Title :
An Affine Invariant Region Detector Using the 4th Differential Invariant
Author :
Cai, Hongping ; Lei, Lin ; Su, Yi
Author_Institution :
Nat. Univ. of Defense Technol., Changsha
Abstract :
A novel affine invariant region detector based on the 4th differential invariant (DI4) is proposed in this paper. The detector combines scale-space theory with an autocorrelation matrix. Since it is proved that DI4 is a scale- space selection function, feature points and their characteristic scales are first detected by the local maxima of the normalized DI4 over scale-space. Then, the auto-correlation matrices, which are used to describe the affine shapes, are estimated on the characteristic scales of the feature points. The ellipse regions given by the auto-correlation matrices are affine invariant. In order to verify the affine invariance, we build up a simulation experiment to test affine invariance using two single- parameter transforms. The experimental results show the detected regions are invariant to rotation, scale and affine transforms as well as robust to illumination changes.
Keywords :
affine transforms; feature extraction; matrix algebra; 4th differential invariant; affine invariant region detector; affine transforms; autocorrelation matrix; scale-space selection function; scale-space theory; single-parameter transforms; Artificial intelligence; Autocorrelation; Detectors; Gaussian processes; Kernel; Laplace equations; Lighting; Robustness; Shape; Testing;
Conference_Titel :
Tools with Artificial Intelligence, 2007. ICTAI 2007. 19th IEEE International Conference on
Conference_Location :
Patras
Print_ISBN :
978-0-7695-3015-4
DOI :
10.1109/ICTAI.2007.112