Breathing Rate Estimation from Video Feature Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating breathing rate from video images are inadequate in uncontrolled environments, such as secure rooms, homes, or fitness settings, where subjects may be freely moving, clothed, and under varying lighting conditions, leading to noisy signals and difficulty in distinguishing breathing movements from other activities.
Innovation Solution
A method that detects and tracks image feature points, forms time series signals from their coordinates, selects strongly correlated signals, applies principal component analysis, and calculates the frequency of these signals to estimate breathing rate, even when the subject is covered or in a less controlled environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional video-based breathing detection methods are used in uncontrolled environments, then the system can operate in diverse settings (home, secure rooms, fitness centers), but the detection reliability deteriorates due to subject movement, variable lighting, and environmental noise
Solution Approach 1:
The patent segments the video signal into multiple feature point tracks across the image frame sequence. By tracking individual feature points and their coordinates over time, the system divides the complex breathing detection task into multiple manageable signal streams, allowing robust breathing rate estimation even when parts of the subject are obscured or moving.
Solution Approach 2:
The patent introduces an intermediary signal processing pipeline that includes coordinate extraction, signal correlation analysis, and principal component analysis. This intermediary processing layer transforms raw video data into reliable breathing rate estimates by filtering out environmental noise and isolating the periodic breathing signal from other movements and disturbances.
2Reliability
If the system tracks multiple feature points to improve breathing signal detection, then the signal robustness improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent applies partial action by selectively processing only the most relevant feature points and their coordinate signals. Rather than analyzing every pixel or feature in the video frame, the system focuses on tracking specific feature points that provide sufficient information for breathing rate estimation, reducing computational complexity while maintaining robustness.
Solution Approach 2:
The patent transforms spatial coordinate data into temporal signals and applies parameter transformations through correlation analysis and principal component analysis. This parameter transformation converts complex multi-dimensional feature point data into simplified temporal frequency characteristics that directly reveal breathing rate, reducing processing complexity.
3Measurement precision
If the system processes video frames at high resolution to improve detection accuracy, then the measurement precision improves, but the processing speed and frame rate decrease
Solution Approach 1:
The patent extracts only the essential coordinate information from video frames rather than processing complete high-resolution images. By taking out just the feature point coordinates and their temporal variations, the system achieves sufficient measurement precision for breathing rate while dramatically reducing computational load and increasing processing speed.
Solution Approach 2:
The system applies partial action by processing only the necessary portions of video data (feature point coordinates) rather than entire frames. This selective processing maintains measurement precision for breathing rate detection while avoiding the computational burden of full-frame analysis, thereby improving processing speed.
Data Source
AI summary
A method and apparatus for extracting a breathing rate estimate from video images of a respiring subject. Signals corresponding to the spatial coordinates of feature points tracked through the video sequence are filtered and excessively large changes are attenuated to reduce movement artefacts. The signals are differentiated and signals which correlate most strongly with other signals are selected. The selected signals are subject to principal component analysis and the best quality of the top five principal components is selected and its frequency is used to calculate and output a breathing rate estimate. The method is particularly suitable for detecting respiration in subject in secure rooms where the video image is of substantially the whole room and the subject is only a small part of the image, and maybe covered or uncovered.