Fall Detection Using Multi-Sensor Corroboration and Privacy Obfuscation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wearable fall detectors, such as accelerometer-based devices and pendants, are unreliable in detecting falls, especially when users are alone or unconscious, and may pose privacy and safety risks, as they can be removed or become hazards.
Innovation Solution
A fall detection system incorporating an image capture device, microphone, and processor with machine-readable code and AI algorithms to capture and analyze visual and audio data, corroborating potential falls and providing corroborative evidence to ensure accurate detection and notification, while protecting user privacy and safety.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If wearable fall detectors (accelerometer-based devices) are used, then portability and ease of operation are improved, but detection accuracy and reliability deteriorate
Solution Approach 1:
The patent combines multiple detection modalities (image capture device, microphone, accelerometer) into a single fall detection system. The image capture device provides visual confirmation of falls, the microphone captures audio cues like screams or impacts, and the accelerometer detects motion patterns. This multi-sensory approach compensates for the limitations of accelerometer-only devices, particularly for gradual falls or when users are unconscious.
Solution Approach 2:
The system introduces an image capture device as an intermediary between the user and the detection algorithm. Instead of relying solely on accelerometer data, the image capture device provides direct visual evidence of fall events, serving as an intermediary that validates or refutes accelerometer-based detections and reduces false positives.
2Reliability
If continuous monitoring is implemented, then detection coverage is improved, but privacy concerns worsen
Solution Approach 1:
The system applies different processing qualities to different regions of the captured data. Privacy-sensitive regions (like bathrooms or bedrooms) are identified and processed with higher obfuscation, while public areas maintain standard monitoring quality. This allows continuous monitoring coverage while protecting privacy in specific local zones where users have reasonable expectation of privacy.
Solution Approach 2:
The system extracts only the essential information needed for fall detection from the full video and audio streams. Instead of storing or transmitting complete high-resolution video footage, the system extracts key features such as fall detection events, audio cues, and summarized metadata. This extraction approach maintains detection coverage while minimizing privacy exposure by removing unnecessary detailed visual information.
3Measurement precision
If image capture device with high resolution is used, then detection accuracy is improved, but privacy protection deteriorates
Solution Approach 1:
The system dynamically changes the resolution parameter of the image capture device based on the detection context. During normal monitoring, lower resolution is used to protect privacy. When a potential fall event is detected by the accelerometer or audio sensors, the system temporarily switches to higher resolution to accurately capture and analyze the fall event, then returns to lower resolution for privacy protection.
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
Enhances the accuracy of fall detection by using multi-sensory data analysis and AI to confirm falls, reduces false negatives, and ensures timely alerts without compromising user privacy or safety, even when users are alone or unconscious.
Implementation Method 1
an image capture device, a microphone, and a processor connected to a memory that includes machine-readable code defining an algorithm for controlling the image capture device
Implementation Method 2
capture sound data from the microphone to supplement information not captured by the image capture device and to corroborate potential falls detected by the image capture device
Data Source
AI summary
A method and system for detecting a person falling, confirming a potential fall, and taking action to mitigate the effects of a fall.