Search
Menu
Cognex Corp. - Smart Sensor 3-24 GIF LB

Industrial Imaging and Vision

Apr 10, 2013
Facebook X LinkedIn Email
TO VIEW THIS WEBINAR:
Login  Register
About This Webinar
Additional Questions & Answers from the webinar are below

Speakers:

3-D Vision for Industrial Robot Guidance


Clay Flannigan
Manager, Robotics and Automation Engineering
Southwest Research Institute



The presentation will discuss current trends in the use of 3-D sensor data for industrial robot guidance. A brief background on 3-D sensor technologies will be provided, coupled with discussion of some of the challenges with data analysis in real-time applications. Several classes of problems will be presented including pose estimation, object recognition, and unstructured part processing. General techniques and specific software tools will be presented along with real-world application examples.

About the speaker: Clay Flannigan manages the Robotics & Automation Engineering Section of the Manufacturing Systems Department in the Automation and Data Systems Div. at Southwest Research Institute (SwRI). For more than 20 years, the Robotics and Automation Engineering Section has been developing innovative automation and robotics solutions. Flannigan is among the world-class experts and experienced engineers that comprise the automation engineering staff. Automation engineering facilities include state-of-the-art laboratories and large prototype areas for development.

1. Is there any value in using polarization imaging to improve contrast in machine vision?

Yes, in some circumstances it can help improve contrast or reduce glare, especially for specular (shiny) reflections. See here for some examples: http://www.edmundoptics.com/learning-and-support/technical/learning-center/application-notes/illumination/successful-light-polarization-techniques/

2. Bin Picking seems to be one of the typical applications where manipulating objects in cluttered scenes is required. What would you suggest to work in these kind of Bin Picking applications?

Bin picking is a classic and difficult problem for robotics. It is hard to evaluate the effectiveness of various solutions because they are so dependent on the particular application (part geometry, material, bin geometry, lighting, gripper, etc.). Several of the robot OEMs have solutions and I would start there. From what I've seen, Fanuc seems to have a well integrated solution: http://www.fanucrobotics.com/robot-applications/M-710iC_Bin_Picking.aspx. You will notice that most bin picking applications deal with relatively simple, large, symmetric parts because of the difficulty of the problem. If you are interested in trying to roll your own solution, I suggest you take a look at the Halcon library: http://www.mvtec.com/halcon/. If you need a system integrator or have a difficult problem, we might be able to help, but I'd want to understand your needs better as a first step.

3. I am working on technology for improved depth-of-field in imaging systems, and I would be curious to know whether limitations in depth-of-field are an issue in 3-D vision for current robotic applications (this one came in before the webinar).

Most of the industrial applications we see use relatively short focal length optics (wide field of view), small sensors (compared to DSLR for example), and small apertures (plenty of light available). These three factors reduce the problem of depth-of-field. I have seen cases where the geometry is constrained or lighting is limited where it could be a problem, but these tend to be edge cases. For example we did some 3D imaging of printed circuit boards where we used relative high magnification that limited our depth of field. I assume industries like silicon fab and computer/electronics run into these problems more often because of the scale that they operate at. Most of our work is on macro systems, and so it isn't a driving factor for us.

2013 North American Machine Vision Market Update


Alex Shikany
Director of Market Analysis
Association for Advancing Automation



Alex Shikany is the new Director of Market Analysis for the Association for Advancing Automation (A3). His responsibilities with A3 include producing quarterly statistical reports on robotics, vision, & motion control sales, publishing updates for A3’s websites with recent economic data & reports, developing articles on new market opportunities, and supporting A3’s branding campaign.

About the speaker: Alex is a recent graduate of Michigan State University’s Broad MBA program where he majored in Marketing and Finance. He also graduated with a Bachelor’s degree in Business Administration from Michigan State University. He has experience in both Marketing (Brand Strategy) and Market Research. Before coming to A3 Alex worked for Belden Wire and Cable, where he conducted an in depth market analysis of the Canadian market and developed a revised brand strategy for one of their main product lines. Prior to Belden Alex held a position with Pierce Education Properties, a property development company in the greater Lansing area, where he headed up their social media marketing team and coordinated numerous events across the area to grow the Pierce brand.

Questions & Answers

1. Where do the mechanical motion components / systems fall into (or break out of) the various markets?

Mechanical motion components / systems would fall into the “Other Machine Vision” category. AIA doesn’t have a dedicated category for those components at this time.

2. In your experience, what will be your forecast for the better industries where somebody can invest in next year?

For 2013, I would look for Computer and Electronic Products to be a hot-button industry for Machine Vision sales. Historically, it has been consistently one of the stronger industries. Additionally, Aerospace and Transportation equipment has been hot as of late. I would look for that to continue into 2013. Finally, I would look for Motor Vehicles and Parts to be a strong industry in 2013. With the increasing use of vision guided robotics and health of the automobile industry, one would expect it to be strong for machine vision as well.

3. What about the job scope in this area?

Jobs in Machine Vision are classified under manufacturing jobs. As such, experts believe that 2013 will be a strong year for them. We are seeing an increase in education for vision technologies at the university level as well. All else equal, historically we have seen that when manufacturing stays healthy, machine vision sales follows suit. In that case, companies will likely be looking to hire more talent and increase their number of employees.
industrialImagingVision SpectraSensors & Detectorsroboticsautomation
We use cookies to improve user experience and analyze our website traffic as stated in our Privacy Policy. By using this website, you agree to the use of cookies unless you have disabled them.