A layered architecture for regularization vision chips
- 1 January 1991
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1007-1020 vol.2
- https://doi.org/10.1109/ijcnn.1991.170530
Abstract
The authors propose a layered architecture for regularization problems with higher order smoothness constraints which requires only immediate neighborhood wiring and demands no negative conductance. They describe the architecture and show how the network naturally solves regularization problems. They also present an application to the smoothing-contrast enhancement filter for image processing. The authors explain how the architecture has been inspired by physiological experiments on lower vertebrate retina. A CMOS circuitry is given which has been confirmed to work at the SPICE level.<>Keywords
This publication has 13 references indexed in Scilit:
- Stability of image processing neuro chips: spatial and temporalPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- An active resistor network for Gaussian filtering of imagesIEEE Journal of Solid-State Circuits, 1991
- An analog CMOS network for Gaussian convolution with embedded image sensingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1990
- Dynamic model of dual layer neural network for vertebrate retinaPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1989
- Computing motion using analog and binary resistive networksComputer, 1988
- A silicon model of early visual processingNeural Networks, 1988
- Interaction between the soma and the axon terminal of retinal horizontal cells in Cyprinus carpio.The Journal of Physiology, 1986
- On the dynamics of electrical networksJournal of Differential Equations, 1976
- Electrical properties of the rod syncytium in the retina of the turtle.The Journal of Physiology, 1976
- Receptive fields of cones in the retina of the turtleThe Journal of Physiology, 1971