Fragmentation in the vision of scenes

Natural images are highly structured in their spatial configuration. Where one would expect a different spatial distribution for every image, as each image has a different spatial layout, we show that the spatial statistics of recorded images can be explained by a single process of sequential fragmentation. The observation by a resolution limited sensory system turns out to have a profound influence on the observed statistics of natural images. The power-law and normal distribution represent the extreme cases of sequential fragmentation. Between these two extremes, spatial detail statistics deform from power-law to normal through the Weibull type distribution as receptive field size increases relative to image detail size.

[1]  W. Weibull.  A Statistical Distribution Function of Wide Applicability , 1951 .

[2]  K. Chung,et al.  Limit Distributions for Sums of Independent Random Variables. , 1955 .

[3]  M. Stephens.  EDF Statistics for Goodness of Fit and Some Comparisons , 1974 .

[4]  G. Matheron.  Random Sets and Integral Geometry , 1976 .

[5]  Alex Pentland,et al.  Fractal-Based Description of Natural Scenes , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  S. Rigdon.  Testing goodness-of-fit for the power law process , 1989 .

[7]  W. K. Brown.  A theory of sequential fragmentation and its astronomical applications , 1989 .

[8]  Stéphane Mallat,et al.  A Theory for Multiresolution Signal Decomposition: The Wavelet Representation , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  William Bialek,et al.  Statistics of Natural Images: Scaling in the Woods , 1993, NIPS.

[10]  W. K. Brown,et al.  Derivation of the Weibull distribution based on physical principles and its connection to the Rosin–Rammler and lognormal distributions , 1995 .

[11]  Shree K. Nayar,et al.  Reflectance and texture of real-world surfaces , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[12]  Terrence J. Sejnowski,et al.  The “independent components” of natural scenes are edge filters , 1997, Vision Research.

[13]  J. H. Hateren,et al.  Independent component filters of natural images compared with simple cells in primary visual cortex , 1998 .

[14]  Eero P. Simoncelli.  Modeling the joint statistics of images in the wavelet domain , 1999, Optics & Photonics.

[15]  Anuj Srivastava,et al.  Probability Models for Clutter in Natural Images , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[16]  Arnold W. M. Smeulders,et al.  Color Invariance , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  Ann B. Lee.  Occlusion Models for Natural Images : A Statistical Study of a Scale-Invariant Dead Leaves Model , 2001 .

[18]  A. Smeulders,et al.  A Physical Explanation for Natural Image Statistics , 2002 .

[19]  Anuj Srivastava,et al.  Universal Analytical Forms for Modeling Image Probabilities , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[20]  Alan Smeaton,et al.  NIST/SEMATECH Engineering Statistics Handbook Chapter 8: Reliability , 2003 .

[21]  J. Koenderink.  The structure of images , 2004, Biological Cybernetics.