Title :
A comparative study of moment invariants and Fourier descriptors in planar shape recognition
Author :
Ezer, Nejat ; Anarim, Emin ; Sankur, Bulent
Author_Institution :
Dept. of Electr. & Electron. Eng., Bogazici Univ., Istanbul, Turkey
Abstract :
Recent developments in the area of pattern recognition have brought up various methodologies to the problem of planar shape recognition. In this study, two popular feature sets, namely moment invariants and Fourier descriptors are applied to the problem of classification of 2-D images of airplanes, and their performance is compared vis-a-vis computational load and accuracy. An alternative form of the moment invariant technique, Zernike moments, is also compared with ordinary moment invariants
Keywords :
Fourier series; feature extraction; image classification; method of moments; military aircraft; 2-D images; Fourier descriptors; Zernike moments; accuracy; airplane images; classification; computational load; feature sets; military airplanes; moment invariants; pattern recognition; performance; planar shape recognition; Airplanes; Character recognition; Lighting; Marine vehicles; Military aircraft; Military computing; Pattern recognition; Postal services; Reflection; Shape;
Conference_Titel :
Electrotechnical Conference, 1994. Proceedings., 7th Mediterranean
Conference_Location :
Antalya
Print_ISBN :
0-7803-1772-6
DOI :
10.1109/MELCON.1994.380923
