Floor Vibration Counting via Neural Network Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for tracking individuals across complex spaces, such as train stations, face limitations due to issues like obscured movements in crowds, lighting challenges, and privacy concerns with camera usage, and require sophisticated pattern recognition to accurately count and differentiate between entry and exit points.
Innovation Solution
A device utilizing vibration sensors and machine learning models, specifically trained in stages with neural networks, to analyze floor vibrations and identify individuals' movements across multiple passageways, enabling accurate tracking and counting without visual surveillance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If camera-based systems are used to track persons, then visual monitoring capability is improved, but privacy concerns and vulnerability to sabotage increase
Solution Approach 1:
The patent replaces optical camera-based detection systems with vibration-based sensing systems. Vibration sensors embedded in the floor detect mechanical vibrations caused by person movements, transforming the detection mechanism from optical to mechanical domain. This substitution eliminates privacy concerns associated with visual surveillance and removes the vulnerability to camera sabotage while maintaining person tracking capability.
Solution Approach 2:
The patent introduces vibration sensors as an intermediary detection medium between persons and the tracking system. Instead of directly observing persons through cameras, the system detects indirect mechanical vibrations transmitted through the floor structure. This intermediary approach provides equivalent tracking information without requiring direct visual observation, thereby addressing privacy and security concerns.
2Object-affected harmful factors
If vibration sensors are used to track persons, then privacy concerns are reduced, but measurement precision in complex spaces deteriorates
Solution Approach 1:
The patent divides the complex space into multiple monitoring zones with distributed vibration sensors. Instead of relying on a single detection point, the system segments the floor area into grid-like zones, each equipped with sensors. This segmentation allows the system to track person movements through sequential detection across multiple zones, maintaining measurement precision even in complex multi-passageway environments like train stations.
Solution Approach 2:
The patent creates a universal vibration detection system that can handle multiple tracking scenarios simultaneously. The same vibration sensor network serves various functions: detecting person presence, determining movement direction, counting entries and exits across different passageways, and tracking individuals through complex paths. This multi-functional approach maintains measurement precision across diverse spatial configurations without requiring specialized systems for each scenario.
3Ease of operation
If manual pattern recognition rules are developed for vibration analysis, then interpretability is improved, but system complexity and development time increase
Solution Approach 1:
The patent replaces manual pattern recognition rule development with machine learning-based automated pattern recognition. Instead of requiring experts to manually create and maintain complex vibration pattern interpretation rules, the system uses trained machine learning models that automatically learn optimal patterns from training data. This substitution significantly reduces system development complexity while maintaining or improving interpretation accuracy in complex vibration environments.
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
This solution effectively tracks individuals' movements and counts entries and exits across complex spaces with high accuracy, overcoming the limitations of traditional camera-based systems by using vibration analysis and machine learning to differentiate patterns without the need for continuous visual monitoring.
Implementation Method 1
sense the vibrations made by persons passing
Data Source
AI summary
A system and method for counting persons using passages to an area by analyzing vibrations in the floor or the air above the floor with sensors and a machine learning system. The machine learning system uses a model, usually implemented as a neural network on a processor. The network is trained in levels and implemented in layers. Different levels classify and analyze vibrations by timing and frequency, by movements of persons, and by identity of persons The same person is identified by patterns in the vibrations and the vibrations are correlated to determine and count when a person uses a combination of passages. Location information for the person is used to identify persons in places and doing activities of interest. The model may be trained on one processor and downloaded to another processor for evaluation. Additional sensors and levels of training may be implemented on the latter processor.