Submarine cable anchor damage detection method and device, electronic equipment and computer medium

Abstract

The application provides a submarine cable anchor damage detection method and device, electronic equipment and computer medium. The submarine cable anchor damage detection method comprises: acquiring target space-time domain data, the target space-time domain data representing phase difference data generated by vibration of the submarine cable in the space-time domain; calculating a mask matrix of the target space-time domain data, and taking the mask matrix as target mask data; obtaining a target mask image based on the target mask data; determining a plurality of target contour images from the target mask image, the plurality of target contour images representing edge distribution of the submarine cable under external force in the space-time domain; obtaining a plurality of morphological feature curves based on the plurality of target contour images, wherein one morphological feature curve matches one target contour image, and the morphological feature curve represents edge features of the submarine cable in the space-time domain; and generating an anchor damage detection result of the submarine cable according to the plurality of morphological feature curves. The application can improve timely alarm of submarine cable anchor damage and avoid submarine cable damage.

Description

Technical Field

[0001] This application relates to the field of submarine cable anchor damage detection technology, and in particular to a submarine cable anchor damage detection method, device, electronic equipment and computer medium. Background Technology

[0002] Submarine optical cables (or simply submarine cables) are vital communication infrastructure supporting international exchange and trade, and are the cornerstone of national participation in global economic and social activities in the digital economy era. Submarine cables carry the enormous demand for intercontinental communication as well as existing intercontinental and island-to-island cross-water communications. Damage to submarine cables would have a severe impact on people's lives and livelihoods.

[0003] Various vessels frequently navigate the sea. The operation of ship anchors can be divided into three main phases: dropping anchor, dragging anchor, and retrieving anchor. All three phases can potentially damage submarine cables. Dropping anchor can strike the cable, causing damage such as cracked outer sheaths, broken optical fibers, and exposed cable cores leading to short circuits and electrical leaks. Dragging anchor refers to the anchor snagging and pulling on the cable while moving on the seabed, which is highly likely to break the cable and has the greatest probability of occurrence and damage. Retrieving anchor refers to the cable breaking during the retrieving process, even if it wasn't broken during the dragging phase. Therefore, anchor intrusion is a major factor in submarine cable damage.

[0004] Currently, all detection methods for anchor damage intrusion are post-incident. This involves checking for leaks in current and voltage, and whether optical fibers are broken, after the submarine cable has been damaged. Post-incident detection of submarine cables suffers from high repair costs and communication disruptions due to the damaged cable itself. Summary of the Invention

[0005] This invention proposes a method, device, electronic equipment, and computer medium for detecting submarine cable anchor damage, in order to solve the problems of high detection difficulty in detecting submarine cables at the time of an incident, and high maintenance costs and communication disruption caused by damage to submarine cables after an incident.

[0006] The technical solution of this invention is a method for detecting anchor damage to submarine cables. The method includes: acquiring target spatiotemporal domain data, wherein the target spatiotemporal domain data characterizes the phase difference data generated by vibration of the submarine cable in the spatiotemporal domain; calculating a mask matrix of the target spatiotemporal domain data and using the mask matrix as target mask data; obtaining a target mask image based on the target mask data, wherein the target mask image is a binary image; determining multiple target contour images from the target mask image, wherein the multiple target contour images characterize the edge distribution of the submarine cable under external force in the spatiotemporal domain; obtaining multiple morphological feature curves based on the multiple target contour images, wherein one morphological feature curve matches one target contour image, and the morphological feature curve characterizes the edge features of the submarine cable in the spatiotemporal domain; and generating an anchor damage detection result for the submarine cable based on the multiple morphological feature curves.

[0007] Compared with related technologies, the embodiments of this application have at least the following advantages: This application calculates a mask matrix of the target's spatiotemporal data to obtain a target mask image that characterizes the feature regions of the submarine cable. Based on the target mask image, multiple target contour images characterizing the edge distribution of the submarine cable are then obtained. By analyzing and processing the target mask image, the edge distribution of the submarine cable can be accurately determined. Finally, multiple morphological feature curves matching the multiple target contour images are analyzed. Based on the edge features within the morphological feature curves, accurate anchor damage detection results are obtained to avoid anchor damage to the submarine cable.

[0008] In some embodiments, calculating the mask matrix of the target spatiotemporal domain data and using the mask matrix as target mask data includes: obtaining mean feature data and variance feature data based on the target spatiotemporal domain data; calculating the mean mask of the mean feature data and using the mean mask as mean mask data; calculating the variance mask of the variance feature data and using the variance mask as variance mask data; and using the mean mask data and the variance mask data as the target mask data.

[0009] In some embodiments, calculating the mean mask of the mean feature data and using the mean mask as mean mask data includes: performing filtering and closing operations on the mean feature data to obtain first mean data; calculating a threshold for the first mean data to obtain a mean threshold; and using the mean threshold to binarize the mean feature data to obtain the mean mask data. Calculating the variance mask of the variance feature data and using the variance mask as variance mask data includes: performing filtering and closing operations on the variance feature data to obtain first variance data; calculating a threshold for the first variance data to obtain a variance threshold; and using the variance threshold to binarize the variance feature data to obtain the variance mask data.

[0010] In some embodiments, determining multiple target contour images from the target mask image includes: determining multiple initial edge contours from the target mask image; calculating the shortest distance between any two of the multiple initial edge contours; detecting whether the shortest distance is less than a shortest distance threshold, and merging two initial edge contours when the shortest distance is detected to be less than the shortest distance threshold; repeating the above process of calculating the shortest distance between any two of the multiple initial edge contours and detecting whether the shortest distance is less than the shortest distance threshold until the shortest distance between all two initial edge contours within the multiple initial edge contours has been calculated, thereby obtaining the multiple target contour images.

[0011] In some embodiments, after acquiring the target spatiotemporal domain data, the method further includes: acquiring target stress data, wherein both the target stress data and the target spatiotemporal domain data are two-dimensional matrices, and the target stress data and the target spatiotemporal domain data have the same dimension; the target stress data characterizes the stress value generated by the submarine cable under external influence in the spatiotemporal domain; calculating the mask matrix of the target spatiotemporal domain data and using the mask matrix as target mask data includes: obtaining mean feature data and variance feature data based on the target spatiotemporal domain data; calculating the mean mask of the mean feature data to obtain mean mask data; calculating the variance mask of the variance feature data to obtain variance mask data; calculating the stress mask of the target stress data to obtain stress mask data; and obtaining the target mask data based on the mean mask data, the variance mask data, and the stress mask data.

[0012] In some embodiments, obtaining multiple morphological feature curves based on multiple target contour images includes: calculating the contour feature value of each target contour image at second preset time intervals, wherein each target contour image corresponds to multiple contour feature values; and obtaining the morphological feature curve matching the target contour image based on the multiple contour feature values ​​corresponding to the target contour image.

[0013] In some embodiments, the anchor damage detection result includes an anchor damage alarm result; generating the anchor damage detection result of the submarine cable based on multiple morphological feature curves includes: obtaining the minimum contour feature value corresponding to each morphological feature curve within a preset detection time, wherein one minimum contour feature value corresponds to one morphological feature curve; if any minimum contour feature value is greater than a contour threshold, generating the anchor damage alarm result.

[0014] One embodiment of this application also provides a submarine cable anchor damage detection device, comprising: a first acquisition module for acquiring target spatiotemporal domain data, wherein the target spatiotemporal domain data characterizes the phase difference data generated by vibration of the submarine cable in the spatiotemporal domain; a calculation module for calculating the mask matrix of the target spatiotemporal domain data and using the mask matrix as target mask data; a conversion module for obtaining a target mask image based on the target mask data, wherein the target mask image is a binary image; a determination module for determining multiple target contour images from the target mask image, wherein the multiple target contour images characterize the edge distribution of the submarine cable under external force in the spatiotemporal domain; a second acquisition module for obtaining multiple morphological feature curves based on the multiple target contour images, wherein one morphological feature curve matches one target contour image, and the morphological feature curve characterizes the edge features of the submarine cable in the spatiotemporal domain; and a processing module for generating anchor damage detection results of the submarine cable based on the multiple morphological feature curves.

[0015] One embodiment of this application also provides an electronic device, including a processor and a memory, wherein the memory is used to store instructions, and the processor is used to call the instructions in the memory to cause the electronic device to perform the submarine cable anchor damage detection method as described above.

[0016] One embodiment of this application also provides a computer-readable storage medium that stores computer instructions that, when executed on an electronic device, cause the electronic device to perform the above-described submarine cable anchor damage detection method.

[0017] Compared with existing technologies, the above-mentioned submarine cable anchor damage detection method, electronic equipment, and computer storage medium firstly calculate the target spatiotemporal data to obtain mean and variance feature data that better reflect the characteristics of the submarine cable. Then, a mask based on the mean and variance feature data is calculated to obtain a target mask image that characterizes the feature regions of the submarine cable. Next, based on the target mask image, multiple initial edge contours matching the target mask image are calculated. Two initial edge contours that meet the merging requirements are combined to obtain multiple target contour images characterizing the edge distribution of the submarine cable. By analyzing and processing the target mask images, the edge distribution of the submarine cable can be accurately determined. Finally, multiple morphological feature curves matching the multiple target contour images are analyzed. Based on the edge features within the morphological feature curves, accurate anchor damage detection results are obtained, allowing for timely alarms against anchor damage and thus preventing damage to the submarine cable. Attached Figure Description

[0018] Figure 1 This is a flowchart of the steps of a method for detecting anchor damage to submarine cables according to one embodiment of this application.

[0019] Figure 2 It shows Figure 1 A schematic diagram of the morphological characteristic curves in the image.

[0020] Figure 3 This is a flowchart illustrating the steps of creating a target mask image according to one embodiment of this application.

[0021] Figure 4 This is a flowchart illustrating another step of a target mask image according to one embodiment of this application.

[0022] Figure 5 This is a schematic diagram of the structure of a submarine cable anchor damage detection device according to one embodiment of this application.

[0023] Figure 6 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application.

[0024] Explanation of main component symbols The following detailed description, in conjunction with the accompanying drawings, will further illustrate this application. Detailed Implementation

[0025] To better understand the above-mentioned objectives, features, and advantages of this application, the application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0026] The following description sets forth many specific details to provide a full understanding of this application. The described embodiments are only some, not all, of the embodiments of this application.

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the specification of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application.

[0028] It should be further noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0029] In this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and drawings of this application are used to distinguish similar objects, not to describe a specific order or sequence.

[0030] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0031] The method for detecting cable anchor damage in this application is set in a cable anchor damage detection system, which can be applied to one or more electronic devices or installed in other devices that are communicatively connected to the electronic devices.

[0032] like Figure 1 The diagram shown is a flowchart illustrating the steps of an embodiment of the submarine cable anchor damage detection method of this application. Depending on different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted.

[0033] See Figure 1As shown, the method for detecting anchor damage to submarine cables may include the following steps.

[0034] Step S10: Obtain target spatiotemporal domain data.

[0035] In some embodiments, the target spatiotemporal domain data characterizes the phase difference data generated by the vibration of the submarine cable in the spatiotemporal domain. Furthermore, the target spatiotemporal domain data is two-dimensional phase difference data after noise removal. In this embodiment, the target spatiotemporal domain data is denoted as T(t1,s1), where t1 is the number of time points and s1 is the number of spatial points. Simultaneously, t1 and s1 are also the row and column numbers of a two-dimensional matrix, and the phase difference data represents the vibration signal.

[0036] To obtain the target spatiotemporal domain data, the following steps are included: (1) Acquiring initial spatiotemporal domain data. In this embodiment, a distributed fiber optic acoustic sensing vibration monitoring device can be used to acquire initial spatiotemporal domain data based on a time sampling rate. The time sampling rate is denoted as sr Hz, and the initial spatiotemporal domain data is denoted as H(t1,s1). The time sampling rate can be 1500Hz, 1200Hz, or 2000Hz. This application does not limit the specific value of the time sampling rate.

[0037] The initial spatiotemporal domain data also characterizes the phase difference data of the submarine cable's vibration caused by external influences in the spatiotemporal domain. Specifically, the initial spatiotemporal domain data includes the phase difference data of the submarine cable's vibration caused by the external force of the ship's anchor and the phase difference data of vibration caused by noise such as seawater erosion, ship noise, and submarine earthquakes. In other embodiments, the initial spatiotemporal domain data may also include other phase difference data, and the initial spatiotemporal domain data may also be obtained using devices other than distributed fiber optic acoustic sensing vibration monitoring equipment.

[0038] In this embodiment, the vibration signal typically acquired by the distributed fiber optic acoustic sensing vibration monitoring device is one-dimensional, that is, a one-dimensional vibration signal that changes with time at a single spatial point. Specifically, it represents the disturbance of the probe light phase by the vibration source. In order to unify the vibration signal with the electrical signal, the DC component of the phase change data of the vibration signal is subtracted to obtain the final phase change data (which presents an alternating positive and negative form).

[0039] Furthermore, to obtain richer data, one-dimensional vibration signals from multiple spatial points can be combined into a two-dimensional vibration signal. This two-dimensional vibration signal can be represented as a two-dimensional grayscale image, where one dimension represents time and the other represents spatial location. That is, the two-dimensional data corresponding to the grayscale image is the initial spatiotemporal domain data. Using spatiotemporal domain vibration data can spatially expand the one-dimensional time-domain vibration signal, enhancing its information-representing capability.

[0040] (2) Perform Fourier transform and inverse transform on the initial spatiotemporal domain data to obtain the target spatiotemporal domain data.

[0041] Specifically, a two-dimensional Fourier transform is used to calculate the two-dimensional spectrum of the initial spatiotemporal domain data. Frequency data in the two-dimensional spectrum below a preset frequency threshold are set to 0. Then, an inverse transform technique is used to process the initial spatiotemporal domain data after the two-dimensional Fourier transform to obtain the target spatiotemporal domain data. In this embodiment, the preset frequency threshold can be 8 Hz, 10 Hz, 12 Hz, or 15 Hz. This application does not limit the specific value of the preset frequency threshold; it can be set according to actual processing requirements.

[0042] For example, suppose the initial spatiotemporal domain data is H(15000, 1700), the time sampling rate is 1500Hz, and the preset frequency threshold is 8Hz. A two-dimensional Fourier transform is used to calculate the two-dimensional spectrum of H(15000, 1700). Frequency data below 8Hz in the two-dimensional spectrum are set to 0. Then, an inverse transform technique is used to process the initial spatiotemporal domain data after the two-dimensional Fourier transform, resulting in the target spatiotemporal domain data T(15000, 1700).

[0043] By using Fourier transform and inverse transform to process the initial spatiotemporal domain data, noise data within the initial spatiotemporal domain data can be removed, resulting in phase difference data that accurately reflects the vibration of the submarine cable caused by external influences in the spatiotemporal domain.

[0044] Step S20: Calculate the mask matrix of the target spatiotemporal domain data and use the mask matrix as the target mask data.

[0045] In this embodiment, the Otsu binarization method is used to process the target spatiotemporal domain data to obtain target mask data. The specific processing steps are described in detail later and will not be repeated here.

[0046] Otsu's binarization method can convert grayscale images into binary images. For example, pixels with grayscale values ​​greater than a certain threshold are set as grayscale maxima, and pixels with grayscale values ​​less than this threshold are set as grayscale minima, thus achieving binarization.

[0047] Step S30: Based on the target mask data, obtain the target mask image, which is a binarized image.

[0048] In this embodiment, the target mask data is converted into a target mask image. There are many ways to convert data into an image, and all of these are existing technologies; this application does not limit the conversion method.

[0049] Step S40: Determine multiple target contour images from the target mask image. The multiple target contour images represent the edge distribution of the submarine cable in the spatiotemporal domain when it is subjected to external force.

[0050] Specifically, multiple initial edge contours are determined from the target mask image. The shortest distance between any two initial edge contours is calculated. It is then checked whether the shortest distance is less than a shortest distance threshold; if so, the two initial edge contours are merged. This process of calculating the shortest distance between any two initial edge contours and checking whether the shortest distance is less than the threshold is repeated until the shortest distances between all pairs of initial edge contours within the multiple initial edge contours have been calculated, resulting in multiple target contour images.

[0051] Assume there are five initial edge contours, labeled H1, H2, H3, H4, and H5. Calculate the shortest distances between H1 and H2, H3, H4 and H5; between H2 and H3; between H4 and H5; between H3 and H4 and H5; and between H4 and H5. Then, compare each of these shortest distances with a shortest distance threshold. Finally, merge the two initial edge contours whose shortest distances are less than the shortest distance threshold to obtain multiple target contour images. For example, if the shortest distances between H1 and H2, and H3 and H5 are both less than the shortest distance threshold, then merge H1 and H2, and merge H3 and H5. Ultimately, three target contour images are obtained: the image after merging H1 and H2, the image after merging H3 and H5, and the image corresponding to H4.

[0052] In this embodiment, the shortest distance between two initial edge contours is calculated using the Euclidean distance method. For example, the Euclidean distance between any point in one initial edge contour and any point in another initial edge contour is calculated. The shortest Euclidean distance obtained is then taken as the shortest distance between the two initial edge contours.

[0053] It should be noted that this application does not limit the use of Euclidean distance to calculate the shortest distance between two initial edge contours. In other embodiments, other methods may also be used to calculate the shortest distance between two initial edge contours.

[0054] In this embodiment, the `findContours()` function in OpenCV is used to process the target mask image to obtain multiple initial edge contours. The area of ​​each initial edge contour is calculated. If the area of ​​an initial edge contour is less than a preset area threshold, it is removed from the multiple initial edge contours. Alternatively, the calculated areas are sorted from largest to smallest, and multiple initial edge contours matching a preset extraction number of areas are extracted. The preset area threshold and preset extraction number are set according to actual conditions. For example, the preset extraction number can be set to 50, 60, or 80.

[0055] First, the target mask image is processed using the `findContours()` function in OpenCV to filter out multiple initial edge contours. Then, Euclidean distance is used to calculate the shortest distance between two initial edge contours. This method can reduce the time spent calculating the shortest distance between two contours, thereby improving the efficiency of the entire submarine cable detection.

[0056] Step S40: Based on multiple target contour images, obtain multiple morphological feature curves, wherein one morphological feature curve is matched with one target contour image.

[0057] In some embodiments, the morphological feature curve characterizes the edge features of the submarine cable in the spatiotemporal domain. To obtain morphological feature curves matching multiple target contour images, the contour feature values ​​of each target contour image are calculated sequentially at second preset time intervals, with each target contour image corresponding to multiple contour feature values. Then, based on the multiple contour feature values ​​corresponding to the target contour images, the morphological feature curve matching the target contour images is obtained. One target contour image corresponds to one morphological feature curve.

[0058] The contour feature values ​​include contour width, contour area, contour centroid, or contour perimeter. For example, the contour width can be calculated using the boundingRect() feature calculation function in OpenCV. In other embodiments, other functions or methods can also be used to calculate the contour width, and this application is not limited in this regard.

[0059] In this embodiment, as Figure 2 The morphological feature curves shown represent the contour feature image with the largest contour area among multiple target contour images, where the contour feature value is the contour width. The horizontal axis of the morphological feature curve represents time information, and the vertical axis represents the specific numerical value of the contour width.

[0060] Step S50: Generate anchor damage detection results for submarine cables based on multiple morphological feature curves.

[0061] In some embodiments, the anchor damage detection results include anchor damage alarm results and anchor damage no-alarm results. To provide timely warnings of anchor damage events, it is necessary to obtain the minimum contour feature value corresponding to each morphological feature curve within a preset detection time period, where each minimum contour feature value corresponds to one morphological feature curve. If any minimum contour feature value is greater than a contour threshold, an anchor damage alarm result is generated.

[0062] Specifically, a minimum contour feature value is determined within a preset detection time for each morphological feature curve. Since each morphological feature curve corresponds to one minimum contour feature value within the preset detection time, multiple minimum contour feature values ​​can be obtained. These multiple minimum contour feature values ​​are compared with a contour threshold. If any minimum contour feature value is greater than the contour threshold, an anchor damage alarm result is generated. The submarine cable anchor damage detection system controls the alarm device to sound an alarm, alerting engineers to handle the situation promptly. If all minimum contour feature values ​​are less than the contour threshold, an anchor damage no-alarm result is generated.

[0063] Furthermore, while generating an anchor damage alarm, the submarine cable anchor damage detection system will also output the timestamp and spatial location information of the anchor damage event. The system can also communicate with external warning devices to issue alarms. These external warning devices include, but are not limited to, external displays, mobile phones, computers, or speakers.

[0064] Compared to existing technologies, the embodiments of this application, in order to obtain phase difference data that better reflects the vibration of the submarine cable caused by external influences in the spatiotemporal domain, perform Fourier transform and inverse transform on the collected initial spatiotemporal domain data to remove low-frequency noise data and obtain accurate target spatiotemporal domain data. Then, a mask of the target spatiotemporal domain data is calculated to obtain a target mask image that accurately reflects the edge region of the submarine cable. This facilitates the acquisition of multiple target contour images of the submarine cable. Finally, the morphological feature curves matched by multiple target contour images are analyzed, and based on the different contour feature values ​​corresponding to each morphological feature curve at different time points, timely warnings of anchor damage events are provided to avoid damage to the submarine cable caused by untimely warnings of anchor damage events.

[0065] Please refer to the following: Figure 3 This is a flowchart illustrating a step in creating a target mask image according to an embodiment of this application. The specific details are as follows: Step S201: Obtain mean feature data and variance feature data based on the target spatiotemporal domain data.

[0066] In some embodiments, data for all spatial points in the target spatiotemporal domain data are calculated at first preset time intervals, based on data from all time points in the target spatiotemporal domain data. The first preset time is denoted as ta. The mean feature data is calculated using the following formula: Where sr is the time sampling rate mentioned above, For target spatiotemporal domain data. This is data with mean characteristics. The variance characteristic data can be calculated using the following formula: in, These are variance characteristic data. The mean value of the target spatiotemporal domain data. The second preset time is greater than the first preset time. Step S202: Filter and close the mean feature data and variance feature data respectively to obtain the first mean data and the first variance data.

[0067] Specifically, the mean feature data is filtered and closed to obtain the first mean data. The variance feature data is then filtered and closed to obtain the first variance data.

[0068] In this embodiment, median filtering is applied to both the mean and variance feature data to remove salt-and-pepper noise. Median filtering is a non-linear smoothing technique that sets the grayscale value of each pixel to the median of the grayscale values ​​of all pixels within a neighborhood window. The median filtering method uses a two-dimensional sliding template of a certain structure to sort the pixels within the template according to their pixel values, generating a monotonically increasing (or decreasing) two-dimensional data sequence. In other embodiments, other methods can also be used to filter the mean and variance feature data to remove salt-and-pepper noise.

[0069] The mean and variance feature data after removing salt-and-pepper noise are then processed using a closing operation to obtain the first mean and first variance data. The closing operation is a mathematical morphology operation aimed at filling small holes while preserving the overall shape and contour features of the image. This process is achieved by first performing a dilation operation on the image, followed by an erosion operation. Since median filtering and the closing operation are existing technologies, this application will not provide a detailed explanation of median filtering and the closing operation.

[0070] Step S203: Calculate the thresholds for the first mean data and the first variance data respectively to obtain the mean threshold and the variance threshold.

[0071] Specifically, calculate the threshold for the first mean data to obtain the mean threshold. Calculate the threshold for the first variance data to obtain the variance threshold.

[0072] In this embodiment, the threshold for the first mean data and the threshold for the first variance data are calculated using the Otsu binarization method. The threshold corresponding to the first mean data is used as the mean threshold, and the threshold corresponding to the first variance data is used as the variance threshold.

[0073] Furthermore, the mean threshold and variance threshold are compared with preset thresholds respectively. If the mean threshold or variance threshold is less than the preset threshold, then the mean threshold or variance threshold is set to the preset threshold.

[0074] Step S204: Use the mean threshold to binarize the mean feature data to obtain the mean mask data.

[0075] In some embodiments, data in the mean feature data that are greater than or equal to the mean threshold are set to 1 or 255, and data in the mean feature data that are less than the mean threshold are set to 0. The mean feature data after processing with the mean threshold is recorded as mean mask data.

[0076] Step S205: Use a variance threshold to binarize the variance feature data to obtain variance mask data.

[0077] In some embodiments, data in the variance feature data that are greater than or equal to the variance threshold are set to 1 or 255, and data in the variance feature data that are less than the variance threshold are set to 0. The variance feature data after processing with the variance threshold is denoted as variance mask data.

[0078] In this embodiment, a preset threshold is set. When there is no phase difference data of the submarine cable vibrating due to external influence in the initial spatiotemporal domain data collected, the problem of excessively large mask areas in the mean mask image or variance mask image caused by the mean threshold or variance threshold calculated by the Otsu binarization method can be avoided, which would affect the accuracy of the anchor damage detection results.

[0079] Step S206: Use the mean mask data and variance mask data as the target mask data.

[0080] In some embodiments, bitwise AND operations are used to process the mean mask data and variance mask data to obtain the target mask data.

[0081] Compared to existing technologies, this embodiment uses filtering and closing operations to process the mean and variance feature data, removing salt-and-pepper noise. This makes the processed data more reflective of the submarine cable's information. The processed mean and variance feature data are then binarized to obtain mask data reflecting the feature regions of the submarine cable. These feature regions are the areas where the submarine cable is located. Based on this target mask data, a target mask image is obtained, facilitating subsequent image processing and analysis to achieve accurate submarine cable anchor damage detection results.

[0082] Please refer to the following: Figure 4 This is another flowchart illustrating the steps involved in creating a target mask image according to an embodiment of this application. The details are as follows: Step S211: Obtain target spatiotemporal domain data and target stress data.

[0083] In some embodiments, the target stress data and the target spatiotemporal domain data have the same dimension, and the target stress data characterizes the stress value of the submarine cable caused by external influences in the spatiotemporal domain.

[0084] In this embodiment, a stress monitoring device can be used to collect target stress data. For example, the stress monitoring device includes a spontaneously distributed Brillouin fiber sensor and a stimulated distributed Brillouin fiber sensor. In other embodiments, other components may also be used to collect target stress data. This application is not limited in this regard.

[0085] Furthermore, to ensure that the dimensionality of the target stress data is the same as that of the target spatiotemporal domain data, linear interpolation or spline interpolation can be used to collect the target stress data. Linear interpolation refers to an interpolation method where the interpolation function is a first-order polynomial, resulting in zero interpolation error at the interpolation nodes. Compared to other interpolation methods, such as parabolic interpolation, linear interpolation is simple and convenient. Spline interpolation fits the data through a series of interpolation segments, generating smoother new samples between minority class samples.

[0086] Step S212: Obtain mean feature data and variance feature data based on the target spatiotemporal domain data.

[0087] Step S213: Calculate the masks for the mean feature data, variance feature data, and target stress data respectively to obtain the mean mask data, variance mask data, and stress mask data.

[0088] Specifically, the mean mask of the mean feature data is calculated to obtain the mean mask data; the variance mask of the variance feature data is calculated to obtain the variance mask data; and the stress mask of the target stress data is calculated to obtain the stress mask data.

[0089] In this embodiment, the method for calculating the mask of the target stress data and obtaining the stress mask image is the same as described above, and will not be repeated here.

[0090] Step S214: Obtain target mask data based on mean mask data, variance mask data, and stress mask data.

[0091] In some embodiments, bitwise AND operations are used to process the mean mask data, variance mask data, and stress mask data to obtain the target mask data.

[0092] Compared to existing technologies, this embodiment adds target stress data that reflects the stress values ​​of the submarine cable caused by external influences. Then, mean mask data, variance mask data, and stress mask data are calculated separately. These three types of mask data are fused to obtain target mask data. Furthermore, this target mask data more comprehensively reflects the characteristic information of the submarine cable. Compared to the above embodiments, this embodiment enables more accurate anchor damage detection results after subsequent image processing and analysis of the target mask image matched with the target mask data.

[0093] In some embodiments, please refer to Figure 5 This application also discloses a submarine cable anchor damage detection device 200. The submarine cable anchor damage detection device 200 includes a first acquisition module 210, a calculation module 220, a conversion module 230, a determination module 240, a second acquisition module 250, and a processing module 260.

[0094] The first acquisition module 210 acquires target spatiotemporal domain data, which represents the phase difference data generated by vibration of the submarine cable in the spatiotemporal domain. The calculation module 220 calculates the mask matrix of the target spatiotemporal domain data and uses the mask matrix as target mask data. The conversion module 230 obtains a target mask image based on the target mask data; the target mask image is a binarized image. The determination module 240 determines multiple target contour images from the target mask image; these multiple target contour images represent the edge distribution of the submarine cable under external force in the spatiotemporal domain. The second acquisition module 250 obtains multiple morphological feature curves based on the multiple target contour images, wherein one morphological feature curve matches one target contour image; the morphological feature curve represents the edge features of the submarine cable in the spatiotemporal domain. The processing module 250 generates anchor damage detection results for the submarine cable based on the multiple morphological feature curves.

[0095] Please refer to Figure 6 This is a schematic diagram of the hardware structure of the electronic device 1000 provided in an embodiment of this application. Figure 6 As shown, the electronic device 1000 may include a processor 1001 and a memory 1002. The memory 1002 is used to store one or more computer programs 1003. The one or more computer programs 1003 are configured to be executed by the processor 1001. The one or more computer programs 1003 include instructions that can be used to implement the methods described above in the electronic device 1000.

[0096] It is understood that the structure illustrated in this embodiment does not constitute a specific limitation on the electronic device 1000. In other embodiments, the electronic device 1000 may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements.

[0097] Processor 1001 may include one or more processing units, such as: application processor (AP), modem, graphics processing unit (GPU), image signal processor (ISP), controller, video codec, digital signal processor (DSP), baseband processor, and / or neural network processing unit (NPU), etc. Different processing units may be independent devices or integrated into one or more processors.

[0098] The processor 1001 may also include a memory for storing instructions and data. In some embodiments, the memory in the processor 1001 is a cache memory. This memory can store instructions or data that the processor 1001 has just used or that are used repeatedly. If the processor 1001 needs to use the instruction or data again, it can retrieve it directly from this memory. This avoids repeated accesses, reduces the waiting time of the processor 1001, and thus improves the efficiency of the system.

[0099] In some embodiments, the processor 1001 may include one or more interfaces. Interfaces may include an inter-integrated circuit (I2C) interface, an inter-integrated circuit sound (I2S) interface, a pulse code modulation (PCM) interface, a universal asynchronous receiver / transmitter (UART) interface, a mobile industry processor interface (MIPI), a general-purpose input / output (GPIO) interface, a SIM interface, and / or a USB interface, etc.

[0100] In some embodiments, memory 1002 may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0101] This embodiment also provides a computer-readable storage medium storing computer instructions. When the instructions are executed on an electronic device, the electronic device performs the aforementioned method steps to implement the methods described in the above embodiments.

[0102] In this embodiment, the electronic device and computer storage medium are used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects of the corresponding methods provided above, and will not be repeated here.

[0103] In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0104] In the several embodiments provided in this application, the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are illustrative. For instance, the division of modules or units is a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0105] The unit described as a separate component may or may not be physically separate. The component shown as a unit can be one physical unit or multiple physical units, that is, it can be located in one place or distributed in multiple different places. Some or all of the units can be selected to achieve the purpose of the solution in this embodiment according to actual needs.

[0106] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0107] If the integrated unit is implemented as a software functional unit and sold or used as an independent printed object, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software printed object. This software printed object is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0108] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the scope of protection of this application.

Claims

1. A method for detecting anchor damage to submarine cables, characterized in that, The method includes: Acquire target spatiotemporal domain data, which characterizes the phase difference data generated by the vibration of the submarine cable in the spatiotemporal domain; Calculate the mask matrix of the target spatiotemporal domain data, and use the mask matrix as the target mask data; Based on the target mask data, a target mask image is obtained, and the target mask image is a binarized image; Multiple target contour images are determined from the target mask image, and the multiple target contour images represent the edge distribution of the submarine cable in the spatiotemporal domain when it is subjected to external force. Based on multiple target contour images, multiple morphological feature curves are obtained, wherein one morphological feature curve is matched with one target contour image. The morphological feature curve is obtained by calculating the change relationship of the contour feature value of each target contour image over time at preset time intervals. The contour feature value includes at least one of contour width, contour area, contour centroid and contour perimeter. Based on multiple morphological feature curves, the anchor damage detection results of the submarine cable are generated.

2. The method for detecting submarine cable anchor damage as described in claim 1, characterized in that, The calculation of the mask matrix of the target spatiotemporal domain data, using the mask matrix as the target mask data, includes: Based on the target spatiotemporal domain data, mean feature data and variance feature data are obtained; The mean feature data is filtered and closed to obtain first mean data. The threshold of the first mean data is calculated to obtain the mean threshold. The mean feature data is then binarized using the mean threshold to obtain mean mask data. The variance feature data is filtered and closed to obtain first variance data. The threshold of the first variance data is calculated to obtain the variance threshold. The variance feature data is then binarized using the variance threshold to obtain variance mask data. The mean mask data and the variance mask data are used as the target mask data.

3. The method for detecting submarine cable anchor damage as described in claim 1, characterized in that, Determining multiple target contour images from the target mask image includes: Multiple initial edge contours are determined from the target mask image; Calculate the shortest distance between any two of the multiple initial edge contours; Detect whether the shortest distance is less than the shortest distance threshold, and when the shortest distance is detected to be less than the shortest distance threshold, merge the two initial edge contours; Repeat the above process of calculating the shortest distance between any two initial edge contours in the plurality of initial edge contours, and detecting whether the shortest distance is less than the shortest distance threshold, until the shortest distance between all two initial edge contours in the plurality of initial edge contours has been calculated, and then obtain the plurality of target contour images.

4. The method for detecting submarine cable anchor damage as described in claim 1, characterized in that, After acquiring the target spatiotemporal domain data, the method further includes: Acquire target stress data, wherein the target stress data and the target spatiotemporal domain data are both two-dimensional matrices, and the target stress data and the target spatiotemporal domain data have the same dimension. The target stress data characterizes the stress value generated by the submarine cable under external influence in the spatiotemporal domain. The calculation of the mask matrix of the target spatiotemporal domain data, using the mask matrix as the target mask data, includes: Based on the target spatiotemporal domain data, mean feature data and variance feature data are obtained; The mean feature data is filtered and closed to obtain first mean data. The threshold of the first mean data is calculated to obtain the mean threshold. The mean feature data is then binarized using the mean threshold to obtain mean mask data. The variance feature data is filtered and closed to obtain first variance data. The threshold of the first variance data is calculated to obtain the variance threshold. The variance feature data is then binarized using the variance threshold to obtain variance mask data. The target stress data is filtered and closed to obtain first stress data. The threshold of the first stress data is calculated to obtain the stress threshold. The target stress data is then binarized using the stress threshold to obtain stress mask data. The target mask data is obtained based on the mean mask data, the variance mask data, and the stress mask data.

5. The method for detecting submarine cable anchor damage as described in claim 1, characterized in that, The process of obtaining multiple morphological feature curves based on multiple target contour images includes: Obtain the contour feature value of each target contour image calculated at each preset time interval, wherein each target contour image corresponds to multiple contour feature values; Based on the multiple contour feature values ​​corresponding to the target contour image, the morphological feature curve matching the target contour image is obtained.

6. The method for detecting submarine cable anchor damage as described in claim 1, characterized in that, The anchor damage detection results include anchor damage alarm results; the generation of anchor damage detection results for the submarine cable based on multiple morphological feature curves includes: Obtain the minimum contour feature value corresponding to each morphological feature curve within a preset detection time, wherein one minimum contour feature value corresponds to one morphological feature curve; If any of the minimum contour feature values ​​is greater than the contour threshold, the anchor damage alarm result is generated.

7. A device for detecting anchor damage to submarine cables, characterized in that, include: First acquisition module: used to acquire target spatiotemporal domain data, wherein the target spatiotemporal domain data characterizes the phase difference data generated by the vibration of the submarine cable in the spatiotemporal domain; Calculation module: used to calculate the mask matrix of the target spatiotemporal domain data, and use the mask matrix as the target mask data; Conversion module: used to obtain a target mask image based on the target mask data, wherein the target mask image is a binarized image; Determination module: used to determine multiple target contour images from the target mask image, wherein the multiple target contour images characterize the edge distribution of the submarine cable in the spatiotemporal domain when it is subjected to external force; The second acquisition module is used to obtain multiple morphological feature curves based on multiple target contour images, wherein one morphological feature curve is matched with one target contour image. The morphological feature curve is obtained by calculating the change relationship of the contour feature value of each target contour image with time at preset intervals. The contour feature value includes at least one of contour width, contour area, contour centroid and contour perimeter. Processing module: used to generate anchor damage detection results for the submarine cable based on multiple morphological feature curves.

8. An electronic device, characterized in that, The electronic device includes a processor and a memory, the memory being used to store instructions, and the processor being used to call the instructions in the memory, causing the electronic device to execute the submarine cable anchor damage detection method according to any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when executed on an electronic device, cause the electronic device to perform the submarine cable anchor damage detection method as described in any one of claims 1 to 6.

Images