The CMAC coding schemes applied on FPGA for image reconstruction

Traditional hardware memory cannot generalize or learn from neighborhood information. Many software neural networks and fuzzy algorithms can solve above problems, but their processes are complicated. Therefore, a simple firmware memory combines learning ability with generalization ability, which applies the CMAC coding schemes on FPGA, is proposed in this paper. The firmware memory includes two parts: (1) the software algorithm is designed by the CMAC coding schemes; (2) the hardware structure is implemented by FPGA. The VHDL is utilized to program FPGA as the firmware memory. Finally, the example of image reconstruction illustrates the good learning effects of the proposed schemes.