Detection-Based Wakeup for Battery-Powered Surveillance Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Detection systems with battery-powered detection devices face reduced efficacy due to higher power consumption, necessitating frequent battery replacements or external power sourcing, especially when devices with higher computational demands are used in surveillance systems.
Innovation Solution
A method involving a first detection device generating a mapping that associates its results with those of a second detection device, allowing the second device to transition to a lower power mode until specific object types are detected, at which point it is awakened to perform actions, thereby conserving power and extending operational duration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If detection devices perform more operations and computationally intensive operations, then detection capability and object identification accuracy are improved, but power consumption increases
Solution Approach 1:
The mapping between detection results and object classifications is pre-computed and stored during a training phase. This preliminary action allows the detection device to perform simple lookups during operation rather than executing computationally intensive classification algorithms in real-time, significantly reducing power consumption while maintaining detection accuracy
Solution Approach 2:
The patent creates a simplified representation (mapping) that copies the essential relationship between detection results and object types. Instead of performing full classification computations, the device uses this copied mapping structure to quickly determine object classifications, reducing computational burden and power consumption
2Reliability
If battery-powered detection devices operate continuously with high computational demands, then detection system efficacy is maintained, but battery life is reduced requiring frequent replacements
Solution Approach 1:
The mapping is pre-computed during a training phase before deployment. This preliminary computation transfers the computational burden to an initial setup phase, allowing the battery-powered device to operate with minimal processing during extended monitoring periods, thereby extending battery life while maintaining detection reliability
Solution Approach 2:
The detection device transitions between sleep mode and active detection mode periodically. It performs simple detection operations continuously but only activates full processing capabilities when objects of interest are detected, based on the pre-computed mapping, thus extending battery life while maintaining system efficacy
3Measurement precision
If computationally intensive detection devices are deployed in remote locations, then detection accuracy is improved, but accessibility for power supply and maintenance is reduced
Solution Approach 1:
The complex mapping relationships are pre-computed during a training phase that can be performed in accessible locations or during initial setup. This preliminary computation allows the remote device to operate with simple lookup operations, maintaining high detection accuracy while reducing the need for frequent human intervention for maintenance and power management
Solution Approach 2:
The detection device autonomously uses the pre-computed mapping to perform accurate object classification without requiring continuous human intervention or complex processing resources. This self-service capability maintains high detection accuracy while minimizing the need for accessible maintenance, making remote deployment more practical
Data Source
AI summary
Techniques are disclosed for detection-based wakeup of detection devices. A method may include generating a mapping that includes entries. Each entry may associate a respective detection result of a first detection device with a respective detection result of a second detection device. The method may further include transitioning the second detection device to a first power mode upon completing the generating. The method may further include determining, by the first detection device, that a first object of a first object type is detected based on the mapping. The method may further include transitioning the second detection device out of the first power mode when the first object of the first object type is determined to be detected. The method may further include performing, by the second detection device, at least one action based on the first object. Related systems and devices are also disclosed.