Article Info ABSTRACT Article history:

The fact of using the classic descriptors such as Zernike Moment and Gist for a large data base has never been a satisfying method for perfect recognition rates. In this paper, we came up with a different approach based on the combination of the different descriptors already mentioned, it is the result of a comparative study of the different descriptors and the different combinations (Zernike + Gist, Zernike + PCA, Gist + PCA) in terms of recognition rate. Eventually, we have deduced that the combination of Zernike moment with Gist descriptors ended up to be the best hybrid description. For the recognition process, we opted for support vector machine (SVM) and Neural Networks (NN). We illustrate the proposed method on 3D objects using a 2D/3D approach based on characteristic views. Keyword: Gist descriptor Zernike moments PCA hybrid descriptors SV