Video-Based Fall Detection Using Optical Flow Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current falling detection systems for elderly individuals are inconvenient, have varying accuracy, and are limited by battery life, as they require wearable devices and rely on accelerometer data analysis.
Innovation Solution
A method and electronic system that uses image analysis to detect falling events by recognizing human body features, calculating movement trends via optical flow algorithms, and inputting these features into a falling classifier to determine if a fall has occurred, sending an alarm if necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a wearable device with three-axis accelerometer is used for falling detection, then falling detection function is provided, but user convenience deteriorates due to requirement of continuous wearing
Solution Approach 1:
The patent replaces the mechanical wearable accelerometer system with an optical-based video analysis system. The image capturing device records video of the monitoring area, and the processor analyzes the video frames to detect falling events through human body feature recognition and movement trend calculation, eliminating the need for wearable mechanical sensors.
Solution Approach 2:
The patent introduces video footage as an intermediary medium between the monitoring target and the detection system. Instead of directly measuring acceleration from a wearable device, the system captures visual information through an image capturing device and processes it to infer falling events, using video as the mediating carrier of information.
2Adaptability or versatility
If wearable device position varies, then detection scenarios increase, but measurement precision deteriorates due to position-dependent accuracy
Solution Approach 1:
The patent creates a universal detection system that functions across multiple scenarios without requiring adjustment for position. The image capturing device captures video from a fixed location, and the processor universally applies human body feature recognition and optical flow analysis to detect falls regardless of where the person is positioned within the monitoring area, providing consistent accuracy across all scenarios.
3Reliability
If continuous monitoring is performed to improve detection reliability, then falling detection reliability is improved, but energy consumption increases due to battery limitations
Solution Approach 1:
The patent implements continuous monitoring through periodic capture and analysis of video frames. The image capturing device continuously records video stream, and the processor periodically analyzes frames to detect changes in human body position and movement trends, maintaining reliable detection coverage without the energy constraints of battery-powered wearable devices.
Data Source
AI summary
A falling detection method and an electronic system using the same are provided. The falling detection method includes: obtaining a video stream, and performing recognition on a person in an image of the video stream to obtain at least one human body feature; calculating based on the human body feature to obtain at least one falling related feature; calculating at least one movement trend in a plurality of directions of the person in the video stream by using an optical flow algorithm; inputting the falling related feature and the movement trend to a falling classifier, such that the falling classifier generates an output result indicating whether a falling event happens to the person; and sending an alarm according to the output result.