NEW RESULTS AND MEASUREMENTS RELATED TO SOME TASKS IN OBJECT-ORIENTEDDYNAMIC IMAGE-CODING USING GNN UNIVERSAL CHIPS

Cellular neural/nonlinear networks (CNN) are considered here for effic ient Implementation of the most computationally intensive steps of dyn amic image coding, Several analogic CNN algorithms are presented for t he generation of binary image masks and image decomposition, Measureme nt results for the first CNN universal chips executing an analogic alg orithm for a reconstruction operator are also presented, Based on meas ured execution times, the viability of the CNN implementation of effic ient but computationally expensive compression algorithms such as dyna mic image coding is assessed.