9 January 1984 An Industrial Vision System Recognising Overlapping Industrial Parts Using Grey Scale Images Under A Wide Range Of Lighting Conditions
Author Affiliations +
Abstract
We describe the design of a computer vision system, which has been implemented and tested in software. The design is aimed at eventual hardware implementation using a video-stream processor approach, and to achieve recognition times of 1 or 2 seconds. The system uses 256x256@8 bit grey scale images of industrial parts which may overlap or touch. It is based upon Nevatia-Babu line extraction, the exhaustive matching of each pair of segments of each stored model separately and the weighing of evidence and extraction of identity and location of each part in the scene. The project is a university-industry collaboration, part of the British Government's SERC Robotics Initiative.
© (1984) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Alan H. Bond, Roger S. Brown, and Chris Rowbury "An Industrial Vision System Recognising Overlapping Industrial Parts Using Grey Scale Images Under A Wide Range Of Lighting Conditions", Proc. SPIE 0432, Applications of Digital Image Processing VI, (9 January 1984); doi: 10.1117/12.936650; https://doi.org/10.1117/12.936650
PROCEEDINGS
9 PAGES


SHARE
Advertisement
Advertisement
RELATED CONTENT

Automatic distribution of vision-tasks on computing clusters
Proceedings of SPIE (January 25 2011)
Edge Detection In Real-Time
Proceedings of SPIE (December 03 1984)
An Overview Of Computer Vision
Proceedings of SPIE (March 28 1988)
Contribution of edges and regions to range image segmentation
Proceedings of SPIE (February 29 1992)

Back to Top