Abstract
This paper describes a straightforward method of analysing the colour spectra of natural images using a hue histogramming technique. Examples of post-processed first moment data from these histograms axe then analysed using simple feedforward neural networks. These networks are shown to provide a good level of generalisation and can therefore be used for classification.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
D. H. Ballard and C. H. Brown. Computer Vision. Prentice Hall Inc., New Jersey, 1982.
J. J. Gibson. The Senses Considered as Perceptual Systems. Houghton-Mifflin, 1966.
S. Haykin. Neural Networks — A Comprehensive Foundation. MacMillan College Publishing Company, New York, 1994.
M. Merleau-Ponty. Eye and Mind. NorthWestern University Press, Evanston, 1964.
Y. Ohta. Knowledge Based Interpretation of Outdoor Natural Colour Scenes. Pitman, London, 1985.
C. Robertson and G. M. Megson. Parallel segmentation network with topological post-processor. Technical Report RUCS/96/TR/003/A, University of Reading, 1996.
A. Watt. Fundamentals of Computer Graphics. Addison-Wesley, Wokingham, UK, 1989.
A. Zell et al. Snns manual version 4.1. Technical Report 6/95, Institute of Parallel and Distributed High Performance Systems (IPVR), 1996. available by ftp from University of Stuttgart.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1998 Springer-Verlag Wien
About this paper
Cite this paper
Robertson, C., Megson, G.M. (1998). Neural Network Analysis of Hue Spectra from Natural Images. In: Artificial Neural Nets and Genetic Algorithms. Springer, Vienna. https://doi.org/10.1007/978-3-7091-6492-1_29
Download citation
DOI: https://doi.org/10.1007/978-3-7091-6492-1_29
Publisher Name: Springer, Vienna
Print ISBN: 978-3-211-83087-1
Online ISBN: 978-3-7091-6492-1
eBook Packages: Springer Book Archive