GENERATION OF AERIAL IMAGES

• US 20110090337A1
• Filed: 01/30/2009
• Published: 04/21/2011
• Est. Priority Date: 02/01/2008
• Status: Active Grant

• Current Assignee: IMINT Image Intelligence AB
• Original Assignee: IMINT Image Intelligence AB
• Inventors: Klomp, Harald, Sandstrom, Jakob
• Granted Patent: US 8,723,953 B2
• Time in Patent Office: Days
• US Class Current: 348/144
• CPC Class Codes: G01C 11/025 by scanning the object
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Abstract

The method according to the invention gene rates an aerial image mosaic viewing a larger area than a single image from a camera can provide using a combination of computer vision and photogrammetry. The aerial image mosaic is based on a set of images acquired from a camera. Selective matching and cross matching of consecutive and non-consecutive images, respectively, are performed and three dimensional motion and structure parameters are calculated and implemented on the model to check if the model is stable. Thereafter the parameters are globally optimised and based on these optimised parameters the serial image mosaic is generated. The set of images may be limited by removing old image data as new images are acquired. The method makes it is possible to establish images in near real time using a system of low complexity and small size, and using only image information.

Claims

1-15. -15. (canceled)

16. A method of generation of an aerial image mosaic viewing at least a portion of a scene, wherein the method comprises the steps of:

acquiring a set of images from the scene using a camera;

selective matching of consecutive image pairs from the set of images using tie points;

cross matching of overlapping non-consecutive images pairs from the set of images using feature point descriptors;

selecting a motion and structure model using a priori assumption on camera motion and scene properties;

calculating three dimensional motion and structure parameters between image pairs;

implementing the three dimensional motion and structure parameters on the motion and structure model;

globally optimizing the three dimensional motion and structure parameters to make the three dimensional motion and structure parameters converge to one consistent solution for the set of images; and

generating the aerial image mosaic.

17. The method according to claim 16, wherein the step of selective matching further comprises the steps of:

selecting tie points in each image;

finding sub pixel coordinates for each tie point position; and

tracking corresponding tie points in each pair of consecutive images using a feature tracking algorithm.

18. The method according to claim 16, wherein the step of calculating the three dimensional motion and structure parameters comprises calculating first three dimensional motion and structure parameters between consecutive images using the tie points.

19. The method according to claim 16, wherein the step of cross matching comprises the step of assigning feature point descriptors.

20. The method according to claim 19, wherein the step of calculating the three dimensional motion and structure parameters comprises calculating second three dimensional motion and structure parameters between non-consecutive images using the feature point descriptors.

21. The method according to claim 20, further comprising the step of removing outliers preceding the step of calculating three dimensional motion and structure parameters.

22. The method according to claim 16, wherein the step of globally optimizing comprises bundle block adjustment.

23. The method according to claim 16, further comprising changing the motion and structure model.

24. The method according to claim 16, further comprising the step of calibrating the camera by:

acquiring images of a checkerboard;

finding corners of the checkerboard;

estimating intrinsic and extrinsic parameters;

estimating coefficients of distortion; and

refining the intrinsic and extrinsic parameters.

25. The method according to claim 16, further comprising the step of providing camera position and/or orientation.

26. The method according to claim 17, wherein the step of selecting tie points is performed using a texturedness criteria.

27. The method according to claim 17, wherein the step of tracking corresponding tie points is performed using image pyramids.

28. The method according to claim 16, wherein the step of calculating the three dimensional motion and structure parameters comprises Random Sample Consensus (RANSAC).

29. The method according to claim 21, wherein the step of removing outliers comprises the step of computing the homography of the tie points.

30. The method according to claim 16, further comprising the step of limiting the number of images by removing old image data.