Automated metric information network

• US 9,875,404 B2
• Filed: 12/30/2015
• Issued: 01/23/2018
• Est. Priority Date: 02/07/2013
• Status: Active Grant

• Current Assignee: Maxar Intelligence Incorporated (Galileo Parent, Inc.)
• Original Assignee: DigitalGlobe, Inc. (Galileo Parent, Inc.)
• Inventors: Johnston, Carolyn P., Sjahputera, Ozy, Bleiler, Laurence C., Bader, Brett W., Smith, II, James Thomas
• Primary Examiner(s): Bitar, Nancy
• Application Number:
  US14/984,613
• Publication Number:
  US 20160117552A1
• Time in Patent Office: 755 Days
• Field of Search: None
• CPC Class Codes: G01C 11/06 by comparison of two or more pictures of the same area
• CPC Class Codes: G06V 20/13 Satellite images
• CPC Class Codes: G06V 30/422 Technical drawings; Geographical maps

Abstract

A Metric Information Network (MIN) with a plurality of Ground Control Points (GCPs) that are selected in an automated fashion. The GCP selection includes clustering algorithms as compared to prior art pair-wise matching algorithms. Further, the image processing that takes place in identifying interest points, clustering, and selecting tie points to be GCPs is all performed before the MIN is updated. By arranging for the processing to happen in this manner, the processing that is embarrassingly parallel (identifying interest points, clustering, and selecting tie points) can be performed in a distributed fashion across many computers and then the MIN can be updated.

Claims

1. A method of selecting Ground Control Points (GCPs) for a geodetic network, comprising:

with a computer, receiving more than one photographic image of the ground in a given area on the Earth;

with a computer, detecting interest points in each photographic image and generating an interest point descriptor and a geo-location for each interest point;

with a computer, clustering the interest points in the more than one photographic image based on geo-location to create clusters of interest points, wherein at least one of the clusters includes at least one interest point from a first one of the more than one photographic image and at least one interest point from a second one of the more than one photographic image;

with a computer, comparing the interest point descriptors for the interest points within each cluster;

with a computer, grouping at least some of the interest points based on the comparison; and

with a computer, selecting certain groups as GCPs that directly tie together two or more received photographic images of the ground in a given area on the ground, wherein the selecting includes considering the similarity of the geo-locations and of the descriptors of the interest points and considering the quantity of interest points within a group.

2. A method as defined in claim 1, wherein each operation is performed without human intervention.

3. A method as defined in claim 1, wherein the selecting operation includes performing a deblundering operation to discard groups that are less worthy than others.

4. A method as defined in claim 1, wherein the geo-location includes earth-centered, earth-fixed coordinates.

5. A method as defined in claim 1, wherein the geo-location includes three dimensions of spatial coordinates.

6. A method as defined in claim 1, wherein the photographic images have been orthorectified prior to detecting of the interest points.

7. A method as defined in claim 1, wherein the photographic images have a normalized pixel resolution.

8. A method as defined in claim 1, wherein the photographic images have been tagged with metadata that can be used to determine the geo-location of each pixel in the photographic images.

9. A method as defined in claim 1, wherein at least one of the interest points include at least one corner of an intersection in at least one of the photographic images.

10. A method as defined in claim 1, wherein the clustering is not limited to pair-wise comparisons.

11. A method as defined in claim 1, wherein the clustering groups together all interest points that are located within a predetermined distance.

12. A method as defined in claim 1, wherein the selecting includes ranking the groups.

13. A method as defined in claim 1, wherein the interest point descriptor is generated automatically by the computer based on feature extraction.

14. A method as defined in claim 1, wherein the geo-location is generated without using a reference image.