Fall Detection Using Triangle Feature Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fall detection systems face challenges in accuracy and speed, particularly in distinguishing falls from other activities like lying down, and may misinterpret user positions, especially when using only acceleration and angular velocity data.
Innovation Solution
A fall detection method and apparatus that calculates a triangle feature using acceleration and angular velocity values, incorporating reference values based on user-specific information such as age, gender, body dimension, and weight, to differentiate between falls and other activities, thereby enhancing detection accuracy and speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only acceleration and angular velocity data are used for fall detection, then the detection speed is fast, but the accuracy is insufficient and false alarms occur when user is lying down
Solution Approach 1:
The patent introduces a triangle feature as an additional dimension beyond traditional acceleration and angular velocity data. This triangle feature is calculated from the acceleration components and provides a new geometric perspective for distinguishing falls from lying-down positions, thereby improving detection accuracy without requiring additional sensors
Solution Approach 2:
The patent transforms acceleration data into a triangle feature parameter that captures the geometric relationship between acceleration components. By changing the parameter representation from raw acceleration values to a derived triangle feature, the system achieves better discrimination between fall and non-fall states
2Reliability
If triangle feature is added to improve accuracy, then false alarms are reduced, but the computational complexity increases
Solution Approach 1:
The patent replaces complex machine learning models with a geometric triangle feature calculation that uses simple arithmetic operations. This substitution maintains high reliability while significantly reducing computational energy consumption, making the system suitable for resource-constrained wearable devices
3Measurement precision
If user-specific reference values are incorporated, then detection accuracy for different users improves, but the system complexity and calibration requirements increase
Solution Approach 1:
The system automatically determines user-specific reference values through self-calibration procedures that require minimal user input. The device performs automatic measurements and establishes personalized thresholds without requiring manual configuration, thereby improving detection accuracy while maintaining ease of operation
Data Source
AI summary
Provided are a fall detection apparatus and method. The fall detection method may include measuring first to third accelerations corresponding to a user, measuring first to third angular velocities corresponding to the user, determining an acceleration value for the user based on at least one of the first to third accelerations, determining an angular velocity value for the user based on at least one of the first and second angular velocities, determining an angle value for the user based on at least one of the first to third accelerations, calculating a triangle feature using the acceleration value, and detecting a fall of the user based on at least one of the acceleration value, the angular velocity value, the angle value, and the triangle feature.