Patent · US Active

System and method for dense image registration using Markov Random Fields and efficient linear programming

US8126291B2 · kind B2 · utility

6Cited by

2References

27Claims

0Family size

Assignee

Ecole Centrale de Paris · FR

Inventors

Nikos Paragios · Paris, FR
Benjamin Michael Glocker · Munich, DE
Nikos Komodakis · Avenida de la Hispanidad, GR

Key dates

Filing date	Jul 8, 2008
Grant date	Feb 28, 2012
Priority date	—
Expiry date	Dec 19, 2030

Classification

Technology area (CPC G)Physics
CPC primaryG06T2207/30016
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

A method for registering digitized images using Markov Random Fields (MRFs) includes providing a source image f and a target image g, defining a deformation grid of control points, defining a coordinate transformation aswhere x is a point on the source image, p is a position vector of control point p, dp is a displacement vector for each control point, up is a label for point p associated with displacement dp, and η( ) is a weighting function for the displacement vector, defining an MRF energy functional to be minimized by T aswherein |G| is a number of control points, |E| is a number of pairs of neighboring control points on a neighborhood system, t is an iteration counter, and associating the MRF with a primary linear program and solving the primary linear program using a fast primal-dual algorithm to yield a coordinate transformation that minimizes the energy functional.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.