Video-Based Vibration Measurement Using Eulerian Motion Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for measuring vibrations using accelerometers are cumbersome and time-consuming, especially for small structures, and video-based methods require extensive processing time and the use of targets, which can be impractical for many applications.
Innovation Solution
A video-based system that extracts pixel-wise Eulerian motion signals from video frames without the need for sensors or targets, allowing for rapid estimation of resonant frequencies and mode shapes, enabling near real-time analysis by downselecting signals based on local contrast and performing end-to-end processing significantly faster than digital image correlation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If accelerometers are used to measure vibrations, then measurement precision is improved, but device complexity and time consumption increase
Solution Approach 1:
The patent replaces mechanical accelerometers with a video-based optical system. A camera captures video frames of the vibrating structure, and pixel-wise Eulerian motion signals are extracted through digital processing. This substitution eliminates the need for physical sensor attachment while maintaining measurement capability, directly resolving the contradiction between measurement precision and device complexity.
2Measurement precision
If accelerometers are attached to small structures, then vibration measurements can be obtained, but the added mass negates the measurements
Solution Approach 1:
The patent replaces physical accelerometers with a non-contact video-based measurement system. By using a camera to capture visual information and extracting motion signals through Eulerian derivative methods, the system eliminates the need for attaching any mass to the structure being measured, thus preserving the natural vibration characteristics of small structures.
3Ease of operation
If video-based methods with pattern matching are used, then sensor attachment is avoided, but processing time increases significantly
Solution Approach 1:
The patent transforms the video processing approach by computing Eulerian motion signals directly from pixel intensity changes over time, rather than using computationally intensive pattern matching or digital image correlation. This parameter change in the processing method reduces computational complexity and enables rapid analysis while maintaining non-contact measurement advantages.
Solution Approach 2:
The patent extracts only the essential motion information needed for vibration analysis by computing Eulerian derivatives at each pixel location. This extraction of critical motion parameters from the video stream allows for efficient processing without the overhead of full pattern matching, thus reducing processing time while maintaining measurement accuracy.
4Ease of operation
If existing video processing methods are used, then non-contact measurement is achieved, but targets with known patterns are required
Solution Approach 1:
The patent enables the video system to extract motion signals directly from the structure's surface features without requiring externally applied targets or patterns. The Eulerian motion signal extraction method utilizes the natural visual information already present in the video frames, allowing the system to be self-sufficient and eliminating the need for additional target placement equipment or procedures.
Data Source
AI summary
A method and corresponding device for identifying operational mode shapes of an object in a video stream includes extracting pixel-wise Eulerian motion signals of an object from an undercomplete representation of frames within a video stream. Pixel-wise Eulerian motion signals are downselected to produce a representative set of Eulerian motion signals of the object. Operational mode shapes of the object are identified based on the representative set. Resonant frequencies can also be identified. Embodiments enable vibrational characteristics of objects to be determined using video in near real time.