Share Email Print
cover

Proceedings Paper

Statistical analysis of infrared image sequences
Author(s): Wilhelm Meier; Heinz-Dieter vom Stein
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

This paper addresses the problem of focusing the attention of image processing steps to the relevant regions and structures in image sequences. This technique is especially useful in the presence of a high amount of noise and clutter, which is very often the situation in infrared image sequences. It helps to save computations and increases the reliability of segmentation/classification steps. For this purpose we propose a scene-independent hierarchical correlation analysis procedure. It uses a pyramidal image structure together with a top-down search strategy. Therefore it is capable to deal with moving, scaled, and deformed objects. The result is expressed as a feature vector for small image regions. Additionally, we present a segmentation technique based on the autocorrelation coefficient as a test statistic. We further extend this technique to cope with various other statistics and introduce a local variance quotient in image sequences to be used in a two sample Kolmogorov-Smirnov test.

Paper Details

Date Published: 17 August 1994
PDF: 7 pages
Proc. SPIE 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision, (17 August 1994); doi: 10.1117/12.182864
Show Author Affiliations
Wilhelm Meier, Univ. der Bundeswehr Hamburg (Germany)
Heinz-Dieter vom Stein, Univ. der Bundeswehr Hamburg (Germany)


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