Title :
A neural network accelerator for image analysis
Author :
Cosatto, Eric ; Craf, H.P.
Author_Institution :
Dept. of Adaptive Syst. Res., AT&T Bell Labs., Holmdel, NJ, USA
fDate :
6/1/1995 12:00:00 AM
Abstract :
At the heart of this image analysis system are two NET32K analog neural network chips, computing over 100 billion multiply-accumulates per second. The system simultaneously scans sixty-four 16×16-pixel templates over bi-level images, producing feature maps that mark matches between the image and the templates. When we code simple, generic shapes into the templates, the feature maps allow us to make quick, robust analyses of complex, noisy images
Keywords :
image processing; neural chips; neural net architecture; NET32K analog neural network chips; feature maps; image analysis; neural network accelerator; noisy images; templates; Analog computers; Image analysis; Image resolution; Image texture analysis; Neural networks; Neurons; Noise shaping; Pixel; Robustness; Shape;
Journal_Title :
Micro, IEEE
