Title :
Generalised complex moment descriptors
Author :
Zaid, A. Abo ; Horne, E.
Author_Institution :
Res. Center, Cairo, Egypt
Abstract :
It is pointed out that, although complex moments are generally dependent, they have good noise immunity and simple rotational characteristics. However, due to the methods of normalization used previously, the results reported in the literature indicate that the complex moments approach is less efficient. The CMD (complex moment descriptor) provides a complete invariant set without any significant calculation for normalization with respect to rotation. A generalized algorithm for calculating the CMD coefficients is proposed. The calculation is simple and the output is a moments feature vector which is invariant to object transformation. Experimental results show the efficiency of the CMD in comparison with the combination approach
Keywords :
image recognition; complete invariant set; generalized algorithm; moments feature vector; noise immunity; rotational characteristics; Chromium; Error analysis;
Conference_Titel :
Circuits and Systems, 1991., Proceedings of the 34th Midwest Symposium on
Conference_Location :
Monterey, CA
Print_ISBN :
0-7803-0620-1
DOI :
10.1109/MWSCAS.1991.252147