Dr. Derek J. Walvoord Profile

Dr. Derek J. Walvoord

AI Architect at L3Harris Technologies Inc

SPIE Involvement:

Author

Publications (8)

Proceedings Article | 7 May 2019 Presentation + Paper

Deep learning architecture advancements for accurate and robust image registration

Derek Walvoord, Doug Couwenhoven

Proceedings Volume 11018, 1101812 (2019) https://doi.org/10.1117/12.2522044

KEYWORDS: Image registration, Network architectures, Convolutional neural networks, Neural networks, Image processing

Read Abstract +

Registration of image collections and video sequences is a critical component in algorithms designed to extract actionable intelligence from remotely sensed data. While methodologies for registration continue to evolve, the accuracy of alignment remains dependent on how well the approach tolerates changes in capture geometry, sensor characteristics, and scene content. Differences in imaging modality and field-of-view present additional challenges. Registration techniques have progressed from simple, global correlation-based algorithms, to higher-order model fitting using salient image features, to two-stage approaches leveraging high-fidelity sensor geometry, to new methods that exploit high-performance computing and convolutional neural networks (ConvNets). The latter offers important advantages by removing model assumptions and learning feature extraction directly through the minimization of a registration cost function. Deep learning approaches to image registration are still relatively unexplored for overhead imaging, and their ability to accommodate a large problem domain offers potential for several new developments. This work presents a new network architecture that improves accuracy and generalization capabilities over our modality-agnostic deep learning approach to registration that recently advanced the state of the art. A thoroughly tested ConvNet pyramid remains the core of our network approach, and has been optimized for registration and generalized to begin addressing derivative applications such as mosaic generation. Further modifications, such as objective function masking and reduced interpolation, have also been implemented to improve the overall registration process. As before, the trained network ingests image frames, applies a vector field, and returns a version of the input image that has been warped to the reference. Qualitative and quantitative performance of the new architecture is evaluated using several overhead still and full-motion video (FMV) data sets.

Proceedings Article | 27 April 2018 Presentation + Paper

Stabilization and registration of full-motion video data using deep convolutional neural networks

Derek Walvoord, Doug Couwenhoven, Michael Bayer

Proceedings Volume 10646, 1064612 (2018) https://doi.org/10.1117/12.2305072

KEYWORDS: Image registration, Video, Convolutional neural networks, Sensors, Network architectures, Infrared imaging, Infrared sensors

Read Abstract +

SPIE Journal Paper | 27 November 2014

Assessing geoaccuracy of structure from motion point clouds from long-range image collections

David Nilosek, Derek Walvoord, Carl Salvaggio

OE, Vol. 53, Issue 11, 113112, (November 2014) https://doi.org/10.1117/12.10.1117/1.OE.53.11.113112

KEYWORDS: Cameras, Atomic force microscopy, Clouds, Error analysis, Image processing, Sensors, Global Positioning System, 3D modeling, 3D image processing, Imaging systems

Read Abstract +

Proceedings Article | 23 May 2013 Paper

Geoaccurate three-dimensional reconstruction via image-based geometry

Derek Walvoord, Adam Rossi, Bradley Paul, Bernie Brower, Matthew Pellechia

Proceedings Volume 8747, 874706 (2013) https://doi.org/10.1117/12.2015974

KEYWORDS: Clouds, Atomic force microscopy, Sensors, Cameras, Image processing, Matrices, 3D image reconstruction, Imaging systems, Data modeling, Error analysis

Read Abstract +

Proceedings Article | 18 March 2008 Paper

Multispectral processing of combined visible and x-ray fluorescence imagery in the Archimedes palimpsest

Derek Walvoord, Allison Bright, Roger Easton

Proceedings Volume 6810, 681005 (2008) https://doi.org/10.1117/12.766665

KEYWORDS: Image processing, Image registration, Principal component analysis, X-ray fluorescence spectroscopy, Ultraviolet radiation, Light emitting diodes, X-rays, Multispectral imaging, Sensors, Visible radiation

Read Abstract +

Proceedings Article | 29 January 2007 Paper

Adding contextual information to improve character recognition on the Archimedes Palimpsest

Derek Walvoord, Roger Easton, Roxanne Canosa

Proceedings Volume 6500, 650008 (2007) https://doi.org/10.1117/12.704471

KEYWORDS: Image filtering, Optical character recognition, Pattern recognition, Photography, Nonlinear filtering, Image classification, Associative arrays, Data modeling, Mathematics, Electronic filtering

Read Abstract +

Proceedings Article | 17 January 2005 Paper

Enhancement and character recognition of the erased colophon of a 15th-century Hebrew prayer book

Derek Walvoord, Roger Easton, Keith Knox, Matthew Heimbueger

Proceedings Volume 5676, (2005) https://doi.org/10.1117/12.587848

KEYWORDS: Image filtering, Ultraviolet radiation, Nonlinear filtering, Optical character recognition, Visible radiation, Composites, Image processing, Feature extraction, Phase only filters, Spatial filters

Read Abstract +

Proceedings Article | 28 May 2004 Paper

Variations of matched filtering for reduced noise amplification

Derek Walvoord, Roger Easton

Proceedings Volume 5298, (2004) https://doi.org/10.1117/12.525833

KEYWORDS: Image filtering, Electronic filtering, Spatial filters, Fourier transforms, Optical filters, Algorithm development, Signal detection, Detection and tracking algorithms, Nonlinear filtering, Feature extraction

Read Abstract +

Showing 5 of 8 publications

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years