Shift invariant neural net for machine vision
- 1 January 1990
- journal article
- Published by Institution of Engineering and Technology (IET) in IEE Proceedings I Communications, Speech and Vision
- Vol. 137 (3) , 183-187
- https://doi.org/10.1049/ip-i-2.1990.0026
Abstract
A multi-layer network is described which is able to recognise simple shapes in a shift, size, and rotation invariant manner. The use of layers of units to smooth and then to shift the image eliminates the need for the very large numbers of cells which are often proposed in shift invariant networks. The network was trained using back-propagation and is not intended to be plausible as a model of biological vision at the level of cell and connection detail. Some interesting parallels with human vision are noted in the emergent behaviour of the network.Keywords
This publication has 3 references indexed in Scilit:
- On the Acquisition of Object Concepts from Sensory DataPublished by Springer Nature ,1989
- Why Cortices ? Neural Computation in the Vertebrate Visual SystemPublished by Springer Nature ,1989
- Parallel Distributed ProcessingPublished by MIT Press ,1986