Virtual Model Security Rule Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating security rules for monitoring systems require direct programming or manual annotation of maps or images, which can be time-consuming and inefficient, especially for large commercial properties, and may not effectively handle complex scenarios like occluded areas or blind spots.
Innovation Solution
A virtual reality environment is used to create a 3D model of a property where users can demonstrate acceptable and unacceptable behaviors, allowing the system to generate generalized security rules based on these interactions, which can then be enforced by the monitoring system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If direct programming or manual annotation methods are used to generate security rules, then rule generation can be achieved, but the process becomes time-consuming and inefficient especially for large commercial properties
Solution Approach 1:
The patent creates virtual copies of real-world properties in a simulated environment. These virtual models replicate the physical space, cameras, and sensor positions, allowing rule generation to occur in the virtual copy rather than requiring direct programming or manual annotation of the actual property. This copying approach enables faster iteration and testing without affecting the real system.
Solution Approach 2:
The system performs preliminary rule generation and testing in the virtual environment before deploying rules to the actual monitoring system. By conducting simulations and generating candidate rules in advance within the virtual model, the system prepares optimized rules that can be directly applied to the real property, reducing the time needed for on-site configuration and adjustment.
2Reliability
If manual annotation of maps or images is used, then security rules can be generated, but the complexity increases especially for handling occluded areas or blind spots
Solution Approach 1:
The patent transforms the rule generation problem from image-space annotations to 3D spatial parameters. By converting 2D image coordinates to 3D world coordinates and using depth information from multiple camera views, the system accurately represents occluded areas and blind spots through spatial parameters rather than complex 2D annotations. This parameter transformation simplifies the handling of occlusion scenarios.
Solution Approach 2:
The system transitions from 2D image annotation to 3D virtual space modeling. By adding the depth dimension and creating three-dimensional representations of the property space, the system naturally handles occluded areas and blind spots through spatial relationships rather than requiring complex 2D map annotations. The virtual model incorporates camera fields of view, occlusion volumes, and blind spot regions in three-dimensional space.
3Productivity
If generalized rules are generated using virtual demonstrations, then deployment efficiency across multiple cameras and locations improves, but the system must process and analyze virtual action data
Solution Approach 1:
The patent creates universal security rules in the virtual environment that can be applied across multiple cameras and locations simultaneously. By generating rules that work in the virtual model, the same rules can be deployed to multiple real-world camera systems without requiring separate configuration for each device or location. This multi-functionality approach enables single-rule deployment across entire property networks.
Solution Approach 2:
The system automatically processes virtual demonstrations and generates rules without requiring manual intervention for each rule creation. The virtual environment self-generates training data through simulated actions, and the system automatically analyzes this data to produce security rules, reducing the need for human experts to manually program each rule while optimizing computational resource usage through automated workflows.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for generating rules for a monitoring system. In some implementations, a method includes presenting a virtual model of a monitoring system of a property; obtaining action data from a computer generated avatar in the virtual model; generating a security rule for the monitoring system using the action data from the computer generated avatar in the virtual model; and providing, to the monitoring system, the security rule to cause the monitoring system to generate, using the security rule and sensor data from the property received by the monitoring system, a security alert.