Deconvolution filter design of transmission channel: application to 3D objects using features extraction from orthogonal descriptor

The proposed work focuses on new transmission method of 3D objects modelized by Roesser local state-space model and moment theory which is considered as an excellent descriptor for 3D objects. The orthogonal Hahn polynomial with optimal parameters and their moment correspond are used to generate the features extraction of 3D objects to generated the input system according to the order defined in advance. Practically this method is based on two pillars. The first is the exact selection of Hahn polynomial parameters for \(\alpha = 20, \beta = 0\) to take full advantage of the benefits in order to build moment matrix. The second one is the correlation of these generated matrices with Roesser local state-space model for transmitting the features vectors instead the full 3D object with, and without noises. As a result, the mean square error curve has been used to measure the performance of the proposed method. The PSNR curve is used to evaluate the validity of the proposed model. Finally, we show that the computational cost of \(\hbox {ETIR} = 80\%\) for the transmission, and comparison of the proposed approach with different kinds of noises generated by the environment.