A detailed analysis of different CNN implementations for a real-time image processing system

Author

Wiehler, K. ; Perezowsky, M. ; Grigat, R.-R.

Author_Institution

Image Process. Syst., Tech. Univ. Hamburg-Harburg, Germany

fYear

2000

fDate

2000

Firstpage

351

Lastpage

356

Abstract

A detailed analysis for different implementations of a real-time CNN signal processing systems is presented. The algorithm for signal reconstruction has been realized both in hardware (analog VLSI multi-FPGA system) and in software (TriMedia VLIW Intel Pentium processor). All implementations are fully functional and embedded in a system environment. Due to the high computational complexity which is needed to solve the nonlinear CNN-equations and the requirements which are different for each application, an efficient implementation has to be tailor-made. In this paper we analyze different realized implementations regarding prototypical pre-requisites

Keywords

CMOS analogue integrated circuits; VLSI; cellular neural nets; image processing; neural chips; real-time systems; CMOS analogue IC; VLSI; cellular neural network; computational complexity; image processing system; real-time systems; Cellular neural networks; Hardware; Real time systems; Signal analysis; Signal processing algorithms; Signal reconstruction; Software algorithms; Software systems; VLIW; Very large scale integration;

fLanguage

English

Publisher

ieee

Conference_Titel

Cellular Neural Networks and Their Applications, 2000. (CNNA 2000). Proceedings of the 2000 6th IEEE International Workshop on

Conference_Location

Catania

Print_ISBN

0-7803-6344-2

Type

conf

DOI

10.1109/CNNA.2000.877354

Filename

877354