RFC: Machine Vision

Technology Reserve seeks know-how deposits and Technology Packages in the field of machine vision, optical character recognition, and controlled or structured illumination.

 

Machine vision is a key automation technology improving the quality and productivity of manufacturing lines. Today’s factories require rapid, automated machine reading of product markings for quality assurance and inventory control. Automated processes reduce costs, decrease production times, and increase reliability. The global industrial machine vision market is expected to reach $2 billion by 2018. 1

While machine vision has conquered automation, one continuing challenge is accurately reading and processing low-contrast features, especially molded objects, with lettering composed of the same materi- als as the background. For example, a camera reading the molded, black-on-black features on a tire sidewall can require ten seconds to scan the image and another five seconds to process the image data. This is obviously unsuitable for production environments. 2

Several potential solutions exist to improve machine vision capabilities for molded lettering, including improvements to detection algorithms, illumination systems, and cameras and data processing hardware. A potential solution includes the use of structured light (e.g. Kinect systems) to enable 2D cameras to produce 3D vision. Since this requires only 2D cameras and optical processing systems, processing can be performed in less than a second. 3

Technology Reserve’s client seeks expertise in machine vision hardware and optical character recognition algorithms, including variable or structured illumination systems. Solutions must enable sub-second recognition of low-contrast features, and be cost effective for application within industrial production environments.

 

Peach, Matthew: “Global Industrial Machine Vision Sales to Hit $2bn by 2018,” Optics.org, 27 June, 2013. http://optics.org/news/4/6/45

Wilson, Andrew: “3-D Vision System Checks Tires,” Vision Systems Design (vision-systems. com). http://www.vision-systems.com/articles/print/volume-11/issue-2/features/industrial- automation-products/3-d-vision-system-checks-tires.html

Chen, S. Y.: “Vision Processing for Realtime 3-D Data Acquisition Based on Coded Structured Light,” IEEE Transactions on Image Processing, Volume 17, No. 2, February 2008. http://www.dmi. usherb.ca/~jodoin/cours/imn559/doc/SpacialPatternChen08.pdf

Key Capabilities:

  • Machine vision experience and optical character recognition expertise (optionally optical modeling capability for structured or variable illumination on 3D surfaces)
  • Familiarity with industrial production environments
  • A successful track record of design for manufacturing

An ideal expert partner will:

  • Deposit the necessary IP and produce a solution integrated with the client’s existing capabilities
  • Participate in the solving, packaging, assembly, or management of this project
 

 For inquiries, contact Astha Vijay, astha@techreserve.com