DocumentCode :
2229198
Title :
A multiplier-free fixed-task digital CNN array for video segmentation system
Author :
Paasio, Ari ; Paakkulainen, Jani ; Isoaho, Jouni
Author_Institution :
Lab. of Electron. & Inf. Technol., Turku Univ., Finland
Volume :
3
fYear :
2000
fDate :
2000
Firstpage :
710
Abstract :
In this paper an alternative approach is given for implementing a fixed template cellular nonlinear network processor. The new concept is based on calculations in the digital domain to achieve accuracy not easily available with analog circuitry. A particular task, namely a noise removal operation, is considered here. Simplifications to the integration update rule are introduced that make the hardware realization simple. Moreover, results are given where the number of iteration steps in the integration show the achieved accuracy in the processed image with different word lengths. The design of the processor cell is given in a block level and both the processing time and the size of the hardware realization of this topology are estimated
Keywords :
CMOS digital integrated circuits; cellular neural nets; digital signal processing chips; image segmentation; interference suppression; neural chips; video signal processing; cellular nonlinear network processor; fixed template nonlinear CNN processor; fixed-task digital CNN array; integration update rule; multiplier-free digital CNN array; noise removal operation; processor cell design; video segmentation system; Algorithm design and analysis; Cellular neural networks; Electronic circuits; Gray-scale; Hardware; Image processing; Image resolution; Information technology; Laboratories; Pixel;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Circuits and Systems, 2000. Proceedings. ISCAS 2000 Geneva. The 2000 IEEE International Symposium on
Conference_Location :
Geneva
Print_ISBN :
0-7803-5482-6
Type :
conf
DOI :
10.1109/ISCAS.2000.856159
Filename :
856159
Link To Document :
بازگشت