Publications

Results 26–45 of 45
Skip to search filters

Polarimetric synthetic-aperture-radar change-type classification with a hyperparameter-free open-set classifier

IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops

Koch, Mark W.; West, Roger D.; Riley, Robert; Quach, Tu-Thach Q.

Synthetic aperture radar (SAR) is a remote sensing technology that can truly operate 24/7. It's an all-weather system that can operate at any time except in the most extreme conditions. Coherent change detection (CCD) in SAR can identify minute changes such as vehicle tracks that occur between images taken at different times. From polarimetric SAR capabilities, researchers have developed decompositions that allow one to automatically classify the scattering type in a single polarimetric SAR (PolSAR) image set. We extend that work to CCD in PolSAR images to identify the type change. Such as change caused by no return regions, trees, or ground. This work could then be used as a preprocessor for algorithms to automatically detect tracks.

More Details

Detecing Surface Change Created by an Underground Chemical Explosion Using Fully Polarimetric VideoSAR

Yocky, David A.; West, Roger D.; Riley, Robert; Calloway, Terry C.; Wahl, Daniel E.; Garley, Leroy G.; Bolin, Samuel A.

Sandia National Laboratories (SNL) flew its Facility for Advanced RF and Algorithm Development (FARAD) X-Band (9.6 GHz center frequency), fully-polarimetric synthetic aperture radar (PolSAR) in VideoSAR-mode to collect complex-valued SAR imagery before, during, and after the fifth and sixth Source Physics Experiment's (SPE-5 and SPE-6) underground explosion. The results from the fifth Source Physics Experiment (SPE-5) used single-polarimetric VideoSAR data while SPE-6 used single and fully-polarimetric VideoSAR data. We show that SAR can provide surface change products indicative of disturbances caused by the underground chemical explosions. These are surface coherence measures, Po1SAR change signatures, and differential interferometric SAR (InSAR) height change.

More Details

A statistical approach to combining multisource information in one-class classifiers

Statistical Analysis and Data Mining

Simonson, Katherine M.; West, Roger D.; Hansen, Ross L.; LaBruyere, Thomas E.; Van Benthem, Mark V.

A new method is introduced for combining information from multiple sources to support one-class classification. The contributing sources may represent measurements taken by different sensors of the same physical entity, repeated measurements by a single sensor, or numerous features computed from a single measured image or signal. The approach utilizes the theory of statistical hypothesis testing, and applies Fisher's technique for combining p-values, modified to handle nonindependent sources. Classifier outputs take the form of fused p-values, which may be used to gauge the consistency of unknown entities with one or more class hypotheses. The approach enables rigorous assessment of classification uncertainties, and allows for traceability of classifier decisions back to the constituent sources, both of which are important for high-consequence decision support. Application of the technique is illustrated in two challenge problems, one for skin segmentation and the other for terrain labeling. The method is seen to be particularly effective for relatively small training samples.

More Details

Creating Fantastic PI Workshops

Perkins, David N.; Biedermann, Laura B.; Clark, Blythe C.; Thayer, Rachel C.; Dagel, Amber L.; Gupta, Vipin P.; Hibbs, Michael R.; West, Roger D.

The goal of this SAND report is to provide guidance for other groups hosting workshops and peerto-peer learning events at Sandia. Thus this SAND report provides detail about our team structure, how we brainstormed workshop topics and developed the workshop structure. A Workshop “Nuts and Bolts” section provides our timeline and check-list for workshop activities. The survey section provides examples of the questions we asked and how we adapted the workshop in response to the feedback.

More Details

Semi-supervised classification of terrain features in polarimetric SAR images using H/A/α and the general four-component scattering power decompositions

Conference Record - Asilomar Conference on Signals, Systems and Computers

Dauphin, Stephen M.; West, Roger D.; Riley, Robert; Simonson, Katherine M.

In an effort to enhance image classification of terrain features in fully polarimetric SAR images, this paper explores the utility of combining the results of two state-of-the-art decompositions along with a semi-supervised classification algorithm to classify each pixel in an image. Each pixel is labeled either with a pre-determined classification label, or labeled as unknown.

More Details

Superpixel segmentation using multiple SAR image products

Proceedings of SPIE - The International Society for Optical Engineering

Moya, Mary M.; Koch, Mark W.; Perkins, David N.; West, Roger D.

Sandia National Laboratories produces copious amounts of high-resolution, single-polarization Synthetic Aperture Radar (SAR) imagery, much more than available researchers and analysts can examine. Automating the recognition of terrains and structures in SAR imagery is highly desired. The optical image processing community has shown that superpixel segmentation (SPS) algorithms divide an image into small compact regions of similar intensity. Applying these SPS algorithms to optical images can reduce image complexity, enhance statistical characterization and improve segmentation and categorization of scene objects. SPS algorithms typically require high SNR (signal-to-noise-ratio) images to define segment boundaries accurately. Unfortunately, SAR imagery contains speckle, a product of coherent image formation, which complicates the extraction of superpixel segments and could preclude their use. Some researchers have developed modified SPS algorithms that discount speckle for application to SAR imagery. We apply two widely-used SPS algorithms to speckle-reduced SAR image products, both single SAR products and combinations of multiple SAR products, which include both single polarization and multi-polarization SAR images. To evaluate the quality of resulting superpixels, we compute research-standard segmentation quality measures on the match between superpixels and hand-labeled ground-truth, as well as statistical characterization of the radar-cross-section within each superpixel. Results of this quality analysis determine the best input/algorithm/parameter set for SAR imagery. Simple Linear Iterative Clustering provides faster computation time, superpixels that conform to scene-relevant structures, direct control of average superpixel size and more uniform superpixel sizes for improved statistical estimation which will facilitate subsequent terrain/structure categorization and segmentation into scene-relevant regions. © 2014 SPIE.

More Details
Results 26–45 of 45
Results 26–45 of 45