Share Email Print

Proceedings Paper

Design of an aid to visual inspection workstation
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Visual Inspection is the most common means for inspecting manufactured parts for random defects such as pits, scratches, breaks, corrosion or general wear. The reason for the need for visual inspection is the very random nature of what might be a defect. Some defects may be very rare, being seen once or twice a year, but May still be critical to part performance. Because of this random and rare nature, even the most sophisticated image analysis programs have not been able to recognize all possible defects. Key to any future automation of inspection is obtaining good sample images of what might be a defect. However, most visual check take no images and consequently generate no digital data or historical record beyond a simple count. Any additional tool to captures such images must be able to do so without taking addition time. This paper outlines the design of a potential visual inspection station that would be compatible with current visual inspection methods, but afford the means for reliable digital imaging and in many cases augmented capabilities to assist the inspection. Considerations in this study included: resolution, depth of field, feature highlighting, and ease of digital capture, annotations and inspection augmentation for repeatable registration as well as operator assistance and training.

Paper Details

Date Published: 19 May 2016
PDF: 8 pages
Proc. SPIE 9868, Dimensional Optical Metrology and Inspection for Practical Applications V, 98680B (19 May 2016); doi: 10.1117/12.2225152
Show Author Affiliations
Robert Tait, GE Global Research (United States)
Kevin Harding, GE Global Research (United States)

Published in SPIE Proceedings Vol. 9868:
Dimensional Optical Metrology and Inspection for Practical Applications V
Kevin G. Harding; Song Zhang, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?