Aerial image registration incorporating GPS/IMU data

Keith A. Redmill; John I. Martin; Umit Ozguner

doi:10.1117/12.820201

29 April 2009 Aerial image registration incorporating GPS/IMU data

Keith A. Redmill, John I. Martin, Umit Ozguner

Proceedings Volume 7347, Evolutionary and Bio-Inspired Computation: Theory and Applications III; 73470H (2009) https://doi.org/10.1117/12.820201
Event: SPIE Defense, Security, and Sensing, 2009, Orlando, Florida, United States

Abstract

We describe a methodology for multiframe image registration of airborne high resolution, multi-camera imagery. In the absence of predetermined camera and lens models, parameters are optimally determined from imagery and known ground reference locations. GPS and IMU data collected from the sensor platform and the identified camera model parameters are used to perform an initial orthorectification and georeferencing of each image. Multiple KLT, Sift, or featureless point-match correspondences are identified and validated using RANSAC techniques. Affine transform hypothesis are then generated, inconsistent hypothesis are removed using a RANSAC approach, and a final optimal transform is generated as the least squares optimal fit of the remaining correspondences. To eliminate long-term drift, key frames are selected and cross-registered. Performance improvements can also be demonstrated using a mask to eliminate correspondences not on the ground plane. This approach is illustrated using the 2007 AFRL Columbus Large Image Format dataset.

Citation Download Citation

Keith A. Redmill, John I. Martin, and Umit Ozguner "Aerial image registration incorporating GPS/IMU data", Proc. SPIE 7347, Evolutionary and Bio-Inspired Computation: Theory and Applications III, 73470H (29 April 2009); https://doi.org/10.1117/12.820201

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available