Prior Learning and Gibbs Reaction-Diffusion

DSpace/Manakin Repository

Prior Learning and Gibbs Reaction-Diffusion

Citable link to this page

. . . . . .

Title: Prior Learning and Gibbs Reaction-Diffusion
Author: Zhu, Song Chun; Mumford, David Bryant

Note: Order does not necessarily reflect citation order of authors.

Citation: Zhu, Song Chun, and David Bryant Mumford. 1997. Prior learning and Gibbs reaction-diffusion. IEEE Transactions on Pattern Analysis and Machine Intelligence 19(11): 1236-1250.
Full Text & Related Files:
Abstract: This article addresses two important themes in early visual computation: it presents a novel theory for learning the universal statistics of natural images, and, it proposes a general framework of designing reaction-diffusion equations for image processing. We studied the statistics of natural images including the scale invariant properties, then generic prior models were learned to duplicate the observed statistics, based on minimax entropy theory. The resulting Gibbs distributions have potentials of the form U(I; Λ, S)=Σα=1kΣx,yλ (α)((F(α)*I)(x,y)) with S={F(1) , F(2),...,F(K)} being a set of filters and Λ={λ(1)(),λ(2)(),...,λ (K)()} the potential functions. The learned Gibbs distributions confirm and improve the form of existing prior models such as line-process, but, in contrast to all previous models, inverted potentials were found to be necessary. We find that the partial differential equations given by gradient descent on U(I; Λ, S) are essentially reaction-diffusion equations, where the usual energy terms produce anisotropic diffusion, while the inverted energy terms produce reaction associated with pattern formation, enhancing preferred image features. We illustrate how these models can be used for texture pattern rendering, denoising, image enhancement, and clutter removal by careful choice of both prior and data models of this type, incorporating the appropriate features
Published Version: doi:10.1109/34.632983
Other Sources:
Terms of Use: This article is made available under the terms and conditions applicable to Other Posted Material, as set forth at
Citable link to this page:

Show full Dublin Core record

This item appears in the following Collection(s)

  • FAS Scholarly Articles [6463]
    Peer reviewed scholarly articles from the Faculty of Arts and Sciences of Harvard University

Search DASH

Advanced Search