19 May 2006 Smart unattended sensor networks with scene understanding capabilities
Author Affiliations +
Abstract
Unattended sensor systems are new technologies that are supposed to provide enhanced situation awareness to military and law enforcement agencies. A network of such sensors cannot be very effective in field conditions only if it can transmit visual information to human operators or alert them on motion. In the real field conditions, events may happen in many nodes of a network simultaneously. But the real number of control personnel is always limited, and attention of human operators can be simply attracted to particular network nodes, while more dangerous threat may be unnoticed at the same time in the other nodes. Sensor networks would be more effective if equipped with a system that is similar to human vision in its abilities to understand visual information. Human vision uses for that a rough but wide peripheral system that tracks motions and regions of interests, narrow but precise foveal vision that analyzes and recognizes objects in the center of selected region of interest, and visual intelligence that provides scene and object contexts and resolves ambiguity and uncertainty in the visual information. Biologically-inspired Network-Symbolic models convert image information into an 'understandable' Network-Symbolic format, which is similar to relational knowledge models. The equivalent of interaction between peripheral and foveal systems in the network-symbolic system is achieved via interaction between Visual and Object Buffers and the top-level knowledge system.
© (2006) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Gary Kuvich, Gary Kuvich, "Smart unattended sensor networks with scene understanding capabilities", Proc. SPIE 6231, Unattended Ground, Sea, and Air Sensor Technologies and Applications VIII, 62311D (19 May 2006); doi: 10.1117/12.662240; https://doi.org/10.1117/12.662240
PROCEEDINGS
12 PAGES


SHARE
Back to Top