Share Email Print

Proceedings Paper

Principal Components Analysis Of Images Via Back Propagation
Author(s): Garrison W. Cottrell; Paul Munro
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The recent discovery of powerful learning algorithms for parallel distributed networks has made it possible to program computation in a new way, by example rather than algorithm. The back propagation algorithm is a gradient descent technique for training such networks. The problem posed to the researcher using such an algorithm is discovering how it did solve the problem if a solution is found. In this paper we apply back propagation to a well understood problem in image analysis, i.e., bandwidth compression, and analyze the internal representation developed by the network. The network used consists of nonlinear units that compute a sigmoidal function of their inputs. It is found that the learning algorithm produces a nearly linear transformation of a Principal Components Analysis of the image, and the units in the network tend to stay in the linear range of the sigmoid function. The particular transform found departs from the standard Principal Components solution in that near-equal variance of the coefficients results, depending on the encoding used. While the solution found is basically linear, such networks can also use the nonlinearity to solve encoding problems where the Principal Components solution is degenerate.

Paper Details

Date Published: 25 October 1988
PDF: 8 pages
Proc. SPIE 1001, Visual Communications and Image Processing '88: Third in a Series, (25 October 1988); doi: 10.1117/12.969060
Show Author Affiliations
Garrison W. Cottrell, University of California San Diego (United States)
Paul Munro, University of Pittsburgh (United States)

Published in SPIE Proceedings Vol. 1001:
Visual Communications and Image Processing '88: Third in a Series
T. Russell Hsing, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?