Occupancy Estimation via Sensor Fusion and Event Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining occupancy in spaces, such as queue lengths or vehicle numbers in car parks, face inaccuracies when using image data alone, particularly when groups of individuals or objects are involved, leading to errors in estimating the number of objects or groups.
Innovation Solution
A method that combines first data on occupancy with second data indicative of incrementing or decrementing events to refine estimates of object or group numbers, using algorithms to minimize error and provide improved occupancy information over time, allowing for real-time processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image data alone is used to determine occupancy, then the system is simple and fast, but measurement precision deteriorates due to inability to resolve groups and identify individuals accurately
Solution Approach 1:
The patent combines multiple data sources (image data from cameras, sensor data from detectors, and transaction data from point-of-sale systems) to create a comprehensive occupancy estimation system. By merging these different types of data, the system overcomes the limitations of image data alone and achieves more accurate occupancy measurements without requiring a single complex system.
Solution Approach 2:
The patent introduces an intermediary processing layer that correlates occupancy data with transaction data and temporal information. This intermediary layer matches image/sensor detections with actual transaction events, serving as a mediator that reconciles the imperfect image data with ground truth transaction information to produce accurate occupancy estimates.
2Measurement precision
If image data processing is enhanced to resolve groups and improve accuracy, then measurement precision improves, but processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary matching of occupancy data with transaction data in advance, creating correlated datasets that can be quickly queried. By pre-processing and correlating the data before it is needed for final analysis, the system reduces the computational burden during real-time operations and enables faster processing when occupancy information is required.
Solution Approach 2:
The patent segments the occupancy estimation problem into distinct components: initial detection from image/sensor data, correlation with transaction events, and final occupancy determination. This segmentation allows each component to be processed independently and efficiently, reducing overall processing time while maintaining accuracy.
3Reliability
If only image data is used, then the system is simple to implement, but reliability deteriorates due to errors in resolving groups and counting individuals
Solution Approach 1:
The patent implements a feedback mechanism where occupancy estimates are continuously refined by comparing image/sensor data with actual transaction data. The system uses transaction events as feedback to validate and correct occupancy measurements, creating a self-correcting system that improves reliability over time while managing complexity through iterative refinement.
Data Source
AI summary
A method of determining information relating to the occupancy of a space that includes receiving first data relating to the occupancy of the space over a time period, receiving second data indicative of either times of decrementing events or times of incrementing events affecting the occupancy of the space during the time period, and either, for each decrementing event time, using the first data to obtain an estimated incrementing event time, or, for each incrementing event time, using the first data to obtain an estimated decrementing event time. The method has particular application in the determination of waiting times, for example in the context of queue monitoring.