Simulation, Signal Extraction, and Augmented Visualization for 3D BOSS Data: APL-UW Buried Object Scanning Sonar (BOSS) Data Processing Framework Overview

Abstract

The goal of this code is to:-Parse binary data files captured by a buried object scanning sonar into a simple Matlab structure containing raw sonar data, pulse and system information.-Perform basic operations like pulse compression and real-aperture beamforming to improve the SNR and detect region's of interest in the data-Geo-locate detected regions in latitude, so that different looks of the some target captured by different scans can be compared-Perform phase-preserving SAS processing on regions of interest so that acoustic color spectra can be observed, allowing feature extraction for machine learning and target classification.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 10, 2020
Accession Number
AD1169175

Entities

People

  • Timothy Marston

Organizations

  • University of Washington

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Buried Objects
  • Classification
  • Compression
  • Computational Processes
  • Computing-Related Activities
  • Dimensionality Reduction
  • Extraction
  • Feature Extraction
  • Image Processing
  • Latitude
  • Machine Learning
  • Pulse Compression
  • Scanning
  • Scanning Sonar
  • Sonar
  • Target Classification

Readers

  • Acoustical Oceanography.
  • Image Processing and Computer Vision.

Technology Areas

  • AI & ML