Title :
A scalable and programmable simplicial CNN digital pixel processor architecture
Author :
Mandolesi, Pablo S. ; Julián, Pedro ; Andreou, Andreas G.
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. Nacional del Sur, Bahia Blanca, Argentina
fDate :
5/1/2004 12:00:00 AM
Abstract :
We propose a programmable architecture for a single instruction multiple data image processor that has its foundation on the mathematical framework of a simplicial cellular neural networks. We develop instruction primitives for basic image processing operations and show examples of processing binary and gray scale images. Fabricated in deep submicron CMOS technologies, the complexity of the digital circuits and wiring in each cell is commensurate with pixel level processing.
Keywords :
CMOS image sensors; cellular neural nets; digital signal processing chips; image processing; neural net architecture; CMOS image sensor; CNN digital pixel; binary images; cellular neural networks; digital circuits; gray scale images; image processing; image processor; pixel level processing; processor architecture; programmable architecture; vision chips; CMOS image sensors; CMOS process; CMOS technology; Cellular neural networks; Charge-coupled image sensors; Computer architecture; Evolution (biology); Image processing; Pixel; Sensor arrays; CMOS imager sensor; CNNs; Cellular neural networks; digital pixel; image processing; pixel level processing; vision chips;
Journal_Title :
Circuits and Systems I: Regular Papers, IEEE Transactions on
DOI :
10.1109/TCSI.2004.827626
