Scenario-based query processing for video-surveillance archives

Abstract

Scenario-based querying and retrieval Visual query specification Event-based querying After-the-fact analysis a b s t r a c t Automated video surveillance has emerged as a trendy application domain in recent years, and accessing the semantic content of surveillance video has become a challenging research area. The results of a considerable amount of research dealing with automated access to video surveillance have appeared in the literature; however, significant semantic gaps in event models and content-based access to surveillance video remain. In this paper, we propose a scenario-based query-processing system for video surveillance archives. In our system, a scenario is specified as a sequence of event predicates that can be enriched with object-based low-level features and directional predicates. We introduce an inverted tracking scheme, which effectively tracks the moving objects and enables view-based addressing of the scene. Our query-processing system also supports inverse querying and view-based querying, for after-the-fact activity analysis. We propose a specific surveillance query language to express the supported query types in a scenario-based manner. We also present a visual queryspecification interface devised to facilitate the query-specification process. We have conducted performance experiments to show that our query-processing technique has a high expressive power and satisfactory retrieval accuracy in video surveillance.

A multi-agent architecture for supporting distributed normality-based intelligent surveillance

“…In this field, the use of intelligent techniques may contribute to solve this kind of problems and to reduce the cost associated to the security staff by semi-automating the tasks that are currently performed by them (Collins et al, 2000). Nowadays, there are non-commercial artificial systems able to infer anomalous behavior (Haritaoglu et al, 2000), to identify suspicious or lost objects (Stringa and Regazzoni, 2000), to analyze object movements (Hu et al, 2004), to recognize the gait of people (Tao et al, 2007;Xu et al, 2006), to detect crowds , or to retrieve relevant information (Saykol et al, 2010) among other abilities (see Valera and Velastin, 2005 for a detailed revision).…”

Hierarchical spatiotemporal modeling for dynamic video trajectory analysis

“…Instead of using discrete value for the weight parameter, 43,44,72,73 we control the sensitivity of interaction, though based on W = [ω i, j ], called the m × m interaction matrix, which has unity and off-diagonal at (i, j) is−ω i, j . High sensitivity of interaction between nodes i and j is achieved by setting the weight parameter ω i, j as high, and it accounts for the smoothness between neighboring nodes based on their direction similarity.…”

“…In Ref. 44, the key frames are detected based on a grid-based motion representation of the moving regions, used to detect unusual behaviors based on finite state automata. In goal-based trajectory analysis, 45 unusual behavior is detected by evaluating the explicability of the agent's trajectory with respect to known spatial goals.…”