A flexible structure vibration detection method and related apparatus

Abstract

The application discloses a flexible structure vibration detection method and related equipment. First, the laser emitting device fixed to the flexible structure forms a vibration transmission mark point on a near-field imaging background plate, converts the far-field micro-vibration into the near-field significant spot movement, solves the problem that the micro-vibration is difficult to identify under long-distance shooting, and reduces the requirement for the shooting equipment precision. Secondly, the motion trajectories of at least three non-collinear static reference points are identified, the mark point trajectory is corrected in coordinates by using affine transformation, the shooting equipment shaking, translation and other motion interferences are compensated in real time, and the influence of the non-stationary equipment on the detection is eliminated. Furthermore, the principal component analysis is used to determine the vibration dominant direction and project to generate a one-dimensional signal, and the dominant frequency is accurately extracted in combination with the spectrum analysis. The process enhances the effective signal by dimension reduction, avoids the noise interference of the multi-dimensional trajectory, and provides a quantitative basis for the structure health diagnosis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vibration detection, more particularly, to a flexible structure vibration detection method and related equipment. BACKGROUND

[0002] In the field of engineering structure health monitoring, the vibration state of flexible structures such as bridge stay cables, building suspension cables, and power transmission lines is a core indicator for evaluating their service safety. Traditional vibration detection is usually achieved through contact methods such as pasting strain gauges and deploying acceleration sensors. Although high-precision data can be obtained, there are limitations such as complex sensor wiring, high maintenance costs, and susceptibility to environmental interference. When the structure vibration amplitude is relatively significant, non-contact identification technology based on video capture has become a promising alternative due to its convenience and low cost.

[0003] However, existing technologies face double technical bottlenecks in practical applications: first, in field operation environments such as strong winds, terrain undulations, and human disturbances, it is difficult for cameras, mobile phones, and other shooting devices to remain absolutely still. Small vibrations of the device will be recorded in the video sequence, causing the vibration trajectory of the target structure to be superimposed with the motion noise of the shooting device, making it difficult for traditional algorithms to effectively separate the real vibration signal from the device motion interference. Second, taking cable structures as an example, their inherent vibration amplitude is usually only millimeters. If the complete scene including the background reference is to be captured, the shooting distance often needs to reach tens of meters or even hundreds of meters. At this time, the slight vibration under far-field conditions will be severely weakened due to insufficient pixel resolution, and conventional shooting devices are difficult to capture effective vibration features, forming a technical contradiction between long-distance shooting and low vibration signal-to-noise ratio.

[0004] The above-mentioned defects cause the detection reliability and precision of the existing method for flexible structure vibration in complex environments to decrease significantly. Therefore, there is an urgent need for a flexible structure vibration detection scheme that can overcome the above-mentioned defects. SUMMARY

[0005] The present application provides a flexible structure vibration detection method and related equipment, which realizes the effects of eliminating shooting device motion interference, improving micro-vibration recognition, and accurately detecting the dominant frequency of flexible structure vibration by laser emission device vibration transfer, static reference point affine transformation coordinate correction, principal component analysis to determine the dominant direction and projection, and frequency spectrum analysis.

[0006] A flexible structure vibration detection method, comprising:

[0007] A dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points is collected by a camera device, and the vibration transmission marker point is formed on an imaging background plate by a laser emission device fixed to a flexible structure;

[0008] identifying a motion trajectory of the static reference point in the dynamic video sequence, and correcting the motion trajectory of the vibration transmission marker point through affine transformation according to inter-frame motion offset of the static reference point;

[0009] extracting time sequence data of the corrected motion trajectory of the vibration transmission marker point, and determining a vibration dominant direction vector through principal component analysis;

[0010] projecting the corrected motion trajectory onto the vibration dominant direction vector to generate a one-dimensional vibration signal;

[0011] performing spectral analysis on the one-dimensional vibration signal to determine a dominant frequency of the vibration of the flexible structure.

[0012] Optionally, the correcting the motion trajectory of the vibration transmission marker point through affine transformation according to the inter-frame motion offset of the static reference point comprises:

[0013] calculating an affine transformation matrix based on the inter-frame motion offset of the static reference point;

[0014] applying the affine transformation matrix to the motion trajectory of the vibration transmission marker point for coordinate reverse transformation to realize the coordinate correction.

[0015] Optionally, a calculation formula of the applying the affine transformation matrix to the motion trajectory of the vibration transmission marker point for coordinate reverse transformation is:

[0016]

[0017] a calculation formula of the affine transformation matrix is:

[0018]

[0019] wherein, is an original coordinate of the vibration transmission marker point, is a corrected coordinate of the vibration transmission marker point, M and T are a linear transformation matrix and a translation vector respectively for constituting the affine transformation matrix, is a coordinate of the i th static reference point in a reference frame, is a coordinate of the i th static reference point in a current frame.

[0020] Optionally, the laser emitting device comprises a laser emitter fixed on the flexible structure and a developing background plate arranged in a laser projection area corresponding to the laser emitter;

[0021] a surface of the developing background plate is coated with fluorescent material or high-reflective material to form a high-contrast light spot as the vibration transmission marker point through projection of the laser emitter;

[0022] The distance between the imaging background plate installation position and the laser emitter is less than the distance from the flexible structure to the camera device.

[0023] The camera device captures a dynamic video sequence of the vibration transmission marker point and at least three non-collinear static reference points pre-set on the imaging background plate.

[0024] Optionally, the determination of the vibration dominant direction vector through principal component analysis comprises:

[0025] The time sequence data of the corrected motion trajectory is subjected to centering processing, and the centroid coordinates of the trajectory point set are calculated.

[0026] The covariance matrix is calculated based on the centered trajectory data.

[0027] The target eigenvector corresponding to the maximum eigenvalue of the covariance matrix is extracted, and the target eigenvector is unitized to serve as the vibration dominant direction vector.

[0028] Optionally, the one-dimensional vibration signal is subjected to spectrum analysis to determine the dominant frequency of the flexible structure vibration, comprising:

[0029] The one-dimensional vibration signal is subjected to fast Fourier transform.

[0030] The spectrum distribution characteristics after the transformation are analyzed, and the frequency component corresponding to the maximum peak of the spectrum amplitude is identified to serve as the dominant frequency of the flexible structure vibration.

[0031] The calculation formula for the fast Fourier transform of the one-dimensional vibration signal is:

[0032]

[0033] wherein, is the fast Fourier transform result, is the frequency variable, is the amplitude of the one-dimensional vibration signal of the kth frame, is the timestamp of the kth frame, and N is the total frame number.

[0034] A flexible structure vibration detection device, comprising:

[0035] A video sequence acquisition unit is configured to acquire a dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points through a camera device, wherein the vibration transmission marker point is formed on an imaging background plate by a laser emission device fixed to a flexible structure.

[0036] a trajectory coordinate correction unit configured to identify a motion trajectory of the static reference points in the dynamic video sequence, and correct coordinates of the motion trajectory of the vibration transmission marker point through affine transformation according to inter-frame motion offset of the static reference points;

[0037] a dominant direction analysis unit configured to extract time sequence data of the corrected motion trajectory of the vibration transmission marker point, and determine a vibration dominant direction vector through principal component analysis;

[0038] a one-dimensional vibration signal unit configured to project the corrected motion trajectory to the vibration dominant direction vector to generate a one-dimensional vibration signal;

[0039] a dominant frequency determination unit configured to perform spectrum analysis on the one-dimensional vibration signal to determine a dominant frequency of the vibration of the flexible structure.

[0040] A flexible structure vibration detection device, comprising a memory and a processor;

[0041] The memory is configured to store a program.

[0042] The processor is configured to execute the program to implement each step of the flexible structure vibration detection method according to any one of the preceding embodiments.

[0043] A readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement each step of the flexible structure vibration detection method according to any one of the preceding embodiments.

[0044] A computer program product, comprising a computer program, wherein the computer program is executed by a processor to implement each step of the flexible structure vibration detection method according to any one of the preceding embodiments.

[0045] As can be seen from the above technical solutions, the flexible structure vibration detection method and related device provided by the embodiments of the present application, the method comprises the following steps: a camera device is used to collect a dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points, wherein the vibration transmission marker point is formed on an imaging background plate by a laser emitting device fixed to a flexible structure; then a motion trajectory of the static reference points is identified, and coordinates of the trajectory of the vibration transmission marker point are corrected through affine transformation according to inter-frame motion offset of the static reference points; time sequence data of the corrected trajectory is extracted, and a vibration dominant direction vector is determined through principal component analysis; the corrected trajectory is projected to the vector to generate a one-dimensional vibration signal; finally, spectrum analysis is performed on the one-dimensional signal to determine a dominant frequency.

[0046] The present application can effectively solve the defects of the prior art. Firstly, the laser emitting device is used to transfer the flexible structure vibration to the near-field display background plate to form a light spot, the vibration recognition distance is shortened, the far-field micro vibration is converted into a near-field significant light spot movement, the recognition contradiction of long distance and micro vibration is solved, and the ordinary camera equipment can capture clear vibration trajectory; secondly, the affine transformation matrix is constructed through the trajectory analysis of the motion trajectories of the three non-collinear static reference points to correct the coordinates of the vibration transmission marker point, the motion interference such as shaking and translation of the shooting device can be compensated in real time, the purity of the vibration signal in the dynamic video is ensured, and the dependence on the static state of the traditional method is broken; thirdly, the principal component analysis is used to determine the dominant direction of vibration and project to generate a one-dimensional signal, and the spectrum analysis technology is combined to accurately extract the dominant frequency of vibration, the effective signal component is strengthened through dimension reduction processing, noise interference in the multi-dimensional trajectory is avoided, and a quantitative basis is provided for structure health diagnosis. The present application not only solves the problem of trajectory distortion caused by device motion, but also improves the recognition accuracy of micro vibration through physical conversion and signal processing, and realizes efficient detection of flexible structure vibration in complex environment. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of the provided drawings.

[0048] Figure 1 A flow chart of a flexible structure vibration detection method disclosed in an embodiment of the present application;

[0049] Figure 2 A schematic diagram of a laser emitting device disclosed in an embodiment of the present application;

[0050] Figure 3 A schematic diagram of a flexible structure vibration detection device disclosed in an embodiment of the present application;

[0051] Figure 4 A hardware structure block diagram of a flexible structure vibration detection device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0052] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0053] The application is operable in a multitude of generic or specific computing device environments or configurations. For example, a personal computer, a server computer, a handheld or laptop device, a tablet device, a multiprocessor system, a distributed computing environment, etc.

[0054] Next, the scheme of the application is introduced. The application proposes the following technical scheme, which is specifically described below.

[0055] Figure 1 A flowchart of a flexible structure vibration detection method disclosed in an embodiment of the application.

[0056] As shown in the figure, the method can include: Figure 1

[0057] Step S1: acquiring, by a camera device, a dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points, the vibration transmission marker point being formed on an imaging background plate by a laser emitter fixed on a flexible structure.

[0058] As shown in the figure, the laser emitter includes a laser emitter fixed on the flexible structure and an imaging background plate arranged in a laser projection area corresponding to the laser emitter. Figure 2

[0059] The surface of the imaging background plate is coated with a fluorescent material or a highly reflective material to form a high-contrast light spot as the vibration transmission marker point by the laser emitter.

[0060] The distance between the installation position of the imaging background plate and the laser emitter is less than the distance between the flexible structure and the camera device.

[0061] The camera device captures a dynamic video sequence of the vibration transmission marker point and at least three non-collinear static reference points prearranged on the imaging background plate.

[0062] Specifically, when detecting the vibration of the flexible structure, a dynamic video containing a specific marker point is first captured by the camera device. The vibration transmission marker point is not directly captured from the flexible structure itself, but is indirectly formed by a laser emitter fixed on the flexible structure (such as a bridge cable or a building cable) to vibrate synchronously with the structure, and an imaging background plate is arranged in the projection area of the laser emitter. The surface of the background plate is coated with a fluorescent material or a highly reflective material. When the laser beam is projected onto the background plate, a clear and visible high-contrast light spot is formed, as shown in the figure. Figure 2 ​​The red dot on the light blue visual background board is the vibration transmission marker point, and its movement fully reflects the vibration of the flexible structure. In order to clearly capture the vibration by the ordinary camera equipment, the installation position of the visual background board needs to satisfy that the distance between the visual background board and the laser emitter is less than the distance between the flexible structure and the camera equipment, so as to convert the micro vibration of the flexible structure in the far field into the obvious movement of the near field light spot. The camera equipment takes the visual background board as the shooting object, and when recording the video, in addition to capturing the movement of the light spot marker point, at least three non-collinear static reference points on the background board need to be recorded, such as Figure 2 The orange points on the light blue visual background board are the reference points, which are fixed in the environment and will be used for subsequent analysis of the motion state of the camera equipment.

[0063] Step S2, identifying the motion trajectory of the static reference point in the dynamic video sequence, and correcting the motion trajectory of the vibration transmission marker point through affine transformation according to the inter-frame motion offset of the static reference point.

[0064] Specifically, in the obtained video sequence, due to the complexity of the outdoor environment, the camera equipment may vibrate, translate and move, which will affect the accurate identification of the marker point trajectory. Therefore, the information of the static reference point needs to be processed first: analyze each frame of the video, and find the positions of the three non-collinear static reference points in each frame through image recognition technology, such as the coordinates of the three points in the first frame. When the next frame, the coordinates of these points will change due to the movement of the equipment. By using these coordinate changes, a transformation matrix can be calculated, which can describe the motion of the camera equipment from one frame to the next, including scaling, rotation and translation, etc. After obtaining this matrix, it can be used to correct the coordinates of the vibration transmission marker point, and the interference caused by the movement of the equipment is removed, so that the trajectory of the marker point only reflects the real vibration of the flexible structure.

[0065] Step S3, extracting the time sequence data of the corrected motion trajectory of the vibration transmission marker point, and determining the vibration dominant direction vector through principal component analysis.

[0066] Specifically, after removing the interference of the equipment movement, the corrected motion trajectory of the marker point is obtained, that is, a series of coordinate data changing with time. These coordinate data may have certain distribution rules in the two-dimensional plane. In order to find the main direction of vibration, principal component analysis needs to be performed on these data. First, the mean value of these coordinates is calculated, and then the distribution correlation of the data in two dimensions is analyzed to find the direction that can best describe the change of the data. This direction is represented by a vector, which corresponds to the direction with the largest data variance, that is, the dominant direction of vibration, and the subsequent analysis will be carried out around this direction.

[0067] Step S4, projecting the corrected motion trajectory to the vibration dominant direction vector to generate a one-dimensional vibration signal.

[0068] Specifically, after determining the vibration dominant direction, the two-dimensional corrected motion trajectory needs to be converted into a one-dimensional signal for analysis. The projection process is to map each coordinate point along the dominant direction vector to a straight line, and the position of each point on the straight line forms a one-dimensional coordinate, thereby generating a one-dimensional vibration signal varying with time. This processing simplifies the complex planar motion into linear motion along the main direction, effectively highlights the main component of vibration, reduces the noise interference in other directions, and makes the vibration characteristics more concentrated and clear.

[0069] Step S5, performing frequency spectrum analysis on the one-dimensional vibration signal to determine the dominant frequency of the flexible structure vibration.

[0070] Specifically, after obtaining the one-dimensional vibration signal, its frequency characteristics need to be analyzed to obtain the key information of the structure vibration. Through frequency spectrum analysis technology, the vibration signal varying with time is converted into the amplitude distribution of different frequency components. In the frequency spectrum diagram, the frequency point with the maximum amplitude is the dominant frequency of the flexible structure vibration, which reflects the inherent vibration characteristics of the structure. For example, when the bridge cable is damaged, the change of its stiffness will cause the shift of the dominant frequency, and through frequency spectrum analysis, this shift can be accurately detected, providing a quantitative basis for judging the health state of the structure.

[0071] As can be seen from the above technical solutions, the method and related equipment provided by the embodiments of the present application collect a dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points through a camera device, wherein the vibration transmission marker point is formed on the imaging background plate by a laser emitting device fixed on the flexible structure; then the motion trajectory of the static reference point is identified, and the trajectory of the vibration transmission marker point is corrected through affine transformation by using the inter-frame motion offset; the time sequence data of the corrected trajectory is extracted, and the vibration dominant direction vector is determined through principal component analysis; the corrected trajectory is projected to the vector to generate a one-dimensional vibration signal; finally, the one-dimensional signal is subjected to frequency spectrum analysis to determine the dominant frequency.

[0072] The present application can effectively solve the defects of the prior art. Firstly, the flexible structure vibration is transferred to the near-field display background plate by the laser emitting device to form a light spot, the vibration recognition distance is shortened, the far-field micro vibration is converted into a near-field significant light spot movement, the recognition contradiction of long distance and micro vibration is solved, and the ordinary camera equipment can capture clear vibration trajectory; secondly, the affine transformation matrix is constructed through the motion trajectory analysis of the three non-collinear static reference points to correct the coordinates of the vibration transmission marker point, the motion interference such as shaking and translation of the shooting equipment can be compensated in real time, the purity of the vibration signal in the dynamic video is ensured, and the dependence on the static state of the equipment in the traditional method is broken; thirdly, the principal component analysis is used to determine the vibration dominant direction and project to generate a one-dimensional signal, and the vibration dominant frequency can be accurately extracted by combining the spectrum analysis technology, the effective signal component is strengthened through dimension reduction processing, noise interference in the multi-dimensional trajectory is avoided, and a quantitative basis is provided for structure health diagnosis. The present application not only solves the trajectory distortion problem caused by equipment motion, but also improves the recognition accuracy of micro vibration through physical conversion and signal processing, and realizes efficient detection of flexible structure vibration in complex environment.

[0073] In some embodiments of the present application, the process of step S2 of correcting the motion trajectory of the vibration transmission marker point through affine transformation according to the inter-frame motion offset of the static reference point is introduced, which can specifically include:

[0074] Step S21, calculating an affine transformation matrix based on the inter-frame motion offset of the static reference point;

[0075] Step S22, applying the affine transformation matrix to the coordinate reverse transformation of the motion trajectory of the vibration transmission marker point to realize coordinate correction.

[0076] Specifically, during the dynamic video sequence processing process, first, the position coordinates of the three non-collinear static reference points need to be identified frame by frame. Taking two adjacent frames as an example, it is assumed that the coordinates of the three static reference points in the first frame are (x1, y1), (x2, y2) and (x3, y3), and the corresponding coordinates in the second frame become (x1', y1'), (x2', y2') and (x3', y3'). Since these static reference points are fixed in the actual scene, the coordinate change between the video frames is completely caused by the motion of the camera equipment, such as shaking, translation, rotation or scaling of the equipment. At this time, an affine transformation matrix needs to be calculated, which is used to describe the motion state of the camera equipment from the first frame to the second frame, and contains parameters such as horizontal scaling, vertical scaling, rotation angle, horizontal translation and vertical translation. By substituting the coordinates of the static reference points in the two frames into the pre-established linear equation set, the affine transformation matrix which can accurately represent the equipment motion can be obtained through the process of solving the equation set.

[0077] Specifically, in the dynamic video sequence processing process, first, the position coordinates of three non-collinear static reference points need to be identified frame by frame. Taking two adjacent frames of images as an example, it is assumed that the coordinates of the three static reference points in the first frame are (x1, y1), (x2, y2), and (x3, y3), and the corresponding coordinates in the second frame become (x1', y1'), (x2', y2'), and (x3', y3'). Since these static reference points are fixed in the actual scene, the change of their coordinates between video frames is completely caused by the motion of the camera device, such as device jitter, translation, rotation, or scaling, etc. At this time, an affine transformation matrix needs to be calculated, which is used to describe the motion state of the camera device from the first frame to the second frame, and contains parameters such as horizontal scaling, vertical scaling, rotation angle, horizontal translation amount, and vertical translation amount. By substituting the coordinates of the static reference points in the two frames into the linear equation set established in advance, and through the process of solving the equation set, the affine transformation matrix that can accurately represent the device motion can be obtained.

[0078] The calculation formula for the coordinate reverse transformation of the motion trajectory of the vibration transmission marker point by applying the affine transformation matrix is:

[0079]

[0080] The calculation formula of the affine transformation matrix is:

[0081]

[0082] wherein, is the original coordinate of the vibration transmission marker point, is the corrected coordinate of the vibration transmission marker point, M and T are a linear transformation matrix and a translation vector, respectively, which constitute the affine transformation matrix, is the coordinate of the i-th static reference point in the reference frame, is the coordinate of the i-th static reference point in the current frame.

[0083] In some embodiments of the present application, the process of determining the vibration dominant direction vector by principal component analysis in step S3 is introduced, which can specifically include:

[0084] Step S31, centering processing is performed on the time sequence data of the corrected motion trajectory, and the centroid coordinates of the trajectory point set are calculated.

[0085] Specifically, after obtaining the corrected vibration transmission marker point motion trajectory timing data, first, the centering processing of the data set is needed. The trajectory point set is usually composed of coordinates of multiple time points, and each coordinate point contains horizontal and vertical position values. The average value of the horizontal coordinates of all trajectory points and the average value of the vertical coordinates are calculated, and the two average values together constitute the centroid coordinates of the trajectory point set. The centering processing is to subtract the horizontal average value from the horizontal coordinate of each trajectory point and subtract the vertical average value from the vertical coordinate, so that the processed data is distributed with the centroid as the origin, thereby eliminating the influence of the overall data offset and facilitating the subsequent analysis of the distribution characteristics and direction trend of the data.

[0086] Step S32, calculate the covariance matrix based on the centered trajectory data.

[0087] Specifically, after completing the centering processing of the trajectory data, the covariance matrix of the data set needs to be calculated. The covariance matrix is used to measure the correlation and dispersion degree of the centered trajectory data in the horizontal and vertical directions. For each centered trajectory point, the product of its horizontal coordinate and vertical coordinate constitutes the off-diagonal element of the covariance matrix, reflecting the correlation degree of the data in the two directions; the square mean of the horizontal coordinate and the square mean of the vertical coordinate constitute the diagonal elements of the covariance matrix, respectively, reflecting the dispersion degree of the data in the respective direction. By calculating the correlation data of all trajectory points and statistically averaging, a second-order covariance matrix can be obtained, which can comprehensively describe the distribution characteristics of the trajectory data in the two-dimensional plane.

[0088] Step S33, extract the target eigenvector corresponding to the maximum eigenvalue of the covariance matrix, and unitize the target eigenvector as the vibration dominant direction vector.

[0089] Specifically, after obtaining the covariance matrix, the eigenvalues and eigenvectors of the matrix need to be solved. The eigenvalue reflects the dispersion degree of the data in the corresponding eigenvector direction, and the larger the eigenvalue, the more dispersed the data distribution in that direction, i.e., the more significant the vibration component in that direction. Finding the maximum eigenvalue in the covariance matrix, the corresponding eigenvector is the target eigenvector, which represents the direction with the most concentrated energy and the most significant change in the vibration trajectory. In order to make the vector more standardized and comparable, it needs to be unitized, i.e., dividing each component of the eigenvector by its module length, so that its length becomes unit length. The unitized eigenvector is the vibration dominant direction vector, which is used to indicate the main motion direction of the flexible structure vibration and provides a key direction reference for subsequent signal projection and frequency analysis.

[0090] In some embodiments of the present application, the process of step S5, frequency spectrum analysis of the one-dimensional vibration signal to determine the dominant frequency of the flexible structure vibration, can specifically include:

[0091] Step S51, performing fast Fourier transform on the one-dimensional vibration signal.

[0092] Step S52, analyzing the transformed frequency spectrum distribution characteristics, and identifying the frequency component corresponding to the maximum peak value of the frequency spectrum amplitude as the dominant frequency of the flexible structure vibration.

[0093] Specifically, after obtaining the one-dimensional vibration signal projected to the dominant vibration direction, it is necessary to convert the signal from the time domain to the frequency domain for analysis. Fast Fourier transform is an efficient algorithm that can decompose the time-varying vibration signal into a superposition of sinusoidal waves of different frequencies. The one-dimensional vibration signal is composed of a series of sampling points arranged in time sequence, each sampling point representing the displacement value of the vibration in the dominant direction at that time. By calculating and processing these sampling points through the fast Fourier transform algorithm, the amplitude and phase information corresponding to each frequency component can be obtained, thereby converting the frequency characteristics that are difficult to identify directly in the time domain into the intuitive frequency distribution in the frequency spectrum, providing a data basis for subsequent determination of the dominant frequency.

[0094] The calculation formula for performing fast Fourier transform on the one-dimensional vibration signal is:

[0095]

[0096] wherein, is the fast Fourier transform result, is the frequency variable, is the amplitude of the one-dimensional vibration signal of the kth frame, is the timestamp of the kth frame, and N is the total number of frames.

[0097] After fast Fourier transform, a frequency spectrum distribution describing the correspondence between frequency and amplitude will be obtained. In the frequency spectrum graph, each frequency point corresponds to an amplitude value, and the amplitude size reflects the energy proportion of the frequency component in the vibration signal. The vibration of the flexible structure is usually composed of multiple frequency components, but the frequency component with the strongest energy corresponds to the inherent vibration characteristics of the structure, and this component appears as the peak with the largest amplitude in the frequency spectrum graph. By traversing all the frequency points in the frequency spectrum graph, the position of the peak with the largest amplitude is found, and the frequency value corresponding to this position is the dominant frequency of the flexible structure vibration. For example, when the bridge cable is in a healthy state, its dominant frequency has a specific value, and if the cable is damaged or loose, the change in structural stiffness will cause the dominant frequency to shift, and by identifying this shift through frequency spectrum analysis, the quantitative diagnosis of the structural health state can be realized.

[0098] A flexible structure vibration detection device provided by an embodiment of the present application is described below. The flexible structure vibration detection device described below can be mutually corresponding to the flexible structure vibration detection method described above.

[0099] Referring to Figure 3 , Figure 3 FIG. 1 is a schematic diagram of a flexible structure vibration detection device according to an embodiment of the present application.

[0100] As shown in Figure 3 , the flexible structure vibration detection device can include:

[0101] a video sequence acquisition unit 110 configured to acquire a dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points by a camera device, the vibration transmission marker point being formed on a display background plate by a laser emitting device fixed to a flexible structure;

[0102] a trajectory coordinate correction unit 120 configured to identify a motion trajectory of the static reference points in the dynamic video sequence, and correct a coordinate of a motion trajectory of the vibration transmission marker point by affine transformation according to an inter-frame motion offset of the static reference points;

[0103] a dominant direction analysis unit 130 configured to extract time sequence data of the corrected motion trajectory of the vibration transmission marker point, and determine a vibration dominant direction vector by principal component analysis;

[0104] a one-dimensional vibration signal unit 140 configured to project the corrected motion trajectory to the vibration dominant direction vector to generate a one-dimensional vibration signal;

[0105] a dominant frequency determination unit 150 configured to perform spectral analysis on the one-dimensional vibration signal to determine a dominant frequency of the vibration of the flexible structure.

[0106] As can be seen from the above technical solution, the flexible structure vibration detection method and related device provided by the present application acquires a dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points by a camera device, wherein the vibration transmission marker point is formed on a display background plate by a laser emitting device fixed to a flexible structure; then identifies a motion trajectory of the static reference points, and corrects a coordinate of a motion trajectory of the vibration transmission marker point by affine transformation according to an inter-frame motion offset of the static reference points; further extracts time sequence data of the corrected motion trajectory, and determines a vibration dominant direction vector by principal component analysis; projects the corrected motion trajectory to the vector to generate a one-dimensional vibration signal; and finally performs spectral analysis on the one-dimensional signal to determine a dominant frequency.

[0107] The present application can effectively solve the defects of the prior art. Firstly, the laser emitting device is used to transfer the flexible structure vibration to the near-field display background plate to form a light spot, the vibration recognition distance is shortened, the far-field micro vibration is converted into a near-field significant light spot movement, the recognition contradiction of long distance and micro vibration is solved, and the ordinary camera equipment can capture clear vibration trajectory; secondly, the affine transformation matrix is constructed through the motion trajectory analysis of the three non-collinear static reference points to correct the coordinates of the vibration transmission marker point, the motion interference such as shaking and translation of the shooting equipment can be compensated in real time, the purity of the vibration signal in the dynamic video is ensured, and the dependence of the traditional method on the static state of the equipment is broken; thirdly, the principal component analysis is used to determine the dominant direction of vibration and project to generate a one-dimensional signal, and the spectrum analysis technology is combined to accurately extract the dominant frequency of vibration, the effective signal component is strengthened through dimension reduction processing, the noise interference in the multi-dimensional trajectory is avoided, and a quantitative basis is provided for structure health diagnosis. The present application not only solves the problem of trajectory distortion caused by equipment motion, but also improves the recognition accuracy of micro vibration through physical conversion and signal processing, and realizes efficient detection of flexible structure vibration in complex environment.

[0108] Optionally, the motion trajectory of the vibration transmission marker point is corrected in coordinates by affine transformation according to the inter-frame motion offset of the static reference point, including:

[0109] An affine transformation matrix is calculated based on the inter-frame motion offset of the static reference point.

[0110] The motion trajectory of the vibration transmission marker point is reversely transformed in coordinates by the affine transformation matrix, so as to realize coordinate correction.

[0111] Optionally, the calculation formula of reversely transforming the motion trajectory of the vibration transmission marker point in coordinates by the affine transformation matrix is:

[0112]

[0113] The calculation formula of the affine transformation matrix is:

[0114]

[0115] wherein, is the original coordinate of the vibration transmission marker point, is the corrected coordinate of the vibration transmission marker point, M and T are respectively a linear transformation matrix and a translation vector for constituting the affine transformation matrix, is the coordinate of the i-th static reference point in the reference frame, is the coordinate of the i-th static reference point in the current frame.

[0116] Optionally, the laser emitting device comprises a laser emitter fixed on the flexible structure and a visualization background plate arranged in a laser projection area corresponding to the laser emitter.

[0117] The visualization background plate is coated with a fluorescent material or a high-reflective material to form a high-contrast light spot as the vibration transmission marker point by the laser emitter.

[0118] The distance between the installation position of the visualization background plate and the laser emitter is less than the distance between the flexible structure and the camera device.

[0119] The camera device captures a dynamic video sequence of the vibration transmission marker point and at least three non-collinear static reference points pre-set on the visualization background plate with the visualization background plate as the shooting object.

[0120] Optionally, the determination of the vibration dominant direction vector through principal component analysis comprises:

[0121] The time sequence data of the corrected motion trajectory is subjected to a centering process to calculate the centroid coordinates of the trajectory point set.

[0122] A covariance matrix is calculated based on the centered trajectory data.

[0123] A target eigenvector corresponding to the maximum eigenvalue of the covariance matrix is extracted, and the target eigenvector is unitized to serve as the vibration dominant direction vector.

[0124] Optionally, the one-dimensional vibration signal is subjected to a spectrum analysis to determine the dominant frequency of the flexible structure vibration, comprising:

[0125] The one-dimensional vibration signal is subjected to a fast Fourier transform.

[0126] The spectrum distribution characteristics after the transformation are analyzed to identify the frequency component corresponding to the maximum peak value of the spectrum amplitude as the dominant frequency of the flexible structure vibration.

[0127] The calculation formula for the fast Fourier transform of the one-dimensional vibration signal is:

[0128]

[0129] wherein, is the fast Fourier transform result, is a frequency variable, is the amplitude of the one-dimensional vibration signal of the kth frame, is the timestamp of the kth frame, and N is the total number of frames.

[0130] The flexible structure vibration detection device provided by the embodiments of the present application can be applied to a flexible structure vibration detection device. Figure 4A hardware structure block diagram of the flexible structure vibration detection device is shown with reference to Figure 4 The hardware structure of the flexible structure vibration detection device can include at least one processor 1, at least one communication interface 2, at least one memory 3 and at least one communication bus 4;

[0131] In the embodiments of the present application, the number of the processor 1, the communication interface 2, the memory 3 and the communication bus 4 is at least one, and the processor 1, the communication interface 2 and the memory 3 complete the communication with each other through the communication bus 4;

[0132] The processor 1 can be a central processing unit CPU, or an application specific integrated circuit ASIC, or one or more integrated circuits configured to implement the embodiments of the present application, etc.

[0133] The memory 3 can include a high-speed RAM memory, and can also include a non-volatile memory, etc., such as at least one disk memory;

[0134] The memory stores a program, and the processor can call the program stored in the memory, and the program is used for:

[0135] Collecting a dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points through the camera device, the vibration transmission marker point is formed on the imaging background plate through the laser emitting device fixed to the flexible structure;

[0136] Identifying the motion trajectory of the static reference point in the dynamic video sequence, and correcting the motion trajectory of the vibration transmission marker point through affine transformation according to the inter-frame motion offset of the static reference point;

[0137] Extracting the time sequence data of the corrected motion trajectory of the vibration transmission marker point, and determining a vibration dominant direction vector through principal component analysis;

[0138] Projecting the corrected motion trajectory to the vibration dominant direction vector to generate a one-dimensional vibration signal;

[0139] Performing spectral analysis on the one-dimensional vibration signal to determine the dominant frequency of the vibration of the flexible structure.

[0140] Optionally, the detailed functions and extended functions of the program can refer to the description above.

[0141] The embodiments of the present application also provide a readable storage medium, which can store a program suitable for the processor to execute, and the program is used for:

[0142] acquiring a dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points by a camera device, the vibration transmission marker point being formed on a display background board by a laser emitting device fixed to a flexible structure;

[0143] identifying a motion trajectory of the static reference points in the dynamic video sequence, and correcting a motion trajectory of the vibration transmission marker point by affine transformation according to inter-frame motion offset of the static reference points;

[0144] extracting time sequence data of the corrected motion trajectory of the vibration transmission marker point, and determining a vibration dominant direction vector by principal component analysis;

[0145] projecting the corrected motion trajectory to the vibration dominant direction vector to generate a one-dimensional vibration signal;

[0146] performing spectral analysis on the one-dimensional vibration signal to determine a dominant frequency of the vibration of the flexible structure.

[0147] Optionally, the refinement function and the extension function of the program can refer to the description above.

[0148] The embodiment of the application further provides a computer program product, comprising a computer program, which, when executed by a processor, performs the method.

[0149] acquiring a dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points by a camera device, the vibration transmission marker point being formed on a display background board by a laser emitting device fixed to a flexible structure;

[0150] identifying a motion trajectory of the static reference points in the dynamic video sequence, and correcting a motion trajectory of the vibration transmission marker point by affine transformation according to inter-frame motion offset of the static reference points;

[0151] extracting time sequence data of the corrected motion trajectory of the vibration transmission marker point, and determining a vibration dominant direction vector by principal component analysis;

[0152] projecting the corrected motion trajectory to the vibration dominant direction vector to generate a one-dimensional vibration signal;

[0153] performing spectral analysis on the one-dimensional vibration signal to determine a dominant frequency of the vibration of the flexible structure.

[0154] Optionally, the refinement function and the extension function of the program can refer to the description above.

[0155] Finally, it should be noted that the terms "first", "second", and the like, herein do not denote any order, quantity, combination, or importance, but rather are used to distinguish one element from another, and are not intended to denote the presence of any such actual relationship or order. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0156] The various embodiments in the specification are described with progression in this order of description. Embodiments having the same or similar descriptions are referenced by the same reference numerals, and an overlapping description is not repeated.

[0157] The above description of disclosed embodiments provides enabling disclosure sufficient for one of ordinary skill in the art to practice the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method of flexible structure vibration detection, characterized by, The method comprises the following steps: acquiring a dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points by a camera device, the vibration transmission marker point being formed on a visualization background plate by a laser emitter fixed to a flexible structure; identifying the motion trajectory of the static reference points in the dynamic video sequence, and correcting the motion trajectory of the vibration transmission marker point by affine transformation according to the inter-frame motion offset of the static reference points; extracting the time sequence data of the corrected motion trajectory of the vibration transmission marker point, and determining a vibration dominant direction vector by principal component analysis; projecting the corrected motion trajectory onto the vibration dominant direction vector to generate a one-dimensional vibration signal; performing spectral analysis on the one-dimensional vibration signal to determine the dominant frequency of the vibration of the flexible structure; the step of determining the vibration dominant direction vector by principal component analysis comprises the following steps: performing centralization processing on the time sequence data of the corrected motion trajectory, and calculating the centroid coordinates of the trajectory point set; calculating a covariance matrix based on the centralized trajectory data; extracting a target eigenvector corresponding to the maximum eigenvalue of the covariance matrix, and unitizing the target eigenvector as the vibration dominant direction vector; the step of performing spectral analysis on the one-dimensional vibration signal to determine the dominant frequency of the vibration of the flexible structure comprises the following steps: performing fast Fourier transform on the one-dimensional vibration signal; analyzing the spectral distribution characteristics after transformation, and identifying the frequency component corresponding to the maximum peak of the spectral amplitude as the dominant frequency of the vibration of the flexible structure; the calculation formula of performing fast Fourier transform on the one-dimensional vibration signal is: ； wherein, is the fast Fourier transform result, is the frequency variable, is the one-dimensional vibration signal amplitude of the kth frame, is the kth frame timestamp, N is the total number of frames, and j is the imaginary unit in the fast Fourier transform.

2. The method of claim 1, wherein, the step of correcting the motion trajectory of the vibration transmission marker point by affine transformation according to the inter-frame motion offset of the static reference points comprises the following steps: calculating an affine transformation matrix based on the inter-frame motion offset of the static reference points; applying the affine transformation matrix to the motion trajectory of the vibration transmission marker point for coordinate reverse transformation to realize coordinate correction.

3. The method of claim 2, wherein, the calculation formula of applying the affine transformation matrix to the motion trajectory of the vibration transmission marker point for coordinate reverse transformation is: ； the calculation formula of the affine transformation matrix is: ； wherein, is the original coordinate of the vibration transmission marker point, is the corrected coordinate of the vibration transmission marker point, M, T are a linear transformation matrix and a translation vector respectively constituting an affine transformation matrix, is the coordinate of the i-th static reference point in the reference frame, is the coordinate of the i-th static reference point in the current frame.

4. The method of claim 1, wherein, the laser emitter, and a visualization background plate arranged in a laser projection area corresponding to the laser emitter; the surface of the visualization background plate is coated with fluorescent material or high-reflective material to form high-contrast light spots as the vibration transmission marker points by the laser emitter; the distance between the installation position of the visualization background plate and the laser emitter is less than the distance between the flexible structure and the camera device; the camera device captures the dynamic video sequence of the vibration transmission marker points and at least three non-collinear static reference points prearranged on the visualization background plate.

5. A flexible structure vibration detection apparatus, characterized by comprising: The method comprises the following steps: a video sequence acquisition unit is configured to acquire a dynamic video sequence containing a vibration transmission marker point and at least three non-collinear static reference points by a camera device, the vibration transmission marker point being formed on a visualization background plate by a laser emitter fixed to a flexible structure; A trajectory coordinate correction unit is configured to identify a motion trajectory of the static reference point in the dynamic video sequence, and correct a motion trajectory of the vibration transmission marker point through an affine transformation according to an inter-frame motion offset of the static reference point; A dominant direction analysis unit is configured to extract time-series data of the corrected motion trajectory of the vibration transmission marker point, and determine a vibration dominant direction vector through principal component analysis; A one-dimensional vibration signal unit is configured to project the corrected motion trajectory onto the vibration dominant direction vector to generate a one-dimensional vibration signal; A dominant frequency determination unit is configured to perform spectrum analysis on the one-dimensional vibration signal to determine a dominant frequency of the flexible structure vibration; The vibration dominant direction vector determined through principal component analysis includes: Centralizing the time-series data of the corrected motion trajectory to calculate a centroid coordinate of a trajectory point set; Calculating a covariance matrix based on the centralized trajectory data; Extracting a target eigenvector corresponding to a maximum eigenvalue of the covariance matrix, and unitizing the target eigenvector as the vibration dominant direction vector; The spectrum analysis on the one-dimensional vibration signal to determine the dominant frequency of the flexible structure vibration includes: Performing fast Fourier transform on the one-dimensional vibration signal; Analyzing the spectrum distribution characteristics after the transformation to identify a frequency component corresponding to a maximum peak of a spectrum amplitude as the dominant frequency of the flexible structure vibration; The calculation formula of performing fast Fourier transform on the one-dimensional vibration signal is: ； wherein, is the fast Fourier transform result, is the frequency variable, is the one-dimensional vibration signal amplitude of the kth frame, is the kth frame timestamp, N is the total number of frames, and j is the imaginary unit in the fast Fourier transform.

6. A flexible structure vibration detection apparatus characterized by comprising: A computer device includes a memory and a processor; The memory is configured to store a program; The processor is configured to execute the program to implement each step of the flexible structure vibration detection method according to any one of claims 1-4.

7. A readable storage medium, having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement each step of the flexible structure vibration detection method according to any one of claims 1-4.

8. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement each step of the flexible structure vibration detection method according to any one of claims 1-4.

Images