Share Email Print
cover

Proceedings Paper

Photogrammetric analysis of image invariance
Author(s): Hazem F. Barakat; Peter Doucette; Edward M. Mikhail
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Basic invariance theory for frame photography has existed in the photogrammetric literature for over a century. It has been recently rediscovered and significantly extended by researchers in computer vision (CV) and image understanding (IU). It is applied to problems involved in image transfer and object recognition and reconstruction, among others. Research is in progress at our Photogrammetric Analysis Laboratory, to analyze various image invariance techniques, for both 2D and 3D objects, particularly with regard to their accuracy and reliability. Existing well established photogrammetric techniques, as well as modifications thereto, are evaluated as both equivalent and complementary methods to invariance. Relationships are investigated between the imaging parameters and the variables involved in invariance, which usually combine such parameters together. Rigorous unified least squares is used in cases of redundancy and constraints. Results show significant improvement in accuracy and robustness as compared to direct linear techniques generally used in the CV/IU literature.

Paper Details

Date Published: 17 August 1994
PDF: 10 pages
Proc. SPIE 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision, (17 August 1994); doi: 10.1117/12.182850
Show Author Affiliations
Hazem F. Barakat, Purdue Univ. (United States)
Peter Doucette, Purdue Univ. (United States)
Edward M. Mikhail, Purdue Univ. (United States)


Published in SPIE Proceedings Vol. 2357:
ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision
Heinrich Ebner; Christian Heipke; Konrad Eder, Editor(s)

© SPIE. Terms of Use
Back to Top