Occupant Counting via Sensor Array Clustering

Solution Overview

Problem

There is a need for an efficient and accurate method to automatically determine the number of people in a specific area, such as a conference room, to manage resources like HVAC, lighting, and scheduling, while distinguishing between people and inanimate objects using sensor data.

Innovation Solution

A sensor array system, including thermopile arrays, RADAR, and image sensors, processes sensor data to filter out noise, estimate the background environment, and use k-means clustering to identify and count occupants by determining the probability of human presence based on heat signatures, effectively distinguishing between moving and stationary heat sources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If sensor data processing is used to automatically determine the number of people in an area, then resource management efficiency is improved, but the ability to distinguish between people and inanimate objects becomes insufficient

Engineering Contradiction:

Improveresource management efficiencyVSAvoidoccupant detection accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent segments the sensor data processing into multiple independent analysis components: thermal signature analysis, motion detection analysis, and pattern recognition analysis. Each component processes specific aspects of the sensor data separately and contributes to the final occupant determination, allowing the system to distinguish between people and objects by comparing results across multiple segmentation channels

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary processing layer that acts as a mediator between raw sensor data and final occupant counting. This intermediary layer applies multiple filtering and analysis techniques to resolve ambiguities in the sensor data, particularly in distinguishing heat signatures of people from those of inanimate objects before reaching the final count determination

Inventive Principle:

Principle #24Intermediary (Mediator)

2Device complexity

If simple sensor data analysis is used to count occupants, then system complexity is reduced, but the accuracy of distinguishing people from objects deteriorates

Engineering Contradiction:

Improvesystem complexityVSAvoidoccupant detection accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent implements dynamic processing that adapts the level of analysis based on the situation. The system uses motion detection as a dynamic filter that activates more sophisticated thermal analysis only when motion is detected, allowing the system to maintain high accuracy while reducing complexity in static conditions where simple thresholding suffices

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent applies partial action by using different levels of analysis for different sensor data channels. Not all data channels undergo the same degree of processing - thermal data receives extensive analysis while other channels use simpler processing, optimizing the balance between accuracy and complexity by applying excessive action only where necessary

Inventive Principle:

Principle #16Partial or excessive action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The system accurately counts occupants with up to 95% accuracy, providing efficient and reliable data for resource management, capable of differentiating between people and objects, and adaptable to various environments and sensor configurations.

Implementation Method 1

A sensor array system, including thermopile arrays, RADAR, and image sensors, processes sensor data to filter out noise, estimate the background environment, and use k-means clustering to identify and count occupants by determining the probability of human presence based on heat signatures

Methodology Applied

Scientific EffectThermopile: Thermopile

Implementation Method 2

A sensor array system, including thermopile arrays, RADAR, and image sensors, processes sensor data to filter out noise, estimate the background environment, and use k-means clustering to identify and count occupants

Methodology Applied

Scientific EffectRADAR: Radar

Data Source

PatentUS11107491B2Sensor data array and method of counting occupants

Publication Date: 2021.08.31 GE LIGHTING SOLUTIONS LLC

AI summary

A system and method including receiving, from a plurality of sensors, a sequence of data indicative of a presence of a person in an area of interest within a field of view of the plurality of sensors; determining an estimate of background level information for the area of interest in an instance of an absence of a person in the area of interest; generating a probability of a person being located in the area of interest based on a combination of the determined background level information and the sequence of data indicative of a presence of a person in the area of interest; determining a number of centroids in the area of interest based on an execution of clustering executed to determine an optimized total number of centroids for a dataset; and generating a count of persons in the area of interest based on determined total number of centroids.