US 6,788,827 B1Granted

Image processing method and system for following a moving object from one image to an other image sequence

AssigneeKoninklijke Philips Electronics N.V.

InventorShérif Makram-Ebeid

Life of this patent

Estimating the term…

Loading patent…

Abstract

An image processing method for processing the images of an image sequence (S) comprising steps of determining image data related to first points of an Object of Interest observed in a first image, said Object of Interest having possible movements, and image data related to correlated points found in a second image of the sequence, and based on said image data, of estimating parameters of sets of transformation functions (400), which transformation functions transform said first points into said correlated points and, from said parameters, of determining one Warping Law (500) that automatically transforms said given Object of Interest of the first image into the same object in the second image of the sequence for following and locating (600) said Object of Interest in said second image of the sequence. Application: Medical imaging, X-ray examination apparatus

Claims13

Dependent claims

An image processing method for processing images of an image sequence representing an Object of Interest having possible movements, wherein the possible movements of the Object of Interest from a first image to a second image of the sequence is described, and the Object of Interest is located in the second image of the sequence, the method comprising the steps of:

defining first points of the Object of Interest in the first image,
determining points correlated to said first points, referred to as second points, found in the second image,
acquiring image data of the first and second points,
determining a set of a given number of parametric functions, which transform first points into correlated second points?
calculating the parameters of said parametric functions, and
estimating one general parametric law, referred to as Warping Law, using said parameters to automatically transform said given Object of Interest of the first image into the same object in the second image of the sequence.

3 more elements

Dependencies are inferred from the claim text.