People Counting System with Local Privacy Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing people counting systems do not adequately address user privacy concerns, as they may inadvertently reveal information about individuals' presence, necessitating a mechanism for users to opt-out without compromising the overall functionality of the system.
Innovation Solution
A people counting system that allows users to opt-out by indicating their position or location using various methods, such as beacon signals, floor plan marking, or explicit signaling, ensuring that location information is not provided to the central processing unit, thereby maintaining system-level counting functionality while respecting user privacy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system aggregates sensor data from multiple sensors to extract information relating to people in the area, then the people counting functionality is improved, but user privacy is compromised
Solution Approach 1:
The system segments the processing of sensor data by introducing local processing units at each luminaire that perform initial analysis and filtering before data is aggregated centrally. This segmentation allows privacy-sensitive operations to occur locally while maintaining centralized counting functionality, thus resolving the contradiction between accurate people counting and user privacy protection.
Solution Approach 2:
The patent introduces an intermediary processing layer between the sensors and the central aggregation system. This intermediary layer processes and anonymizes data locally, acting as a mediator that protects user privacy while still enabling accurate people counting through aggregated metrics, thus resolving the privacy-accuracy contradiction.
2Measurement precision
If the system processes image data centrally to extract people count information, then the counting accuracy is improved, but the system complexity and data transmission requirements increase
Solution Approach 1:
The system divides the processing architecture into distributed local processing units and a central aggregation layer. Each luminaire contains a local processor that performs initial image analysis and extracts relevant metrics, reducing the complexity burden on the central system while maintaining counting accuracy through distributed intelligence.
Solution Approach 2:
Instead of performing complete image processing centrally, the system performs partial processing locally at each luminaire, extracting only the necessary metrics (such as presence detection and basic counting information). This partial action approach reduces central processing complexity while maintaining sufficient accuracy for people counting applications.
Data Source
Figure 1~3
Figure 3A
Figure 4~4A
AI summary
In a people counting system, a plurality of vision sensors is arranged to provide sensor coverage of an area. Each is arranged to provide individual sensor coverage of a portion of the area within its field of view. Each of a plurality of local image processors is connected to a respective one of the vision sensors. Each of the local image processors is configured to apply a local person detection algorithm to at least one image captured by its respective vision sensor, thereby generating a local presence metric representative of a number of people detected in the at least one image. A central processor is configured to estimate the total number of people in the area covered by the vision sensors by applying an aggregation algorithm to the local presence metrics generated by the local image processors. As it is critical that user privacy be taken into account when utilising such people counting technology, an opt-out is enabled. For instance, there may be users who do not want to reveal any information that may be perceived give away information related to their presence. In this context, a user is a person in the environment where people counting may be implemented. The disclosure applies not only to people counting, but to other contexts where people may be monitored, for example, a CCTV environment.