Colour Object recognition combining Motion Descriptors, Zernike Moments and Support Vector Machine
Fethi Smach, Cédric Lemaître, Johel Mitéran, Jean Paul Gauthier, Mohamed Saleh Abid · Proceedings of the Annual Conference of the IEEE Industrial Electronics Society · 2006
Fourier descriptors have been used successfully in the past to grey-level images, rigid bodied object. Here we used motion descriptors (MD) introduced recently by Gauthier et al., combined with Zernike Moments (ZM), in order to perform a recognition task in colour images. The feature vector for the MD obtained for each object appears to be unique and can be used for shape recognition. The MD, alone or combined with ZM, are used as an input of a support vector machine (SVM) based classifier. We illustrate results on three available datasets: ORL faces database, COIL-100, which consists of 3D objects and A R faces