Composite Event Detection in Multi-Camera Surveillance Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video processing systems face challenges in detecting complex events across multiple camera views without pre-defined event definitions, requiring a generic and user-friendly interface to define and combine primitive events into composite alerts efficiently.
Innovation Solution
A system and method for monitoring events using an event manager to combine primitive event definitions into composite alert events, allowing users to define and track events across multiple cameras with a graphical user interface, saving event definitions in a structured format like XML for parsing by tracking engines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple cameras are used to increase field of view and coverage, then the ability to detect complex events is improved, but the system complexity and difficulty of event definition increase
Solution Approach 1:
The system segments complex multi-camera events into primitive event types (motion detection, directional motion, abandoned object, object removal, trip wire crossing) that can be defined and detected independently in each camera view. These primitives are then combined using operators to form composite events, allowing complex multi-camera scenarios to be built from simple, reusable components without requiring complete redefinition for each camera configuration.
Solution Approach 2:
The event definition structure implements nesting where primitive events are contained within composite events, which can themselves be nested within higher-level composite events. This hierarchical organization allows events to be defined at multiple levels of abstraction, where a composite event can combine multiple primitive events from different cameras, and these composites can be further combined to create increasingly complex event scenarios across multiple camera views.
2Ease of operation
If pre-defined event definitions are hard-coded into the system, then the system operation is simplified, but the adaptability to different applications and users is reduced
Solution Approach 1:
The system transitions from static, hard-coded event definitions to dynamic, user-configurable event definitions. Users can define custom primitive events and composite events through a graphical interface without requiring programming knowledge. The system adapts to different applications and users by allowing flexible configuration of event parameters, regions of interest, and combination operators, while maintaining simplified operation through automated parsing and engine execution.
Solution Approach 2:
The event definition scheme creates a universal framework that can express diverse events across multiple applications. The same primitive event types and operators can be used to define events for different users and applications (surveillance, traffic monitoring, statistics gathering) without requiring application-specific customization of the core system. The structured format allows the same definition mechanism to serve multiple functions and domains.
3Measurement precision
If complex event definitions with multiple parameters are allowed, then the detection capability is improved, but the user interface complexity and difficulty of definition increase
Solution Approach 1:
The system introduces an intermediary layer between the user and the complex detection parameters. The graphical user interface serves as a mediator that presents simplified event definition options to users, automatically translating user-friendly selections into detailed parameter configurations. Users define events through high-level concepts (primitive types and operators) without directly managing complex parameters, while the system handles the parameter extraction, validation, and configuration automatically.
Solution Approach 2:
The system uses template-based copying where predefined primitive event types serve as reusable templates. Once a primitive event type is defined, it can be copied and combined multiple times in different configurations to create composite events. This eliminates the need for users to redefine similar events from scratch and reduces interface complexity by providing standardized building blocks that can be instantiated repeatedly with different parameters.
4Productivity
If event definitions are saved in a structured format for parsing, then the system efficiency is improved, but the requirement for standardized formats and processing increases complexity
Solution Approach 1:
The system changes the state of event definitions from unstructured or semi-structured input to a standardized structured format (such as XML) through automated parameter extraction and validation. The graphical interface collects user input and transforms it into a consistent structured format that can be efficiently parsed by tracking engines. This parameter transformation occurs once during definition, enabling efficient processing during event detection without requiring complex processing during runtime.
Data Source
AI summary
A system and method for monitoring events includes defining alert events based on primitive event types in an environment monitored by monitoring devices. Alert parameters are set for each alert event, and alert events are combined to create a composite alert event based on a plurality of alert events. A plurality of engines is monitored to determine if multiple events comprising the composite alert event have occurred.