19th IEEE International Conference on Tools with Artificial Intelligence - Vol.1 (ICTAI 2007)
An Affine Invariant Region Detector Using the 4th Differential Invariant
Paris, France
October 29-October 31
ISBN: 0-7695-3015-X
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 auto- correlation 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.
Citation:
Hongping Cai, Lin Lei, Yi Su, "An Affine Invariant Region Detector Using the 4th Differential Invariant," ictai, vol. 1, pp.540-543, 19th IEEE International Conference on Tools with Artificial Intelligence - Vol.1 (ICTAI 2007), 2007