Full motion video (FMV) is increasingly being used for maritime surveillance. The immense amount of FMV data produced, lack of automation, and dependency on human analysis make processing and extracting intelligence from FMV challenging. Maritime characteristics such as sun glint, white caps, ship wakes, and ship size further complicate object detection and tracking in FMV. There is a need for an automated process that can quickly and efficiently analyze maritime FMV and extract meaningful data to assist analysts.
RAPIER® FMV is a maritime target detection, tracking, and identification solution that quickly and automatically analyzes video, alerts analysts of important targets, and outputs target information. At the core of RAPIER® FMV is an object detection system that uses basic signal processing and machine learning approaches. The system maintains a high level of performance without making prior assumptions about foreground-background characteristics. It is robust to variations in illumination, ever-changing background characteristics, and video quality; as well as abrupt changes in the perspective, size, appearance, and number of targets. RAPIER® FMV also uses a unique horizon detection method that aids in improving processing times. Applications for this technology may include search and rescue, vessel tracking, counter-piracy, and harbor or port security.
- Processes 1080p video in real-time, raw video streaming from unmanned aerial vehicles (UAVs) in real-time, and 640x480 video up to 2x faster
- Supports real-time video feeds from manned and unmanned aircraft, satellite, ship-mounted, and shore-based sensors; supports recorded data
- Acquires and tracks multiple maritime targets simultaneously and displays detected and recognized targets
- Provides simple and easy to read output formats for end users: HTML tip sheet, KML (Google Earth™), shape files, XML, OTH-G, RIT, etc.
- Annotates video streams with metadata including sensor positioning, coordinates, time, and date
- US patent 8,958,602 available for license
- Potential for collaboration with Navy researchers