Voosen, product manager, machine vision, and Nicolas Vazquez, principal engineer, National Instruments . They looked at the basics of machine vision at NIWeek 2004. It’s worth noting that systems vary according to need, application, and equipment used. In some cases, for example, no separate lighting or controller is needed; lighting or logic might be included within the machine vision hardware and be enough for a particular application. Some products combine, automate, or simplify these steps.

At bottom are links to other Control Engineering machine vision coverage.

An industrial machine vision system needs to acquire the image, inspect the image, integrate with logic, configure the human-machine interface (HMI), and archive results. The system can include integrated or separate lighting and optics, camera or sensor, frame grabber or vision device, and application software.

1. Image acquisition generally requires configuring triggers, setting up cameras and lighting, and then actually acquiring images.
2. Image inspection involves enhancing the image, locating and measuring features, checking for presence (or absence), and/or identifying parts.

3. Integration with control logic means transferring results to a programmable logic controller (PLC) [or PC, for that matter], setting inspection parameters, closing control loops, toggling digital inputs and outputs, reading encoders, and sending and receiving RS-232 [or other communications].

4. Configuration of the human-machine interface can include viewing inspection results, customizing an overlay, selecting inspection, and adjusting inspection and related setpoints.

5. Archiving of results can occur on local storage, on a removable hard drive, or via network FTP (file transfer protocol).

NI’s site has a technical tutorial called “Machine Vision and Scientific Image Processing,” said to help with developing a machine vision and image processing application. (Viewing is free with registration.)

Related reading from Control Engineering includes:

• Application Update: Machine vision. A foundry gives vision to robotics; inspecting transmissions and jar lids

• Vision: Robotic patent, products

• Product Focus: Machine vision investment increases

• ‘See’ the Results: Vision and code readers feed data acquisition systems, analyzing more applications with greater consistency and easier setup than ever before. Here’s how it’s done.

—Mark T. Hoske, editor-in-chief, Control Engineering, MHoske@cfemedia.com

Do you have experience and expertise with the topics mentioned in this content? You should consider contributing to our WTWH Media editorial team and getting the recognition you and your company deserve. Click here to start this process.