THIS GRANT IS A CONTINUATION OF N00014-13-1-0403 Inversion, uncertainties, and multiple scattering in synthetic aperture
Abstract
Summary The objective of the proposed work is to rethink the imaging methods of synthetic aperture radar (SAR) and synthetic aperture sonar (SAS) and explore the potential gains from a more comprehensive inversion approach. This e ort is inspired by recent progress in other disciplines, such as seismic imaging where large-scale optimization is becoming commonplace, modern signal/data processing and the quest for the implications of sparsity, applied probability and how it raises new questions in matrix computation, and the geometry of high-frequency wave propagation for structured computation. Anticipated outcomes. The project should lead to a better understanding of the feasi- bility and potential of deploying optimization tools on large-scale radar and sonar imaging problems. Success would mean obtaining more faithful images with the same data, or being able to invert complex datasets that no other method could previously handle. A compre- hensive cross-modality large-scale inversion software environment will be developed with the goal of inviting discussion and collaboration on problems of interest to all parties. Impact on DoN Capabilities. First, SAR is an important tool for non-cooperative target detection and identi cation. Creating higher-quality SAR images would be a direct way to improve interpretation and minimize unintended casualties. Second, better SAR and SAS capabilities would have implications for the detection of landmines and underwater mines. Third, the availability of adequate 3D inversion software could help create volumetric maps of urban scenes and complex (e.g. coastal) landscapes from SAR data, and better bathymetry elevation maps from SAS data. Fourth, the same synthetic-aperture inversion problem arises in tomographic ultrasound | an appealing speculative modality for nondestructive testing of cracks and defects in stressed materials. 1
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jun 03, 2016
- Source ID
- N000141612122
Entities
People
- Laurent Demanet
Organizations
- Massachusetts Institute of Technology
- Office of Naval Research
- United States Navy