Video Analytics Execution Control for Cost Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The continuous execution of video analytics for proactive incident detection in security systems drains computing resources and incurs significant costs, which can be dynamically high due to varying activity levels and environmental factors, necessitating a method to selectively enable or disable video analytics based on cost considerations.
Innovation Solution
An electronic computing device accesses an incident database to compare the average costs of incident resolution between human-reported and video analytics-reported incidents, enabling video analytics when the human-reported costs exceed video analytics-reported costs by a predefined threshold, thereby optimizing resource usage and cost savings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If video analytics is continuously executed to proactively detect incidents, then incident detection capability is improved, but computing resource consumption and costs increase
Solution Approach 1:
The system dynamically adjusts the execution state of video analytics based on real-time cost evaluations and incident patterns. The video analytics execution is switched between enabled and disabled states according to whether the evaluated cost meets a threshold, allowing the system to adapt to varying operational conditions and optimize resource usage while maintaining detection capability when beneficial
Solution Approach 2:
The system changes the operational parameter of video analytics execution by evaluating costs associated with different execution states. By comparing the cost of continuous execution against alternative approaches and determining whether costs meet a predefined threshold, the system optimizes resource allocation while maintaining incident detection effectiveness
2Reliability
If video analytics is continuously executed to proactively detect incidents, then proactive detection capability is improved, but operational costs increase
Solution Approach 1:
The system dynamically switches between proactive detection modes by evaluating operational costs in real-time. When cost evaluation indicates that continuous execution exceeds the threshold, the system transitions to a disabled state, thereby reducing operational costs while maintaining the capability to activate proactive detection when cost-effective
Solution Approach 2:
The system optimizes operational costs by changing the execution parameter of video analytics based on cost evaluations. By determining whether evaluated costs meet a predefined threshold, the system adjusts operational expenditure while preserving proactive detection capability under optimal cost conditions
3Loss of energy
If video analytics execution is selectively enabled or disabled based on cost, then cost savings are improved, but incident detection responsiveness may worsen
Solution Approach 1:
The system incorporates feedback mechanisms by continuously evaluating costs associated with video analytics execution and comparing them against predefined thresholds. This feedback loop enables the system to make informed decisions about enabling or disabling analytics execution, balancing cost savings with incident detection needs based on real-time cost evaluations and historical incident patterns
Solution Approach 2:
The system performs preliminary cost evaluations before executing video analytics to determine whether the anticipated costs meet a predefined threshold. By evaluating costs in advance and making proactive decisions about execution state, the system avoids unnecessary computational expenditure while ensuring analytics are enabled when cost-effective and incident detection is prioritized
Data Source
AI summary
A process of selectively enabling execution of video analytics on videos captured by cameras. In operation, an electronic computing device accesses an incident database identifying incidents resolved by one or more agencies. The incidents include a first set of incidents that were first reported to the agencies by a human source and a second set of incidents that were first reported to the agencies by a video analytics system. The electronic computing device then estimates a first average cost incurred in resolving the first set of incidents and a second average cost incurred in resolving the second set of incidents. The electronic computing device enables the video analytics system to execute video analytics on videos captured by the one or more cameras to proactively detect and report incidents when the first average cost is higher than the second average cost by at least the predefined threshold.