5 May 2014 Invariant correlation filters comparison for multiclass recognition of scaled objects
Author Affiliations +
Abstract
There are presented results of scaled image recognition modeling with the help of optical correlator with invariant correlation filters like MACE, GMACE, MINACE and DCCF. The image database for testing include images of true and false classes. There are presented qualitative and quantitative characteristics of output correlation peaks. Also there is provided an analysis of positive and negative side of each filter's type as for single class only scaled images recognition so for multiclass one. Also there are shown results for modeling of images with more complex distortions recognition using all mentioned types of filters.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Petr A. Ivanov, Petr A. Ivanov, } "Invariant correlation filters comparison for multiclass recognition of scaled objects", Proc. SPIE 9094, Optical Pattern Recognition XXV, 90940L (5 May 2014); doi: 10.1117/12.2053865; https://doi.org/10.1117/12.2053865
PROCEEDINGS
6 PAGES


SHARE
Back to Top