Video Phase Analysis and Motion Magnification for Pulse Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting and magnifying subtle motions in videos, such as cardiovascular pulses and breathing, are invasive or inefficient, particularly in applications like measuring jugular venous pressure (JVP), which requires catheterization.
Innovation Solution
A method using image phase information generated through multiple filter applications at different spatial scales, followed by temporal filtering and scaling, to create a motion-magnified video stream that selectively amplifies motions at specific frequencies, such as cardiovascular pulses, without the need for invasive techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If catheterization is used to measure jugular venous pressure, then measurement precision is improved, but device complexity and invasiveness increase
Solution Approach 1:
The patent replaces the mechanical catheter-based measurement system with an optical-based motion magnification system. Video cameras capture subtle motions of the jugular vein, and image processing algorithms extract physiological signals without requiring physical insertion of catheters into the patient's body.
Solution Approach 2:
The patent introduces video recording and image phase information as an intermediary between the physiological phenomenon (JVP) and the measurement process. Instead of directly measuring pressure through catheterization, the system captures visual manifestations of vein motion and translates them into physiological data through computational analysis.
2Measurement precision
If motion magnification is applied to detect subtle motions, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the image processing into distinct stages: generating image phase information through filtering, applying temporal filters to extract specific frequency components, and separately magnifying the extracted signals. This segmentation of the processing pipeline makes the complex task more manageable and systematic.
Solution Approach 2:
The patent transforms spatial image data into a different dimensional representation by extracting phase information and temporal frequency components. This dimensionality change from spatial pixels to temporal-spectral domain enables selective amplification of specific motion frequencies while filtering out noise and irrelevant variations.
3Loss of information
If full video stream is transmitted, then information completeness is improved, but transmission bandwidth requirements increase
Solution Approach 1:
The patent extracts only the essential motion information from the full video stream by generating image phase information and applying temporal filters. Instead of transmitting every pixel of the original video, the system extracts and transmits only the relevant physiological signal components, significantly reducing data volume while preserving diagnostic information.
Data Source
Figure 1A~2C
Figure 3A~3E
Figure 4A~5C
AI summary
Example embodiments allow for fast, efficient motion-magnification of video streams by decomposing image frames of the video stream into local phase information at multiple spatial scales and/or orientations. The phase information for each image frame is then scaled to magnify local motion and the scaled phase information is transformed back into image frames to generate a motion-magnified video stream. Scaling of the phase information can include temporal filtering of the phase information across image frames, for example, to magnify motion at a particular frequency, ln some embodiments, temporal filtering of phase information at a frequency of breathing, cardiovascular pulse, or some other process of interest allows for motion-magnification of motions within the video stream corresponding to the breathing or the other particular process of interest. The phase information can also be used to determine time-varying motion signals corresponding to motions of interest within the video stream.