A short rope micro-vibration visual detection method and system

Abstract

This application discloses a visual detection method and system for minor vibrations of short cables, belonging to the field of short cable vibration detection technology. The method includes: acquiring a vibration video of the short cable; determining the normal phase displacement time history of each candidate measurement point in the short cable based on the vibration video; determining the normal phase synchronization score of each candidate measurement point pair based on the vibration synchronicity between the normal phase displacement time histories of the two candidate measurement points in each pair; determining the effective measurement point pairs from each pair of candidate measurement point pairs based on the normal phase synchronization score; and determining the vibration parameters of the short cable based on the normal phase displacement time history of at least one candidate measurement point in the effective measurement point pair. This approach can improve the accuracy of visual detection of minor vibrations in short cables.

Description

Technical Field

[0001] This application relates to the field of short cable vibration detection technology, and more specifically, to a visual detection method and system for minor vibrations of short cables. Background Technology

[0002] Short cable structures (such as bridge suspenders, short cables, and tensioning components) will produce slight vibrations during service due to factors such as wind loads, vehicle loads, and temperature changes. The frequency and amplitude of these vibrations can reflect the cable tension and the health status of the structure.

[0003] In related technologies, non-contact vibration detection methods based on video images characterize subpixel-level displacement by extracting local phase information from the image. However, since short cables appear as elongated strips with a large aspect ratio in images, effective texture is scarce, and the phase signals of randomly extracted measurement points are easily affected by noise, occlusion, and local image anomalies, resulting in a large amount of interference in the phase displacement time history signal. At the same time, linear textures, shadow edges, and illumination changes in complex backgrounds can also produce pseudo-phase changes, making it difficult for existing methods to effectively distinguish between real short cable edge measurement points and pseudo-measurement points in the background, thus affecting the accuracy of vibration detection. Summary of the Invention

[0004] Therefore, it is necessary to provide a visual detection method and system for short cable micro-vibrations to address the above problems, which can improve the accuracy and stability of visual detection of short cable micro-vibrations.

[0005] This application provides a visual detection method for small vibrations in short cables, including: Obtain vibration video of the short cable; Based on the vibration video of the short cable, the normal phase displacement time history of each candidate measuring point in the short cable is determined; the candidate measuring points constitute several pairs of candidate measuring points, each pair of candidate measuring points includes two candidate measuring points located in the bilateral edge regions of the short cable; in at least one frame of the vibration video, the positions of the two candidate measuring points correspond along the normal direction of the short cable. Based on the vibration synchronicity between the normal phase displacement time histories of the two candidate measuring points in each pair of candidate measuring points, the normal phase synchronization score of each pair of candidate measuring points is determined; vibration synchronicity is used to characterize the degree of coordination between the two candidate measuring points in terms of vibration frequency, vibration phase, and normal motion direction. Based on the normal phase synchronization score, valid measurement point pairs are determined from each group of candidate measurement point pairs; The vibration parameters of the short cable are determined based on the normal phase displacement time history of at least one candidate measuring point in the effective measuring point pair.

[0006] In one optional implementation, the candidate measurement point pairs are determined by: Edge detection is performed on the short cable image in the vibration video to obtain the centerline of the short cable and the first and second edge regions located on both sides of the centerline. The axial direction of the short cable is determined based on the centerline, and the direction perpendicular to the axial direction is determined as the normal direction of the short cable. Multiple candidate measurement points were selected in the first edge region and the second edge region respectively; Two candidate measurement points located in the first edge region and the second edge region respectively, and whose positions correspond to the positions along the normal direction of the short cable, are determined as a pair of candidate measurement points.

[0007] In one optional implementation, the two candidate measurement points are located on both sides of the center line, and the distance between the projected positions of the two candidate measurement points in the axial direction of the short cable is less than a preset distance threshold.

[0008] In one optional implementation, the normal phase displacement time history of each candidate measuring point in the short cable is determined based on the vibration video of the short cable, including: The candidate measurement points and their neighborhoods in the vibration video are filtered to obtain the complex response corresponding to each candidate measurement point. Based on the real and imaginary parts of the complex-valued response, the local phase information of each candidate measurement point is determined; Based on the local phase change of the same candidate measurement point in adjacent video frames and the gradient of the local phase information in the short cable normal direction, the normal displacement of the candidate measurement point in the short cable normal direction is determined. Based on the normal displacement corresponding to each video frame, the normal phase displacement time history of the candidate measurement point is obtained.

[0009] In one optional implementation, the normal phase synchronization score of each candidate measuring point pair is determined based on the vibration synchronization between the normal phase displacement time histories of the two candidate measuring points in each pair, including: Based on the normal phase displacement time history of the two candidate measurement points, determine the consistency of the main frequency, the stability of the phase difference, the consistency of the normal motion direction, the rationality of the amplitude ratio, and the prominence of the spectral peak between the two candidate measurement points. The normal phase synchronization score of candidate measurement point pairs is determined based on the consistency of the main frequency, the stability of the phase difference, the consistency of the normal motion direction, the rationality of the amplitude ratio, and the prominence of the spectral peak.

[0010] In one optional implementation, based on the normal phase displacement time history of two candidate measurement points, the consistency of the dominant frequency, the stability of the phase difference, the consistency of the normal motion direction, the rationality of the amplitude ratio, and the prominence of the spectral peaks between the two candidate measurement points are determined, including: Frequency domain analysis was performed on the normal phase displacement time history of the two candidate measurement points to determine the candidate frequencies; The consistency of the main frequency is determined based on the difference between the main frequencies corresponding to the normal phase displacement time histories of the two candidate measurement points at the candidate frequencies. The stability of the phase difference is determined based on the degree of fluctuation of the phase difference between two candidate measurement points near the candidate frequency within a preset time window; The consistency of the normal motion direction is determined based on the correlation between the normal phase displacement time histories of the two candidate measurement points. The rationality of the amplitude ratio is determined based on the ratio between the vibration amplitudes of the two candidate measurement points at the candidate frequency. The salience of the spectral peak is determined based on the difference between the spectral peak at the candidate frequency and the adjacent noise band.

[0011] In one optional implementation, valid measurement point pairs are determined from each group of candidate measurement point pairs based on normal phase synchronization scoring, including: Based on the normal phase synchronization score of each group of candidate measurement point pairs, the candidate measurement point pairs that meet the preset valid conditions are determined as valid measurement point pairs; Preset valid conditions include: The normal phase synchronization score of the candidate measurement point pair is greater than or equal to the preset scoring threshold; or, The normal phase synchronization score of the candidate measurement point pair is within the preset sorting range of the normal phase synchronization score sorting results of each group of candidate measurement point pairs.

[0012] In one optional implementation, the vibration parameters of the short cable are determined based on the normal phase displacement time history of at least one candidate measuring point in the effective measuring point pair, including: Based on the normal phase synchronization score of the effective measuring point pair and the mass parameters of each candidate measuring point in the effective measuring point pair, the measuring point confidence level of each candidate measuring point is determined; the mass parameters are used to characterize the reliability of the vibration data provided by each candidate measuring point. Based on the confidence level of the measurement point, the target measurement point is determined from the candidate measurement points in at least one set of valid measurement point pairs; The vibration parameters of the short cable are determined based on the normal phase displacement time history of the target measuring point.

[0013] In one optional implementation, determining the target measurement point from candidate measurement points in at least one set of valid measurement point pairs based on the measurement point confidence level includes: Based on the confidence level of the measurement points, multiple target measurement points are determined from the candidate measurement points in at least one set of valid measurement point pairs; Based on the normal phase displacement time history of the target measuring point, determine the vibration parameters of the short cable, including: Based on the confidence level of each target measuring point, the normal phase displacement time histories of multiple target measuring points are weighted and fused to obtain the fused vibration time histories. The vibration parameters of the short cable are determined based on the fusion vibration time history.

[0014] Furthermore, this application also provides a visual inspection system for short cable minute vibrations, used to implement the above-mentioned visual inspection method for short cable minute vibrations. The system includes: The video acquisition module is used to acquire vibration videos of the short cable; The phase displacement calculation module is used to determine the normal phase displacement time history of each candidate measurement point in the short cable based on the vibration video of the short cable. The candidate measurement points constitute several pairs of candidate measurement points, and each pair of candidate measurement points includes two candidate measurement points located in the two edge regions of the short cable respectively. In at least one frame of the vibration video, the positions of the two candidate measurement points correspond along the normal direction of the short cable. The synchronization score calculation module is used to determine the normal phase synchronization score of each candidate measurement point pair based on the vibration synchronization between the normal phase displacement time histories of the two candidate measurement points in each pair; vibration synchronization is used to characterize the degree of coordination between the two candidate measurement points in terms of vibration frequency, vibration phase, and normal motion direction. The measurement point screening module is used to determine the effective measurement point pairs from each group of candidate measurement point pairs based on the normal phase synchronization score. The vibration parameter calculation module is used to determine the vibration parameters of the short cable based on the normal phase displacement time history of at least one candidate measuring point in the effective measuring point pair.

[0015] Compared to existing technologies, the advantages of this application are as follows: Firstly, this application's technical solution selects multiple pairs of candidate measurement points in the bilateral edge regions of the short cable, ensuring that the two candidate measurement points in each pair are located in the bilateral edge regions of the short cable and their axial positions correspond. Then, the normal phase displacement time history of each candidate measurement point is obtained, and the corresponding normal phase synchronization score is determined based on the vibration synchronicity between the normal phase displacement time histories of the two candidate measurement points in each pair. Since the bilateral edges of a real short cable typically exhibit strong coordination in vibration frequency, vibration phase, and normal motion direction when experiencing minor normal vibrations, and since pseudo-motion signals generated by background textures, shadow edges, local noise, or non-cable edges rarely simultaneously satisfy these synchronization characteristics, this application's technical solution can filter out effective measurement point pairs from the candidate measurement point pairs based on the normal phase synchronization score, reducing the influence of pseudo-measurement points on the short cable vibration detection results. Finally, the vibration parameters of the short cable are determined based on the normal phase displacement time history of at least one candidate measuring point in the effective measuring point pair. This ensures that the calculation of the vibration parameters is based on the effective measuring points after the synchronization of both sides of the edge is verified, thereby improving the accuracy and stability of visual detection of small vibrations in the short cable. Attached Figure Description

[0016] Figure 1 This is a flowchart of a visual detection method for minor vibrations of short cables provided in an embodiment of this application; Figure 2This is a schematic diagram of the bilateral edge region of a short cable provided in an embodiment of this application; Figure 3 This is a schematic diagram showing the partial cross-section and orientation definition of a short cable according to an embodiment of this application; Figure 4 A candidate measurement point and a top view of the candidate measurement point are provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0017] To make the above-mentioned objectives, features, and advantages of this application more apparent and understandable, the specific embodiments of this application are described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0018] Short cable structures, such as bridge hangers, short cables, and tensioning members, are crucial load-bearing components in bridges and buildings. By detecting the minute vibration frequencies and amplitudes of these short cables, cable tension can be calculated, thereby assessing the structural health. However, in real-world testing scenarios, the short cable images captured by cameras typically occupy only a small pixel width, lacking texture information. Furthermore, the background often contains interfering elements such as linear textures and shadowed edges, making it difficult for conventional phase analysis methods based on single points or edges to distinguish between genuine short cable vibrations and background pseudo-motions, thus compromising detection accuracy.

[0019] Therefore, in order to solve the above problems, this application provides a visual detection method for small vibrations of short cables. Figure 1 A flowchart of a visual detection method for minor vibrations of a short cable according to an embodiment of this application is shown. (Refer to...) Figure 1 As shown, the method includes the following steps: Step 101: Obtain the vibration video of the short cable.

[0020] In this embodiment of the application, an industrial camera, high-speed camera, high-definition camera, or other image acquisition device with continuous image acquisition capability can be used to acquire video of the short cable to be detected, thereby obtaining the vibration video of the short cable.

[0021] Among them, short cables can be bridge hangers, short cables, tension members, cable-like members in cable net structures, or other slender members that can generate minute vibrations.

[0022] Specifically, when acquiring vibration video of a short cable, at least a portion of the short cable can be placed within the field of view of the image acquisition device, ensuring the cable remains visible during vibration. The image acquisition device can be positioned at a preset distance from the short cable to acquire continuous images of the cable under environmental or artificial excitation in a non-contact manner.

[0023] The vibration video can include multiple video frames arranged in chronological order, each representing the image state of the short cable at a corresponding acquisition time. The frame rate of the image acquisition device can be set according to the possible vibration frequency of the short cable so that the vibration video can reflect the positional changes of the short cable during vibration. The image resolution of the image acquisition device can be set according to factors such as the imaging width of the short cable in the image, the amplitude of the vibration to be detected, and the image acquisition distance.

[0024] In one optional implementation, after acquiring the vibration video of the short cable, a region of interest (ROI) containing the short cable to be detected can be selected from the vibration video. The ROI may include the main body of the short cable and some background areas located on both sides of the main body of the short cable. By selecting the ROI, the image areas irrelevant to the vibration detection of the short cable can be reduced, thereby reducing the amount of data in subsequent image processing.

[0025] Furthermore, each video frame in the vibration video can be subjected to grayscale conversion, noise reduction, brightness adjustment, contrast enhancement, or image stabilization to improve the distinction between the short cable and the background. It should be noted that the above image preprocessing methods are common techniques in the field, and this application does not limit the specific image preprocessing methods used.

[0026] Step 102: Determine the normal phase displacement time history of each candidate measuring point in the short cable based on the vibration video of the short cable.

[0027] Among them, the candidate measurement points constitute several pairs of candidate measurement points, and each pair of candidate measurement points includes two candidate measurement points located in the bilateral edge regions of the short cable respectively; in at least one frame of the vibration video, the two candidate measurement points are positioned correspondingly along the normal direction of the short cable.

[0028] In this embodiment, the short cable can be represented as a strip-shaped image region with a certain length and width in a video frame. (Refer to...) Figure 2 As shown, the bilateral edge regions of the short cable can be image regions located on both sides of the center region of the short cable, each containing the contour edges of the short cable on both sides. Specifically, the bilateral edge regions can include a first edge region and a second edge region, which are located on both sides of the center line of the short cable, respectively.

[0029] Specifically, at least one frame can be selected from the vibration video as a reference image, and the outline of the short cable in the reference image can be detected to obtain the center line, axial direction, and side edge regions of the short cable in the image. Edge detection can employ Canny edge detection, Sobel gradient detection, Hough line detection, morphological thinning, image segmentation, or other image processing methods capable of recognizing the edges of the short cable outline.

[0030] Reference Figure 3 As shown, the axial direction of the short cable can be the direction in which the short cable extends in the image plane, and the normal direction of the short cable can be the direction perpendicular to the axial direction in the image plane. The movement of the small transverse vibration of the short cable in the image is usually mainly manifested as a change in position along the normal direction. Therefore, the normal phase displacement time history of each candidate measurement point can be obtained by determining the change in displacement of each candidate measurement point along the normal direction of the short cable over time.

[0031] The first and second edge regions are strip-shaped image regions located on either side of the center line and encompassing the edges of the short cables, respectively, within a certain width. Multiple candidate measurement points are determined within each of the first and second edge regions, and two candidate measurement points corresponding to each other along the normal direction of the short cables are grouped into a pair of candidate measurement points. (Refer to...) Figure 4 As shown, the positions along the normal direction of the short cable correspond, which can be understood as the two candidate measurement points being located on the two sides of the short cable respectively, and their positions in the axial direction of the short cable are the same or similar, thus being able to approximately characterize the motion state of the two sides of the short cable at the same local position.

[0032] In one optional implementation, one or more frames can be selected from the vibration video as reference images, and candidate measurement point pairs can be determined based on the spatial positional relationship of the two edges of the short cable in the reference images. Thus, the determination of candidate measurement point pairs is a preliminary spatial matching of the measurement points on both edges, which is only used to determine the combination of measurement points for subsequent vibration synchronization comparison, and does not indicate that the candidate measurement point pair can accurately characterize the actual vibration of the short cable.

[0033] For each candidate measurement point, local image information of the candidate measurement point in multiple video frames can be obtained, and the normal phase displacement time history of the candidate measurement point along the short cable normal direction can be determined based on the change of the local image information over time.

[0034] The phase displacement time history can be a displacement time series determined by the change in local image phase over time corresponding to the candidate measurement point. The local image phase can be obtained using complex-valued filtering, local frequency analysis, or other image processing methods capable of extracting local image phase information. Determining minute displacements based on local image phase changes is a common technique in this field, and this application does not specifically limit it.

[0035] Specifically, the local phase of the candidate measurement point in each video frame can be obtained sequentially, and the displacement change of the candidate measurement point along the normal direction of the short cable in the corresponding time period can be determined based on the local phase change between adjacent video frames. Furthermore, the displacement changes corresponding to each time period are arranged or accumulated according to the time sequence of the video frames to obtain the normal phase displacement time history of the candidate measurement point.

[0036] Each displacement value in the normal phase displacement time history can correspond to a video frame or a sampling moment in the vibration video, and is used to characterize the positional change of the candidate measurement point along the normal direction of the short cable at the corresponding moment. Therefore, the normal motion process of the candidate measurement point during the vibration video acquisition can be reflected by the normal phase displacement time history of the candidate measurement point.

[0037] Step 103: Determine the normal phase synchronization score of each candidate measuring point pair based on the vibration synchronization between the normal phase displacement time histories of the two candidate measuring points in each candidate measuring point pair.

[0038] Vibration synchronization is used to characterize the degree of coordination between two candidate measurement points in terms of vibration frequency, vibration phase, and normal motion direction.

[0039] In this embodiment of the application, after the candidate measurement point pair has completed the initial matching through the spatial position relationship in the image, the dynamic vibration relationship between the two candidate measurement points in the candidate measurement point pair can be further verified by utilizing the time series information corresponding to the vibration video.

[0040] It is understandable that for two candidate measuring points actually located on both sides of the short cable and corresponding to the same local position of the short cable, when the short cable undergoes normal vibration, these two candidate measuring points usually move together with the main body of the short cable. Therefore, the normal phase displacement time histories of the two candidate measuring points should have a certain degree of coordination in terms of vibration frequency, vibration phase, and normal motion direction.

[0041] Conversely, when candidate measurement points are affected by background texture, shadow edges, local occlusion, lighting changes, image noise, or other non-short-term motion factors, even if the spatial positions of two candidate measurement points in a single frame of an image meet the initial matching relationship, their normal phase displacement time histories in consecutive video frames may not be able to maintain a stable vibration coordination relationship.

[0042] Specifically, the vibration frequency characteristics, vibration phase characteristics, and normal motion direction characteristics of the two candidate measuring points can be obtained from the normal phase displacement time histories of the two candidate measuring points in the candidate measuring point pair. Based on the similarity or consistency between the above characteristics, the normal phase synchronization score of the corresponding candidate measuring point pair can be determined.

[0043] Among them, the degree of coordination in vibration frequency can be used to characterize whether the main periodic changes of the two candidate measuring points are consistent; the degree of coordination in vibration phase can be used to characterize whether the relative relationship of the periodic changes of the two candidate measuring points in time is stable; and the degree of coordination in the normal motion direction can be used to characterize whether the displacement change trends of the two candidate measuring points along the normal direction of the short cable are consistent.

[0044] In one optional implementation, the normal phase displacement time histories of the two candidate measurement points can be analyzed in the time domain or frequency domain, and the vibration characteristics obtained from the analysis can be converted into corresponding synchronization evaluation results. Then, different synchronization evaluation results can be combined to obtain the normal phase synchronization score of the candidate measurement point pair.

[0045] Step 104: Based on the normal phase synchronization score, determine the valid measurement point pairs from each group of candidate measurement point pairs.

[0046] In this embodiment of the application, the vibration synchronization degree of each group of candidate measurement point pairs can be determined based on the normal phase synchronization score corresponding to each group of candidate measurement point pairs, and the candidate measurement point pairs whose vibration synchronization degree meets the detection requirements are determined as valid measurement point pairs.

[0047] Among them, the effective measurement point pair can be the combination of measurement points that can stably characterize the true normal vibration of the short cable from the candidate measurement point pair. Correspondingly, other candidate measurement point pairs that are not determined as effective measurement point pairs can be combinations of measurement points that are affected by background interference, edge recognition error, local noise, illumination changes, occlusion or other abnormal factors.

[0048] Specifically, the normal phase synchronization score of each candidate measurement point pair can be compared with pre-set judgment conditions, and the valid measurement point pair can be determined based on the comparison results. Alternatively, the candidate measurement point pair with a relatively higher normal phase synchronization score can be selected as the valid measurement point pair based on the relative relationship between the normal phase synchronization scores of each candidate measurement point pair. This application does not limit the specific method of determining the valid measurement point pair based on the normal phase synchronization score.

[0049] Understandably, the spatial correspondence of candidate measurement point pairs only indicates that the two candidate measurement points may be located on the bilateral edges of the same local location in the short cable, and cannot guarantee that neither candidate measurement point is affected by image interference. By further determining the effective measurement point pairs based on normal phase synchronization scoring, the dynamic cooperative relationship between the bilateral edges in the actual short cable vibration process can be used to further screen the spatially initially matched candidate measurement point pairs.

[0050] Therefore, candidate measurement point pairs that appear to correspond in spatial location but whose normal phase displacement time histories cannot stably reflect the true vibration of the short cable can be eliminated, reducing the influence of false measurement points, abnormal measurement points, or disturbed measurement points on the subsequent vibration parameter calculation results.

[0051] Step 105: Determine the vibration parameters of the short cable based on the normal phase displacement time history of at least one candidate measuring point in the effective measuring point pair.

[0052] The vibration parameters may include at least one of the following: vibration displacement, vibration amplitude, vibration frequency, vibration period, or other parameters that can characterize the vibration state of the short cable.

[0053] Specifically, the displacement of the candidate measuring point along the normal direction of the short cable can be obtained over time based on the normal phase displacement time history of the candidate measuring point. Furthermore, the vibration amplitude of the short cable can be determined based on the displacement extremes, displacement range, or periodic variation amplitude in the normal phase displacement time history; time-domain or frequency-domain analysis can also be performed on the normal phase displacement time history to determine the vibration frequency or vibration period of the short cable.

[0054] In one optional implementation, when the displacement represented by the normal phase displacement time history is a pixel displacement in the image coordinate system, the pixel displacement can be converted into the physical displacement of the short cable in actual space according to a predetermined image scale relationship. The image scale relationship can be determined based on the calibration parameters of the image acquisition device, the known dimensions of the short cable, reference objects in the image, or other commonly used calibration methods.

[0055] It should be noted that step 105 only requires determining the vibration parameters of the short cable based on the normal phase displacement time history of at least one candidate measuring point in the effective measuring point pair. Therefore, one candidate measuring point from a set of effective measuring point pairs can be selected, or two candidate measuring points from a set of effective measuring point pairs can be selected, or multiple candidate measuring points from multiple sets of effective measuring point pairs can be selected. This application embodiment does not limit the specific number of candidate measuring points selected or the signal processing method for multiple candidate measuring points.

[0056] In summary, this embodiment first determines candidate measurement point pairs and the normal phase displacement time history of each candidate measurement point in the bilateral edge region of the short cable based on the vibration video; then, based on the vibration synchronicity between the normal phase displacement time histories of the two candidate measurement points in each pair, a normal phase synchronization score is determined, and valid measurement point pairs are selected based on the normal phase synchronization score; finally, the vibration parameters of the short cable are determined based on the normal phase displacement time history of at least one candidate measurement point in the valid measurement point pair. Since the actual vibration of the short cable typically exhibits a stable dynamic coordination relationship at the bilateral edges, while pseudo-motions caused by background texture, lighting changes, local noise, or other interference are unlikely to continuously satisfy the aforementioned vibration synchronization relationship, this embodiment can reduce the influence of abnormal measurement points on the short cable vibration parameter detection results, improving the accuracy and stability of visual detection of minor vibrations in short cables.

[0057] For example, this application also provides another embodiment of a visual detection method for minor vibrations of short cables, which may include the following steps: Step 201: Obtain the vibration video of the short cable.

[0058] Please see details Figure 1 Step S101 of the illustrated embodiment will not be described again here.

[0059] Step 202: Determine the normal phase displacement time history of each candidate measuring point in the short cable based on the vibration video of the short cable.

[0060] Among them, the candidate measurement points constitute several pairs of candidate measurement points, and each pair of candidate measurement points includes two candidate measurement points located in the bilateral edge regions of the short cable respectively; in at least one frame of the vibration video, the two candidate measurement points are positioned correspondingly along the normal direction of the short cable.

[0061] Furthermore, in each pair of candidate measurement points, the two candidate measurement points are located on opposite sides of the centerline, and the distance between the projected positions of the two candidate measurement points in the axial direction of the short cable is less than a preset distance threshold. In other words, the two candidate measurement points in each pair can be located at the same or similar axial positions of the short cable, thereby approximately representing the bilateral edge motion state at the same local position of the short cable.

[0062] Optionally, candidate measurement point pairs can be determined as follows: perform edge detection on the short cable image in the vibration video to obtain the centerline of the short cable and the first edge region and the second edge region located on both sides of the centerline; determine the axial direction of the short cable based on the centerline, and determine the direction perpendicular to the axial direction as the normal direction of the short cable; select multiple candidate measurement points in the first edge region and the second edge region respectively; and determine two candidate measurement points located in the first edge region and the second edge region respectively, and whose positions correspond to the positions along the normal direction of the short cable, as a set of candidate measurement point pairs.

[0063] After selecting candidate measurement points, they can be paired in the first edge region and the second edge region based on their positional relationship in the axial and normal directions of the short cable. Specifically, two candidate measurement points whose axial projection positions are less than a preset distance threshold and located on opposite sides of the centerline can be identified as a pair of candidate measurement points. Optionally, the normal phase displacement time history of each candidate measurement point in the short cable can be determined as follows: filter each candidate measurement point and its neighborhood in the vibration video to obtain the complex response corresponding to each candidate measurement point; determine the local phase information of each candidate measurement point based on the real and imaginary parts of the complex response; determine the normal displacement of the candidate measurement point in the normal direction of the short cable based on the local phase change of the same candidate measurement point in adjacent video frames and the gradient of the local phase information in the normal direction of the short cable; and obtain the normal phase displacement time history of the candidate measurement point based on the normal displacement corresponding to each video frame.

[0064] Specifically, complex-valued filters can be used to filter the candidate measurement point and its surrounding local image region. A complex-valued filter outputs a complex response, including both real and imaginary parts. Based on the real and imaginary parts of the complex-valued response, the local phase of the candidate measurement point in each video frame can be calculated. The local phase can be used to characterize the positional state of the local image structure surrounding the candidate measurement point. When a candidate measurement point undergoes sub-pixel displacement, its local grayscale change may not be significant, but the local phase will usually change accordingly. Therefore, by analyzing the changes in local phase across consecutive video frames, minute motion information of the candidate measurement point can be extracted.

[0065] The complex-valued filter can be a Gabor filter, a complex-valued controllable pyramid filter, a Riesz pyramid filter, or other complex-valued filters capable of extracting local phase information. In this embodiment, a multi-directional, multi-scale Gabor filter can be selected to perform phase decomposition on the neighborhood of the candidate measurement point.

[0066] For the position in the t-th frame image The complex-valued filter response of the candidate measurement point at the location is denoted as . The local phase is calculated based on the real and imaginary parts of the complex-valued response. The calculation formula is as follows: ; in, Let represent the local phase at position (x, y) in the t-th frame of the image. This represents the complex response after convolution of a complex-valued filter. The imaginary part of the complex response is represented. Represents the real part of the complex response.

[0067] Furthermore, the local phase change of the same candidate measurement point in adjacent video frames can be calculated, along with the gradient of the local phase along the short cable normal direction. Based on the relationship between the local phase change and the normal gradient, the normal displacement of the candidate measurement point along the short cable normal direction between adjacent video frames can be determined. By arranging the normal displacements of the candidate measurement point at corresponding times in each video frame in chronological order, or by accumulating the normal displacements between adjacent video frames, the time history of the normal phase displacement of the candidate measurement point can be obtained.

[0068] The normal phase displacement time history can include multiple normal displacement values ​​arranged in time. Each normal displacement value corresponds to a video frame or a sampling time in the vibration video, and is used to characterize the position change of the candidate measurement point along the normal direction of the short cable at the corresponding time.

[0069] In one specific embodiment, the axial direction of the short cable is represented as a unit vector. The normal direction of the short cable is represented as a unit vector. or The specific orientation can be determined based on the image coordinate system and the positions of the two edges of the short cable.

[0070] First, calculate the local phase change between adjacent frames: ; in, This indicates the position (x, y) at the th Frame relative to the first The amount of phase change in a frame.

[0071] To avoid phase crossing or When a transition occurs, phase wrapping can be applied to the phase change between adjacent frames: ; in, This indicates that the phase difference is limited to Phase wrap function within the interval.

[0072] Then, the small displacement of the candidate measurement point in the normal direction is estimated based on the phase change and the gradient of the local phase along the normal direction of the short cable. The gradient of the local phase along the normal direction can be expressed as: ; For the normal direction The normal displacement of the candidate measuring point can be expressed as:

[0073] in, Indicates the candidate measurement point at the th The normal displacement of the frame.

[0074] When the phase difference after phase wrapping is used, the normal displacement of the candidate measurement point can be further expressed as: ; in, To prevent the denominator from approaching zero, a stable term is used.

[0075] The normal displacement of each candidate measurement point in consecutive frames is accumulated to obtain the time history of the normal phase displacement of the candidate measurement point: ; in, Indicates the first The candidate measurement points at the th The normal phase displacement time history at the corresponding frame time.

[0076] Step 203: Determine the normal phase synchronization score of each candidate measuring point pair based on the vibration synchronization between the normal phase displacement time histories of the two candidate measuring points in each candidate measuring point pair.

[0077] Optionally, the normal phase synchronization score for each pair of candidate measurement points can be determined as follows: based on the normal phase displacement time histories of the two candidate measurement points in the pair, determine the consistency of the dominant frequency, the stability of the phase difference, the consistency of the normal motion direction, the rationality of the amplitude ratio, and the prominence of the spectral peak between the two candidate measurement points; based on the consistency of the dominant frequency, the stability of the phase difference, the consistency of the normal motion direction, the rationality of the amplitude ratio, and the prominence of the spectral peak, determine the normal phase synchronization score of the pair of candidate measurement points.

[0078] Optionally, based on the normal phase displacement time histories of two candidate measurement points, determining the consistency of the dominant frequency, the stability of the phase difference, the consistency of the normal motion direction, the rationality of the amplitude ratio, and the prominence of the spectral peak between the two candidate measurement points may include the following steps: performing frequency domain analysis on the normal phase displacement time histories of the two candidate measurement points to determine the candidate frequencies; determining the consistency of the dominant frequency based on the difference between the dominant frequencies corresponding to the normal phase displacement time histories of the two candidate measurement points at the candidate frequencies; determining the stability of the phase difference based on the fluctuation of the phase difference of the two candidate measurement points near the candidate frequencies within a preset time window; determining the consistency of the normal motion direction based on the correlation between the normal phase displacement time histories of the two candidate measurement points; determining the rationality of the amplitude ratio based on the ratio between the vibration amplitudes of the two candidate measurement points at the candidate frequencies; and determining the prominence of the spectral peak based on the difference between the spectral peak at the candidate frequencies and the adjacent noise frequency bands.

[0079] In this embodiment, the process of obtaining the above-mentioned normal phase synchronization score is further explained.

[0080] First, when the vibration frequency of the short cable is unknown or may drift, broadband filtering and phase pre-enhancement are performed on the normal phase displacement time history corresponding to the candidate measurement points.

[0081] Specifically, based on the camera frame rate The broadband filter frequency range is set according to the range of possible vibration frequencies of the short cable. The broadband filter frequency range is not limited to a single fixed frequency, but covers multiple candidate vibration frequencies that the short cable may exhibit. For example, an initial broadband frequency range can be set based on the sampling frequency, short cable length, linear density, and tension prediction range.

[0082] A broadband time-domain filter is applied to the normal phase displacement time history corresponding to the candidate measurement points to retain the phase change components within the possible vibration frequency range of the short cable, and the filtered phase signal is then pre-enhanced. Specifically, the broadband-filtered normal phase displacement signal of the measurement points can be expressed as: ; in, This represents the normal phase displacement signal of the measurement point after broadband filtering. Indicates passband is Broadband filters; and These represent the lower limit frequency and the upper limit frequency of the broadband filter, respectively.

[0083] Subsequently, frequency domain analysis is performed on the time history of the pre-enhanced normal phase shift to obtain a set of candidate frequencies. Frequency domain analysis methods include Fast Fourier Transform (FFT), Welch power spectrum estimation, Short-Time Fourier Transform (SFT), or Wavelet Transform. In this embodiment, Fast Fourier Transform is used as an example.

[0084] Specifically, the power spectrum corresponding to the candidate measurement point can be expressed as: ; in, This represents the power spectrum corresponding to the candidate measurement point; Indicates frequency; This represents the Fast Fourier Transform.

[0085] The candidate frequency set can be represented as: ; in, Represents the set of candidate frequencies; This indicates the peak value screening threshold.

[0086] After obtaining the candidate frequency set, for each candidate frequency, the synchronization characteristics between candidate measurement point pairs are calculated.

[0087] Specifically, clock frequency consistency can be expressed as: ; in, Indicates clock frequency consistency; and They represent the first The dominant frequencies of the first and second edge measuring points in the candidate measuring point pair; It is the main frequency difference scale parameter.

[0088] Furthermore, the phase difference stability is calculated based on the instantaneous phase difference between two candidate measurement points near the candidate frequency. The phase difference stability can be expressed as: ; in, Indicates phase difference stability; and They represent the first The instantaneous phase of the first and second edge measuring points near the target frequency in the candidate measuring point pair. Indicates standard deviation, This is the phase difference fluctuation scale parameter.

[0089] Furthermore, the consistency of the normal motion direction is determined based on the correlation between the time histories of the corresponding normal phase displacements of the two candidate measurement points: ; in, Indicates consistency in the direction of normal motion; Represents the correlation coefficient. and These represent the normal vibration time history signals of the first edge measuring point and the second edge measuring point, respectively.

[0090] Furthermore, the rationality of the amplitude ratio is calculated based on the vibration amplitude values ​​corresponding to the two candidate measurement points: ; in, This indicates the reasonableness of the amplitude ratio; and These represent the vibration amplitudes of the first and second edge measuring points at the candidate frequencies, respectively. This is the amplitude scaling parameter. .

[0091] Furthermore, the salience of the spectral peak is determined based on the difference between the spectral peak corresponding to the candidate frequency and the adjacent noise frequency band: ; in, Indicates the salience of the spectral peaks; Indicates candidate frequency, This represents the noise band adjacent to the candidate frequency but excluding the candidate frequency peak. This represents the mean.

[0092] Furthermore, the above synchronization features are weighted and fused to obtain the normal phase synchronization score corresponding to the candidate measurement point pairs: ; in, Indicates the first Simultaneous scoring of candidate measurement points on both sides of the group. Indicates the clock frequency consistency score. This indicates the phase difference stability score. The consistency score for representing the normal direction of motion. The score indicates the reasonableness of the amplitude ratio. Indicates the peak salience score of the spectrum. These are the corresponding weighting coefficients. Each weighting coefficient can be set according to actual detection needs or obtained through training on historical samples. A higher normal phase synchronization score indicates a higher degree of vibration coordination between the two candidate measurement points; a lower normal phase synchronization score indicates a lower degree of vibration coordination between the two candidate measurement points.

[0093] Optionally, each synchronization feature can be normalized before weighted fusion to avoid the impact of dimensional differences between different evaluation indicators on the normal phase synchronization score. A synchronization score threshold can be set. ,when When the corresponding measuring point pair is identified as an unreliable measuring point or a spurious vibration measuring point, then... When the time is right, the corresponding measuring point pair will be determined as a valid short cable vibration measuring point.

[0094] In summary, by comprehensively evaluating the consistency of the dominant frequency, the stability of the phase difference, the consistency of the normal motion direction, the rationality of the amplitude ratio, and the prominence of the spectral peak of the candidate measurement point pair at the candidate frequency, and by using a weighted fusion method to obtain the normal phase synchronization score, the degree of vibration synchronization between the candidate measurement point pairs can be more accurately reflected.

[0095] Step 204: Based on the normal phase synchronization score, determine the valid measurement point pairs from each group of candidate measurement point pairs.

[0096] Optionally, valid measurement point pairs can be determined as follows: based on the normal phase synchronization score of each group of candidate measurement point pairs, candidate measurement point pairs that meet the preset valid conditions are determined as valid measurement point pairs.

[0097] The preset effective conditions can be set according to the detection accuracy requirements, the number of candidate measurement point pairs, or the distribution of normal phase synchronization scores. By setting the effective conditions, candidate measurement point pairs with low vibration coordination can be screened out, and only candidate measurement point pairs that can stably characterize short cable vibration can be retained.

[0098] Optionally, the preset valid conditions may include: the normal phase synchronization score of the candidate measurement point pair is greater than or equal to the preset score threshold; or, the normal phase synchronization score of the candidate measurement point pair is within the preset sorting range of the normal phase synchronization score sorting results of each group of candidate measurement point pairs.

[0099] In this embodiment, to further improve the accuracy of short cable target vibration frequency identification, after obtaining the normal phase synchronization score of each group of candidate measurement points in step 203, the frequency dimension expansion of the normal phase synchronization score can be performed by combining the spectral characteristics corresponding to the normal phase displacement time history obtained in step 202, so that the normal phase synchronization score is associated with the corresponding candidate frequency, and the target vibration frequency band is adaptively determined according to the synchronization score and spectral characteristics corresponding to the candidate frequency, and the normal phase signal in the target vibration frequency band is enhanced in a narrow band.

[0100] Specifically, for each candidate frequency obtained in the broadband pre-enhancement stage, the bilateral edge synchronization score, spectral peak prominence, and phase stability corresponding to that candidate frequency are calculated, and a target frequency selection score is constructed based on the above evaluation indicators: ; in, Indicates the target frequency selection score; Represents candidate frequency Corresponding bilateral edge-synchronized scoring; Represents candidate frequency Corresponding spectral peak prominence; Represents candidate frequency Corresponding time stability; , as well as These are the weighting coefficients.

[0101] Based on the target frequency corresponding to each candidate frequency, a score is selected to determine the target dominant frequency: ; in, Indicates the target clock frequency; This represents the set of candidate frequencies.

[0102] Subsequently, with the target main frequency as the center, the target narrowband filtering range is adaptively determined based on the spectral peak width, sampling frequency, and preset bandwidth coefficient. ; in, Indicates the peak width or frequency resolution of the candidate frequency; This represents the bandwidth coefficient.

[0103] Furthermore, narrowband filtering and secondary phase enhancement are performed on the short cable edge phase signal within the target frequency band to suppress non-target frequency components, background texture motion, and broadband noise, thereby improving the signal-to-noise ratio of the phase change corresponding to the target vibration frequency.

[0104] When the vibration frequency of the short cable drifts, the synchronization score and target frequency selection score corresponding to the candidate frequency can be recalculated according to the preset time window, and the target narrowband filtering range can be dynamically updated, thereby avoiding missed detections or false detections caused by fixed filtering frequency bands and improving the stability of vibration frequency tracking.

[0105] Furthermore, to reduce the impact of common disturbances such as background texture, lighting changes, and camera micro-vibrations on the short cable edge phase signal, background phase compensation can be performed on the narrowband enhanced short cable edge phase signal using a background reference area.

[0106] Specifically, one or more background reference regions are selected near the region of interest of the short cable, and these background reference regions do not contain the short cable target. The same phase calculation method as that used for the short cable edge is used to obtain the background phase time history signal corresponding to the background reference region.

[0107] Subsequently, the correlation between the phase signal of the background reference region and the phase signal of the short cable edge is calculated. When a phase change occurs simultaneously in both the background reference region and the short cable edge region, and does not satisfy the phase synchronization constraint of the bilateral edge normals, it is determined to be a common disturbance caused by camera micro-vibration, illumination changes, or background texture movement.

[0108] The common disturbance component is estimated based on the phase signal of the background reference region, and the common disturbance component is subtracted or suppressed from the phase signal of the short cable edge to obtain the short cable edge phase time history signal after background phase compensation.

[0109] In one embodiment, background phase compensation can be expressed as: ; in, Indicates the first The original normal phase displacement time history of each candidate measurement point; This represents the normal phase displacement time history after background compensation; Indicates the first Background phase time history corresponding to each background reference region; This represents the background perturbation weight.

[0110] The background perturbation weights can be obtained using least squares estimation: ; in, This represents a matrix composed of phase time histories from multiple background reference regions.

[0111] To avoid mistakenly subtracting actual short-cable vibrations as background disturbances, the background difference can be further calculated: ; in, Indicates the first Background difference corresponding to each candidate measurement point. When A larger value indicates that the phase change at the measurement point differs significantly from the background common disturbance, thus possessing higher reliability.

[0112] Furthermore, the confidence level of each candidate measurement point is calculated based on the time history of the normal phase displacement after background compensation and the mass parameters of each candidate measurement point. The mass parameters are used to characterize the reliability of the vibration data provided by the candidate measurement points, and may include edge intensity, spectral peak prominence, local signal-to-noise ratio, and background difference, etc.

[0113] No. The confidence level of a candidate measurement point can be expressed as: ; in, Indicates the first The confidence level of each candidate measurement point; This indicates the normal phase synchronization score corresponding to the candidate measurement point; Indicates edge strength; Indicates the salience of the spectral peaks; Indicates the local signal-to-noise ratio; Indicates the degree of background difference; , , , as well as These are the weighting coefficients.

[0114] Subsequently, based on the confidence level of the measurement points and the preset valid conditions, candidate measurement point pairs with low confidence levels or whose normal phase synchronization scores do not meet the preset valid conditions are eliminated, and the remaining candidate measurement point pairs are determined as valid measurement point pairs.

[0115] Step 205: Determine the vibration parameters of the short cable based on the normal phase displacement time history of at least one candidate measuring point in the effective measuring point pair.

[0116] Optionally, the vibration parameters of the short cable can be determined as follows: the measurement point confidence level of each candidate measurement point is determined based on the normal phase synchronization score of the effective measurement point pair and the mass parameters of each candidate measurement point in the effective measurement point pair; the mass parameters are used to characterize the reliability of the vibration data provided by each candidate measurement point; the target measurement point is determined from the candidate measurement points in at least one set of effective measurement point pairs based on the measurement point confidence level; and the vibration parameters of the short cable are determined based on the normal phase displacement time history of the target measurement point.

[0117] Optionally, determining target measuring points from candidate measuring points in at least one set of valid measuring point pairs based on measuring point confidence can include the following steps: determining multiple target measuring points from candidate measuring points in at least one set of valid measuring point pairs based on measuring point confidence; determining the vibration parameters of the short cable based on the normal phase displacement time history of the target measuring points, including: weighting and fusing the normal phase displacement time histories of multiple target measuring points based on the measuring point confidence of each target measuring point to obtain a fused vibration time history; and determining the vibration parameters of the short cable based on the fused vibration time history.

[0118] In this embodiment, the normal phase displacement time histories corresponding to multiple target measuring points are weighted and fused according to the measuring point confidence level of each target measuring point to obtain the overall vibration phase time histories of the short cable.

[0119] In one embodiment, weighted fusion can be performed based on the confidence levels of the measurement points corresponding to each target measurement point. The overall vibration phase time history of the short cable after weighted fusion can be expressed as: ; in, This represents the overall vibration phase time history of the short cable after fusion; Indicates the first The confidence level of each target measurement point; Indicates the first Time history of normal phase displacement corresponding to each target measuring point; Indicates the number of target measurement points.

[0120] Furthermore, the confidence scores of each target measurement point can be normalized to obtain the fusion weights corresponding to each target measurement point. ; in, Indicates the first The normalized fusion weights corresponding to each target measurement point.

[0121] Subsequently, based on the normalized fusion weights corresponding to each target measurement point, the time histories of the normal phase displacement after background compensation for multiple target measurement points are weighted and fused to obtain the fused overall normal phase displacement time histories of the short cable: ; in, This represents the time history of the overall normal phase displacement of the merged short cable; Indicates the first The time history of the normal phase displacement after background compensation for each target measuring point.

[0122] Furthermore, when it is necessary to convert phase displacement into actual physical displacement, the scaling factor can be calculated based on the camera imaging parameters. When the imaging plane is approximately perpendicular to the plane of motion of the short cable, the scaling factor can be expressed as: ; in, Indicates the scaling factor; This indicates the distance from the camera to the target point on the short cable; Indicates pixel size; Indicates the camera's focal length.

[0123] When there is an angle between the camera's optical axis and the plane of motion of the short cable, the scaling factor can be further corrected: ; in, This indicates the corrected scaling factor; This indicates the angle between the camera's optical axis and the plane of motion of the short cable.

[0124] Based on the scaling factor, the fused normal phase displacement time history is converted into the actual normal vibration displacement time history of the short cable: ; in, This represents the time history of the actual normal vibration displacement of the short cable; This represents the pixel displacement time history calculated from the fused phase time history signal.

[0125] Subsequently, frequency domain analysis was performed on the actual normal vibration displacement time history of the short cable, and the dominant frequency of the short cable can be expressed as: ; in, Indicates the effective analysis frequency range.

[0126] The vibration amplitude of the short cable at the dominant frequency can be expressed as: ; in, Indicates the short-range main frequency; This represents the vibration amplitude of the short cable at the dominant frequency.

[0127] Furthermore, when cable force identification is required, the cable force of the short cable can be calculated based on the effective length, linear density, and the identified dominant frequency. In the case of neglecting bending stiffness and approximately hinged boundary support, the first... The cable force corresponding to the first frequency can be expressed as: ; in, Indicates according to the first Cable force calculated from first-order frequency; Indicates the density of short cable lines; Indicates the effective length of the short cable.

[0128] Furthermore, based on actual engineering needs, the vibration displacement time history, vibration spectrum, and cable force identification results of the short cable can be visualized and displayed. The vibration amplitude, vibration frequency, frequency change rate, or cable force change can be compared with preset alarm thresholds. When the corresponding alarm threshold is exceeded, the short cable abnormality warning result is output, providing a basis for short cable operation status assessment and maintenance decision-making.

[0129] In summary, based on the effective measurement point screening completed in step 204, this embodiment performs weighted fusion of multiple target measurement points according to the measurement point confidence level, and completes the conversion of normal phase displacement to actual physical displacement. Furthermore, it extracts vibration parameters such as short cable vibration frequency, vibration amplitude, and cable force, thereby effectively reducing the impact of single measurement point anomalies, local occlusion, and background noise on the detection results, and improving the accuracy and stability of short cable vibration parameter identification.

[0130] This application provides a visual detection system for minor vibrations in short cables, the system comprising: The video acquisition module is used to acquire vibration videos of the short cable; The phase displacement calculation module is used to determine the normal phase displacement time history of each candidate measurement point in the short cable based on the vibration video of the short cable. The candidate measurement points constitute several pairs of candidate measurement points, and each pair of candidate measurement points includes two candidate measurement points located in the two edge regions of the short cable respectively. In at least one frame of the vibration video, the positions of the two candidate measurement points correspond along the normal direction of the short cable. The synchronization score calculation module is used to determine the normal phase synchronization score of each candidate measurement point pair based on the vibration synchronization between the normal phase displacement time histories of the two candidate measurement points in each pair; vibration synchronization is used to characterize the degree of coordination between the two candidate measurement points in terms of vibration frequency, vibration phase, and normal motion direction. The measurement point screening module is used to determine the effective measurement point pairs from each group of candidate measurement point pairs based on the normal phase synchronization score. The vibration parameter calculation module is used to determine the vibration parameters of the short cable based on the normal phase displacement time history of at least one candidate measuring point in the effective measuring point pair.

[0131] Reference Figure 5 As shown, Figure 5 This is a schematic diagram of the structure of an electronic device provided in an optional embodiment of the present invention. This electronic device can be a computer device used to execute the above-described method. Figure 5 As shown, the electronic device includes one or more processors 10, a memory 20, and interfaces for connecting the various components, including high-speed interfaces and low-speed interfaces. The various components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processor can process instructions executed within the electronic device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces).

[0132] The processor 10 may further include a hardware chip. This hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0133] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0134] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the use of the electronic device based on the display of a mini-program landing page. Furthermore, the memory 20 may include high-speed random access memory (RAM), and may also include non-transient memory, such as at least one disk storage device, flash memory device, or other non-transient solid-state storage device. The memory 20 may include volatile memory, such as RAM; the memory may also include non-volatile memory, such as flash memory, hard disk, or solid-state drive; the memory 20 may also include combinations of the above types of memory.

[0135] The electronic device also includes a communication interface 30 for communicating with other devices or communication networks.

[0136] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0137] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0138] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A visual detection method for minute vibrations of short cables, characterized in that, The method includes: Obtain vibration video of the short cable; Based on the vibration video of the short cable, the normal phase displacement time history of each candidate measuring point in the short cable is determined; the candidate measuring points constitute several pairs of candidate measuring points, and each pair of candidate measuring points includes two candidate measuring points located in the bilateral edge regions of the short cable; in at least one frame of the vibration video, the positions of the two candidate measuring points correspond along the normal direction of the short cable. Based on the vibration synchronicity between the normal phase displacement time histories of the two candidate measuring points in each pair of candidate measuring points, the normal phase synchronization score of each pair of candidate measuring points is determined; the vibration synchronicity is used to characterize the degree of coordination between the two candidate measuring points in terms of vibration frequency, vibration phase, and normal motion direction. Based on the normal phase synchronization score, valid measurement point pairs are determined from each group of candidate measurement point pairs. The vibration parameters of the short cable are determined based on the normal phase displacement time history of at least one candidate measuring point in the effective measuring point pair.

2. The visual detection method for minor vibrations of short cables according to claim 1, characterized in that, The methods for determining the candidate measurement point pairs include: Edge detection is performed on the short cable image in the vibration video to obtain the centerline of the short cable and the first edge region and the second edge region located on both sides of the centerline; The axial direction of the short cable is determined based on the centerline, and the direction perpendicular to the axial direction is determined as the normal direction of the short cable. Multiple candidate measurement points are selected in the first edge region and the second edge region, respectively; Two candidate measurement points located in the first edge region and the second edge region respectively, and corresponding to the positions along the normal direction of the short cable, are determined as a pair of candidate measurement points.

3. The visual detection method for minor vibrations of short cables according to claim 2, characterized in that, The two candidate measuring points are located on both sides of the center line, and the distance between the projected positions of the two candidate measuring points in the axial direction of the short cable is less than a preset distance threshold.

4. The visual detection method for minor vibrations of short cables according to claim 1, characterized in that, The step of determining the normal phase displacement time history of each candidate measuring point in the short cable based on the vibration video of the short cable includes: The candidate measurement points and their neighborhoods in the vibration video are filtered to obtain the complex response corresponding to each candidate measurement point. Based on the real and imaginary parts of the complex-valued response, the local phase information of each candidate measurement point is determined; The normal displacement of the candidate measurement point in the normal direction of the short cable is determined based on the local phase change of the same candidate measurement point in adjacent video frames and the gradient of the local phase information in the normal direction of the short cable. The normal phase displacement time history of the candidate measurement point is obtained based on the normal displacement corresponding to each video frame.

5. The visual detection method for minor vibrations of short cables according to claim 1, characterized in that, The step of determining the normal phase synchronization score for each pair of candidate measuring points based on the vibration synchronization between the normal phase displacement time histories of the two candidate measuring points in each pair includes: Based on the normal phase displacement time history of the two candidate measurement points, determine the consistency of the main frequency, the stability of the phase difference, the consistency of the normal motion direction, the rationality of the amplitude ratio, and the prominence of the spectral peak between the two candidate measurement points. The normal phase synchronization score of the candidate measurement point pair is determined based on the main frequency consistency, the phase difference stability, the normal motion direction consistency, the amplitude ratio rationality, and the spectral peak prominence.

6. The visual detection method for minor vibrations of short cables according to claim 5, characterized in that, The step of determining the consistency of the dominant frequency, the stability of the phase difference, the consistency of the normal motion direction, the rationality of the amplitude ratio, and the prominence of the spectral peaks between the two candidate measurement points based on the normal phase displacement time history of the two candidate measurement points includes: Frequency domain analysis was performed on the normal phase displacement time history of the two candidate measurement points to determine the candidate frequencies; The consistency of the main frequency is determined based on the difference between the main frequencies corresponding to the normal phase displacement time histories of the two candidate measurement points at the candidate frequencies. The stability of the phase difference is determined based on the degree of fluctuation of the phase difference between the two candidate measurement points near the candidate frequency within a preset time window; The consistency of the normal motion direction is determined based on the correlation between the normal phase displacement time histories of the two candidate measurement points. The rationality of the amplitude ratio is determined based on the ratio between the vibration amplitudes of the two candidate measurement points at the candidate frequency. The salience of the spectral peak is determined based on the difference between the spectral peak at the candidate frequency and the adjacent noise band.

7. The visual detection method for minute vibrations of short cables according to claim 1, characterized in that, The step of determining valid measurement point pairs from each group of candidate measurement point pairs based on the normal phase synchronization score includes: Based on the normal phase synchronization score of each group of candidate measurement point pairs, the candidate measurement point pairs that meet the preset valid conditions are determined as valid measurement point pairs; The preset valid conditions include: The normal phase synchronization score of the candidate measurement point pair is greater than or equal to a preset scoring threshold; or... The normal phase synchronization score of the candidate measurement point pair is within the preset sorting range of the normal phase synchronization score sorting results of each group of candidate measurement point pairs.

8. The visual detection method for minute vibrations of short cables according to claim 1, characterized in that, The step of determining the vibration parameters of the short cable based on the normal phase displacement time history of at least one candidate measuring point in the effective measuring point pair includes: Based on the normal phase synchronization score of the effective measurement point pair and the mass parameters of each candidate measurement point in the effective measurement point pair, the measurement point confidence of each candidate measurement point is determined; the mass parameters are used to characterize the reliability of the vibration data provided by each candidate measurement point. Based on the confidence level of the measurement point, the target measurement point is determined from the candidate measurement points in at least one set of valid measurement point pairs; The vibration parameters of the short cable are determined based on the normal phase displacement time history of the target measuring point.

9. The visual detection method for minute vibrations of short cables according to claim 8, characterized in that, The step of determining the target measurement point from candidate measurement points in at least one set of valid measurement point pairs based on the measurement point confidence level includes: Based on the confidence level of the measurement point, multiple target measurement points are determined from candidate measurement points in at least one set of valid measurement point pairs; Determining the vibration parameters of the short cable based on the normal phase displacement time history of the target measuring point includes: Based on the measurement point confidence of each target measurement point, the normal phase displacement time history of multiple target measurement points is weighted and fused to obtain the fused vibration time history; The vibration parameters of the short cable are determined based on the fusion vibration time history.

10. A visual detection system for minor vibrations of short cables, used to implement the visual detection method for minor vibrations of short cables according to any one of claims 1 to 9, characterized in that, The system includes: The video acquisition module is used to acquire vibration videos of the short cable; The phase displacement calculation module is used to determine the normal phase displacement time history of each candidate measurement point in the short cable based on the vibration video of the short cable; the candidate measurement points constitute several pairs of candidate measurement points, and each pair of candidate measurement points includes two candidate measurement points located in the bilateral edge regions of the short cable; in at least one frame of the vibration video, the two candidate measurement points are positioned correspondingly along the normal direction of the short cable. The synchronization scoring calculation module is used to determine the normal phase synchronization score of each candidate measuring point pair based on the vibration synchronization between the normal phase displacement time histories of the two candidate measuring points in each pair; the vibration synchronization is used to characterize the degree of coordination between the two candidate measuring points in terms of vibration frequency, vibration phase, and normal motion direction. The measurement point screening module is used to determine the effective measurement point pairs from each group of candidate measurement point pairs based on the normal phase synchronization score. The vibration parameter calculation module is used to determine the vibration parameters of the short cable based on the normal phase displacement time history of at least one candidate measuring point in the effective measuring point pair.

Images