A computer vision-based bridge cable force measurement method and system

Abstract

The application discloses a bridge cable force measuring method and system based on computer vision, and the method comprises the following steps: collecting a vibration video of a bridge cable containing a target point; tracking the target point in the vibration video through a preset tracking algorithm to obtain motion time history data of the cable; processing the motion time history data through an EEMD method to obtain a plurality of intrinsic mode functions corresponding to the motion time history data; removing the intrinsic mode functions irrelevant to the vibration of the cable according to the vibration frequency characteristics of the cable, and reconstructing signals of the remaining intrinsic mode functions to obtain reconstructed motion time history data; determining the vibration frequency of the cable based on the reconstructed motion time history data; and calculating the cable force of the cable through a preset frequency method according to the vibration frequency. Thus, the physical contact requirement for equipment in the traditional measuring method is avoided, remote operation is realized, the safety measures and equipment investment required for on-site operation are reduced, and the labor cost and equipment maintenance cost are reduced.

Description

Technical Field

[0001] This application relates to the field of bridge health monitoring technology, and in particular to a method and system for measuring bridge cable tension based on computer vision. Background Technology

[0002] Driven by rapid socio-economic development and technological advancements, the field of health monitoring for transportation infrastructure is booming at an unprecedented pace and scale. Cable-stayed bridges, in particular, with their superior spanning capacity, aesthetically pleasing design, and cost-effectiveness, have become the preferred structure for long-span bridge construction, often serving as crucial transportation hubs. In these cable-stayed bridges, the stay cables play a vital role, acting not only as the key "link" connecting the main girder to the towers but also as the core components bearing the main force of the entire structure. The stay cables are responsible for effectively transferring the weight and live load of the main girder to the towers, generating necessary pressure, and significantly enhancing the axial crack resistance of the main girder by applying lateral preload. Furthermore, the cable tension is a key parameter controlling the condition of the cable-stayed bridge, directly affecting the internal force distribution and geometric shape of the structure.

[0003] Currently, there are various methods for monitoring the force of stay cables, including pressure sensor methods, frequency methods, magnetic flux methods, and fiber optic grating force measurement methods. Among these, the frequency method is particularly widely used in practice due to its flexible equipment selection and accurate measurement results. In actual engineering operations, the natural frequency of stay cables can be obtained through equipment such as vibration pickups, acoustic emission sensors, or laser vibrometers. However, most of these methods are contact measurements, which have problems such as difficult installation and high costs. Summary of the Invention

[0004] This application provides a computer vision-based method and system for measuring bridge cable tension, which avoids the physical contact requirements of equipment in traditional measurement methods, enables remote operation, reduces the safety measures and equipment investment required for on-site operations, and lowers labor and equipment maintenance costs.

[0005] To achieve the above objectives, the technical solution of this invention is as follows:

[0006] In a first aspect, embodiments of the present invention provide a computer vision-based method for measuring bridge cable force, comprising: acquiring a vibration video of a bridge cable containing a target point; tracking the target point in the vibration video using a preset tracking algorithm to obtain motion time history data of the cable; processing the motion time history data using the EEMD method to obtain multiple corresponding intrinsic mode functions; eliminating intrinsic mode functions unrelated to cable vibration based on the vibration frequency characteristics of the cable, and reconstructing the remaining intrinsic mode functions to obtain reconstructed motion time history data; determining the vibration frequency of the cable based on the reconstructed motion time history data; and calculating the cable force of the cable using a preset frequency method based on the vibration frequency.

[0007] In some possible implementations, before tracking the target point in the vibration video using a preset tracking algorithm to obtain the motion time history data of the cable, the method further includes: performing frame-by-frame image processing on the vibration video to obtain a continuous image sequence; extracting and amplifying the motion signal from the image sequence based on the phase Euler motion amplification algorithm, and reconstructing the digital image data to obtain the amplified motion video.

[0008] In some possible implementations, frame-by-frame image processing is performed on the vibration video to obtain a continuous image sequence, including: converting the vibration video to a new format, cropping the image, and decomposing it to obtain a continuous image sequence; and removing environmental noise from each image in the image sequence using grayscale averaging and probability statistics methods.

[0009] In some possible implementations, a phase-Eulerian motion amplification algorithm is used to extract and amplify motion signals from an image sequence and reconstruct digital image data to obtain an amplified motion video. This includes: spatial domain filtering, which uses a complex-adjustable pyramid to perform spatial domain filtering on the input image sequence to obtain high-pass residuals, low-pass residuals, and image sequences at different scales and orientations; temporal domain filtering, which manually sets the frequency range and performs temporal bandpass filtering on the image sequence to extract the phase difference information of the motion signal of interest; motion signal amplification, which manually sets the amplification factor to linearly amplify the phase difference of the motion signal of interest extracted by temporal filtering to obtain an amplified motion signal; and digital image data reconstruction, which uses the image sequence synthesized from the amplified motion signal, the high-pass residuals and low-pass residuals obtained from spatial domain filtering to reconstruct the motion-amplified digital image data, thus obtaining the amplified motion video.

[0010] In some possible implementations, a pre-defined tracking algorithm is used to track target points in vibration videos to obtain the time-history data of the cable's motion. This includes: determining the position and size of the target point based on the initial frame image information in the vibration video and drawing candidate boxes; extracting HOG features from the candidate box region, generating a training sample set by cyclically shifting the feature data, training the labels of the training samples based on the shift distance and Gaussian function of the training sample set, and training the kernel correlation filter model using the ridge regression method; reading the next frame image, extracting image features from the search region, generating a candidate sample set based on the extracted features, and using the updated filter to perform correlation filtering on the candidate sample set to obtain a correlation response map; determining the position with the largest response value as the optimal prediction position, and updating the target position and candidate box in the next frame based on the peak point of the response map; adding bilinear interpolation to the target model update to accommodate changes in the appearance model; after the model update is complete, using the updated filter classifier template and the search region of subsequent frame images to perform response calculations and predict the target position in subsequent frames until all video sequences are detected; calculating the centroid coordinates of the target point in each frame image to obtain the time-history data of the cable's motion.

[0011] In some possible implementations, the vibration frequency characteristics of the cable are determined based on relevant information about the cable, including at least: cable type, material, diameter, calculated length, linear density, inclination angle, boundary conditions, service time, serial number, and / or installation location.

[0012] In some possible implementations, the vibration frequency of the cable is determined based on the reconstructed motion time history data, including: performing a fast Fourier transform on the reconstructed motion time history data to obtain the corresponding spectral information; and extracting spectral peaks from the spectral information using the direct peak method to obtain the natural frequencies of the cable.

[0013] Secondly, embodiments of the present invention provide a bridge cable force measurement system based on computer vision, comprising: an image acquisition module for acquiring vibration videos of bridge cables containing target points; a target point tracking module for tracking target points in the vibration videos using a preset tracking algorithm to obtain motion time history data of the cables; a first processing module for processing the motion time history data using the EEMD method to obtain multiple corresponding intrinsic mode functions; a second processing module for eliminating intrinsic mode functions unrelated to cable vibration based on the vibration frequency characteristics of the cables, and reconstructing the remaining intrinsic mode functions to obtain reconstructed motion time history data; a first determination module for determining the vibration frequency of the cables based on the reconstructed motion time history data; and a second determination module for calculating the cable force of the cables using a preset frequency method based on the vibration frequency.

[0014] Thirdly, embodiments of the present invention provide an electronic device, including: a memory for storing executable instructions; and a processor for implementing the method provided in the first aspect of the present invention when executing the executable instructions or computer program stored in the memory.

[0015] Fourthly, the present invention provides a computer-readable storage medium storing executable instructions for implementing the method provided in the first aspect of the present invention when executed by a processor.

[0016] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:

[0017] In this embodiment of the invention, vibration videos of bridge cables containing target points are acquired; the target points in the vibration videos are tracked using a preset tracking algorithm to obtain the motion time history data of the cables; the motion time history data is processed using the EEMD method to obtain multiple corresponding intrinsic mode functions (IMFs); based on the vibration frequency characteristics of the cables, IMFs unrelated to cable vibration are eliminated, and the remaining IMFs are reconstructed to obtain reconstructed motion time history data; the vibration frequency of the cables is determined based on the reconstructed motion time history data; and the cable force is calculated using a preset frequency method based on the vibration frequency. This avoids the physical contact requirements of traditional measurement methods, reduces potential human error and equipment wear, and enables remote operation, greatly improving the safety of the measurement process. Furthermore, the convenience and efficiency of this method are significant advantages, enabling the acquisition and analysis of large amounts of data in a short time, providing fast and reliable cable force information for bridge maintenance and management. It is suitable for non-contact measurement of bridge cable forces, establishing a non-contact, remote, convenient, and efficient cable force measurement method. Attached Figure Description

[0018] To more clearly illustrate the embodiments of the present invention, the accompanying drawings used in the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A schematic flowchart illustrating an embodiment of a computer vision-based bridge cable force measurement method provided for the implementation of this invention;

[0020] Figure 2 This is a photographic illustration of the cable in an embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram of the signal amplification structure using the Euler motion amplification algorithm in an embodiment of the present invention;

[0022] Figure 4 This is a time history diagram of cable displacement in an embodiment of the present invention;

[0023] Figure 5 The spectrum is the result of direct fast Fourier transform in existing technologies;

[0024] Figure 6 The spectrum obtained by the method provided in the embodiments of the present invention;

[0025] Figure 7 This is a schematic diagram of a bridge cable force measurement system based on computer vision, as described in an embodiment of the present invention.

[0026] Figure 8 This is a schematic diagram of an electronic device structure according to an embodiment of the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0028] In the relevant descriptions of this embodiment, the terms "including," "containing," and "possessing" are all open terms and are generally understood to include but not be limited to; the term "at least one" is generally understood to mean one or more, where "multiple" refers to two or more; the term "at least one of the following" or similar expressions refer to any combination of these items, including any combination of single or multiple items, for example, "at least one of a, b, or c", or "at least one of a, b, and c", which can all mean: a, b, c, ab (i.e., a and b), ac, bc, or abc, where a, b, and c can be single or multiple; the symbol "A / B" is used to describe the selection relationship of associated objects, generally indicating an "or" relationship.

[0029] In the following description of the embodiments, the terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms "a" and "the" as used in the embodiments of this application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0030] Those skilled in the art should understand that, in the following description of the embodiments of this application, the sequence of numbers does not imply the order of execution. Some or all steps may be executed in parallel or sequentially. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0031] Those skilled in the art will understand that the numerical ranges in the embodiments of this application should be understood to specifically disclose each intermediate value between the upper and lower limits of the range. Any stated value or intermediate value within a stated range, as well as any other stated value or each smaller range between intermediate values ​​within a range, are also included within this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0032] Unless otherwise stated, the technical / scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. While this application describes only preferred methods and materials, any methods and materials similar or equivalent to those described herein may be used in the implementation or testing of this application. All references to this specification are incorporated by way of citation to disclose and describe the methods and / or materials associated with those references. In the event of any conflict with any incorporated reference, the content of this specification shall prevail.

[0033] To illustrate the technical solution of the present invention, specific embodiments are described below.

[0034] Driven by rapid socio-economic development and technological advancements, the field of health monitoring for transportation infrastructure is booming at an unprecedented pace and scale. Cable-stayed bridges, in particular, with their superior spanning capacity, aesthetically pleasing design, and cost-effectiveness, have become the preferred structure for long-span bridge construction, often serving as crucial transportation hubs. In these cable-stayed bridges, the stay cables play a vital role, acting not only as the key "link" connecting the main girder to the towers but also as the core components bearing the main force of the entire structure. The stay cables are responsible for effectively transferring the weight and live load of the main girder to the towers, generating necessary pressure, and significantly enhancing the axial crack resistance of the main girder by applying lateral preload. Furthermore, the cable tension is a key parameter controlling the condition of the cable-stayed bridge, directly affecting the internal force distribution and geometric shape of the structure.

[0035] Currently, there are various methods for monitoring the force of stay cables, including pressure sensor methods, frequency methods, magnetic flux methods, and fiber optic grating force measurement methods. Among these, the frequency method is particularly widely used in practice due to its flexible equipment selection and accurate measurement results. In actual engineering operations, the natural frequency of stay cables can be obtained through equipment such as vibration pickups, acoustic emission sensors, or laser vibrometers. However, most of these methods are contact measurements, which have problems such as difficult installation and high costs.

[0036] There are various methods for monitoring cable tension in cable stays. The commonly used cable tension measurement methods in current engineering projects typically include: hydraulic jack reading method, pressure sensor testing method, magnetic flux method, and cable vibration frequency method.

[0037] Hydraulic gauge method: Suitable for measuring cable force during construction. Force is applied by jacks and the reading from a hydraulic gauge is converted into cable force. It is easy to operate, but the equipment is heavy, the accuracy is affected by operation, and it requires regular calibration.

[0038] Pressure sensor method: Used during construction, this method determines cable force by measuring strain through sensors installed at the anchoring end. It offers high accuracy and simple operation, but is costly, requires calibration, and is not suitable for widespread use.

[0039] Vibration frequency method: Applicable to both construction and operation phases, it uses vibration sensors to collect signals and process them to obtain the fundamental frequency, thereby calculating cable force. It is a mature and highly accurate technology, but installation is difficult and costly. Cable-stayed bridges typically use dense cable systems to increase span and reduce costs. Contact measurement methods usually employ wired accelerometers for rapid measurements of a few cables during construction or for temporary testing during operation. However, this method involves complex wiring, numerous power connection issues, and the cables may disrupt traffic. Wireless accelerometers improve measurement flexibility but are susceptible to environmental signal interference. Limitations of the accelerometer measurement method include: difficulty in installing in the middle of the cables, impact on cable force and natural frequency, and difficulty in long-term stable operation in harsh external environments.

[0040] Magnetic flux method: Applicable to the construction and operation phases, it measures cable force by utilizing the change in magnetic flux after a magnetic material is subjected to force. It allows for non-contact measurement, but the equipment is expensive, easily affected by interference from other metallic materials, and requires improved measurement accuracy.

[0041] Given this, the limitations of traditional contact vibration measurement methods in terms of measurement point layout, measurement efficiency, and long-term stability have prompted a shift in vibration measurement technology from contact to non-contact methods. Advances in computer vision technology have made this shift possible, enabling intelligent, rapid, and low-cost remote identification and detection, thus gaining increasing importance in the field of health monitoring of civil engineering structures. In practical applications, vision technology is not only used to detect damage to structural surfaces but is also widely applied to deformation or vibration measurement of various structures such as buildings, bridges, and cable-stayed structures. Among these, computer vision-based non-contact vibration measurement methods (referred to as "visual vibration measurement methods") have attracted widespread attention and are developing rapidly in the field of dynamic displacement measurement of civil engineering structures. This method involves capturing video of a vibrating object from a distance and analyzing the images using computer vision target detection and tracking algorithms or digital image processing technology to extract the object's motion information. Finally, based on the imaging geometric model, the vibration displacement time history curve of the object is calculated. Compared to traditional accelerometer measurement methods, vibration measurement technology based on computer vision not only has the advantages of non-contact, long-distance multi-point measurement, wide measurement range, and high efficiency, but also requires simple equipment, including only a camera, tripod, telephoto lens, and computer, making it less expensive and more flexible in measurement methods.

[0042] Based on this, embodiments of the present invention provide a reliability optimization method for offshore wind power systems to solve the problem of low efficiency in reliability optimization in the prior art.

[0043] Figure 1 A schematic flowchart illustrating an embodiment of a computer vision-based bridge cable force measurement method provided for implementation of the present invention is shown below. Figure 1 As shown, the above-mentioned computer vision-based bridge cable force measurement method may include:

[0044] S101, Collect vibration video of bridge cables containing the target point;

[0045] The target point can be a location of particular interest on the bridge cables, such as the part of the cable experiencing the greatest vibration amplitude, or the area that has the greatest impact on the overall structural stability. After identifying the target point, video recording can be carried out in a targeted manner.

[0046] For example, Figure 2 This is a schematic diagram of the cable in an embodiment of the present invention. See [link / reference]. Figure 2 As shown, it includes a target point (i.e. Figure 2 (Measurement point 2 in the middle).

[0047] In some embodiments, waterproof stickers with a significant color difference from the background can be affixed to suitable locations on the cable, or clear markings on the cable can be used as target points. A rigid tripod combined with a controllable gimbal is used to stabilize the camera. The camera is then positioned appropriately to capture video. The tripod should be stable and have a high load-bearing capacity to ensure the camera does not shake or shift during video capture. The gimbal needs to be controllable, allowing for easy adjustment of the camera's angle and field of view, ensuring the cable components remain within the camera's field of view at all times.

[0048] S102, through a preset tracking algorithm, tracks the target point in the vibration video to obtain the motion time history data of the cable;

[0049] In some embodiments, before performing step S102, the method may further include the following steps:

[0050] S21, perform frame-by-frame image processing on the vibration video to obtain a continuous image sequence;

[0051] S22, based on the phase Euler motion amplification algorithm, extracts and amplifies motion signals from image sequences, and reconstructs digital image data to obtain amplified motion video.

[0052] Understandably, many of the minute or imperceptible vibrations of the cable are not visible to the naked eye, but these changes may contain important information or indications. By amplifying these motion signals, the vibration patterns can be observed and analyzed more clearly. Therefore, after acquiring the aforementioned vibration video, the vibration signals of the cable contained within the video can be amplified.

[0053] In some embodiments, step S21 above may include:

[0054] S211, converts the vibration video in format, crops and decomposes the image to obtain a continuous image sequence;

[0055] S212 removes environmental noise from each image in the image sequence using the gray-scale average method and probability statistics method.

[0056] Specifically, the acquired video of cable vibration is transmitted to a computer, and the MATLAB image processing toolbox is used to perform operations such as image cropping and segment extraction to remove images containing significant noise from the digital image data. Then, the output video is decomposed into a continuous image sequence, and environmental noise is removed using grayscale averaging and probability statistics methods.

[0057] The gray-scale averaging method includes: calculating the gray-scale values ​​of each target area and averaging them, denoted as . Next, calculate the mean gray value of the M regions, denoted as . by The grayscale value benchmark represents the structure. Grayscale value tolerance coefficient is used. A grayscale value range that accurately represents the structure is determined. Pixels within the grayscale value range retain their original grayscale values, while pixels outside the range have their grayscale values ​​replaced with zero. The corresponding theoretical calculations can be achieved using the following formulas (1) to (3):

[0058]

[0059]

[0060]

[0061] Where m and n are the height and width of the image, respectively, H k (i,j) represents the original pixel grayscale value.

[0062] The probability and statistics method includes: assuming that the gray values ​​of the cable structure constitute the population N, and that N follows a normal distribution N(μ, σx) 2 Select M representative cable target areas from the image, and statistically analyze the grayscale values ​​of the target areas as sample X, with the sample mean being... Then the sample mean It is an unbiased estimator of the population mean μ. The confidence interval at a confidence level of α can be calculated using the following formulas (4) to (6):

[0063]

[0064]

[0065]

[0066] Thus, through the above formulas (4) to (6), the confidence interval that can characterize the gray value of the cable structure can be obtained, which can be expressed by the following formula (7):

[0067]

[0068] The grayscale value of each pixel in the entire field of view is retained and replaced using the aforementioned confidence interval. When the pixel grayscale value is within the confidence interval, the original grayscale value is retained. When the pixel grayscale value is outside the confidence interval, the grayscale value of the pixel is replaced with zero.

[0069] In some embodiments, step S22 above may include the following steps:

[0070] S221, Spatial Domain Filtering, uses a complex adjustable pyramid to perform spatial domain filtering on the input image sequence to obtain high-pass residual, low-pass residual, and image sequences of different scales and directions (i.e., amplitude spectrum and phase spectrum).

[0071] S222, time-domain filtering, involves manually setting a frequency range and performing time-domain bandpass filtering on an image sequence to extract the phase difference information of the motion signal of interest;

[0072] S223, motion signal amplification, the amplification factor is manually set, and the phase difference of the motion signal of interest extracted by time-domain filtering is linearly amplified to obtain the amplified motion signal;

[0073] S224, Digital Image Data Reconstruction, uses the image sequence synthesized from the amplified motion signal, the high-pass residual and low-pass residual obtained by spatial domain filtering to reconstruct the amplified digital image data and obtain the amplified motion video.

[0074] See Figure 3 As shown, Figure 3 This is a schematic diagram of the signal amplification structure using the Euler motion amplification algorithm in an embodiment of the present invention.

[0075] In step S221, the complex tunable pyramid is an overcomplete transform that decomposes a digital image according to its spatial scale and orientation. Its basis function is similar to a sine function composed of a Gaussian function envelope. The high-pass and low-pass residuals output from this step are not subjected to temporal filtering or motion signal amplification; instead, they are directly used to reconstruct the motion-amplified digital image data using the pyramid.

[0076] In S222, the image sequence obtained from the complex tunable pyramid decomposition contains motion signals of different frequencies. Since the motion signal of interest is often within a certain frequency range, the phase difference information of the motion signal of interest can be extracted by manually setting the frequency range and applying bandpass filtering to the image sequence. Specifically, to ensure that the amplitude and phase of the motion signal of interest are not distorted while suppressing non-interesting motion signals, an ideal filter can be used in the time domain bandpass filter.

[0077] In some embodiments, after steps S21 to S22 described above, the vibration signal of the cable can be magnified and highlighted. Then, based on the magnified motion video, a tracking algorithm can be used to track the target point and obtain the time history data of the cable's motion.

[0078] Understandably, cable motion time history data refers to the continuous record of the vibration, displacement, or other motion characteristics of bridge cables over a certain period of time, reflecting the motion state and changing patterns of the cables at different moments. Analyzing this data allows us to understand the cable's vibration characteristics, frequency, amplitude, and other key parameters, thereby assessing the overall stability and safety of the bridge. Furthermore, this data can be used to verify the correctness of bridge design theories, optimize bridge structural design schemes, and improve the bridge's load-bearing capacity and service life. However, cable motion time history data can be affected by various factors, such as environmental noise, sensor accuracy, and measurement errors. Therefore, a series of measures are needed to ensure the accuracy and reliability of this data during collection and processing.

[0079] For example, Figure 4 This is a cable displacement time history diagram according to an embodiment of the present invention. In some embodiments, the above step S102 of calculating the cable displacement time history data may include:

[0080] S1021, Based on the initial frame image information in the vibration video, determine the position and size of the target point and draw the candidate box;

[0081] S1022, extract HOG features of candidate bounding box regions, generate training sample set X by cyclically shifting the feature data, and train the labels (i.e., expected outputs) of the samples based on the shift distance of training sample set X and the Gaussian function. i Furthermore, the kernel correlation filter model was trained using the ridge regression method.

[0082] The goal of ridge regression is to find a function f(x) i )=ω T x i , so that the sampled data f(x) i ) and the actual target position y in the next frame i Minimize the distance.

[0083] The ridge regression objective function can be expressed by the following formula (8):

[0084]

[0085] Where, x i For the sample, y i The target being regressed is defined by a Gaussian function. λ is the regularization coefficient, used to prevent overfitting and improve the generalization ability of the fitted function. ω is the regression coefficient.

[0086] To improve the classification performance of the filter tracker, a mapping φ can be used to map sample x. i Mapping to a high-dimensional space, the regression coefficient ω can be expressed as a linear combination of the mapped samples, and ω can be represented by the following formula (9):

[0087]

[0088] Where, α i These are the combination coefficients.

[0089] Substitute formula (9) into f(x) i )=ω T x i The following formula (10) can be obtained from this:

[0090]

[0091] Where α is α i A linear combination of x, and introduce a kernel function k(x) i ,x j )=φ T (x i )φ(x j ).

[0092] The solution to α can be expressed by the following formula (11):

[0093] α=(K+λI) -1 y (11)

[0094] Where K is the kernel autocorrelation matrix of the sample, i.e., K ij =(x i ,x j ).

[0095] Since K is a cyclic matrix, the properties of cyclic matrices are used to organize the filter tracker. It can be expressed by the following formula (12):

[0096]

[0097] Where, the symbol ^ represents the DFT transformation of the variable; y represents the Gaussian regression label y i The vector formed, k xx The first row of the kernel autocorrelation matrix of sample x; This is a filter tracker.

[0098] S1023, read the next frame image, extract the image features of the search area, generate a candidate sample set based on the extracted features, and use the updated filter to perform correlation filtering on the candidate sample set to obtain the correlation response map f(z); f(z) can be expressed by the following formula (13):

[0099]

[0100] Where, the symbol ^ represents the DFT transform of the variable; F-1 k is the inverse Fourier transform matrix. xz Refers to the kernel matrix K z The first line; f(z) is the set of response values ​​for the test sample set z.

[0101] S1024, determine the position with the largest response value as the best prediction position, and update the position of the target and the candidate box in the next frame according to the peak point of the response map;

[0102] S1025, In response to changes in the appearance model, the bilinear interpolation method is added to the update of the target model; the filter coefficients α and the target observation model x can be expressed by the following formulas (14) to (15):

[0103]

[0104]

[0105] Where 0 < γ < 1 is the learning factor, and t is the image frame number. It should be noted that the updated model is used in the process of detecting the target in the (t+1)th frame image.

[0106] S1026 After the model is updated, the updated filter classifier template and the search region of the subsequent frame image are used to perform response calculations to predict the target position of the subsequent frame. Repeat the above operations from S1021 to S1026 until all video sequences are detected.

[0107] S1027, calculate the centroid coordinates of the target point in each frame of the image to obtain the motion time history data of the cable.

[0108] Specifically, let the coordinates of the top-left corner of the candidate box in the initial frame be (x... l0 ,y l0 If the height of the candidate box is h0 and the width is w0, then the center coordinates (x0, y0) of the candidate box can be expressed by the following formula (16):

[0109]

[0110] Let the coordinates of the top-left corner of the candidate box in frame t be (x... lt ,y lt The height of the candidate box is h. t Width is w t Then the center coordinates of the candidate box (x t ,y t ) can be expressed by the following formula (17):

[0111]

[0112] Then the displacement Δx of the cable in frame tt It can be expressed by the following formula (18):

[0113] Δx t =[(x0-x t ) 2 +(y0-y t ) 2 ] 1 / 2 (18)

[0114] The displacement-time history curve x(t) of the cable is obtained, with time on the horizontal axis and displacement on the vertical axis.

[0115] S103, the motion time history data is processed by the EEMD method to obtain the corresponding multiple intrinsic mode functions;

[0116] It should be noted that the essence of the Ensemble Empirical Mode Decomposition (EEMD) method is a multiple empirical mode decomposition with superimposed Gaussian white noise. It utilizes the statistical characteristic that Gaussian white noise has a uniform frequency distribution. By adding different white noises of the same amplitude each time, the extreme point characteristics of the signal are changed. Then, the corresponding Intrinsic Mode Functions (IMFs) obtained from multiple EMDs are averaged to cancel out the added white noise, thereby effectively suppressing the generation of mode mixing.

[0117] In some embodiments, EEMD decomposition may include the following steps:

[0118] (1) Add N times Gaussian white noise n(t) with zero mean and constant standard deviation to the signal x(t) to be decomposed to obtain a new population x. i (t) can be expressed by the following formula (19):

[0119] x i (t)=x(t)+n i (t) (19)

[0120] Where i = 1 to N, i represents the i-th additive white noise sequence, and represents the additional noise signal of the i-th experiment;

[0121] (2) For x i (t) Perform EMD decomposition to obtain K IMF components and a remainder r. i (t) can be expressed by the following formula (20):

[0122]

[0123] Where: c ij(t) represents the j-th IMF component obtained after the i-th addition of white noise, where j = 1 to K.

[0124] (3) Based on the characteristic that the mean of an irrelevant random sequence is zero, the IMF components obtained in the second step are averaged to eliminate the influence of multiple additions of white noise on the decomposition results. The IMF components and remainders obtained from the EEMD decomposition can be expressed by the following formulas (21) to (22):

[0125]

[0126]

[0127] Where: c j (t) represents the j-th IMF component obtained by EEMD decomposition of the original signal.

[0128] The original signal is finally decomposed into the following two parts, represented by the following formula (23):

[0129]

[0130] Two important parameters in EEMD decomposition are the number of times white noise is added, N, and its standard deviation σ. N is generally taken between 100 and 300, and σ is generally taken between 0.01 and 0.5.

[0131] S104. Based on the vibration frequency characteristics of the cable, the intrinsic mode functions that are unrelated to the cable vibration are eliminated, and the remaining intrinsic mode functions are reconstructed to obtain the reconstructed motion time history data.

[0132] Understandably, considering that the vibration frequency of bridge cables has a specific range, factors such as bridge deck vibration and vibration of other cables may cause non-cable vibration components to be mixed into the collected data. For example, during bridge deck photography, any vibration of the bridge deck will reduce the stability of the camera image, and the vibration of the cables is affected by bridge deck vibration and vibration of other cables. These vibrations will mix with the cable vibration, thereby reducing the accuracy and reliability of the detection. Alternatively, when there are multiple cables, the vibration of multiple cables will also have a certain influence on each other, resulting in vibrations of a certain frequency. Therefore, in this embodiment of the invention, these IMFs can be screened according to the vibration frequency characteristics of the cables to remove those components unrelated to cable vibration.

[0133] Specifically, this embodiment of the invention analyzes the frequency and characteristics of each IMF (Integrated Motion Frame) and compares them with the frequency characteristics of the cable's own vibration. IMFs whose frequencies or characteristics do not match the cable's vibration frequency are considered irrelevant to the cable's vibration and are discarded. After this discarding, the remaining IMFs will primarily contain components related to the cable's vibration. Next, these remaining IMFs are reconstructed to obtain reconstructed motion time history data. This reconstruction process aims to recombine the filtered components related to cable vibration to reflect the cable's own vibration characteristics as accurately as possible. Thus, the resulting reconstructed motion time history data will be cleaner, primarily containing the cable's own vibration information. This helps to more accurately analyze the cable's vibration behavior, thereby improving the accuracy of its force calculation.

[0134] In some embodiments, the vibration frequency characteristics of the cable are determined based on relevant information about the cable; such relevant information may include at least: cable type, material, diameter, calculated length, linear density, tilt angle, boundary conditions, service time, serial number, and / or installation location.

[0135] Based on the information obtained above, the vibration frequency characteristics of the corresponding cable can be calculated, thereby enabling the aforementioned optimization of the motion time history data.

[0136] Understandably, the IMF components obtained from EEMD decomposition are arranged in descending order of frequency. Therefore, the noise component in the signal is usually decomposed into the high-frequency IMF components. Due to interpolation errors and boundary effects, there will be some spurious components in the IMF components, and the frequencies of these spurious components may coincide with the frequencies of the real signal. The principle of EEMD denoising is to remove the high-frequency noise components, spurious components, and residual terms from the original signal, and then reconstruct the remaining IMF components to obtain the denoised real signal.

[0137] The key to EEMD noise reduction is selecting appropriate IMF components to ensure optimal noise reduction without losing useful components. As the autocorrelation function is known, the autocorrelation function of a periodic signal is also a periodic function, and has the same period as the original signal. Therefore, by calculating the correlation coefficient between the autocorrelation function of the IMF component and the autocorrelation function of the original signal, useful components and spurious components can be effectively distinguished. The specific algorithm is as follows:

[0138] The first step is to calculate the autocorrelation function of the original signal and each component according to the following formula (24):

[0139]

[0140] The second step is to calculate R according to the following formula (25). IMF1 ..., RIMFK With R x Correlation coefficient:

[0141]

[0142] Where N is the number of signal points, and IMFj represents the j-th IMF component. In this embodiment of the invention, IMF components with ρ(j) > 0.5 can be selected for signal reconstruction.

[0143] S105, Determine the vibration frequency of the cable based on the reconstructed motion time history data;

[0144] In some embodiments, step S105 may include:

[0145] S1051, Perform a fast Fourier transform on the reconstructed motion time history data to obtain the corresponding spectral information;

[0146] S1052 extracts spectral peaks from the spectral information using the direct peak method to obtain the natural frequencies of the cable.

[0147] The Fast Fourier Transform (FFT) is an efficient method for calculating the Discrete Fourier Transform and its inverse, which can quickly calculate the spectrum of a signal. Through FFT, we can obtain the frequency domain representation of each IMF (Integrated Frequency Flow), i.e., the spectrum. By extracting the peak values ​​of the spectrum using the direct peak method, we can identify the natural frequencies of the cable-stayed bridge.

[0148] It should be noted that during bridge deck photography, various other related vibrations are mixed with the cable vibrations, reducing the accuracy and reliability of the detection. Furthermore, visual vibration measurement methods often produce spectrograms with numerous spikes, poor low-frequency performance, and difficulty in identifying higher-order cable frequencies. Existing methods typically calculate the cable vibration frequency by directly performing a Fast Fourier Transform (FFT) on the displacement time history map of the target point. However, due to factors such as numerous spikes in the spectrogram, the cable vibration frequency obtained by this method contains significant errors. For example, Figure 5 The spectrum is the result of directly performing a Fast Fourier Transform in existing technologies. Figure 6 This is a spectrum obtained by the method provided in the embodiments of the present invention. See also: Figures 5 to 6 As shown, the EEMD+FFT method in this embodiment of the invention has a better noise reduction effect than the prior art, reduces spectrogram glitches and highlights the cable vibration frequency, and the higher-order frequencies are also more prominent, eliminating the influence of other vibrations on the cable vibration.

[0149] S106, calculate the cable force based on the vibration frequency using a preset frequency method.

[0150] For example, the preset frequency method can be represented by the following formula (26):

[0151]

[0152] Where T is the cable force, l is the length of the stay cable, m is the mass per unit length of the stay cable, and f is the mass per unit length of the stay cable. n Let n be the nth natural frequency of the cable, n be the order of the natural frequency, EI be the bending stiffness of the cable, and π be pi.

[0153] Thus, by establishing the correspondence between frequency and cable force, the calculation of cable force using computer vision in this embodiment of the invention is achieved.

[0154] Understandably, traditional cable inspection methods require workers to operate at heights, posing safety risks such as falls and collisions, and potentially leading to equipment damage or personal injury due to improper operation. The method provided in this invention reduces close contact between personnel and the cables, significantly improving on-site safety. Furthermore, this method employs the KCF tracking algorithm and Fast Fourier Transform, which not only accelerates tracking speed and improves accuracy but also demonstrates strong robustness to target deformation, lighting changes, and occlusion. It can quickly process and analyze image data, automatically completing cable force measurement, greatly improving inspection efficiency and shortening the inspection cycle. Regarding accuracy, methods such as gray-scale averaging and probabilistic statistics effectively eliminate environmental noise, highlighting the cable structure and reducing the computational difficulty for subsequent target localization. Simultaneously, the combined algorithm of EEMD (Empirical Mode Decomposition) and FFT (Fast Fourier Transform) removes noise interference from the signal, identifies cable mechanical faults, and addresses the shortcomings of visual vibration measurement methods, such as numerous spikes in the spectrum, poor performance in the low-frequency region, and difficulty in identifying the cable's fundamental frequency, thereby improving the accuracy of cable force measurement. In terms of cost, the camera-based method eliminates the need for safety scaffolding and lifting equipment required by traditional methods, saving on equipment purchase and maintenance costs. It also reduces labor costs and potential insurance expenses associated with working at heights, resulting in cost savings. Furthermore, this method enables continuous automatic monitoring of cable vibration, collecting a large amount of real-time data that provides more comprehensive and detailed information on the bridge's health status. Simultaneously, combined with modern data processing and communication technologies, it allows for real-time data transmission and analysis, providing technical support for building a smart bridge management system and enhancing the scientific and intelligent level of bridge management.

[0155] The following specific example illustrates the effects achieved by the embodiments of the present invention:

[0156] A hybrid steel box girder cable-stayed bridge, 485 meters long and 29.6 meters wide, with six lanes in both directions. The main span is 240 meters long, a double-cable-stayed arch bridge with a single inclined tower and a semi-floating system. Cable measurements were performed using only a Hikvision industrial camera, computer, tripod, and gimbal, with an HK1000 / 50mm lens, a data acquisition frequency of 60Hz, and a frame rate stabilized at approximately 80fps. Cable forces were measured on four cables (S2, S3, S4, and S5) on the northeast side. Measurement points were selected at two flat areas approximately 30 meters above and below the cables on the bridge. Simultaneously, a cable force dynamic testing instrument identified the vibration frequencies of the cables for comparison with the results of the visual frequency identification method. Specific comparison results are shown in Table 1.

[0157] Table 1

[0158]

[0159] Given that the linear density of cables No. 2 and No. 3 is m = 35.7 kg / m, and their bending stiffness EI = 280.3 kN·m 2 The calculated length of cable No. 2 is L2 = 96.92m, and the calculated length of cable No. 3 is L3 = 92.31m.

[0160] At this point, the results of the cable force calculation are shown in Table 2:

[0161] Table 2

[0162]

[0163] In this embodiment of the invention, vibration videos of bridge cables containing target points are acquired; the target points in the vibration videos are tracked using a preset tracking algorithm to obtain the motion time history data of the cables; the motion time history data is processed using the EEMD method to obtain multiple corresponding intrinsic mode functions (IMFs); based on the vibration frequency characteristics of the cables, IMFs unrelated to cable vibration are eliminated, and the remaining IMFs are reconstructed to obtain reconstructed motion time history data; the vibration frequency of the cables is determined based on the reconstructed motion time history data; and the cable force is calculated using a preset frequency method based on the vibration frequency. This avoids the physical contact requirements of traditional measurement methods, reduces potential human error and equipment wear, and enables remote operation, greatly improving the safety of the measurement process. Furthermore, the convenience and efficiency of this method are significant advantages, enabling the acquisition and analysis of large amounts of data in a short time, providing fast and reliable cable force information for bridge maintenance and management. It is suitable for non-contact measurement of bridge cable forces, establishing a non-contact, remote, convenient, and efficient cable force measurement method.

[0164] Based on the same inventive concept, this application also provides a bridge cable force measurement system based on computer vision. Figure 7This is a schematic diagram of a computer vision-based bridge cable force measurement system according to an embodiment of the present invention. See also... Figure 7 As shown, the computer vision-based bridge cable force measurement system 700 may include: an image acquisition module 701 for acquiring vibration videos of bridge cables containing target points; a target point tracking module 702 for tracking target points in the vibration videos using a preset tracking algorithm to obtain motion time history data of the cables; a first processing module 703 for processing the motion time history data using the EEMD method to obtain multiple corresponding intrinsic mode functions; a second processing module 704 for removing intrinsic mode functions unrelated to cable vibration based on the vibration frequency characteristics of the cables, and reconstructing the remaining intrinsic mode functions to obtain reconstructed motion time history data; a first determination module 705 for determining the vibration frequency of the cables based on the reconstructed motion time history data; and a second determination module 706 for calculating the cable force based on the vibration frequency using a preset frequency method.

[0165] In some possible implementations, the system may also include: a signal amplification module for performing frame-by-frame image processing on the vibration video to obtain a continuous image sequence; extracting and amplifying motion signals from the image sequence based on the phase Euler motion amplification algorithm, and reconstructing digital image data to obtain the amplified motion video.

[0166] In some possible implementations, the signal amplification module is also used to convert the vibration video format, crop the image, and decompose it to obtain a continuous image sequence; and to remove environmental noise from each image in the image sequence using the grayscale averaging method and probability statistics method.

[0167] In some possible implementations, the signal amplification module is also used for spatial domain filtering, employing a complex-adjustable pyramid to perform spatial domain filtering on the input image sequence to obtain high-pass residuals, low-pass residuals, and image sequences at different scales and directions; temporal domain filtering, by manually setting the frequency range and performing temporal bandpass filtering on the image sequence to extract the phase difference information of the motion signal of interest; motion signal amplification, by manually setting the amplification factor to linearly amplify the phase difference of the motion signal of interest extracted by temporal filtering to obtain the amplified motion signal; and digital image data reconstruction, using the image sequence synthesized from the amplified motion signal, the high-pass residuals and low-pass residuals obtained from spatial domain filtering, to reconstruct the motion-amplified digital image data, thus obtaining the amplified motion video.

[0168] In some possible implementations, the target point tracking module 702 is also used to determine the position and size of the target point and draw candidate boxes based on the initial frame image information in the vibration video; extract HOG features of the candidate box region, generate a training sample set by cyclically shifting the feature data, train the labels of the training samples based on the shift distance and Gaussian function of the training sample set, and train the kernel correlation filter model by combining the ridge regression method; read the next frame image, extract the image features of the search region, generate a candidate sample set based on the extracted features, and use the updated filter to perform correlation filtering on the candidate sample set to obtain a correlation response map; determine the position with the largest response value as the best prediction position, and update the position of the target and the candidate box of the next frame based on the peak point of the response map; add the bilinear interpolation method to the update of the target model to accommodate changes in the appearance model; after the model is updated, use the updated filter classifier template and the search region of the subsequent frame image to perform response calculations to predict the target position of the subsequent frame until all video sequence detections are completed; calculate the centroid coordinates of the target point in each frame image to obtain the motion time history data of the cable.

[0169] In some possible implementations, the vibration frequency characteristics of the cable are determined based on relevant information about the cable, including at least: cable type, material, diameter, calculated length, linear density, inclination angle, boundary conditions, service time, serial number, and / or installation location.

[0170] In some possible implementations, the first determining module 705 is also used to perform a fast Fourier transform on the reconstructed motion time history data to obtain the corresponding spectral information; and to extract the spectral peaks from the spectral information using the direct peak method to obtain the natural frequencies of the cable.

[0171] Based on the same inventive concept, this application provides an electronic device that can be consistent with the computer vision-based bridge cable force measurement method in one or more of the above embodiments. Figure 8 This is a schematic diagram of an electronic device structure according to an embodiment of the present invention. See also... Figure 8 As shown, the electronic device 800 can use general computer hardware, including a processor 801 and a memory 802.

[0172] In some possible implementations, at least one processor can constitute any physical device having circuitry that performs logical operations on one or more inputs. For example, at least one processor may include one or more integrated circuits, including application-specific integrated circuits, microchips, microcontrollers, microprocessors, all or part of a central processing unit, graphics processing unit, digital signal processor, field-programmable gate array, or other circuitry suitable for executing instructions or performing logical operations. Instructions executed by at least one processor may, for example, be preloaded into memory integrated with or embedded in the controller, or may be stored in separate memory. Memory may include random access memory, read-only memory, hard disk, optical disk, magnetic media, flash memory, other permanent, fixed, or volatile memory, or any other mechanism capable of storing instructions. In some embodiments, at least one processor may include more than one processor. Each processor may have a similar architecture, or processors may have different configurations that are electrically connected or disconnected from each other. For example, processors may be separate circuits or integrated into a single circuit. When more than one processor is used, the processors may be configured to operate independently or collaboratively. Processors may be coupled electrically, magnetically, optically, acoustically, mechanically, or by other means that allow them to interact.

[0173] Based on the same inventive concept, this application provides a computer storage medium storing computer-executable instructions. After being executed by a processor, the computer-executable instructions can realize the computer vision-based bridge cable force measurement method as described in one or more of the above embodiments.

[0174] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, please refer to each other. Each embodiment focuses on describing the differences from other embodiments.

[0175] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of this application.

Claims

1. A method for measuring bridge cable tension based on computer vision, characterized in that, include: Collect vibration videos of bridge cables containing target points; The target point in the vibration video is tracked using a preset tracking algorithm to obtain the motion time history data of the cable; The motion time history data is processed using the EEMD method to obtain multiple corresponding intrinsic mode functions; Based on the vibration frequency characteristics of the cable, eigenmode functions that are unrelated to the cable vibration are eliminated, and the remaining eigenmode functions are reconstructed to obtain the reconstructed motion time history data. The vibration frequency of the cable is determined based on the reconstructed motion time history data; The cable force is calculated using a preset frequency method based on the vibration frequency. The step of tracking the target point in the vibration video using a preset tracking algorithm to obtain the motion time history data of the cable includes: Based on the initial frame image information in the vibration video, the position and size of the target point are determined and a candidate bounding box is drawn; Extract the HOG features of the candidate box region, generate a training sample set by cyclically shifting the feature data, train the labels of the training samples based on the shift distance of the training sample set and the Gaussian function, and train the kernel correlation filter model by combining the ridge regression method. Read the next frame image, extract the image features of the search area, generate a candidate sample set based on the extracted features, and use the updated filter to perform correlation filtering on the candidate sample set to obtain the correlation response map. The location with the largest response value is determined as the best prediction location, and the target's position and candidate box are updated in the next frame based on the peak point of the response map. To address changes in the appearance model, bilinear interpolation is incorporated into the update of the target model. After the model is updated, the updated filter classifier template is used to perform response calculations with the search region of the subsequent frame image to predict the target position of the subsequent frame until all video sequences are detected. The centroid coordinates of the target point in each frame of the image are calculated to obtain the motion time history data of the cable.

2. The method according to claim 1, characterized in that, Before tracking the target point in the vibration video using a preset tracking algorithm to obtain the motion time history data of the cable, the method further includes: The vibration video is processed frame by frame to obtain a continuous image sequence; Based on the phase Euler motion amplification algorithm, motion signals are extracted and amplified from the image sequence, and digital image data is reconstructed to obtain amplified motion video.

3. The method according to claim 2, characterized in that, The step of performing frame-by-frame image processing on the vibration video to obtain a continuous image sequence includes: The vibration video is converted in format, cropped, and decomposed to obtain the continuous image sequence. Environmental noise in each image of the image sequence is removed using the gray-scale average method and probability statistics.

4. The method according to claim 3, characterized in that, The phase Euler motion magnification algorithm extracts and amplifies motion signals from the image sequence and reconstructs digital image data to obtain a magnified motion video, including: Spatial domain filtering uses a complex-oriented adjustable pyramid to perform spatial domain filtering on the input image sequence, resulting in high-pass residuals, low-pass residuals, and image sequences of different scales and orientations. Temporal filtering involves manually setting a frequency range and applying temporal bandpass filtering to an image sequence to extract the phase difference information of the motion signal of interest. Motion signal amplification involves manually setting the amplification factor to linearly amplify the phase difference of the motion signal of interest extracted by time-domain filtering, thus obtaining the amplified motion signal. Digital image data reconstruction involves using the image sequence synthesized from the amplified motion signal, the high-pass residual obtained from spatial domain filtering, and the low-pass residual to reconstruct the motion-amplified digital image data, thereby obtaining the amplified motion video.

5. The method according to claim 1, characterized in that, The vibration frequency characteristics of the cable are determined based on relevant information about the cable; such relevant information includes at least the cable's model, material, diameter, calculated length, linear density, tilt angle, boundary conditions, service time, serial number, and / or installation location.

6. The method according to claim 5, characterized in that, Determining the vibration frequency of the cable based on the reconstructed motion time history data includes: The reconstructed motion time history data is subjected to a fast Fourier transform to obtain the corresponding spectral information; The natural frequencies of the cable are obtained by extracting the spectral peaks from the spectral information using the direct peak method.

7. A bridge cable force measurement system based on the computer vision-based bridge cable force measurement method of claim 1, characterized in that, include: The image acquisition module is used to acquire vibration videos of bridge cables containing target points; The target point tracking module is used to track the target point in the vibration video using a preset tracking algorithm to obtain the motion time history data of the cable. The first processing module is used to process the motion time history data using the EEMD method to obtain multiple corresponding intrinsic mode functions; The second processing module is used to eliminate intrinsic mode functions that are unrelated to the vibration of the cable based on the vibration frequency characteristics of the cable, and to reconstruct the remaining intrinsic mode functions to obtain the reconstructed motion time history data. The first determining module is used to determine the vibration frequency of the cable based on the reconstructed motion time history data; The second determining module is used to calculate the cable force of the cable based on the vibration frequency using a preset frequency method.

8. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; A processor, when executing executable instructions or computer programs stored in the memory, implements the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing executable instructions or a computer program, characterized in that, When the executable instructions are executed by the processor, they implement the method as described in any one of claims 1 to 6.

Images