Occupancy Sensor Calibration via Vision Data Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional occupancy sensors, such as PIR sensors, provide binary information on occupancy/vacancy but lack precise sensing region definition, leading to inaccurate data aggregation and false positives/negatives in lighting control systems, as the actual sensing region is unknown or poorly defined.
Innovation Solution
An occupancy sensor calibration device uses vision sensors to determine the location and extent of occupancy within the sensing region, allowing for precise calibration and improved data aggregation by correlating occupancy data from both types of sensors to define the sensing region accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional occupancy sensors (PIR sensors) are used to provide binary occupancy information, then the system is simple and energy-efficient, but the sensing region is unknown or poorly defined leading to inaccurate data aggregation
Solution Approach 1:
A vision sensor is introduced as an intermediary device to calibrate and determine the actual sensing region of the occupancy sensor. The vision sensor captures images and processes them to identify the precise sensing region, which is then used to correct and refine the occupancy data from the PIR sensor, resolving the contradiction between simplicity and measurement precision.
Solution Approach 2:
The patent replaces the mechanical/physical limitation of unknown sensing regions with an information-based solution using vision sensors and image processing. Instead of relying on fixed mechanical assumptions about sensor fields of view, the system uses computational vision to dynamically determine and correct sensing region boundaries, improving precision without requiring complex hardware modifications.
2Productivity
If occupancy sensors with wide field-of-view are used, then more sensors can detect occupancy in the same distribution, but the sensing region coverage becomes less precise and causes false positives
Solution Approach 1:
The system uses vision sensor data as feedback to continuously calibrate and refine the occupancy sensor's sensing region definition. By comparing vision-based occupancy detection with PIR sensor readings, the system identifies and corrects false positives, adjusting the sensing region boundaries to improve accuracy while maintaining high detection capability.
Solution Approach 2:
The patent dynamically adjusts the sensing region parameters (field of view, detection boundaries) based on real-time vision sensor data. The system changes the effective sensing region parameters to match actual occupancy patterns detected by the vision sensor, allowing the system to adapt to different environmental conditions and maintain both high productivity and precision.
3Ease of manufacture
If the sensing region is defined based on sensor specs and mounting information, then the system is easy to install, but the actual sensing region is unknown leading to false positives and negatives
Solution Approach 1:
The system performs preliminary calibration using the vision sensor to determine the actual sensing region before relying on occupancy data for control decisions. This preliminary action of mapping and calibrating the sensing region ensures that subsequent occupancy detection is reliable, while the calibration process itself is automated and does not complicate the initial installation.
Solution Approach 2:
The vision sensor automatically performs calibration and sensing region determination without requiring manual intervention or complex configuration. The system self-calibrates by processing vision data to identify occupancy patterns and derive the actual sensing region, maintaining ease of installation while significantly improving detection reliability through automated self-service calibration.
Data Source
AI summary
Some embodiments are directed to an occupancy sensor calibration device (100) arranged to repeatedly detect an occupancy in vision data and a concurrent occupancy detection in the occupancy data, determine a location of the detected occupancy in the vision data, and store the location as part of the occupancy sensing region.