Video processing system generating corrected geospatial metadata for a plurality of georeferenced video feeds and related methods

• US 8,933,961 B2
• Filed: 12/10/2009
• Issued: 01/13/2015
• Est. Priority Date: 12/10/2009
• Status: Active Grant

• Current Assignee: Harris Corporation (L3Harris Technologies, Inc.)
• Original Assignee: Harris Corporation (L3Harris Technologies, Inc.)
• Inventors: McDonald, Robert, Dunkel, Christopher T., Heminghous, John, Peterson, Aric, Bakir, Tariq
• Primary Examiner(s): Ma, Tize
• Application Number:
  US12/634,823
• Publication Number:
  US 20110145257A1
• Time in Patent Office: 1,860 Days
• US Class Current: 345/619
• CPC Class Codes: G06T 1/00 General purpose image data processing
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 7/292 Multi-camera tracking
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Abstract

A video processing system may include a video ingest module for receiving a plurality of georeferenced video feeds each including a sequence of video frames and initial geospatial metadata associated therewith, and each georeferenced video feed having a respective different geospatial accuracy level associated therewith. The system may further include a video processor coupled to the video ingest module and configured to perform image registration among the plurality of georeferenced video feeds, and generate corrected geospatial metadata for at least one of the georeferenced video feeds based upon the initial geospatial metadata, the image registration and the different geospatial accuracy levels.

Claims

1. A video processing system comprising:

a video ingest module for receiving a plurality of georeferenced video feeds each comprising a sequence of video frames and initial geospatial coordinates associated with objects in the video feed, and each georeferenced video feed having a respective different geospatial accuracy level associated therewith, and at least one of the georeferenced video feeds missing geospatial coordinates for a given object; and

a video processor coupled to said video ingest module and configured to

perform image registration between overlapping portions of the plurality of georeferenced video feeds, and

generate corrected geospatial coordinates for the missing coordinates of the given object in the at least one georeferenced video feed based upon an average of geospatial coordinates for the given object in other georeferenced video feeds weighted based upon the geospatial accuracy levels of their respective georeferenced video feeds.

2. The video processing system of claim 1 further comprising a display coupled to said video processor;

and wherein said video processor is also configured to display the sequence of video frames of the at least one georeferenced video feed on said display and with the corrected geospatial metadata associated therewith.

3. The video processing system of claim 1 further comprising at least one geospatial database coupled to said video processor for storing the corrected geospatial metadata.

4. The video processing system of claim 1 further comprising at least one geospatial database coupled to said video processor for storing the sequence of video frames for each video feed.

5. The video processing system of claim 1 further comprising at least one geospatial database coupled to said video processor for storing at least one geospatially referenced landmark;

and wherein said video processor is further configured to generate the corrected geospatial metadata based upon the at least one geospatially referenced landmark.

6. The video processing system of claim 1 wherein said video processor generates the corrected geospatial metadata every N number of video frames, where N is greater than 1.

7. The video processing system of claim 1 wherein each of the georeferenced video feeds comprises a live video feed.

8. A video processing system comprising:

a geospatial image database for storing a plurality of geospatial reference images;

a video ingest module for receiving a plurality of georeferenced video feeds each comprising a sequence of video frames and initial geospatial coordinates associated with objects in the video feed, and each georeferenced video feed having a respective different geospatial accuracy level associated therewith, and at least one of the georeferenced video feeds missing geospatial coordinates for a given object;

a display; and

a video processor coupled to said video ingest module, said display, and said at least one geospatial database and configured to

perform image registration between overlapping portions of the plurality of georeferenced video feeds and based upon the geospatial reference images,

generate corrected geospatial coordinates for the missing coordinates of the given object in the at least one georeferenced video feed based upon an average of geospatial coordinates for the given object in other georeferenced video feeds weighted based upon the geospatial accuracy levels of their respective georeferenced video feeds, and

display the sequence of video frames of the at least one georeferenced video feed on said display and with the corrected geospatial metadata associated therewith.

9. The video processing system of claim 8 further comprising at least one geospatial database for storing the corrected geospatial metadata and the sequence of video images for each video feed.

10. The video processing system of claim 8 wherein said video processor generates the corrected geospatial metadata every N number of video frames, where N is greater than 1.

11. A video processor for processing a plurality of georeferenced video feeds each comprising a sequence of video frames and initial geospatial coordinates associated with objects in the video feed, and each georeferenced video feed having a respective different geospatial accuracy level associated therewith, and at least one of the georeferenced video feeds missing geospatial coordinates for a given object, the video processor comprising:

an image registration module configured to perform image registration between overlapping portions of the plurality of georeferenced video feeds; and

a metadata correction module configured to generate corrected geospatial coordinates for the missing coordinates of the given object in the at least one georeferenced video feed based upon an average of geospatial coordinates for the given object in other georeferenced video feeds weighted based upon the geospatial accuracy levels of their respective georeferenced video feeds.

12. The video processor of claim 11 wherein said metadata correction module is also configured to display the sequence of video frames of the at least one georeferenced video feed on said display with the corrected geospatial metadata associated therewith.

13. A video processing method comprising:

receiving a plurality of georeferenced video feeds each comprising a sequence of video frames and initial geospatial coordinates associated with objects in the video feed, and each georeferenced video feed having a respective different geospatial accuracy level associated therewith, and at least one of the georeferenced video feeds missing geospatial coordinates for a given object;

performing image registration between overlapping portions of the plurality of georeferenced video feeds; and

generating corrected geospatial coordinates for the missing coordinates of the given object in the at least one georeferenced video feed based upon an average of geospatial coordinates for the given object in other georeferenced video feeds weighted based upon the geospatial accuracy levels of their respective georeferenced video feeds.

14. The method of claim 13 further comprising displaying the sequence of video frames of the at least one georeferenced video feed on a display and with the corrected geospatial metadata associated therewith.

15. The method of claim 13 wherein generating further comprises generating the corrected geospatial metadata also based upon at least one geospatially referenced landmark.