Automated motion image-based tracking is an increasingly important tool in Intelligence, Surveillance, and Reconnaissance (ISR). Unfortunately, current tracking technology is not up to the performance levels needed to deliver key subtasks in this arena. We postulate that the under-performance of automated trackers derives from the under-exploitation of the rich sets of features related the identification of items being tracked. To address this we propose a formulation of features that supports easy exchange and integration of new features. We believe this approach will provide the foundations for a far wider and more effective exploration of potential features related to tracking, and as a result, significantly better and more sustainable growth in tracker performance. OBJECTIVENGA's Motion Imagery Standards Board (MISB) is engaged in an initiative to establish a standards-based approach to automated tracking. Our intent is to support the R&D community in its endeavor to build better tracking-frommotion-imagery solutions. Our approach is to establish a plug-and-play tracking capability through standards. The objective of this paper is to introduce that plug-and-play approach to a deeper level within tracking -to the plugand-play of the computed features that drive feature-based automated tracking. BACKGROUNDTracking has a long pedigree in Radar systems. With the rapid proliferation of motion imagery sensors such as full motion and wide area surveillance systems, automated trackers are becoming important in passive EO and IR as well. These must complement human operators and achieve something resembling human performance to be fully useful, and feature-based tracking is the best bet to achieve this level of capability. This section provides background on these developments. Radar-Based Tracking and "Tracklets"Automated approaches to tracking have a long tradition in radar-based technologies (see, e.g., [1]). With the proliferation of video surveillance systems and the appearance of wide-area motion imagery for persistent surveillance, it is apparent that automated tracking from passive EO/IR sources will become increasingly important. The computer vision community has been responding with implementations of trackers that generalize the tracking approaches of the radar community. Radar-based tracking has relied on kinetics-based models of probable motion. These are often called "tracklets" in part because they are often just fragments of what the human operator knows to be longer tracks. The radar community has devoted little attention to exploiting descriptive features of the moving objects.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.