Visual Surveillance Metadata Management

Abstract

The paper deals with a solution for visual surveillance metadata management. Data coming from many cameras is annotated using computer vision units to produce metadata representing moving objects in their states. It is assumed that the data is often uncertain, noisy and some states are missing.The solution consists of the following three layers: (a) data cleaning layer -improves quality of the data by smoothing it and by filling in missing states in short sequences referred to as tracks that represent a composite state of a moving object in a spatiotemporal subspace followed by one camera. (b) Data integration layer -assigns a global identity to tracks that represent the same object. (c) Persistence layermanages the metadata in a database so that it can be used for online identification and offline querying, analyzing and mining. A Kalman filter technique is used to solve (a) and a classification based on the moving object's state and its visual properties is used in (b). An object model for layer (c) is presented too.18th International Workshop on Database and Expert Systems Applications 1529-4188/07 $25.00

Metadata extraction and organization for intelligent video surveillance system

“…In addition to this set, appropriate descriptions for the relation between camera and scene were also considered. To summarize, the research papers [1][2][3][4][5][6][7][8] have proposed various methods and structure of metadata organization, and have used them for the event recognition and also end-user queries.…”

“…In this situation, the necessity to produce and benefit from the extracted metadata [1] [2] in an efficient way becomes more and more significant in the video surveillance research community. Metadata that contains object's moving trajectories, object shape, object behaviors, duration time can be extracted from the raw image sequence.…”

“…Metadata that contains object's moving trajectories, object shape, object behaviors, duration time can be extracted from the raw image sequence. [2] It can then be provided for the further event recognition application, as well as for end-user's understanding on the details of the occurred event and human activity. Figure 1 shows the use of metadata extracted from input image sequence for user's query or subsequent event recognition.…”