Automic Change Detection and Classification (ACDC) System
Abstract
The authors are developing an Automated Change Detection and Classification (ACDC) system for the Mine Warfare (MIW) group at the Naval Oceanographic Office (NAVOCEANO). ACDC detects features in sidescan imagery, classifies the features (e.g., as minelike or not), and searches through historical databases of previously detected features to perform change detection (i.e., to determine whether the feature is new or pre-existing, relative to earlier surveys). The completed ACDC system will assist mine countermeasures (MCM) warfighters to more quickly and reliably identify minelike contacts in sidescan, thereby reducing warfighter fatigue, reducing risk to the warfighter, and shortening timelines for completion of MCM objectives. ACDC consists of five major components: (1) clutter detection, (2) feature completion/ classification, (3) geospatial searching, (4) clustering, and (5) scene matching. Depending on the quality of sidescan imagery to be analyzed, each of these steps can be performed autonomously (i.e., with no human intervention) or as a computer-assisted function (in which ACDC suggests statistically likely outcomes to the operator, who makes a final determination).
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2005
- Accession Number
- ADA494240
Entities
People
- G. J. Layne
- M. C. Lohrenz
- M. L. Gendron
Organizations
- United States Naval Research Laboratory