CN116363121A - Computer vision-based inhaul cable force detection method, system and device - Google Patents

Abstract

The present invention discloses a method, system and device for detecting cable force based on computer vision. The method includes: acquiring relevant information of the cable to be detected; Preprocessing is carried out to obtain the target image; the phase-based Euler motion amplification algorithm is used to obtain the enlarged image sequence for the target image, and the enlarged video is reconstructed; the sub-pixel template matching algorithm is used to determine the video after the motion amplification The corresponding displacement time-history response data; based on the displacement time-history response data, the fundamental frequency corresponding to the cable to be detected is obtained, and according to the actual boundary conditions of the cable, an appropriate cable force-frequency relationship is selected to calculate the cable force. It can be seen that the present application realizes the detection of the cable force with a very small amplitude of the bridge cable, and does not require additional sensors, which reduces the cost and has broad application prospects.

Description

A computer vision-based cable force detection method, system and device

technical field

The invention relates to the technical field of bridge health monitoring, in particular to a computer vision-based cable force detection method, system and device.

Background technique

Cable-stayed bridges are widely used in long-span bridges due to their advantages such as strong spanning ability, beautiful appearance, and high cost performance. A cable-stayed bridge is a statically indeterminate structural system that is subjected to a combination of main girders in bending, bridge towers in pressure, and cables in tension. The cables of cable-stayed bridges are often prone to damage and relaxation due to corrosion, fatigue and other reasons. As important stress-bearing components of cable-stayed bridges, cable-stayed cables often produce large displacements due to small stress and strain changes. And lead to its relaxation and stress loss, and the damage of the cable may bring disastrous consequences to the overall structure. It is because of these characteristics that the cable force detection of cable-stayed bridges is of great significance in the construction and use stages of the structure, and the safety and durability evaluation of the bridge structure needs to take the cable force as an important measurement and evaluation index, and then provide Provides basis for bridge condition assessment. Therefore, how to quickly and accurately measure and evaluate the cable force of the bridge is very important, which is the prerequisite for ensuring the safe operation of the bridge and a powerful guarantee for real-time monitoring of the health of the bridge.

At present, the most commonly used cable force detection methods mainly include oil pressure gauge measurement method, pressure sensor measurement method, frequency method, magnetic flux method, resistance sheet measurement method, sag method, elongation method, etc. The oil pressure gauge measurement method, the pressure sensor measurement method and the magnetic flux method are not suitable for the cable force detection in the operation stage. The frequency method usually uses contact sensors such as acceleration sensors to pick up the vibration signals of the cables, and the measurement results are reliable, but there are disadvantages such as difficult instrument installation, high cost, and the introduction of additional mass. In recent years, with the continuous development of computer vision technology and image acquisition equipment, structural displacement monitoring methods based on computer vision have emerged and have been verified in practical engineering applications. At present, the cable force recognition method based on computer vision mainly uses the target tracking algorithm to track the artificial markers or natural markers on the surface of the structure to obtain the displacement time-history response data. However, the vibration amplitude of the cable under environmental excitation is small. It is difficult for the target tracking algorithm to obtain high-precision cable displacement time history data, which affects the cable force detection results; and when extracting cable vibration displacement, the result obtained by digital image correlation template matching is pixel-level displacement, but in practical applications Higher measurement accuracy is required in , so further optimization search is required to reach the sub-pixel level. Therefore, it is necessary to study how to apply high-precision and high-speed photogrammetry technology and equipment to the traditional bridge engineering field. For this purpose, a phase-based Euler motion amplification algorithm and sub-pixel template matching are proposed to establish a set of non-contact A convenient, economical and efficient cable force detection method is of great significance for ensuring the safe operation of the bridge and realizing the long-term stable use of the bridge.

Contents of the invention

Purpose of the invention: Aiming at the deficiencies in the background technology, the present invention proposes a computer vision-based cable force detection method, system and device to realize rapid and accurate detection of the cable force.

Technical solution: In order to achieve the purpose of the present invention, the present invention proposes a computer vision-based cable force detection method, system and device, and its specific scheme is as follows:

The first aspect: the application discloses a computer vision-based cable force detection method, which specifically includes the following steps:

Step 1: Obtain the relevant information of the cable to be detected;

Step 2: Collect the micro-vibration video of the cable to be detected, and preprocess the frame-by-frame image of the video to obtain the target image;

Step 3: using a phase-based Euler motion magnification algorithm for the target image to obtain an amplified image sequence, and reconstructing the amplified video;

Step 4: Using a sub-pixel template matching algorithm to determine the displacement time-history response data corresponding to the video after the motion amplification;

Step 5: Obtain the fundamental frequency corresponding to the cable to be detected based on the displacement time-history response data, and select an appropriate cable force-frequency relationship to calculate the cable force according to the actual boundary conditions of the cable;

Further, in the step 1, the relevant information includes the model, material, diameter, calculated length, linear density, inclination angle, boundary condition, use time, serial number or position of the cable, etc.

Further, the specific process of the step 2 is as follows:

Step 2.1: Select 1/4 of the cable as the detection area, observe whether there are natural markers here on the cable, if not, you need to set the target point here; fix the high-precision industrial camera in a suitable position, and inspect the site According to the lighting conditions, judge whether supplementary light is needed. If supplementary light is required, turn on the supplementary light with its own power supply for lighting; adjust the position of the camera to ensure that the cable target can be completely captured by the camera and will not exceed the limit when vibrating. Video range, adjust the focal length and aperture of the lens to ensure that the cable image in the camera field of view is clearly visible; set the camera sampling frequency to collect tiny vibration videos of the cable;

Step 2.2: Disassemble the micro-vibration video of the cable captured by the camera into continuous frame images, and set the same region of interest for each frame of image; the region of interest includes the target point and the movement area of the cable to be detected;

Step 2.3: Perform frame-by-frame image preprocessing on the continuous frame images obtained by dismantling, including cropping, rotating and scaling the target image sequence, and removing larger noise in the target image to highlight the concerned cable structure and reduce other environmental factors.

Further, the specific method of the step 3 is as follows:

Step 3.1. Spatial domain filtering: use the complex controllable pyramid to perform spatial domain filtering on the input target image, and obtain high-pass residuals, low-pass residuals, and image sequences of different scales and different directions, that is, local amplitude spectrum and local phase spectrum;

Step 3.2, time-domain filtering: calculate the phase difference according to the obtained local phase spectrum, manually set the frequency range and filter, and then time-domain band-pass filtering to extract the phase difference signal in the frequency band of interest;

Step 3.3, amplifying the motion signal: manually setting the amplification factor, linearly amplifying the phase difference of the motion signal of interest extracted by time-domain filtering, to obtain the amplified motion signal;

Step 3.4, video reconstruction and synthesis: use the image sequence synthesized by the amplified motion signal, the high-pass residual and low-pass residual reconstruction obtained by spatial domain filtering to obtain the digital image data after motion amplification, and finally synthesize the amplified video.

Further, in the step 4, the sub-pixel template matching algorithm is used to determine the displacement time-history response data corresponding to the motion-amplified video, including:

The enlarged video is divided into frames to obtain the enlarged video image, using the zero-mean normalized cross-correlation theory in statistics, and the sub-pixel template matching algorithm is used to calculate the first frame of the enlarged video image and other The sub-pixel displacement between the frame-enlarged video images, and then concatenating the image processing results of each frame with time as the baseline, can obtain the vibration displacement time-history response data of the cable of the target object.

Further, in the step 5, based on the displacement time-history response data, the fundamental frequency corresponding to the cable to be detected is obtained, and according to the actual boundary conditions of the cable, an appropriate cable force-frequency relationship is selected to calculate the cable force, including:

Using the obtained displacement time-history response data to obtain the fundamental frequency corresponding to the cable to be detected in the micro-vibration video by using Fast Fourier Transform;

The practical formulas for calculating the cable force by the vibration method are all simplified and derived from various theoretical models, and have different scopes of application. It is necessary to select the appropriate cable force-frequency relationship to calculate the cable force according to the actual situation of the cable.

Second aspect: the application discloses a computer vision-based cable force detection system, including:

Cable-related information acquisition module, used to obtain relevant information about the cable to be detected;

An image acquisition module, configured to acquire a micro-vibration video of the cable to be detected, and acquire a target image based on the micro-vibration video;

The image preprocessing module is used to preprocess the target image, including cropping, rotating and scaling the target image, and removing larger noise in the target image, so as to highlight the structure of the cable concerned and reduce other environmental factors Impact;

The video acquisition module after motion amplification is used to perform space-domain filtering on the input target image by using complex adjustable pyramids to obtain image sequences in different directions and scales, and artificially set the frequency range to perform time-domain filtering on the image sequences to extract the motion signal of interest Phase difference information, the phase difference is amplified through the preset amplification factor to obtain an enlarged image sequence, and then reconstructed through the pyramid inverse process to obtain an enlarged video;

A displacement time-history response data determination module, configured to determine the displacement time-history response data corresponding to the motion-amplified video by using a sub-pixel template matching algorithm;

The cable force calculation module is used to obtain the fundamental frequency corresponding to the cable to be detected based on the displacement time-history response data, and calculate the cable force by selecting an appropriate cable force-frequency relationship according to the actual boundary conditions of the cable.

The third aspect: the present application discloses a cable force detection device based on computer vision, which is characterized in that it includes: a memory and a processor, the memory is used to store a computer program executable by the processor, and the processing When the device is used to execute the computer program, the steps of the computer vision-based cable force detection method of the present invention are realized.

A fourth aspect: the present application discloses a computer-readable storage medium, which is characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, a computer program described in the present invention is implemented. A computer vision-based cable force detection method for the steps of the cable.

Beneficial effects: Compared with the prior art, the present invention has the following beneficial effects:

The computer vision-based cable force detection method, system and device of the present invention overcome the disadvantages of difficult sensor installation, high equipment cost and low test efficiency in the traditional cable force test frequency method. Obtain the relevant information of the cable to be detected; collect the micro-vibration video of the cable to be detected, and preprocess the frame-by-frame image of the video to obtain the target image; use the phase-based Euler motion magnification algorithm to obtain the magnification of the target image After the image sequence is reconstructed, the enlarged video is reconstructed; the sub-pixel template matching algorithm is used to determine the displacement time-history response data corresponding to the video after the motion amplification; based on the displacement time-history response data, the corresponding According to the fundamental frequency of the cable, according to the actual boundary conditions of the cable, select the appropriate cable force-frequency relationship to calculate the cable force, which provides a new technical approach for cable force testing of cable-stayed bridges. It realizes the cable force detection with very small amplitude of the bridge cable, and does not need additional sensors, which reduces the cost. It has the advantages of low cost, convenience, easy operation, high scalability and accurate cable force detection, etc., and has a wide range of applications prospect.

Description of drawings

Fig. 1 is a flow chart of a computer vision-based cable force detection method provided by the application;

Fig. 2 is a schematic diagram of a cable force detection based on computer vision provided by the present application.

FIG. 3 is a schematic structural diagram of a computer vision-based cable force detection system provided by the present application.

Detailed ways

The present invention will be described in further detail below through specific embodiments and in conjunction with the accompanying drawings.

A kind of cable force detection method based on computer vision of the present invention, with reference to Fig. 1, specifically comprises the following steps:

S1: Obtain relevant information about the cable to be detected;

S2: Collect the micro-vibration video of the cable to be detected, and preprocess the video frame by frame to obtain the target image;

S3: using a phase-based Euler motion magnification algorithm on the target image to obtain an amplified image sequence, and reconstructing an amplified video;

S4: Using a sub-pixel template matching algorithm to determine the displacement time-history response data corresponding to the video after the motion amplification;

S5: Obtain the fundamental frequency corresponding to the cable to be detected based on the displacement time-history response data, and calculate the cable force by selecting an appropriate cable force-frequency relationship according to the actual boundary conditions of the cable.

Further, in the step S1, the relevant information includes the type, material, diameter, calculated length, line density, inclination angle, boundary condition, use time, serial number or position of the cable, etc. Obtaining the relevant information of the cable is the basic information for calculating the cable force later.

Further, the specific process of the step S2 is as follows:

S2.1: Select 1/4 of the cable as the detection area, observe whether there is a natural marker here on the cable, if not, you need to set a target point here; fix the high-precision industrial camera at a suitable position, and inspect According to the on-site lighting conditions, judge whether supplementary light is needed. If supplementary light is required, turn on the supplementary light with its own power supply for lighting; adjust the position of the camera to ensure that the cable target can be completely captured by the camera and will not vibrate. Beyond the video range, adjust the lens focal length and aperture to ensure that the cable image is clearly visible in the camera field of view; set the camera sampling frequency to collect tiny vibration videos of the cable;

S2.2: Disassemble the micro-vibration video of the cable captured by the camera into continuous frame images, and set the same region of interest for each frame of image; the region of interest includes the target point and the movement area of the cable to be detected;

S2.3: Perform frame-by-frame image preprocessing on the disassembled continuous frame images, including compressing digital images, cropping, rotating and scaling the target video image sequence, and removing larger noise in the target image to highlight the concerns The cable structure reduces the influence of other environmental factors.

Specifically, compressing digital images means that in order to reduce the calculation amount of computer processing images, the three primary color information is often converted into gray value information according to a certain rule, that is, the original image data is gray transformed, and the gray image data can be obtained. Significantly reduce the amount of computation. The calculation formula for converting the three primary color images into grayscale images according to different weight coefficients is:

H(x,y)＝0.299R(x,y)+0.587G(x,y)+0.114B(x,y)

Further, the specific method of the step S3 is as follows:

S3.1. Spatial domain filtering: use complex controllable pyramids to perform spatial domain filtering on the input target image, and obtain high-pass residuals, low-pass residuals, and image sequences of different scales and directions, that is, local amplitude spectrum and local phase spectrum; Among them, the complex steerable pyramid is a multi-directional, multi-scale, multi-resolution image decomposition algorithm with self-transformation ability, which has the advantages of directional steerability and translation invariance, and its basis function is similar to a Gaussian function envelope Formed sine function. It is worth noting that the high-pass residual and low-pass residual output in this step are not subjected to time-domain filtering and motion signal amplification processing, but are directly used for pyramid reconstruction of image data after motion amplification.

S3.2. Time-domain filtering: Calculate the phase difference according to the obtained local phase spectrum, manually set the frequency range and select the filter, and then time-domain band-pass filter to extract the phase difference signal in the frequency band of interest;

S3.3. Amplify the motion signal: manually set the amplification factor α, and linearly amplify the phase difference of the motion signal of interest extracted by time-domain filtering to obtain the amplified motion signal;

S3.4. Video reconstruction and synthesis: use the image sequence synthesized by the amplified motion signal, the high-pass residual and low-pass residual reconstruction obtained by spatial domain filtering to obtain the digital image data after motion amplification, and finally synthesize the amplified video .

Specifically, motion amplification is to amplify the small motions existing in the video sequence or image sequence, so that the motion range of these small motions is increased so as to extract valuable information from these small motions. The phase-based Euler motion amplification algorithm is proposed based on the Fourier shift theorem. In a two-dimensional image, the phase corresponds to the motion of the object, and the image undergoes Fourier transform to process the contained phase to achieve object motion amplification. This method directly operates on the phase information contained in the image. In terms of noise processing, it only translates the noise without amplifying the noise, reduces the appearance of motion artifacts, and supports a larger magnification.

In order to illustrate the principle of the algorithm intuitively, a one-dimensional image is taken as an example for illustration. Assume that f(x) is a one-dimensional image brightness function, and x is the image pixel coordinates. After time t, assuming that the object translates δ(t) on the image, the image brightness at time t is f(x+δ(t)) . Perform Fourier transform on f(x) and f(x+δ(t)) respectively to get

Where ω is the harmonic frequency, and A _ω is the harmonic amplitude.

Corresponding to a certain harmonic frequency ω, the phase difference of the harmonic components of f(x) and f(x+δ(t)) is

Obviously, the phase difference is directly related to the signal δ(t), which contains motion information.

If the phase is amplified by α times, the corresponding harmonic component is adjusted to, and the brightness function of the reconstructed image is

By comparing the image brightness functions f(x) and f(x+δ(t)), the amplified motion signal (1+αδ(t)) can be obtained. It is worth noting that in order to amplify motion signals in a certain frequency range, δ(t) is actually time-domain filtered.

Further, in the step 4, the sub-pixel template matching algorithm is used to determine the displacement time-history response data corresponding to the motion-amplified video, including:

The enlarged video is divided into frames to obtain the enlarged video image, using the zero-mean normalized cross-correlation theory in statistics, and the sub-pixel template matching algorithm is used to calculate the first frame of the enlarged video image and other The sub-pixel displacement between the frame-enlarged video images, and then concatenating the image processing results of each frame with time as the baseline, can obtain the vibration displacement time-history response data of the cable of the target object. And taking a single stay cable as an example, its vibration in space mainly occurs in three directions: the chord direction in the vertical plane, the vertical chord direction in the vertical plane, and the direction out of the plane. Generally, the vibration amplitude of the cable along the chord direction is much smaller than that of the other two directions. Considering that the aerodynamic damping of bridge stay cables in the vertical chord direction is about half that in the transverse direction, it is considered that the main vibration direction of bridge stay cables is the vertical chord direction in the vertical plane. Therefore, the present invention mainly studies the vibration characteristics of the stay cable in the vertical plane and the vertical chord direction by using the computer vision method.

Specifically, the sub-pixel template matching calculation displacement process is as follows: Suppose there are two images f(x,y) and h(x,y) with the same size (M×N), where h(x,y) is the same as the reference image f(x,y) has relative translation, and the cross-correlation relationship between f(x,y) and h(x,y) after Fourier transform can be defined as:

In the formula: M and N are the dimensions of the image; (x ₀ , y ₀ ) is the amount of coordinate shift; "*" indicates complex conjugate; F(u,v) and H ^* (u,v) respectively indicate f Discrete Fourier Transform of (x,y) and h(x,y).

The expression of F(u,v):

Pixel-level displacements of structural vibrations are extracted by peaks of R _hf . Then, a time-sensitive matrix multiplication-based discrete Fourier transform cross-correlation is performed to extract sub-pixel-level displacements of structural vibrations in the region around the initial peak of R _hf .

In particular, it should be pointed out that only frequency information is needed to estimate the cable force. In other words, there is no need to determine a scale factor to convert the pixel coordinate vibration displacement into the physical coordinate vibration displacement, which will make the vision-based measurement process more efficient and practical.

Further, in the step 5, based on the displacement time-history response data, the fundamental frequency corresponding to the cable to be detected is obtained, and according to the actual boundary conditions of the cable, an appropriate cable force-frequency relationship is selected to calculate the cable force, including:

The obtained displacement time-history response data is obtained by fast Fourier transform to obtain the fundamental frequency corresponding to the cable to be detected in the micro-vibration video; the first-order vibration frequency of the cable is the fundamental frequency, which is the At the same time, engineers who test the cable force on site are used to using the fundamental frequency to calculate the cable force, so the present invention calculates the cable force by obtaining an accurate fundamental frequency.

The practical formulas for calculating the cable force by the vibration method are all simplified and derived from various theoretical models, and have different scopes of application. It is necessary to select the appropriate cable force-frequency relationship to calculate the cable force according to the actual situation of the cable. At present, the cable force calculation models based on vibration are mainly divided into four categories: tension string model theory, simply supported beam model theory, fixed beam theory, and complex boundary model theory.

Tensioned string model theory: The tensioned string theory simplifies the stay cable as a tensioned string with no stiffness and no sag, and ignores the bending stiffness.

F＝4ml ² f ²

Simply supported beam model theory: The simply supported beam model is a simplified horizontal beam in which the stay cables are hinged at one end and simply supported at one end.

The theory of fixed beam model: The fixed beam model adopts fixed support constraints at both ends. Compared with the simply supported beam model, the boundary conditions are different. Considering the influence of bending stiffness, the calculation formula is as follows:

F＝4ml ² f ² (210≤ξ)

Considering the effect of sag, the calculation formula is as follows:

F＝4ml ² f ² (λ ² ≤0.17,4π ² ≤λ ² )

Complex boundary model theory: In the actual application process, the boundary conditions of the stay cables are relatively complex, which may be between simple support and fixed support, and there is an elastic boundary.

λ²为垂度的无量纲量，λ²＝[mgl/F]·(EAl/FL_e)；L_e＝l[(1+(mglcosθ/F²/8))]；

Among them, F is the mass of the unit length of the cable in m; l is the length of the cable; f is the first-order vibration frequency of the cable, that is, the fundamental frequency; EI represents the bending stiffness of the cable; EA is the axis of the cable direction stiffness; ξ is the dimensionless quantity of bending stiffness,

λ ² is the dimensionless quantity of sag, λ ² =[mgl/F]·(EAl/FL _e ); L _e =l[(1+(mglcosθ/F ² /8))];

Obtain the relevant information of the cable to be detected; collect the micro-vibration video of the cable to be detected, and preprocess the frame-by-frame image of the video to obtain the target image; use the phase-based Euler motion magnification algorithm to obtain the magnification of the target image After the image sequence is reconstructed, the enlarged video is reconstructed; the sub-pixel template matching algorithm is used to determine the displacement time-history response data corresponding to the video after the motion amplification; based on the displacement time-history response data, the corresponding According to the fundamental frequency of the cable, according to the actual boundary conditions of the cable, select the appropriate cable force-frequency relationship to calculate the cable force, which provides a new technical approach for cable force testing of cable-stayed bridges.

A kind of cable force detection system based on computer vision of the present invention, with reference to Fig. 2, comprises:

Cable-related information acquisition module, used to obtain relevant information about the cable to be detected;

An image acquisition module, configured to acquire a micro-vibration video of the cable to be detected, and acquire a target image based on the micro-vibration video;

The image preprocessing module is used to preprocess the target image, including cropping, rotating and scaling the target image, and removing larger noise in the target image, so as to highlight the structure of the cable concerned and reduce other environmental factors Impact;

The video acquisition module after motion amplification is used to perform space-domain filtering on the input target image by using complex adjustable pyramids to obtain image sequences in different directions and scales, and artificially set the frequency range to perform time-domain filtering on the image sequences to extract the motion signal of interest Phase difference information, the phase difference is amplified through the preset amplification factor to obtain an enlarged image sequence, and then reconstructed through the pyramid inverse process to obtain an enlarged video;

A displacement time-history response data determination module, configured to determine the displacement time-history response data corresponding to the motion-amplified video by using a sub-pixel template matching algorithm;

The cable force calculation module is used to obtain the fundamental frequency corresponding to the cable to be detected based on the displacement time-history response data, and calculate the cable force by selecting an appropriate cable force-frequency relationship according to the actual boundary conditions of the cable.

It should be noted that, for more specific working processes of the above-mentioned modules, reference may be made to the corresponding contents disclosed in the foregoing embodiments, and details are not repeated here.

The present application discloses a cable force detection device based on computer vision, which is characterized in that it includes: a memory and a processor, the memory is used to store a computer program executable by the processor, and the processor is used to execute the The computer program is used to realize the steps of a computer vision-based cable force detection method according to the present invention.

The present application discloses a computer-readable storage medium, which is characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, a computer vision-based The steps of the cable force detection method of the present invention.

Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the requirements of the present invention and equivalent technologies, the present invention also intends to include these modifications and variations.

Claims (9)

1. The computer vision-based inhaul cable force detection method is characterized by comprising the following steps of:

step 1: acquiring relevant information of a cable to be detected;

step 2: collecting a tiny vibration video of a cable to be detected, and preprocessing a frame-by-frame image of the video to obtain a target image;

step 3: acquiring an amplified image sequence from the target image by adopting a Euler motion amplification algorithm based on phase, and reconstructing to obtain an amplified video;

step 4: determining displacement time-course response data corresponding to the video after the motion amplification by using a sub-pixel template matching algorithm;

step 5: and acquiring the fundamental frequency corresponding to the cable to be detected based on the displacement time response data, and selecting a proper cable force-frequency relation to calculate the cable force of the cable according to the actual boundary condition of the cable.

2. The computer vision-based inhaul cable force detection method according to claim 1, wherein the method comprises the steps of: in the step 1, the related information includes a model, a material, a diameter, a calculated length, a linear density, an inclination angle, a boundary condition, a service time, a number or a position of the inhaul cable, and the like.

3. The computer vision-based inhaul cable force detection method according to claim 1, wherein the method comprises the steps of: the specific process of the step 2 is as follows:

step 2.1: selecting 1/4 of the inhaul cable as a detection area, observing whether natural markers exist in the inhaul cable or not, and setting a target point in the inhaul cable if the natural markers do not exist in the inhaul cable; fixing the high-precision industrial camera at a proper position, inspecting the field illumination condition, judging whether light supplementing is needed, and if the light supplementing is needed, turning on a self-powered light supplementing lamp to light; the position of a camera is adjusted, so that a guy cable target can be completely shot by the camera, the guy cable target can not exceed a video range during vibration, the focal length and the aperture of a lens are adjusted, and the guy cable in the field of view of the camera is enabled to be imaged clearly and visually; setting a sampling frequency of a camera, and collecting a guy cable micro-vibration video;

step 2.2: disassembling a guy cable micro-vibration video shot by a camera into continuous frame images, and setting the same region of interest for each frame image; the region of interest comprises a target spot and a inhaul cable movement region to be detected;

step 2.3: and carrying out frame-by-frame image preprocessing on the continuous frame images obtained by disassembly, wherein the preprocessing comprises cutting, rotating and scaling the target image sequence, and removing larger noise in the target image so as to highlight the inhaul cable structure concerned and reduce the influence of other environmental factors.

4. The computer vision-based inhaul cable force detection method according to claim 1, wherein the method comprises the steps of: the specific method of the step 3 is as follows:

step 3.1, spatial domain filtering: carrying out spatial domain filtering on an input target image by using a complex controllable pyramid to obtain a high-pass residual error, a low-pass residual error and image sequences with different scales and different directions, namely a local amplitude spectrum and a local phase spectrum;

step 3.2, time domain filtering: calculating phase difference according to the obtained local phase spectrum, manually setting a frequency range and a filter, and then performing time domain band-pass filtering to extract phase difference signals in the interested frequency band;

step 3.3, amplifying the motion signal: manually setting an amplification coefficient, and linearly amplifying the phase difference of the interesting motion signal extracted by the time filtering to obtain an amplified motion signal;

step 3.4, video reconstruction and synthesis: and reconstructing the high-pass residual error and the low-pass residual error obtained by utilizing the image sequence synthesized by the amplified motion signals and the spatial domain filtering to obtain the image sequence amplified by the motion, and finally synthesizing the amplified video.

5. The computer vision-based inhaul cable force detection method according to claim 1, wherein the method comprises the steps of: in the step 4, a sub-pixel template matching algorithm is used to determine displacement time-course response data corresponding to the video after the motion amplification, including:

and carrying out framing treatment on the amplified video to obtain an amplified image sequence, respectively calculating sub-pixel displacement between the amplified target image of the first frame and the amplified target images of other frames by using a sub-pixel template matching algorithm by using a zero mean value normalization cross-correlation theory in statistics, and then connecting the frame image treatment results in series by taking time as a base line to obtain vibration displacement time-course response data of the inhaul cable of the target object.

6. The computer vision-based inhaul cable force detection method according to claim 1, wherein the method comprises the steps of: in the step 5, the fundamental frequency corresponding to the cable to be detected is obtained based on the displacement time response data, and a proper cable force-frequency relation is selected to calculate the cable force according to the actual boundary condition of the cable, including:

acquiring the fundamental frequency corresponding to the inhaul cable to be detected in the micro vibration video by utilizing the obtained displacement time response data through fast Fourier transform;

the practical formulas for calculating the cable force by the vibration method are simplified and deduced by various theoretical models, have different application ranges, and need to select proper cable force-frequency relation according to the actual condition of the cable to calculate the cable force of the cable.

7. A computer vision-based guy cable force detection system, comprising:

the cable related information acquisition module is used for acquiring related information of the cable to be detected;

the image acquisition module is used for acquiring a micro vibration video of the inhaul cable to be detected and acquiring a target image based on the micro vibration video;

the image preprocessing module is used for preprocessing the target image, including cutting, rotating and zooming the target image, and removing larger noise in the target image so as to highlight a cable structure concerned and reduce the influence of other environmental factors;

the video acquisition module after the motion amplification is used for carrying out spatial domain filtering on an input target image by using a complex adjustable pyramid so as to acquire image sequences with different dimensions in different directions, carrying out time domain filtering on the image sequences by manually setting a frequency range to extract phase difference information of a motion signal of interest, amplifying the phase difference by a preset amplification coefficient so as to obtain an amplified image sequence, and then reconstructing the amplified image sequence by using a pyramid inverse process so as to obtain an amplified video;

the displacement time-course response data determining module is used for determining displacement time-course response data corresponding to the video after the motion amplification by utilizing a sub-pixel template matching algorithm;

and the cable force calculation module is used for acquiring the fundamental frequency corresponding to the cable to be detected based on the displacement time-course response data, and selecting a proper cable force-frequency relation to calculate the cable force of the cable according to the actual boundary condition of the cable.

8. Computer vision-based inhaul cable force detection device is characterized by comprising: a memory for storing a computer program executable by the processor, and a processor for implementing the steps of a computer vision-based cable force detection method according to any one of claims 1 to 6 when the computer program is executed.

9. A computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the steps of a computer vision-based cable force detection method according to any one of claims 1-6 are implemented.

Images