CN110108348B - Thin-wall part micro-amplitude vibration measurement method and system based on motion amplification optical flow tracking - Google Patents
Thin-wall part micro-amplitude vibration measurement method and system based on motion amplification optical flow tracking Download PDFInfo
- Publication number
- CN110108348B CN110108348B CN201910401791.1A CN201910401791A CN110108348B CN 110108348 B CN110108348 B CN 110108348B CN 201910401791 A CN201910401791 A CN 201910401791A CN 110108348 B CN110108348 B CN 110108348B
- Authority
- CN
- China
- Prior art keywords
- thin
- wall part
- vibration
- pixel
- frequency
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H9/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/207—Analysis of motion for motion estimation over a hierarchy of resolutions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
Abstract
The invention discloses a thin-wall part micro-amplitude vibration measurement method and system based on motion amplification optical flow tracking, which relate to the technical field of thin-wall part micro-amplitude vibration measurement and mainly obtain target vibration frequency through brightness change; processing the gray image sequence by a phase Euler method, and amplifying the vibration amplitude; processing the video after motion amplification by adopting an optical flow tracking algorithm, extracting a vibration signal of a target corner point by taking a pixel as a unit, and compensating attenuation; and establishing a measurement mathematical model, and converting the measurement mathematical model into a vibration signal of an actual unit. The invention combines the phase Euler amplification method and the optical flow tracking method, effectively solves the problem that the micro-amplitude vibration of the thin-wall part cannot be measured, and has the advantages of convenient operation, high spatial resolution, non-contact and visual micro-vibration and the like.
Description
Technical Field
The invention relates to the technical field of thin-wall part micro-amplitude vibration measurement, in particular to a thin-wall part micro-amplitude vibration measurement method and system based on motion amplification optical flow tracking.
Background
The measurement and research of the tiny vibration signal have very important significance in military and civil use. The vibration signal is commonly existed in thin-wall part vibration of the aircraft engine. The relevant research shows that the vibration amplitude of thin-wall parts of the aeroengine is between dozens of micrometers and hundreds of micrometers. Because parts in the aircraft engine often use thin-walled parts as main structural forms, accurately and efficiently acquiring vibration signals of the thin-walled parts of the aircraft engine has important significance for the research on the aspects of vibration characteristic analysis, vibration reduction optimization design, structural damage identification and the like of the aircraft engine. However, at present, efficient and accurate measurement of minute amplitude vibration is not yet achieved.
Currently, there are three main types of techniques for measuring vibration: mechanical measurement techniques, electrical measurement techniques, and optical measurement techniques. The optical measurement technology has the advantages of high precision, high sensitivity, non-contact, use in flammable and explosive severe environments, no damage to the surface and the like, and has become a main means of high-precision vibration measurement, and methods such as laser vibration measurement, visual vibration measurement and the like can independently and objectively complete measurement on a measured object, wherein the displacement measured by the laser vibration measurement method can be as small as several microns, the measurement precision reaches dozens of nanometers, but the spatial resolution is low; the existing visual measurement method has the advantages of high spatial resolution, non-contact, simple operation and the like, but has limited measurement precision. Recent studies have shown that a binocular stereo vision based visual vibration measurement platform can measure 0.1mm displacement, but the measurement error reaches 0.028mm, and the diagonal of the field of view range is only 0.2 m. Meanwhile, for the two methods, since the vibration amplitude is very small, the naked eye cannot judge and select a proper measurement position, and if data of a required measurement position point, such as a maximum amplitude position, needs to be obtained, a large number of measurements need to be performed on different positions of a measured object, and obviously, the method is difficult to implement.
In summary, the existing vibration measurement technology cannot effectively solve the problem of measuring the micro-amplitude vibration. Under the circumstances, aiming at the problem of micro-amplitude vibration measurement, a measurement method with high spatial resolution, high precision and simple and convenient operation is urgently needed to meet the requirement of micro-amplitude vibration measurement.
Disclosure of Invention
In order to solve the problems, the invention provides a thin-wall part micro-amplitude vibration measurement method and system based on motion amplification optical flow tracking, which can better solve the problem of thin-wall part micro-amplitude vibration measurement and realize the accurate measurement of thin-wall part micro-amplitude vibration.
In order to achieve the purpose, the invention provides the following scheme:
a thin-wall part micro-amplitude vibration measurement method based on motion amplification optical flow tracking comprises the following steps:
acquiring a thin-wall part vibration video, and converting the thin-wall part vibration video into a thin-wall part gray level video;
processing the gray level video of the thin-wall part to determine the target vibration frequency and the frequency band bandwidth;
processing the thin-wall part gray-scale video by adopting a phase Euler amplification algorithm according to the target vibration frequency, the frequency band bandwidth and the selected amplification factor to obtain an amplified thin-wall part gray-scale video;
carrying out corner detection on a first frame of picture of the amplified thin-wall part gray-scale video, and determining a target corner and a pixel coordinate corresponding to the target corner on the first frame of picture of the thin-wall part gray-scale video;
acquiring pixel coordinates corresponding to the target corner points from a second frame picture to a last frame picture of the thin-wall part gray-scale video by adopting an optical flow tracking algorithm;
determining a vibration signal of the target corner point by taking a pixel as a unit according to all pixel coordinates corresponding to the target corner point;
establishing a measurement model; the measurement model is a model established according to a camera imaging model and an Euler amplification principle;
and compensating the vibration signals with the pixel as the unit, and converting the compensated vibration signals with the pixel as the unit according to the measurement model to obtain the actually measured vibration signals.
Optionally, the converting the thin-wall part vibration video into a thin-wall part gray scale video specifically includes:
and converting each frame of image in the thin-wall part vibration video from an RGB space to a gray space to obtain the thin-wall part gray video.
Optionally, the processing the gray scale video of the thin-wall part to determine a target vibration frequency and a frequency band bandwidth specifically includes:
performing smooth filtering processing on each frame of image of the thin-wall part gray level video by adopting a Gaussian pyramid algorithm;
performing frequency domain analysis on signals of all pixel points in the processed gray scale video of the thin-wall part on a time sequence, calculating a pixel point power spectrum, and determining a frequency value corresponding to the maximum power in the pixel point power spectrum as a pixel point frequency value;
selecting any one frame of image in the processed gray scale video of the thin-wall part, framing out a rectangular target area to be measured, and extracting frequency values corresponding to all pixel points in the rectangular target area to be measured;
according to the frequency values corresponding to all pixel points in the rectangular target area to be measured, background points are removed, the frequency with the largest number of pixel points with the same frequency is determined as a target vibration frequency according to the remaining pixel point frequency values, and meanwhile, a frequency band bandwidth is determined; the background points are pixel points lower than a set frequency.
Optionally, the measurement model is
Wherein h' is the displacement of the target corner point in the image, h is the actual displacement of the target corner point, alpha is the magnification factor, v is the distance, and u is the object distance.
Optionally, the compensating the vibration signal with the pixel as the unit specifically includes:
and compensating the vibration signal with the pixel as a unit according to the attenuation coefficient of the time domain filter in the phase Euler amplification process.
Optionally, the converting, according to the measurement model, the compensated vibration signal with the pixel as a unit to obtain an actually measured vibration signal specifically includes:
according to the formula
Calculating an actually measured vibration signal;
d is the vibration displacement of the target angular point in the compensated vibration signal with the pixel as the unit, and rho is the physical size of the pixel unit of the camera lens; h is the actual displacement of the target corner point, alpha is the magnification factor, v is the distance, u is the object distance,
representing pixels/mm.
A thin-walled workpiece micro-amplitude vibration measurement system based on motion amplification optical flow tracking comprises:
the thin-wall part gray level video conversion module is used for acquiring a thin-wall part vibration video and converting the thin-wall part vibration video into a thin-wall part gray level video;
the target vibration frequency and frequency band bandwidth determining module is used for processing the gray level video of the thin-wall part and determining the target vibration frequency and frequency band bandwidth;
the thin-wall part gray scale video amplification module is used for processing the thin-wall part gray scale video by adopting a phase Euler amplification algorithm according to the target vibration frequency, the frequency band bandwidth and the selected amplification factor to obtain an amplified thin-wall part gray scale video;
the pixel coordinate first determining module is used for carrying out corner detection on a first frame picture of the amplified thin-wall part gray-scale video, and determining a target corner and pixel coordinates corresponding to the target corner on the first frame picture of the thin-wall part gray-scale video;
the second pixel coordinate determining module is used for acquiring pixel coordinates corresponding to the target corner points from a second frame picture to a last frame picture of the thin-wall part gray-scale video by adopting an optical flow tracking algorithm;
a first vibration signal determining module, configured to determine, according to all pixel coordinates corresponding to the target corner point, a vibration signal of the target corner point in units of pixels;
the measurement model establishing module is used for establishing a measurement model; the measurement model is a model established according to a camera imaging model and an Euler amplification principle;
and the vibration signal second determining module is used for compensating the vibration signal with the pixel as the unit and converting the compensated vibration signal with the pixel as the unit according to the measurement model to obtain the actually measured vibration signal.
Optionally, the thin-wall part grayscale video conversion module specifically includes:
the acquiring unit is used for acquiring a vibration video of the thin-wall part;
and the conversion unit is used for converting each frame of image in the thin-wall part vibration video from an RGB space to a gray space to obtain the thin-wall part gray video.
Optionally, the target vibration frequency and frequency band bandwidth determining module specifically includes:
the smoothing processing unit is used for performing smoothing filtering processing on each frame of image of the gray level video of the thin-wall part by adopting a Gaussian pyramid algorithm;
the frequency value selecting unit is used for carrying out frequency domain analysis on signals of all pixel points in the processed thin-wall part gray-scale video on a time sequence, calculating a pixel point power spectrum, and determining a frequency value corresponding to the maximum power in the pixel point power spectrum as a pixel point frequency value;
the target area framing unit is used for selecting any one frame of image in the processed thin-wall part gray-scale video, framing a rectangular target area to be measured and extracting frequency values corresponding to all pixel points in the rectangular target area to be measured;
the target vibration frequency and frequency band bandwidth determining unit is used for eliminating background points according to the frequency values corresponding to all pixel points in the rectangular target area to be measured, determining the frequency with the largest number of pixel points with the same frequency as the target vibration frequency according to the remaining pixel point frequency values, and determining the frequency band bandwidth; the background points are pixel points lower than a set frequency.
Optionally, the vibration signal second determining module specifically includes:
the compensation unit is used for compensating the vibration signal with the pixel as a unit according to the attenuation coefficient of the time domain filter in the phase Euler amplification process;
a calculation unit for calculating according to a formula
Calculating an actually measured vibration signal;
d is the vibration displacement of the target angular point in the compensated vibration signal with the pixel as the unit, and rho is the physical size of the pixel unit of the camera lens; h is the actual displacement of the target corner point, alpha is the magnification factor, v is the distance, u is the object distance,
representing pixels/mm.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention provides a thin-wall part micro-amplitude vibration measuring method and system based on motion amplification optical flow tracking. The method does not affect the vibration characteristic of the thin-wall part, has flexible structure and convenient operation, solves the problem that the micro-amplitude vibration of the thin-wall part cannot be measured, and can accurately measure the micro-amplitude vibration of the thin-wall part.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a thin-walled workpiece micro-amplitude vibration measurement method based on motion amplification optical flow tracking according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a thin-walled workpiece micro-amplitude vibration measurement device based on motion amplification optical flow tracking according to an embodiment of the present invention;
FIG. 3 is a schematic end view of a thin-walled cylinder to be measured according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a thin-walled workpiece micro-amplitude vibration measurement system based on motion amplification optical flow tracking according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Example 1
As shown in fig. 1, the thin-wall part micro-amplitude vibration measurement method based on motion-amplified optical flow tracking provided by this embodiment includes the following steps:
step 101: the method comprises the steps of obtaining a thin-wall part vibration video, and converting the thin-wall part vibration video into a thin-wall part gray level video.
Step 102: and processing the gray level video of the thin-wall part to determine the target vibration frequency and the frequency band bandwidth.
Step 103: and processing the thin-wall part gray video by adopting a phase Euler amplification algorithm according to the target vibration frequency, the frequency band bandwidth and the selected amplification factor to obtain the amplified thin-wall part gray video.
Step 104: and carrying out corner detection on the first frame picture of the amplified thin-wall part gray-scale video, and determining a target corner and pixel coordinates corresponding to the target corner on the first frame picture of the thin-wall part gray-scale video.
Step 105: and acquiring pixel coordinates corresponding to the target corner points from the second frame picture to the last frame picture of the gray-scale video of the thin-wall part by adopting an optical flow tracking algorithm.
Step 106: and determining the vibration signal of the target corner point by taking the pixel as a unit according to all the pixel coordinates corresponding to the target corner point.
Step 107: establishing a measurement model; the measurement model is a model established according to a camera imaging model and an Euler amplification principle.
Step 108: and compensating the vibration signals with the pixel as the unit, and converting the compensated vibration signals with the pixel as the unit according to the measurement model to obtain the actually measured vibration signals.
and converting each frame of image in the acquired thin-wall part vibration video from an RGB space to a gray space to obtain a thin-wall part gray video.
and performing smooth filtering processing on each frame of image of the gray level video of the thin-wall part by adopting a Gaussian pyramid algorithm.
And performing frequency domain analysis on signals of all pixel points in the processed gray-scale video of the thin-wall part on the time sequence, calculating a pixel point power spectrum, and determining a frequency value corresponding to the maximum power in the pixel point power spectrum as a pixel point frequency value.
Selecting any frame of image in the processed gray-scale video of the thin-wall part, framing out a rectangular target area to be measured, and extracting frequency values corresponding to all pixel points in the rectangular target area to be measured.
According to the frequency values corresponding to all pixel points in the rectangular target area to be measured, background points are removed, the frequency with the largest number of pixel points with the same frequency is determined as a target vibration frequency according to the remaining pixel point frequency values, and meanwhile, a frequency band bandwidth is determined; the background points are pixel points lower than a set frequency.
The measurement model is
Wherein h' is the displacement of the target corner point in the image, h is the actual displacement of the target corner point, alpha is the magnification factor, v is the distance, and u is the object distance.
and compensating the vibration signal with the pixel as a unit according to the attenuation coefficient of the time domain filter in the phase Euler amplification process.
According to the formula
Calculating an actually measured vibration signal;
wherein d is meshThe vibration displacement of the marking point in the compensated vibration signal with the pixel as the unit is determined, rho is the physical size of the pixel unit of the camera lens,
representing pixels/mm.
Example 2
As shown in fig. 2-3, the thin-walled workpiece micro-amplitude vibration measuring device based on motion amplification and optical flow tracking of the invention comprises a tripod, a CCD industrial camera, a clamp, a vibration exciter and a control device thereof, an LED light source, a thin-walled cylinder to be measured, a computer and an elastic rope. One end of the thin-wall cylinder to be measured is fixed by a clamp, a plurality of light-reflecting characteristic points are arranged at the other end of the thin-wall cylinder to be measured at equal intervals, and an exciter is suspended right above the thin-wall cylinder to be measured by an elastic rope and is connected with the thin-wall cylinder to be measured through a rod. The CCD industrial camera is horizontally fixed right in front of the end face of the thin-wall cylinder to be detected through a tripod, the lens is opposite to the end face, and the CCD industrial camera is connected to the control device through a data line. The LED light sources are arranged in front of the end face of the thin-wall cylinder to be measured and are respectively arranged on the left side and the right side of the CCD industrial camera.
The thin-wall part micro-amplitude vibration measurement method based on motion amplification and optical flow tracking comprises the following steps:
the first step is as follows: and calculating the relative pose relationship between the CCD industrial camera and the thin-wall cylinder to be measured, ensuring that the lens of the CCD industrial camera is over against the end surface of the thin-wall cylinder to be measured, and measuring the object distance u.
The second step is that: the method comprises the steps of carrying out fixed-frequency excitation on a thin-wall cylinder to be detected to enable the thin-wall cylinder to be detected to generate stable vibration, collecting all sequence images within a period of time by a CCD industrial camera, transmitting the sequence images to an image processing device through an image collecting device, carrying out sequence image processing, and extracting a vibration signal of a certain light-reflecting characteristic point.
And (2.1) carrying out fixed-frequency excitation on the thin-wall cylinder to be tested by using a vibration exciter, and simulating the micro-amplitude vibration of the thin-wall cylinder to be tested under the actual condition.
And (2.2) converting the original sequence image into a gray scale image.
And (2.3) smoothing filtering each frame of image by adopting a Gaussian pyramid algorithm, wherein the smoothing filtering is to obtain a time waveform of relatively smooth pixel point brightness change in the next step, then carrying out frequency domain analysis on the time waveform to obtain power spectrum representation, and selecting the frequency corresponding to the maximum power as the pixel point frequency.
In order to avoid the influence of other vibrating objects and background noise in the images, a rectangular region ROI is manually and interactively framed on the first frame image or any one frame image of the sequence image, the frequency of pixel points in the ROI region is extracted, the frequency of the pixel points lower than 0.5HZ is regarded as background points, and the background points are removed through a threshold segmentation method.
And finally, selecting the frequency with the most pixel points as a center frequency F, and selecting a bandwidth B of 10 to obtain a frequency band [ F-5, F +5 ].
And (2.4) selecting a magnification factor alpha (generally 20-50), and performing amplification processing on the gray image sequence by adopting a phase Euler amplification method.
And (2.5) carrying out corner detection on the image, selecting a target corner at the reflecting characteristic point, tracking by adopting an optical flow tracking method, and extracting a vibration signal taking a pixel as a unit.
The third step: and compensating the measured vibration amplitude, and converting the vibration signal with the pixel as a unit into a vibration signal of an actual unit according to the established vibration measurement mathematical model.
Because the band-pass filter can attenuate signals to a certain degree in the phase Euler amplification process, the vibration signals need to be compensated, and the attenuation coefficient gamma of the band-pass filter in the time domain is usedkIt can be known that the measured vibration amplitude needs to be compensated by 5%, i.e. the final vibration amplitude is 1.05 times of the original amplitude.
(3.1) combining a camera imaging model and an Euler amplification principle, and establishing a vibration measurement mathematical model as follows:
where h' is the displacement of a point in the image, h is the actual displacement of the point, α is the magnification, v is the distance, and u is the object distance.
In combination with h' ═ d · ρ, the conversion ratio of the displacement in units of pixels to the actual displacement can be obtained as follows:
where d is the displacement of a point in pixels, ρ is the physical size of a pixel unit of the camera lens,
in units, pixels/mm. Thereby, the vibration signal in units of pixels is converted into a vibration signal in actual units.
Example 3
As shown in fig. 4, the thin-wall part micro-amplitude vibration measurement system based on motion-amplified optical flow tracking according to the present embodiment includes:
the thin-wall part gray level video conversion module 100 is used for acquiring a thin-wall part vibration video and converting the thin-wall part vibration video into a thin-wall part gray level video;
the target vibration frequency and frequency band bandwidth determining module 200 is used for processing the gray level video of the thin-wall part and determining the target vibration frequency and frequency band bandwidth;
the thin-wall part gray scale video amplification module 300 is used for processing the thin-wall part gray scale video by adopting a phase Euler amplification algorithm according to the target vibration frequency, the frequency band bandwidth and the selected amplification factor to obtain an amplified thin-wall part gray scale video;
the first pixel coordinate determining module 400 is configured to perform corner detection on a first frame of picture of the amplified thin-walled workpiece grayscale video, and determine a target corner and a pixel coordinate corresponding to the target corner on the first frame of picture of the thin-walled workpiece grayscale video;
the second pixel coordinate determining module 500 is configured to acquire pixel coordinates corresponding to the target corner point from a second frame of picture to a last frame of picture of the thin-wall part grayscale video by using an optical flow tracking algorithm;
a first vibration signal determining module 600, configured to determine, according to all pixel coordinates corresponding to the target corner point, a vibration signal of the target corner point in units of pixels;
a measurement model building module 700 for building a measurement model; the measurement model is a model established according to a camera imaging model and an Euler amplification principle;
and a vibration signal second determining module 800, configured to compensate the vibration signal in units of pixels, and convert the compensated vibration signal in units of pixels according to the measurement model to obtain an actually measured vibration signal.
The thin-wall part gray scale video conversion module 100 specifically comprises:
and the acquisition unit is used for acquiring the vibration video of the thin-wall part.
And the conversion unit is used for converting each frame of image in the thin-wall part vibration video from an RGB space to a gray space to obtain the thin-wall part gray video.
The target vibration frequency and frequency band bandwidth determining module 200 specifically includes:
and the smoothing processing unit is used for performing smoothing filtering processing on each frame of image of the gray-scale video of the thin-wall part by adopting a Gaussian pyramid algorithm.
And the frequency value selecting unit is used for carrying out frequency domain analysis on signals of all pixel points in the processed thin-wall part gray-scale video on the time sequence, calculating a pixel point power spectrum, and determining a frequency value corresponding to the maximum power in the pixel point power spectrum as a pixel point frequency value.
And the target area framing unit is used for selecting any one frame of image in the processed gray-scale video of the thin-wall part, framing out a rectangular target area to be measured, and extracting frequency values corresponding to all pixel points in the rectangular target area to be measured.
The target vibration frequency and frequency band bandwidth determining unit is used for eliminating background points according to the frequency values corresponding to all pixel points in the rectangular target area to be measured, determining the frequency with the largest number of pixel points with the same frequency as the target vibration frequency according to the remaining pixel point frequency values, and determining the frequency band bandwidth; the background points are pixel points lower than a set frequency.
The vibration signal second determining module 800 specifically includes:
and the compensation unit is used for compensating the vibration signal with the pixel as a unit according to the attenuation coefficient of the time domain filter in the phase Euler amplification process.
A calculation unit for calculating according to a formula
The actual measured vibration signal is calculated.
D is the vibration displacement of the target angular point in the compensated vibration signal with the pixel as the unit, and rho is the physical size of the pixel unit of the camera lens; h is the actual displacement of the target corner point, alpha is the magnification factor, v is the distance, u is the object distance,
representing pixels/mm.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (9)
1. A thin-wall part micro-amplitude vibration measurement method based on motion amplification optical flow tracking is characterized by comprising the following steps:
acquiring a thin-wall part vibration video, and converting the thin-wall part vibration video into a thin-wall part gray level video;
processing the gray level video of the thin-wall part to determine the target vibration frequency and the frequency band bandwidth;
processing the thin-wall part gray-scale video by adopting a phase Euler amplification algorithm according to the target vibration frequency, the frequency band bandwidth and the selected amplification factor to obtain an amplified thin-wall part gray-scale video;
carrying out corner detection on a first frame of picture of the amplified thin-wall part gray-scale video, and determining a target corner and a pixel coordinate corresponding to the target corner on the first frame of picture of the thin-wall part gray-scale video;
acquiring pixel coordinates corresponding to the target corner points from a second frame picture to a last frame picture of the thin-wall part gray-scale video by adopting an optical flow tracking algorithm;
determining a vibration signal of the target corner point by taking a pixel as a unit according to all pixel coordinates corresponding to the target corner point;
establishing a measurement model; the measurement model is a model established according to a camera imaging model and an Euler amplification principle;
and compensating the vibration signal with the pixel as the unit according to the attenuation coefficient of the time domain filter in the phase Euler amplification process, and converting the compensated vibration signal with the pixel as the unit according to the measurement model to obtain the actually measured vibration signal.
2. The thin-walled workpiece micro-amplitude vibration measurement method according to claim 1, wherein the converting of the thin-walled workpiece vibration video into a thin-walled workpiece gray scale video specifically comprises:
and converting each frame of image in the thin-wall part vibration video from an RGB space to a gray space to obtain the thin-wall part gray video.
3. The thin-walled workpiece micro-amplitude vibration measurement method according to claim 1, wherein the processing of the gray-scale video of the thin-walled workpiece to determine a target vibration frequency and a frequency band bandwidth specifically comprises:
performing smooth filtering processing on each frame of image of the thin-wall part gray level video by adopting a Gaussian pyramid algorithm;
performing frequency domain analysis on signals of all pixel points in the processed gray scale video of the thin-wall part on a time sequence, calculating a pixel point power spectrum, and determining a frequency value corresponding to the maximum power in the pixel point power spectrum as a pixel point frequency value;
selecting any one frame of image in the processed gray scale video of the thin-wall part, framing out a rectangular target area to be measured, and extracting frequency values corresponding to all pixel points in the rectangular target area to be measured;
according to the frequency values corresponding to all pixel points in the rectangular target area to be measured, background points are removed, the frequency with the largest number of pixel points with the same frequency is determined as a target vibration frequency according to the remaining pixel point frequency values, and meanwhile, a frequency band bandwidth is determined; the background points are pixel points lower than a set frequency.
4. The microform vibration measurement method of claim 1, wherein the measurement model is
Wherein h' is the displacement of the target corner point in the image, h is the actual displacement of the target corner point, alpha is the magnification factor, v is the distance, and u is the object distance.
5. The thin-wall part micro-amplitude vibration measurement method according to claim 1, wherein the step of converting the compensated vibration signal in pixel units according to the measurement model to obtain an actually measured vibration signal specifically comprises:
according to the formula
Calculating an actually measured vibration signal;
wherein d is pixel of the compensated target corner pointVibration displacement in the vibration signal is taken as a unit, and rho is the physical size of a pixel unit of the camera lens; h is the actual displacement of the target corner point, alpha is the magnification factor, v is the distance, u is the object distance,
representing pixels/mm.
6. A thin-walled workpiece micro-amplitude vibration measurement system based on motion amplification optical flow tracking is characterized by comprising:
the thin-wall part gray level video conversion module is used for acquiring a thin-wall part vibration video and converting the thin-wall part vibration video into a thin-wall part gray level video;
the target vibration frequency and frequency band bandwidth determining module is used for processing the gray level video of the thin-wall part and determining the target vibration frequency and frequency band bandwidth;
the thin-wall part gray scale video amplification module is used for processing the thin-wall part gray scale video by adopting a phase Euler amplification algorithm according to the target vibration frequency, the frequency band bandwidth and the selected amplification factor to obtain an amplified thin-wall part gray scale video;
the pixel coordinate first determining module is used for carrying out corner detection on a first frame picture of the amplified thin-wall part gray-scale video, and determining a target corner and pixel coordinates corresponding to the target corner on the first frame picture of the thin-wall part gray-scale video;
the second pixel coordinate determining module is used for acquiring pixel coordinates corresponding to the target corner points from a second frame picture to a last frame picture of the thin-wall part gray-scale video by adopting an optical flow tracking algorithm;
a first vibration signal determining module, configured to determine, according to all pixel coordinates corresponding to the target corner point, a vibration signal of the target corner point in units of pixels;
the measurement model establishing module is used for establishing a measurement model; the measurement model is a model established according to a camera imaging model and an Euler amplification principle;
and the vibration signal second determining module is used for compensating the vibration signal with the pixel as the unit according to the attenuation coefficient of the time domain filter in the phase Euler amplification process, and converting the compensated vibration signal with the pixel as the unit according to the measurement model to obtain the actually measured vibration signal.
7. The thin-walled workpiece micro-amplitude vibration measurement system according to claim 6, wherein the thin-walled workpiece gray scale video conversion module specifically comprises:
the acquiring unit is used for acquiring a vibration video of the thin-wall part;
and the conversion unit is used for converting each frame of image in the thin-wall part vibration video from an RGB space to a gray space to obtain the thin-wall part gray video.
8. The thin-walled workpiece micro-amplitude vibration measurement system according to claim 6, wherein the target vibration frequency and frequency band bandwidth determination module specifically comprises:
the smoothing processing unit is used for performing smoothing filtering processing on each frame of image of the gray level video of the thin-wall part by adopting a Gaussian pyramid algorithm;
the frequency value selecting unit is used for carrying out frequency domain analysis on signals of all pixel points in the processed thin-wall part gray-scale video on a time sequence, calculating a pixel point power spectrum, and determining a frequency value corresponding to the maximum power in the pixel point power spectrum as a pixel point frequency value;
the target area framing unit is used for selecting any one frame of image in the processed thin-wall part gray-scale video, framing a rectangular target area to be measured and extracting frequency values corresponding to all pixel points in the rectangular target area to be measured;
the target vibration frequency and frequency band bandwidth determining unit is used for eliminating background points according to the frequency values corresponding to all pixel points in the rectangular target area to be measured, determining the frequency with the largest number of pixel points with the same frequency as the target vibration frequency according to the remaining pixel point frequency values, and determining the frequency band bandwidth; the background points are pixel points lower than a set frequency.
9. The thin-walled workpiece micro-amplitude vibration measurement system according to claim 6, wherein the vibration signal second determination module specifically comprises:
the compensation unit is used for compensating the vibration signal with the pixel as a unit according to the attenuation coefficient of the time domain filter in the phase Euler amplification process;
a calculation unit for calculating according to a formula
Calculating an actually measured vibration signal;
d is the vibration displacement of the target angular point in the compensated vibration signal with the pixel as the unit, and rho is the physical size of the pixel unit of the camera lens; h is the actual displacement of the target corner point, alpha is the magnification factor, v is the distance, u is the object distance,
representing pixels/mm.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910401791.1A CN110108348B (en) | 2019-05-15 | 2019-05-15 | Thin-wall part micro-amplitude vibration measurement method and system based on motion amplification optical flow tracking |
| NL2024715A NL2024715B1 (en) | 2019-05-15 | 2020-01-21 | Method and system for measuring micro-amplitude vibration of thin-walled part based on motion amplification and optical flow tracking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910401791.1A CN110108348B (en) | 2019-05-15 | 2019-05-15 | Thin-wall part micro-amplitude vibration measurement method and system based on motion amplification optical flow tracking |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110108348A CN110108348A (en) | 2019-08-09 |
| CN110108348B true CN110108348B (en) | 2021-04-23 |
Family
ID=67490138
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910401791.1A Active CN110108348B (en) | 2019-05-15 | 2019-05-15 | Thin-wall part micro-amplitude vibration measurement method and system based on motion amplification optical flow tracking |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110108348B (en) |
| NL (1) | NL2024715B1 (en) |
Families Citing this family (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110617973B (en) * | 2019-04-26 | 2021-08-27 | 深圳市豪视智能科技有限公司 | Vibration detection method and related device |
| CN110595779B (en) * | 2019-04-26 | 2021-10-12 | 深圳市豪视智能科技有限公司 | Bearing abnormity detection method and related equipment |
| CN110631690B (en) * | 2019-08-29 | 2022-01-14 | 深圳市广宁股份有限公司 | Intelligent vibration detection method and device based on digital twins |
| CN110675427B (en) * | 2019-08-30 | 2023-11-03 | 深圳市广宁股份有限公司 | Intelligent vibration detection method and related products |
| CN110553714B (en) * | 2019-08-31 | 2022-01-14 | 深圳市广宁股份有限公司 | Intelligent vibration augmented reality testing method and related product |
| CN110674700B (en) * | 2019-08-31 | 2023-07-21 | 深圳市广宁股份有限公司 | Intelligent health prediction method and device for electronic equipment based on digital twin model |
| CN111307487A (en) * | 2020-03-12 | 2020-06-19 | 智方达(天津)科技有限公司 | Rotating mechanical vibration measurement method based on micro-motion amplification |
| CN111413064A (en) * | 2020-03-27 | 2020-07-14 | 智方达(天津)科技有限公司 | Response measurement method for aircraft model in wind tunnel |
| CN111623942B (en) * | 2020-05-26 | 2022-03-08 | 东南大学 | Displacement measurement method for test structure model of unidirectional vibration table |
| CN111854917B (en) * | 2020-06-05 | 2022-06-28 | 感知矩阵(沈阳)科技有限公司 | Non-contact torsional vibration measurement method based on machine vision |
| CN111784647A (en) * | 2020-06-19 | 2020-10-16 | 西北工业大学 | High-precision structural modal testing method based on video vibration amplification |
| CN112207039B (en) * | 2020-09-28 | 2021-08-31 | 安徽发顺钣金机械有限公司 | Adjustable wind pulling mechanism for deashing of vibrating chaff screening machine |
| CN113063484B (en) * | 2021-03-31 | 2022-10-04 | 中煤科工集团重庆研究院有限公司 | Vibration identification amplification method |
| CN113682342A (en) * | 2021-08-25 | 2021-11-23 | 深圳市永安环保实业有限公司 | Track anomaly detection method and related product |
| CN114821415B (en) * | 2022-04-24 | 2022-11-29 | 中国水利水电科学研究院 | Video micro-vibration amplification calibration and real-time extraction method based on machine learning |
| CN115797335B (en) * | 2023-01-31 | 2023-05-09 | 武汉地震工程研究院有限公司 | Euler movement amplification effect evaluation and optimization method for bridge vibration measurement |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009223113A (en) * | 2008-03-18 | 2009-10-01 | Panasonic Corp | Optical reflection element and image projection apparatus using the same |
| CN102645555A (en) * | 2012-02-22 | 2012-08-22 | 佛山科学技术学院 | Micromotion measuring method |
| CN107679290B (en) * | 2017-09-14 | 2019-03-26 | 北京空间飞行器总体设计部 | A kind of multidisciplinary micro-vibration assessment optimum design method and system |
| CN107730575A (en) * | 2017-10-18 | 2018-02-23 | 西北农林科技大学 | A kind of outdoor tree computer animation generation method based on video drive |
| CN107909575B (en) * | 2017-12-30 | 2023-09-15 | 煤炭科学研究总院唐山研究院 | Binocular vision on-line detection device and detection method for running state of vibrating screen |
| CN108593087A (en) * | 2018-03-29 | 2018-09-28 | 湖南科技大学 | A kind of thin-wall part operational modal parameter determines method and system |
| CN108731788B (en) * | 2018-05-22 | 2020-06-09 | 河海大学常州校区 | Visual detection device and method for low-frequency vibration of aerial work arm |
| CN109632085B (en) * | 2018-12-29 | 2021-04-27 | 中国计量科学研究院 | Monocular vision-based low-frequency vibration calibration method |
-
2019
- 2019-05-15 CN CN201910401791.1A patent/CN110108348B/en active Active
-
2020
- 2020-01-21 NL NL2024715A patent/NL2024715B1/en active
Non-Patent Citations (1)
| Title |
|---|
| Adaptive Eulerian Video Processing of Thermal Video: An Experimental Analysis;Stephanie Bennett, et al;《IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT》;20171031;第66卷(第10期);第2516-2524页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110108348A (en) | 2019-08-09 |
| NL2024715B1 (en) | 2020-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110108348B (en) | Thin-wall part micro-amplitude vibration measurement method and system based on motion amplification optical flow tracking | |
| CN107121093B (en) | Gear measuring device and method based on active vision | |
| CN106989812B (en) | Large fan blade modal testing method based on photogrammetry technology | |
| CN111174961B (en) | Cable force optical measurement method based on modal analysis and measurement system thereof | |
| CN106885622B (en) | A kind of big visual field multiple spot three-dimensional vibrating measurement method | |
| CN109559355B (en) | Multi-camera global calibration device and method without public view field based on camera set | |
| CN105842062A (en) | Real-time crack propagation monitoring device and real-time crack propagation monitoring method | |
| CN105953741B (en) | System and method for measuring local geometric deformation of steel structure | |
| CN110261052B (en) | Structural vibration mode analysis system and method adopting force hammer excitation and photogrammetry | |
| CN106500832A (en) | A kind of low-frequency vibration calibrating installation based on machine vision | |
| CN110533649B (en) | Unmanned aerial vehicle general structure crack identification and detection device and method | |
| CN114187330A (en) | Structural micro-amplitude vibration working mode analysis method based on optical flow method | |
| GB2616322A (en) | Computer vision-based dynamic bridge shape recognition method | |
| CN110532725B (en) | Engineering structure mechanical parameter identification method and system based on digital image | |
| CN116202423A (en) | Line laser two-dimensional positioning method based on laser triangulation ranging | |
| CN113340405B (en) | Bridge vibration mode measuring method, device and system | |
| CN112816120B (en) | Cable force measuring method | |
| CN114062265B (en) | Evaluation method for stability of support structure of vision system | |
| CN114266835A (en) | Deformation monitoring control method and system for non-measuring camera | |
| CN114219768A (en) | Method, device, equipment and medium for measuring inhaul cable force based on pixel sensor | |
| CN113899528A (en) | Liquid level dynamic wave height measuring method based on 3D-DIC principle | |
| CN110779615A (en) | Method and device for measuring object vibration by using optical principle | |
| CN113063367A (en) | Full-field deflection real-time measurement system and measurement method based on oblique optical axis digital image correlation method | |
| CN113532473B (en) | Image pickup measurement error suppression method by arranging near-field stationary points | |
| Zheng et al. | Bridge vibration virtual sensor based on eulerian video magnification and gray mean difference |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |