IP Library › Granted Patent US 8,160,364
Granted Patent B2
US 8,160,364 · App. 11/707,800 · Granted Apr 17, 2012

System and method for image registration based on variable region of interest

Assignee: Raytheon Company
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 8,160,364
App. No.
11/707,800
Granted
Apr 17, 2012
Kind
B2
Abstract

An image registration system for aligning first and second images. The novel system includes a first system for extracting a region of interest (ROI) from each image and a second system for coarsely aligning the regions of interest. The first system determines the size and location of the ROI based on the number of features contained within the region. The size of the ROI is enlarged until a number of features contained in the ROI is larger than a predetermined lower bound or until the size is greater than a predetermined upper bound. The second system computes a cross-correlation on the regions of interest using a plurality of transforms to find a coarse alignment transform having a highest correlation. The image registration system may also include a third system for performing sub-pixel alignment on the regions of interest.

Claims (38)

1. A system for aligning a first image and a second image comprising:

First means for extracting a region of interest sized based on edge features from each of the first image and the second image; and

second means for coarsely aligning said regions of interest using a cross correlation on said region of interest with a plurality of image shifts to find a coarse image shift having a highest correlation; and

Fine alignment means for interpolating said regions of interest and performing a cross-correlation on the interpolated regions of interest using a plurality of image shifts to find a sub-pixel image shift having a highest correlation.

2. The invention of claim 1 wherein said first means includes means for detecting the edge features in one of said images.

3. The invention of claim 2 wherein said first means further includes means for determining a size for said region of interest.

4. The invention of claim 3 wherein said size is dependent on a number of the edge features contained in said region of interest.

5. The invention of claim 4 wherein said size is enlarged until a number of the edge features contained in said region of interest is larger than a predetermined lower bound or until said size is greater than a predetermined upper bound.

6. The invention of claim 3 wherein said first means further includes third means for determining a location for said region of interest.

7. The invention of claim 6 wherein said location is fixed.

8. The invention of claim 7 wherein said location is at a center of said image.

9. The invention of claim 6 wherein said location is dependent on a number of the edge features contained in said region of interest.

10. The invention of claim 9 wherein said third means includes means for counting a number of the edge features in each row and each column of said image.

11. The invention of claim 10 wherein said third means further includes means for finding a first subset of rows having a large number of the edge features.

12. The invention of claim 11 wherein said third means further includes means for finding a second subset of columns having a large number of the edge features.

13. The invention of claim 12 wherein said location is at an intersection of said first and second subsets.

14. The invention of claim 6 wherein said first means further includes means for applying said size and said location to said first and second images to extract said regions of interest.

15. The invention of claim 1 wherein said second means includes means for applying a plurality of transforms to one or both of said regions of interest.

16. The invention of claim 15 wherein said second means further includes means for computing a correlation metric between said regions of interest for each transform.

17. The invention of claim 16 wherein said second means further includes means for selecting a first transform corresponding with a highest correlation for coarse alignment.

18. The invention of claim 17 wherein said second means further includes means for applying said first transform to one or both of said regions of interest to coarsely align said regions of interest.

19. The invention of claim 18 wherein said fine alignment means includes fourth means for determining a sub-pixel alignment from said regions of interest.

20. The invention of claim 19 wherein said fourth means includes means for interpolating each of said regions of interest by an integer N.

21. The invention of claim 20 wherein said fourth means further includes means for applying a plurality of transforms to one or both of said interpolated regions of interest.

22. The invention of claim 21 wherein said fourth means further includes means for computing a correlation metric between said interpolated regions of interest for each transform.

23. The invention of claim 22 wherein said fourth means further includes means for selecting a second transform corresponding with a highest correlation for sub-pixel alignment.

24. The invention of claim 23 wherein said fourth means further includes means for applying said first and second transforms to one or both of said images to align said images.

25. The invention of claim 21 wherein said transforms include translations of a plurality of shifts.

26. An image registration system comprising:

a region of interest extraction system for receiving first and second input images and extracting a region of interest from each image;

a coarse alignment system for performing a cross-correlation on said regions of interest using a plurality of image shifts to find a coarse image shift having a highest correlation; and

a fine alignment system for interpolating said regions of interest and performing a cross-correlation on the interpolated regions of interest using a plurality of image shifts to find a sub-pixel image shift having a highest correlation.

27. The invention of claim 26 wherein said region of interest extraction system includes a system for detecting features in one of said images.

28. The invention of claim 27 wherein said region of interest extraction system further includes a system for determining a size of said region of interest based on a number of features contained within said region of interest in one of said images.

29. The invention of claim 28 wherein said size is enlarged until a number of features contained in said region of interest is larger than a predetermined lower bound or until said size is greater than a predetermined upper bound.

30. The invention of claim 27 wherein said region of interest extraction system further includes a system for determining a location of said region of interest based on a number of features contained within said region of interest in one of said images.

31. The invention of claim 30 wherein said system for determining said location is adapted to count a number of features in each row and each column of said image, find a first subset of rows having a large number of features and a second subset of columns having a large number of features, and choose said location at an intersection of said first and second subsets.

32. The invention of claim 26 wherein a location of said region of interest is at a center of said image.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 16, 2007
From: LAM, ERIC P.; LEDDY, CHRISTOPHER A.; NASH, STEPHEN R.; PARKS, HARRISON A.
To: RAYTHEON COMPANY
Reel/Frame 019003/0558 →
Continuity (1)
Related Publication 20080199078A1 · Aug 21, 2008