IP Library Granted Patent US 10,460,458
Granted Patent B1
US 10,460,458 · App. 15/704,300 · Granted Oct 29, 2019

Method for registration of partially-overlapped aerial imagery using a reduced search space methodology with hybrid similarity measures

Inventors: Mohammad I. Vakil (Beavercreek, OH); John Malas (Kettering, OH)
Assignee: United States of America as represented by the Secretary of the Air Force
G06T7/337G06T3/0068G06T5/002G06T7/0002
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 10,460,458
App. No.
15/704,300
Granted
Oct 29, 2019
Kind
B1
Abstract

A method for registration of partially-overlapped images, comprises (a) performing noise reduction and feature extraction in a reference image and an unregistered image; (b) determining a template size using a phase transition methodology for a sufficiently-sampled finite data set; (c) identifying a template region in the reference image; (d) performing a wide angle estimation of the reference image and the unregistered image; (e) performing orientation and translation of the reference image and the unregistered image; (f) performing a search space reduction of the reference image and the unregistered image; (g) performing a coarse angle estimation of the reference image and the unregistered image; (h) performing orientation of the reference image and the unregistered image of the coarse angle estimation; and (i) performing a fine angle estimation and registration of the reference image and the unregistered image.

Claims (22)

1. A method for registration of partially-overlapped images, comprising:

performing noise reduction and feature extraction in a reference image and an unregistered image;

determining a template size using a phase transition methodology for a sufficiently-sampled finite data set, wherein the phase transition methodology comprises calculating the maximum possible entropic states and actual entropic states as number of the pixels;

identifying a template region in the reference image;

performing a wide angle estimation of the reference image and the unregistered image;

performing orientation and translation of the reference image and the unregistered image;

applying margin, imtrim, and overlap to reduce a search space of the reference image and the unregistered image;

performing a coarse angle estimation of the reference image and the unregistered image;

performing orientation of the reference image and the unregistered image of the coarse angle estimation; and

performing a fine angle estimation and registration of the reference image and the unregistered image.

2. The method for registration of partially-overlapped images of claim 1 , wherein the step of determining a template size using a phase approach for a sufficiently-sampled data set further comprises:

calculating the maximum possible entropic states and actual entropic states as a number of intensities present in the pixels of a given reference and unregistered image.

3. The method for registration of partially-overlapped images of claim 1 , wherein the step of determining a template size using a phase approach for a sufficiently-sampled data set further comprises:

determining template size with an m×m template, where m is a power of 2; wherein the total number of pixels M=m×m is equal to or greater than the number of pixels identified by the phase transition methodology.

4. The method for registration of partially-overlapped images of claim 1 , wherein the step of performing a coarse angle estimation of the reference image and the unregistered image further comprises:

performing sequential normalized cross-correlation in the reduced search space.

5. The method for registration of partially-overlapped images of claim 4 , wherein the step of performing a coarse angle estimation of the reference image and the unregistered image further comprises:

performing a mutual information-based variation of information.

6. The method for registration of partially-overlapped images of claim 1 , wherein the step of performing a fine angle estimation of the reference image and the unregistered image further comprises:

performing sequential normalized cross-correlation in the reduced search space.

7. The method for registration of partially-overlapped images of claim 4 , wherein the step of performing a fine angle estimation of the reference image and the unregistered image further comprises:

performing a mutual information-based variation of information.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Sep 14, 2017
From: MALAS, JOHN; VAKIL, MOHAMMAD I
To: GOVERNMENT OF THE UNITED STATES AS REPRESENTED BY THE SECRETARY OF THE AIR FORCE
Reel/Frame 043592/0462 →