Title :
A Random Projection Imager for Visual Pattern Classification in Analog VLSI
Author :
Bridges, Seth ; Holleman, Jeremy ; Mitros, Ania ; Diorio, Chris ; Figueroa, Miguel
Author_Institution :
Dept. of Comput. Sci. & Eng., Washington Univ., Seattle, WA
Abstract :
In this paper, we present a novel CMOS imager architecture that implements the random projection dimensionality reduction algorithm in the focal plane. We employ analog signal processing techniques to achieve low-power operation and our imager can readily integrate with known low-power VLSI classifiers. We fabricated a 20 times 20 pixel prototype of our 4.2mm2 imager in 0.35 mum CMOS that performs 1GOPS while consuming 1.25mW of power from a 5V supply
Keywords :
CMOS image sensors; VLSI; low-power electronics; pattern classification; signal processing; 0.35 micron; 1.25 mW; 5 V; CMOS imager architecture; analog VLSI; analog signal processing techniques; low-power VLSI classifiers; random projection imager; visual pattern classification; Analog computers; Face detection; Hardware; Pattern classification; Pixel; Principal component analysis; Prototypes; Signal processing algorithms; Vectors; Very large scale integration;
Conference_Titel :
Solid-State Circuits Conference, 2006. ESSCIRC 2006. Proceedings of the 32nd European
Conference_Location :
Montreux
Print_ISBN :
1-4244-0303-0
DOI :
10.1109/ESSCIR.2006.307472
