Kitware to enhance AFRL analysts' FMV exploitation capabilities
The US Air Force Research Laboratory (AFRL) has awarded a broad agency announcement (BAA) contract to Kitware to enhance the air force analysts' full motion video (FMV) exploitation performance.
Under the contract, the company will integrate its video and image retrieval and analysis tool (VIRAT) with the AFRL's Planning and Direction, Collection, Processing and Exploitation, Analysis and Production, and Dissemination Experimentation environment (PCPAD-X) system.
In addition to improving the analyst's productivity, the integration is expected to offer an improved user interface and more standardised communication and recording mechanisms, and an advanced object tracking system for automatic detection of unusual events in the data using normalcy models.
Kitware Computer Vision assistant director and the project principal investigator Dr Amitha Perera said: ''We are confident that our system will transform analyst workflows and provide more accurate and efficient extraction of information from FMV feeds.''
Currently, the FMV exploitation is deemed a very labour-intensive process due to the involvement of teams of people for monitoring a single video feed to extract the value for the primary mission.
The process also has its own drawbacks, as different missions, including video feed monitoring and call-out events monitoring are performed by separate analysts, whichraises concern that analysts focused on finding mission-specific elements of interest might overlook other events and objects of interest to other parties.
Additionally, communication of events between shifts of analysts on a particular mission, or to outside entities, is also likely to result in receipt of non-standard, fragmented descriptions of events and other salient information.
Developed primarily under the Defense Advanced Research Projects Agency (DARPA) funding, VIRAT can automatically process FMV in real-time to extract and record elements of interest, such as moving targets and events, or objects that match a specific appearance model.
Image: Kitware's VIRAT system will enhance AFRL analysts' full-motion video exploitation capabilities. Photo: courtesy of Tech. Sgt. James Harper/Released.