DocumentCode
2687284
Title
On scene-adapted illumination techniques for industrial inspection
Author
Gruna, Robin ; Beyerer, Jürgen
Author_Institution
Vision & Fusion Lab., Karlsruhe Inst. of Technol. (KIT), Karlsruhe, Germany
fYear
2010
fDate
3-6 May 2010
Firstpage
498
Lastpage
503
Abstract
In many machine vision applications for automated inspection the illumination design is crucial to the robustness and speed of the inspection process. Hence, there is need to investigate and to experimentally evaluate new illumination designs and techniques. We briefly review a representative selection of illumination techniques that aim to minimize the effort of defect detection by adapting the illuminating light field to the nominal state of the inspection task. Based on this principle we propose an illumination technique using a projector-camera system which provides inspection images that directly display differences in the reflectance between two scenes. A comparison with image differencing for deviation detection shows that the proposed illumination technique is in many cases advantageous from a signal-to-noise point of view.
Keywords
cameras; computer vision; industrial engineering; lighting; automated inspection; deviation detection; image differencing; industrial inspection; inspection images; light field illumination; machine vision applications; projector-camera system; scene-adapted illumination technique; Digital images; Holographic optical components; Inspection; Layout; Lighting; Machine vision; Optical design; Optical polarization; Shape; Testing;
fLanguage
English
Publisher
ieee
Conference_Titel
Instrumentation and Measurement Technology Conference (I2MTC), 2010 IEEE
Conference_Location
Austin, TX
ISSN
1091-5281
Print_ISBN
978-1-4244-2832-8
Electronic_ISBN
1091-5281
Type
conf
DOI
10.1109/IMTC.2010.5488093
Filename
5488093
Link To Document