Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

How to detect Edgar Allan Poe's 'purloined letter' - Or: Cross correlation algorithms in digitised video images for object identification, movement evaluation and deformation analysis

: Dost, M.; Vogel, D.; Winkler, T.; Vogel, J.; Erb, R.; Kieselstein, E.; Michel, B.


Doctor, S.R.; Bar-Cohen, Y.; Aktan, A.E.:
Nondestructive Detection and Measurement for Homeland Security : San Diego, Tuesday 04 March 2003
Bellingham, WA: SPIE, 2003 (Proceedings of SPIE 5048)
ISBN: 0-8194-4853-2
Nondestructive Detection and Measurement for Homeland Security <2003, San Diego>
Fraunhofer IZM ()

Cross correlation analysis of digitised grey scale patterns is based on - at least - two images which are compared one to each other. Comparison is performed by means of a two-dimensional cross correlation algorithm applied to a set of local intensity submatrices taken from the pattern matrices of the reference and the comparison images in the surrounding of predefined points of interest. Established as an outstanding NDE tool for 2D and 3D deformation field analysis with a focus on micro- and nanoscale applications (microDAC and nanoDAC), the method exhibits an additional potential for far wider applications, that could be used for advancing homeland security. Cause the cross correlation algorithm in some kind seems to imitate some of the "smart" properties of human vision, this "field-of-surface-related" method can provide alternative solutions to some object and process recognition problems that are difficult to solve with more classic "object-related" image processi ng methods. Detecting differences between two or more images using cross correlation techniques can open new and unusual applications in identification and detection of hidden objects or objects with unknown origin, in movement or displacement field analysis and in some aspects of biometric analysis, that could be of special interest for homeland security.