Title: Image analysis by efficient Gegenbauer moments computation for 3D objects reconstruction

Authors: Zaineb Bahaoui; Hakim El Fadili; Khalid Zenkouar; Hassan Qjidaa; Arsalane Zarghili

Addresses: Intelligents Systèmes and Applications Laboratory (LSIA), Faculty of Technical Sciences Fez, Sidi Mohamed Ben Abdellah University, Fez, Morocco ' National School of Applied Sciences Appliquées Fez, Sidi Mohamed Ben Abdellah University, Fez, Morocco ' Intelligents Systèmes and Applications Laboratory (LSIA), Faculty of Technical Sciences Fez, Sidi Mohamed Ben Abdellah University, Fez, Morocco ' Faculty of Sciences Dhar el Mehraz, Sidi Mohamed Ben Abdellah University, Fez, Morocco ' Applications Laboratory (LSIA), Faculty of Technical Sciences Fez, Sidi Mohamed Ben Abdellah University, Fez, Morocco

Abstract: In this paper, we suggest a new technique for fast computation of Gegenbauer orthogonal moments for the reconstruction of 3D images/objects; a typical comparison of the proposed method with the conventional ZOA methods shows significant improvements in term of error reduction, image quality and consumption time. Then we compare our new approach with an existing method using Legendre and Zernike moments in the case of 3D image/object. The obtained results prove that Legendre and Zernike moments are slightly better than Gegenbauer, they are still very efficient and gives very good results in terms of MSE and PSNR. But, Zernike moments have higher computational cost than Gegenbauer moments.

Keywords: Gegenbauer moments computation; Legendre moments; Zernike moments; 3D images/object; computation time.

DOI: 10.1504/IJICT.2019.099111

International Journal of Information and Communication Technology, 2019 Vol.14 No.3, pp.263 - 276

Received: 01 Jul 2016
Accepted: 11 Jun 2017
Published online: 16 Apr 2019 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article