论文信息 - The principal components of natural images

The principal components of natural images

A neural net was used to analyse samples of natural images and text. For the natural images, components resemble derivatives of Gaussian operators, similar to those found in visual cortex and inferred from psychophysics. While the results from natural images do not depend on scale, those from text images are highly scale dependent. Convolution of one of the text components with an original image shows that it is sensitive to inter-word gaps.

Leslie S. Smith | R. Baddeley | P. Hancock | L. Smith

[1] E. Oja. Simplified neuron model as a principal component analyzer , 1982, Journal of mathematical biology.

[2] Professor Dr. Guy A. Orban. Neuronal Operations in the Visual Cortex , 1983, Studies of Brain Function.

[3] R. Linsker,et al. From basic network principles to neural architecture , 1986 .

[4] Ralph Linsker,et al. Self-organization in a perceptual network , 1988, Computer.

[5] Terence D. Sanger,et al. Optimal unsupervised learning in a single-layer linear feedforward neural network , 1989, Neural Networks.

[6] D. Mackay,et al. Analysis of Linsker's application of Hebbian rules to linear networks , 1990 .

[7] R. Baddeley,et al. A statistical analysis of natural images matches psychophysically derived orientation tuning curves , 1991, Proceedings of the Royal Society of London. Series B: Biological Sciences.

[8] Anders Krogh,et al. Introduction to the theory of neural computation , 1994, The advanced book program.

[9] D. Foster,et al. Asymmetries in oriented-line detection indicate two orthogonal filters in early vision , 1991, Proceedings of the Royal Society of London. Series B: Biological Sciences.