EP0587329A2 Expired / Lapsed EP ⏱️ 0.0 years remaining

Image processing system

Assignee: International Business Machines Corp

•

Filing: 1993-08-18

•

Granted: —

•

Expires: 2013-08-18

📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract The present invention provides a system and method for processing a sequence of images representing a moving object or feature of interest. The system comprises entry means for entering a next image from the sequence into the system, descriptor means for providing a current description of the feature, and location means for providing an estimate of the location of the feature in the next image. A feature extraction means extracts an area of data from said next image based on the estimate provided by the location means. A feature identification means programmed to compare tne extracted data with the current feature descriptor then determines the location of the feature in said next image, and a presentation means presents historical data describing past locations of the feature. Then an estimating device estimates the position of the feature in a further image, the estimation being performed on the basis of the location of the feature determined by said feature identification means and said historical data provided by the presentation means.

Inventors

👤 Richard William Turner

Bibliographic Details

Application Number —

Jurisdiction EP

Data Source google_patents

Last Synchronized 2026-10-26 15:15

CPC Classifications

G; G06; G06T; G06T7/00; G06T7/20; G06T7/246; H; H04; H04N; H04N19/00; H04N19/20

Claims Analysis

Total Claims: 8

Claim 1

Independent Claim 173 words

An image processing system for processing a sequence of images representing a moving feature of interest, comprising: entry means (10) for entering a next image from the sequence into the system; descriptor means (30) for providing a current description of the feature; location means (20) for providing an estimate of the location of the feature in the next image; feature extraction means (40) for extracting an area of data from said next image, said area being selected based on the estimate provided by the location means (20); feature identification means (40) programmed to compare the extracted data with the current feature descriptor to determine the location of the feature in said next image; presentation means (70) for presenting historical data describing past locations of the feature; and an estimating device (60) programmed to estimate the position of the feature in a further image, the estimation being performed on the basis of the location of the feature determined by said feature identification means (40) and said historical data provided by the presentation means (70).

Claim 2

Dependent Claim 42 words

A system as claimed in Claim 1, wherein said feature extraction means (40) is further programmed to process said extracted data using a probabilistic canny edge detection algorithm so as to generate an image showing only edges found by the canny algorithm.

Claim 3

Dependent Claim 33 words

A system as claimed in Claim 1 or Claim 2, wherein the output from said feature extraction means (40) is stored in memory to serve as the feature descriptor for said further image.

Claim 4

Independent Claim 30 words

A system as claimed in any of claims 1 to 3, wherein said feature identification means (40) compares the extracted data with the current feature descriptor by fourier domain cross-correlation.

Claim 5

Independent Claim 141 words

A method of processing a sequence of images representing a moving feature of interest, comprising the steps of: (a) entering a next image from the sequence into the system; (b) providing a current description of the feature; (c) providing an estimate of the location of the feature in the next image; (d) extracting an area of data from said next image, said area being selected based on the estimate provided at step (c); (e) comparing the extracted data with the current feature descriptor to determine the location of the feature in said next image; (f) presenting historical data describing past locations of the feature; and (g) estimating the position of the feature in a further image, the estimation being performed on the basis of the location of the feature determined at step (e) and said historical data provided at step (f).

Claim 6

Dependent Claim 40 words

A method as claimed in Claim 5, wherein step (d) further includes the step of processing said extracted data using a probabilistic canny edge detection algorithm so as to generate an image showing only edges found by the canny algorithm.

Claim 7

Dependent Claim 30 words

A method as claimed in Claim 5 or Claim 6, wherein the output from step (d) is stored in memory to serve as the feature descriptor for said further image.

Claim 8

Independent Claim 32 words

A method as claimed in any of claims 5 to 7, wherein the comparison in step (e) of the extracted data with the current feature descriptor is performed by fourier domain cross-correlation.

References Cited (Backward Citations)

No citation records indexed.

USPTO Annuity & Maintenance Windows

Maintenance schedule is calculated for granted US utility patents.

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag

Select Category

Notes (Optional)