Automatic Change Detection and Classification (ACDC) System
Abstract
The Naval Research Laboratory is developing an Automatic Change Detection and Classification (ACDC) system for the Naval Oceanographic Office to use with sidescan sonar imagery (SSI). The ACDC system will automatically detect seafloor features in SSI; classify, catalog, and cluster the features; and compare them with features previously detected (and stored in historical databases), to determine whether each newly detected feature has simply moved, relative to a previously plotted position, or is actually new (change detection). This research project is on going, with an estimated completion date of 2006. The first two components of ACDC - computer aided detection and computer aided classification - have been completed and fully tested. This year, dynamic database components are being implemented and further work is being performed on two wavelet networks for feature and area matching (i.e., change detection).
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 2004
- Accession Number
- ADA495163
Entities
People
- Geary Layne
- Juliette Ioup
- Marlin Gendron
- Maura Lohrenz
Organizations
- United States Naval Research Laboratory