Wireless Analytics Person Count Correction via Video Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wireless analytics systems tend to overestimate the number of people in a given area due to multiple devices carried by individuals and external devices connecting to wireless access points, while also failing to count individuals without devices, leading to inaccurate person counts.
Innovation Solution
Implementing a system that combines video trip wire technology with wireless device counts, using a correction factor based on video counts to refine the accuracy of person counts in areas without video coverage, by dividing video counts by wireless device counts and applying this factor to areas without video coverage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If wireless device counts are used to estimate person counts, then the coverage area is expanded, but the measurement precision deteriorates due to multiple devices per person and external devices
Solution Approach 1:
The patent introduces video camera counts as an intermediary reference to mediate between wireless device counts and actual person counts. The video system serves as a mediator that provides ground-truth data for calibration, allowing the wireless system to correct its overcounting issues while maintaining expanded coverage area.
Solution Approach 2:
The patent changes the parameter relationship by establishing a correction factor (ratio of video count to wireless device count) that transforms the raw wireless device count into a more accurate person count estimate. This parameter transformation allows the system to maintain broad coverage while improving measurement precision through mathematical adjustment.
2Measurement precision
If video cameras are deployed in all locations to achieve accurate person counts, then the measurement precision improves, but the device complexity and cost increase
Solution Approach 1:
The patent segments the monitoring system into two distinct components: video cameras deployed only in areas where accurate person counts are critical, and wireless access points deployed throughout the entire facility. This segmentation allows the system to achieve high measurement precision in key areas without the prohibitive cost and complexity of universal video coverage.
Solution Approach 2:
The patent makes the wireless access point system multi-functional by enabling it to serve both its primary network function and a secondary person-counting function. By calibrating the wireless system against video camera data, the access points become universal devices that provide both network connectivity and analytics capabilities across the entire facility.
3Ease of operation
If wireless analytics are used to count people, then the ease of operation improves, but the reliability deteriorates due to external devices and multiple devices per person
Solution Approach 1:
The patent implements a feedback mechanism where video camera counts provide ground-truth data that feeds back into the wireless analytics system. This feedback loop allows the system to continuously refine its correction factors and improve reliability while maintaining the ease of operation of wireless-based counting.
Data Source
AI summary
Systems, methods, and computer program products to perform an operation comprising receiving, from a video monitoring system, a count of people detected at a first location, wherein the video monitoring system comprises a camera configured to capture video of the first location, receiving a count of wireless devices detected in a second location, and applying a first correction factor to the count of wireless devices detected at the second location to determine a number of people present in the second location, wherein the first correction factor is based on the count of people detected by the video monitoring system and a count of wireless devices in the first location.