Acoustic wave-based optical cable external damage prevention method and system, electronic device, and storage medium

Abstract

The application relates to the field of optical cable operation and maintenance, in particular to an optical cable external damage prevention method and system based on sound waves, an electronic device and a storage medium, which comprises the following steps: acquiring an optical cable sound wave sequence diagram corresponding to optical cable data; marking external damage features of the optical cable sound wave sequence diagram through a pre-trained sound wave feature marking model; determining an external damage event type in the optical cable data according to the marked external damage features; cutting the marked external damage features in the optical cable sound wave sequence diagram according to the external damage event type to obtain a cutting feature segment of the external damage event type; and processing the cutting feature segment through a pre-trained feature recognition model according to the external damage event type to obtain external damage event information of the external damage event. Compared with the prior art, the application cuts out the segment containing the external damage features in the optical cable sound wave sequence diagram for recognition, can effectively reduce the interference of noise, and further improves the precision of external damage event information recognition and effectively prevents and controls the external damage event.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of optical cable operation and maintenance, and more particularly to an optical cable anti-external damage method and system based on sound waves, an electronic device and a storage medium. BACKGROUND

[0002] With the continuous development of optical cable technology, the range of optical cable laying is increasing, and optical cables have been widely used in key infrastructure in the fields of communication, power and transportation. However, since the optical cable is mostly laid in complex environments such as underground, pipeline or open field, it is easily damaged by external construction, excavation, rolling and other human or natural factors, causing communication interruption and even causing major safety accidents. In order to ensure the safe operation of the optical cable, timely and accurate monitoring of external damage events has become an important issue in operation and maintenance.

[0003] In the prior art, the monitoring of the external damage event of the optical cable can be obtained by the sound wave information received by the optical cable, but in the daily use process of the optical cable, since the optical cable is mostly laid in urban environment, due to human activities in urban environment, the optical cable receives a lot of noise, which will affect the judgment of the external damage event and increase the difficulty of external damage event prevention.

[0004] Therefore, there is an urgent need for a method that can effectively reduce the influence of optical cable external damage event noise. SUMMARY

[0005] The present application aims to overcome at least one of the above-mentioned defects (shortcomings) of the prior art, and provides an optical cable anti-external damage method and system based on sound waves, an electronic device and a storage medium, which can effectively reduce the influence of external damage event noise and improve the recognition accuracy of external damage events.

[0006] According to a first aspect of the present application, an optical cable anti-external damage method based on sound waves is provided, the method comprising:

[0007] collecting optical cable data of the optical cable, and obtaining a corresponding optical cable sound wave sequence graph according to the optical cable data;

[0008] labeling external damage features of the optical cable sound wave sequence graph by using a pre-trained sound wave feature labeling model;

[0009] determining the type of external damage event in the optical cable data according to the labeled external damage features; the type of external damage event is divided into transient event type and periodic event type;

[0010] cutting the labeled external damage features in the optical cable sound wave sequence graph according to the type of external damage event, to obtain a cutting feature segment corresponding to the type of external damage event;

[0011] According to the outer damage event type, a pre-trained feature recognition model is selected to process the cut feature segment, and outer damage event information of the outer damage event is obtained.

[0012] Optionally, the determination of the outer damage event type in the optical cable data according to the marked outer damage feature specifically includes:

[0013] An average value of intervals between outer damage features in the optical cable acoustic wave sequence diagram is obtained; the interval between the outer damage features is a time interval between two continuous outer damage features marked in the optical cable acoustic wave sequence diagram;

[0014] If the average value is greater than or equal to a preset interval threshold, the outer damage event type in the optical cable data is the transient event type;

[0015] If the average value is less than the preset interval threshold, the outer damage event type in the optical cable data is the periodic event type.

[0016] Optionally, the marking of the outer damage features in the optical cable acoustic wave sequence diagram by the pre-trained acoustic feature marking model specifically includes:

[0017] The region containing the outer damage features in the optical cable acoustic wave sequence diagram is obtained by the pre-trained acoustic feature marking model;

[0018] The region containing the outer damage features in the optical cable acoustic wave sequence diagram is framed out by an adaptive window.

[0019] Optionally, the cutting of the marked outer damage features in the optical cable acoustic wave sequence diagram according to the outer damage event type to obtain the cut feature segment corresponding to the outer damage event type specifically includes:

[0020] For the optical cable acoustic wave sequence diagram of the transient event type:

[0021] According to a size of a largest window marked by each outer damage feature of the transient event type, a size of a first cutting window is determined;

[0022] According to the first cutting window, each outer damage feature is cut from the optical cable acoustic wave sequence diagram of the transient event type, and a plurality of first feature segments are obtained as the cut feature segments of the transient event type;

[0023] For the optical cable acoustic wave sequence diagram of the periodic event type:

[0024] According to a window size marked by one outer damage feature corresponding to the periodic event type, a range of a second cutting window is determined;

[0025] According to the second cutting window, the one outer break feature is cut out from the period event type optical cable sound wave sequence graph to obtain a second feature segment as a cutting feature segment of the period event type.

[0026] Optionally, the pre-trained feature recognition model is selected according to the outer break event type to process the cutting feature segment to obtain outer break event information of the outer break event, and specifically includes:

[0027] For the first feature segment of the transient event type, the pre-trained multi-branch feature recognition model is used to process the first feature segment to obtain outer break event information of the outer break event.

[0028] For the second feature segment of the period event type, the pre-trained time sequence feature recognition model is used to process the second feature segment to obtain outer break event information of the outer break event.

[0029] Optionally, the multi-branch feature recognition model includes an input unit, a feature splicing unit and a feature recognition unit connected in sequence, and the input unit includes a plurality of input branches.

[0030] The pre-trained multi-branch feature recognition model is used to process the first feature segment to obtain outer break event information of the outer break event, and specifically includes:

[0031] Each of the first feature segments is respectively extracted by each of the input branches in the input unit to obtain corresponding first segment features;

[0032] Each first segment feature is weighted and spliced by the feature splicing unit to obtain a weighted splicing feature;

[0033] The weighted splicing feature is recognized by the feature recognition unit to obtain outer break event information of the outer break event.

[0034] Optionally, the time sequence feature recognition model includes a feature extraction unit, a time sequence feature splicing unit and a time sequence feature recognition unit connected in sequence.

[0035] The pre-trained time sequence feature recognition model is used to process the second feature segment to obtain outer break event information of the outer break event, and specifically includes:

[0036] A plurality of time sequence window segments are obtained by sliding and cutting the second feature segment according to a preset sliding window and a preset step length;

[0037] Each time sequence window segment is processed in time sequence by the feature extraction unit to obtain corresponding time sequence segment features;

[0038] The time sequence feature splicing unit splices the time sequence features in time sequence to obtain time sequence spliced features;

[0039] The time sequence feature recognition unit performs feature recognition on the time sequence spliced features to obtain external damage event information of the external damage event.

[0040] According to a second aspect of the present application, a sound wave-based optical cable external damage prevention system is provided, and the system comprises:

[0041] A data acquisition module is configured to acquire optical cable data of an optical cable, and to obtain a corresponding optical cable sound wave sequence diagram based on the optical cable data.

[0042] A feature marking module is configured to mark external damage features of the optical cable sound wave sequence diagram based on a pre-trained sound wave feature marking model.

[0043] An event type identification module is configured to determine an external damage event type in the optical cable data based on the marked external damage features; the external damage event type is divided into a transient event type and a periodic event type.

[0044] A feature cutting module is configured to cut the marked external damage features in the optical cable sound wave sequence diagram based on the external damage event type to obtain a cut feature segment corresponding to the external damage event type.

[0045] An event information identification module is configured to select a pre-trained feature recognition model to process the cut feature segment based on the external damage event type to obtain external damage event information of the external damage event.

[0046] According to a third aspect of the present application, an electronic device is provided, and the electronic device comprises:

[0047] A memory is configured to store one or more computer programs.

[0048] A processor is configured to implement the sound wave-based optical cable external damage prevention method of the first aspect when the one or more computer programs are executed by the processor.

[0049] According to a fourth aspect of the present application, a computer readable storage medium is provided, and the computer readable storage medium stores computer instructions, which are configured to enable a processor to implement the sound wave-based optical cable external damage prevention method of the first aspect when executed.

[0050] Based on any one of the above aspects, the embodiment of the present application provides a method and system for preventing external damage of an optical cable based on a sound wave, an electronic device and a computer storage medium, wherein the pre-trained feature marking model is used to mark the sound wave sequence diagram of the optical cable, and then the segment containing the external damage feature can be cut out according to the marked external damage feature, the noise interference contained in other redundant segments can be effectively reduced, the accuracy of the external damage event information identification can be improved, and the prevention and treatment of the external damage event can be effectively realized.

[0051] Meanwhile, according to the marked external damage feature, it is determined whether the external damage event type is a transient event type or a periodic event type, a corresponding feature identification model is selected for identification according to different external damage event types, and the accuracy of the external damage event information identification can be further improved. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0053] Figure 1 The flowchart of the method for preventing external damage of an optical cable based on a sound wave is provided for the embodiment.

[0054] Figure 2 The step schematic diagram of the external damage feature marking is provided for the embodiment.

[0055] Figure 3 The step schematic diagram of the external damage event type determination is provided for the embodiment.

[0056] Figure 4 The step schematic diagram of the transient event type feature cutting is provided for the embodiment.

[0057] Figure 5 The step schematic diagram of the periodic event type feature cutting is provided for the embodiment.

[0058] Figure 6 The step schematic diagram of the transient event type information identification is provided for the embodiment.

[0059] Figure 7 The step schematic diagram of the periodic event type information identification is provided for the embodiment.

[0060] Figure 8 The functional module schematic diagram of the system for preventing external damage of an optical cable based on a sound wave is provided for the embodiment.

[0061] Figure 9A structural schematic diagram of an electronic device is provided for the embodiment. DETAILED DESCRIPTION

[0062] The drawings provided in the present application are only for illustrative purposes and should not be construed as limiting the present application. In order to better illustrate the following embodiments, some components in the drawings may be omitted, enlarged or reduced, and do not represent the actual size of the product; it is understandable for those skilled in the art that some well-known structures and their descriptions in the drawings may be omitted.

[0063] In order for those skilled in the art to better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0064] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0065] In the existing technology, the monitoring of the external breakage event of the optical cable mainly relies on the analysis of the received acoustic wave information. The optical cable is an important communication medium, and its safety is directly related to the stable operation of the communication network. However, in actual application, the environment in which the optical cable is located is often complex and changeable, which makes a lot of noise mixed in the received acoustic wave information.

[0066] These noises may come from various factors, such as mechanical vibration in the surrounding environment, traffic flow, wind and rain in nature, etc. These noises and the acoustic signals generated by the real external breakage event are mixed together, which brings great challenges to the accurate judgment of the external breakage event. How to effectively identify the external breakage event from these complex acoustic wave information is one of the key problems to be solved in the current optical cable monitoring technology.

[0067] The embodiment provides a technical solution that can solve the above problems, and the specific embodiments of the present application will be described in detail below in combination with the drawings.

[0068] As Figure 1 shown, the embodiment provides a sound wave-based optical cable external damage prevention method, which can include the following steps:

[0069] S1: Collect optical cable data of an optical cable, and obtain a corresponding optical cable sound wave sequence diagram according to the optical cable data;

[0070] In this embodiment, an optical time domain device equipped with a DAS (Distributed Acoustic Sensing) system can be connected to the optical cable, DAS data of the optical cable can be collected by the optical time domain device, sound wave data of the optical cable can be restored by analyzing the DAS data, and the optical cable sound wave sequence diagram can be constructed according to the sound wave data.

[0071] In an implementation, the optical cable sound wave sequence diagram can be a heat map of optical cable sound wave data. The heat map of optical cable sound wave data can better reflect the feature situation in the optical cable sound wave.

[0072] S2: Label external damage features in the optical cable sound wave sequence diagram by using a pre-trained sound wave feature labeling model;

[0073] In this embodiment, as Figure 2 shown, the external damage feature labeling can include the following steps:

[0074] S21: Obtain a region containing external damage features in the optical cable sound wave sequence diagram by using a pre-trained sound wave feature labeling model;

[0075] S22: Frame the region containing external damage features in the optical cable sound wave sequence diagram by using an adaptive window.

[0076] In this embodiment, the optical cable sound wave sequence diagram can be identified by using a pre-trained sound wave feature labeling model, a region containing external damage features in the optical cable sound wave sequence diagram can be identified, and then the region containing external damage features can be labeled on the optical cable sound wave sequence diagram in a framed manner by using an adaptive window, so that each external damage feature can be better and more intuitively obtained subsequently. The adaptive window can be adaptively adjusted according to the size of the region containing external damage features, so that the adaptive window can contain the corresponding external damage feature.

[0077] In an embodiment, the acoustic feature labeling model can be constructed by a convolutional network, and the acoustic feature labeling model can include an input layer, an encoding layer, a decoding layer and an output layer connected in sequence. The input layer is used to receive an input acoustic sequence diagram of the optical cable, wherein the size and the number of channels of the acoustic sequence diagram of the optical cable are set for the acoustic feature labeling model. The encoding layer includes a plurality of convolutional modules, which gradually reduce the spatial resolution and enhance the feature abstraction level, so as to effectively capture the typical pattern of the external damage event in the acoustic sequence diagram of the optical cable. In an embodiment, an attention module can be arranged in the encoding layer to enhance the attention of the position of the external damage event related features in the acoustic feature labeling model. The decoding layer includes a plurality of deconvolutional modules, which gradually recover the spatial resolution and accurately locate the area of the external damage event in the acoustic sequence diagram by combining the multi-scale features extracted by the encoding layer. The output layer is used to output the position of the external damage feature in the acoustic sequence diagram of the optical cable, so that the adaptive window can frame the external damage feature area.

[0078] The training of the acoustic feature labeling model can include:

[0079] Acoustic sample sequence diagrams containing known external damage events (such as digging, impact) are collected, and the positions of the external damage features are labeled. The labeled acoustic sample sequence diagrams are used to supervise the training of the acoustic feature labeling model, and a trained acoustic feature labeling model is obtained.

[0080] The acoustic feature model of the present embodiment is improved by improving the convolutional network, so that the acoustic feature model only identifies and labels the external damage feature area in the acoustic sequence diagram of the optical cable. However, although the external damage feature area in the acoustic sequence diagram of the optical cable is labeled, the acoustic sequence diagram of the optical cable still contains a lot of noise at this time, so it cannot be directly used for feature recognition.

[0081] S3: determining the type of external damage event in the optical cable data according to the labeled external damage feature;

[0082] In the present embodiment, the type of external damage event is divided into transient event type and periodic event type.

[0083] It can be understood that the characteristic acoustic wave of the external damage event of the transient event type will appear multiple times within a certain time, and each appearance has a certain time interval, and the duration of each appearance is relatively short. Exemplarily, the transient event type can include knocking, shovel digging, etc.

[0084] The characteristic acoustic wave of the external damage event of the periodic event type continuously appears and fluctuates within a period of time. Exemplarily, the periodic event type can include electric grab, excavator operation, etc.

[0085] Therefore, it can be understood that the main difference between the transient event type and the periodic event type is the time interval of the external damage feature, so the specific type of the external damage event can be determined according to the difference, such as Figure 3 As shown in the figure, the determination of the external damage event type can specifically include the following steps:

[0086] S31: Obtain the mean value of the interval of each external damage feature in the optical cable acoustic sequence diagram; the interval of the external damage feature is the time interval between the marked two continuous external damage features in the optical cable acoustic sequence diagram;

[0087] S32: If the mean value is greater than or equal to a preset interval threshold, the external damage event type in the optical cable data is the transient event type;

[0088] S33: If the mean value is less than the preset interval threshold, the external damage event type in the optical cable data is the periodic event type.

[0089] As described above, the main difference between the transient event type and the periodic event type is the time interval of the external damage feature, so the external damage event type can be determined based on the interval of the external damage feature. At the same time, since the external damage feature can be a man-made feature such as knocking, etc., in this embodiment, by obtaining the mean value of the interval of each external damage feature, the external damage event type can be better determined.

[0090] S4: According to the external damage event type, the external damage features marked in the optical cable acoustic sequence diagram are cut to obtain a cut feature segment corresponding to the external damage event type;

[0091] In this embodiment, since the external damage event type can include the two types of transient event type and periodic event type, the acquisition of the cut feature segment can be cut by different cutting strategies for the two types of external damage events respectively, and the specific strategies are as follows:

[0092] As Figure 4 shown, for the optical cable acoustic sequence diagram of the transient event type, the following steps can be used for cutting:

[0093] A1: According to the size of the maximum window marked by each external damage feature of the transient event type, determine the size of the first cutting window;

[0094] A2: According to the first cutting window, cut each external damage feature from the optical cable acoustic sequence diagram of the transient event type to obtain a plurality of first feature segments as the cut feature segments of the transient event type.

[0095] In the embodiment, since the sizes of the regions of the respective external damage features can be different, the sizes of the adaptive windows of the regions of the external damage features are different, and meanwhile, as described above, since the duration of the external damage feature of the transient event type is short, it is difficult to identify the information of the external damage event with only one external damage feature, and therefore, for the external damage event of the transient event type, all the marked external damage features are cut out, and the size of the largest window in the adaptive windows is selected as the size of the first cutting window, so that the respective external damage features can be completely cut out by the first cutting window, and the information of the external damage event can be better identified.

[0096] As shown in Figure 5 For the optical cable acoustic wave sequence graph of the periodic event type, the following steps can be adopted for cutting:

[0097] B1: determining the size of a second cutting window according to the size of the window marked by one external damage feature of the periodic event type;

[0098] B2: cutting the one external damage feature from the optical cable acoustic wave sequence graph of the periodic event type according to the second cutting window to obtain a second feature segment as a cutting feature segment of the periodic event type.

[0099] In the embodiment, the periodic event type is different from the transient event type, and since the acoustic wave (a region containing an external damage feature) of the periodic event type usually lasts for a relatively long time, the external damage event of the periodic event type can be identified from one second feature segment.

[0100] It can be understood that, for the entire optical cable acoustic wave sequence graph, the proportion of the external damage features is very small, and there is a certain interval between the external damage features, and especially for the external damage features of the transient event type, the interval between the external damage features does not help much in identifying the external damage event, and instead, becomes noise for identifying the external damage event. If the external damage features and the interval between the external damage features are identified together, the influence of the environmental noise will be amplified, and the signal-to-noise ratio of the external damage features will be reduced, resulting in a reduction in the identification accuracy. Therefore, the embodiment cuts the regions of the external damage features from the optical cable acoustic wave sequence graph to reduce the influence of the noise outside the external damage features, and thus the accuracy of identifying the external damage event can be improved, and the prevention and control of the external damage event can be effectively realized.

[0101] S5: selecting a pre-trained feature identification model to process the cutting feature segment according to the type of the external damage event to obtain external damage event information of the external damage event.

[0102] It can be understood that, since the external breakage events of the transient event type are intermittent and the external breakage events of the periodic event type are persistent, the characteristics of the two events are different on the optical cable acoustic wave sequence diagram, and thus, if the same model is used for identification, the results of the identification will be biased. Therefore, in the embodiment, different models are used for identification for the transient event type and the periodic event type respectively, and thus, different external breakage events can be accurately identified.

[0103] Specifically, for the first feature segment of the transient event type, the pre-trained multi-branch feature identification model is used to process the first feature segment to obtain the external breakage event information of the external breakage event.

[0104] For the second feature segment of the periodic event type, the pre-trained time sequence feature identification model is used to process the second feature segment to obtain the external breakage event information of the external breakage event.

[0105] In the embodiment, the multi-branch feature identification model can include an input unit, a feature splicing unit and a feature identification unit connected in sequence, the input unit includes a plurality of input branches, and thus, as shown in Figure 6 The processing of the transient event type can include the following steps:

[0106] C1: Each first feature segment is respectively subjected to feature extraction by each input branch in the input unit to obtain corresponding first segment features;

[0107] C2: The first segment features are weighted and spliced by the feature splicing unit to obtain weighted splicing features;

[0108] C3: The weighted splicing features are subjected to feature identification by the feature identification unit to obtain the external breakage event information of the external breakage event.

[0109] In an implementation, the input branch can be set as a one-dimensional convolutional network for feature extraction of the input first feature segment to obtain corresponding first segment features; and the feature identification unit can be constructed based on a convolutional network and include a plurality of convolutional modules stacked in sequence for feature identification of the weighted splicing features to identify the external breakage event information of the corresponding external breakage event.

[0110] It can be understood that the multi-branch feature recognition model can be trained individually for each unit, or can be trained end-to-end based on the multi-branch feature recognition model as a whole. Taking end-to-end training as an example, the training of the multi-branch feature recognition model can include: collecting an optical cable acoustic wave sequence diagram containing a plurality of transient event types, wherein the external damage event corresponding to the transient event type is known, and labeling the optical cable acoustic wave sequence diagram based on the external damage event; cutting the labeled optical cable acoustic wave sequence diagram to obtain a first feature segment; using the obtained first feature segment to supervise the training of the multi-branch feature recognition model, and obtaining a trained multi-branch feature recognition model.

[0111] In the embodiment, the weight of the weighted splicing can be set according to the proportion of the external damage feature in the corresponding adaptive window. The higher the proportion, the higher the weight setting. It can be understood that the higher the proportion of the first segment feature in the adaptive window, the higher the signal-to-noise ratio of the corresponding first segment feature. Therefore, in the embodiment, a higher weight is set for the first segment feature with a higher proportion of external damage features in the adaptive window, which can effectively improve the recognition effect of the weighted splicing feature and further improve the recognition accuracy of the external damage event information.

[0112] It should be noted that since the sizes of the first feature segments can be different, in the embodiment, the dimensions of the first feature segments are unified before feature extraction of the first feature segments. The first feature segments can be unified to the same dimension by scaling, padding or cutting.

[0113] In the embodiment, the time sequence feature recognition model can include a feature extraction unit, a time sequence feature splicing unit and a time sequence feature recognition unit connected in sequence. As shown in Figure 7 The processing of the periodic event type can include the following steps:

[0114] D1: A plurality of time sequence window segments are obtained by sliding and cutting the second feature segment through a preset sliding window according to a preset step size;

[0115] D2: The feature extraction unit processes each time sequence window segment in time sequence to obtain corresponding time sequence segment features;

[0116] D3: The time sequence feature splicing unit splices each time sequence segment feature in time sequence to obtain time sequence splicing features;

[0117] D4: The time sequence feature recognition unit performs feature recognition on the time sequence splicing features to obtain external damage event information of the external damage event.

[0118] In the embodiment, the second feature segment is cut by the sliding window, and a plurality of time window segments with time sequence relationship are obtained from the second feature segment. The size of the sliding window can be set according to the length of the second feature segment, so as to ensure that enough time window segments are obtained for identification. There can be a certain feature overlap between adjacent two time window segments.

[0119] It can be understood that, since the external damage feature of the periodic event type lasts for a long time, the identification of the external damage event information of the periodic event type can be realized by a complete external damage feature.

[0120] In an implementation, the feature extraction unit can be constructed based on a convolution network, and the convolution network is used to extract features of the time window segment. The time sequence feature recognition unit can be constructed based on an LSTM (Long Short-Term Memory network) model. The time sequence feature recognition model constructed based on the LSTM model can effectively recognize the time sequence feature with time sequence relationship, and obtain the external damage event information of the periodic event type.

[0121] It can be understood that, the time sequence feature recognition model can be trained based on each unit separately, or can be trained based on the whole time sequence feature recognition model in an end-to-end manner. Taking the end-to-end training as an example, the training of the time sequence feature recognition model can include: collecting a cable acoustic sequence graph containing a periodic event type, wherein the external damage event corresponding to the periodic event type is known, and the cable acoustic sequence graph is labeled by the external damage event; cutting the labeled cable acoustic sequence to obtain a second feature segment; using the obtained second feature segment to supervise the training of the time sequence feature recognition model, and obtaining a trained time sequence feature recognition model.

[0122] It should be noted that, in the embodiment, the cable can also be a cable containing an optical fiber.

[0123] In the embodiment, the feature cutting feature segment containing the feature in the optical cable acoustic wave sequence graph can be extracted by identifying and marking the feature in the optical cable acoustic wave sequence graph, and the remaining noise segments in the optical cable acoustic wave sequence graph are screened out, the signal-to-noise ratio in the cutting feature segment is effectively improved, when the cutting feature segment is used to identify the external damage event, the interference of noise can be greatly reduced, and the accuracy of the external damage event information identification is effectively improved, and the prevention and treatment of the external damage event are effectively realized. Meanwhile, different feature identification models are used to process the external damage events of the transient event type and the periodic event type respectively, the external damage events of the transient event type and the periodic event type can be accurately identified according to the characteristics of the external damage events of the transient event type and the periodic event type, and the accuracy of the external damage event information identification is further improved.

[0124] As shown in Figure 8 The embodiment of the application also provides an optical cable external damage prevention system based on acoustic waves. Optionally, the identification system can include:

[0125] A data acquisition module 11 is configured to acquire optical cable data of an optical cable, and obtain a corresponding optical cable acoustic wave sequence graph according to the optical cable data.

[0126] In the embodiment, the data acquisition module 11 can be configured to perform step S1 as shown in Figure 1 The specific description of the data acquisition module 11 can be referred to the description of step S1.

[0127] A feature marking module 12 is configured to mark external damage features of the optical cable acoustic wave sequence graph by using a pre-trained acoustic wave feature marking model.

[0128] In the embodiment, the feature marking module 12 can be configured to perform step S2 as shown in Figure 1 The specific description of the feature marking module 12 can be referred to the description of step S2.

[0129] An event type identification module 13 is configured to determine an external damage event type in the optical cable data according to the marked external damage features.

[0130] In the embodiment, the event type identification module 13 can be configured to perform step S3 as shown in Figure 1 The specific description of the event type identification module 13 can be referred to the description of step S3.

[0131] A feature cutting module 14 is configured to cut the marked external damage features in the optical cable acoustic wave sequence graph according to the external damage event type, and obtain a cutting feature segment corresponding to the external damage event type.

[0132] In the embodiment, the feature cutting module 14 can be configured to performFigure 1 As shown in step S4, the specific description of the feature cutting module 14 can refer to the description of step S4.

[0133] The event information identification module 15 is configured to select a pre-trained feature recognition model according to the external damage event type, process the cut feature segment, and obtain external damage event information of the external damage event.

[0134] In this embodiment, the event information identification module 15 can be used to execute Figure 1 As shown in step S5, the specific description of the event information identification module 15 can refer to the description of step S5.

[0135] The electronic device 20 provided in the embodiments of the present application has the structure as Figure 9 shown.

[0136] As Figure 9 shown, the electronic device 20 includes a memory 21, a processor 22, a communication module 23, and an input / output interface 24, etc. Optionally, the memory 21, the processor 22, the communication module 23, and the input / output interface 24 can be connected and communicated through a bus 25.

[0137] The memory 21 is configured to store one or more computer programs and transmit the code of the computer programs to the processor 22; when the one or more computer programs are executed by the processor 22, a sound wave-based optical cable anti-external damage method in the embodiments of the present application is implemented.

[0138] Optionally, the electronic device 20 can be connected to a network through the communication module 23 to communicate with other devices such as terminals or servers through the network to realize the interaction of data. The electronic device 20 can be various forms of digital computers, exemplarily such as desktop computers, servers, workstations, mainframe computers, or other types of computers. The electronic device 20 can also be various forms of mobile terminals, exemplarily such as smart phones, tablet computers, wearable devices (such as helmets, glasses, watches, etc.), and other similar mobile terminals.

[0139] Optionally, the electronic device 20 can connect required input / output devices, such as a keyboard, a display device, etc., through the input / output interface 24. The electronic device 20 itself can have a display device, and can also be externally connected to other display devices through the input / output interface 24. Optionally, the input / output interface 24 can also be connected to a storage device, such as a hard disk, etc., so as to store data in the electronic device 20 into the storage device, or read data in the storage device, and also store data in the storage device into the memory 21. It can be understood that the input / output interface 24 can be a wired interface or a wireless interface. According to different actual application scenarios, the devices connected to the input / output interface 24 can be a component of the electronic device 20, or can be an external device connected to the electronic device 20 when needed.

[0140] Optionally, the memory 21 can be a volatile memory and / or a non-volatile memory. The volatile memory can be a random access memory, etc., and the non-volatile memory can be a read-only memory, a programmable read-only memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, or a flash memory, etc.

[0141] Optionally, the computer program stored in the processor 22 can be divided into one or more modules, which are stored in the memory 21 and executed by the processor 22 to complete the method provided in the embodiment. The one or more modules can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program in the electronic device 20.

[0142] Optionally, the processor 22 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 22 include, but are not limited to, a central processing unit, a graphics processing unit, a digital signal processor, various special-purpose artificial intelligence computing chips, various processors running machine learning model algorithms, and any appropriate controller, microcontroller, processor, etc. The processor 22 executes various methods and processes of the embodiment, exemplarily such as a sound wave-based optical cable anti-external damage method of the embodiment.

[0143] Optionally, the bus 25 can include a path for transmitting information. The bus 25 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. According to different functions, the bus 25 can be divided into an address bus, a data bus, a control bus, etc.

[0144] In an optional implementation, the embodiment of the present application further provides a computer storage medium, which stores a computer program. When the computer program is executed by a computer, the computer can execute the method of the method embodiment. Part or all of the computer program can be loaded and / or installed on the memory 21 of the electronic device 20. When the computer program is executed by the processor 22, one or more steps of the method of the optical cable external damage prevention method based on sound waves can be executed.

[0145] Optionally, the computer readable storage medium can be a random access memory, a read-only memory, a programmable read-only memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, etc.

[0146] Obviously, the above embodiments of the present application are only examples for clearly illustrating the technical solutions of the present application, and are not intended to limit the specific embodiments of the present application. Any modification, equivalent replacement and improvement made within the spirit and principle of the claims of the present application should be included in the protection scope of the claims of the present application.

Claims

1. A method for protecting an optical cable from external damage based on acoustic waves, characterized in that, The method comprises: Collecting optical cable data of an optical cable, and obtaining a corresponding optical cable acoustic wave sequence diagram according to the optical cable data; Performing external damage feature labeling on the optical cable acoustic wave sequence diagram by using a pre-trained acoustic wave feature labeling model; Obtaining a mean value of intervals between external damage features in the optical cable acoustic wave sequence diagram, wherein the interval between external damage features is a time interval between two continuous external damage features labeled in the optical cable acoustic wave sequence diagram; If the mean value is greater than or equal to a preset interval threshold, an external damage event type in the optical cable data is a transient event type; If the mean value is less than the preset interval threshold, the external damage event type in the optical cable data is a periodic event type; According to the external damage event type, cutting the labeled external damage features in the optical cable acoustic wave sequence diagram to obtain a cutting feature segment corresponding to the external damage event type; For a first feature segment of the transient event type, processing the first feature segment by using a pre-trained multi-branch feature recognition model to obtain external damage event information of the external damage event; For a second feature segment of the periodic event type, processing the second feature segment by using a pre-trained time sequence feature recognition model to obtain the external damage event information of the external damage event.

2. The acoustic-based optical cable external breakage prevention method according to claim 1, characterized by, The method of performing external damage feature labeling on the optical cable acoustic wave sequence diagram by using the pre-trained acoustic wave feature labeling model specifically comprises: Obtaining a region containing external damage features in the optical cable acoustic wave sequence diagram by using the pre-trained acoustic wave feature labeling model; By using an adaptive window, the region containing external damage features in the optical cable acoustic wave sequence diagram is framed out.

3. The acoustic-based optical cable breakage prevention method of claim 2, wherein, The method of cutting the labeled external damage features in the optical cable acoustic wave sequence diagram according to the external damage event type to obtain a cutting feature segment corresponding to the external damage event type specifically comprises: For the optical cable acoustic wave sequence diagram of the transient event type: According to a size of a largest window labeled by each external damage feature of the transient event type, determining a size of a first cutting window; According to the first cutting window, cutting each external damage feature from the optical cable acoustic wave sequence diagram of the transient event type to obtain a plurality of first feature segments as the cutting feature segment of the transient event type; For the optical cable acoustic wave sequence diagram of the periodic event type: According to a window size labeled by one external damage feature of the periodic event type, determining a size of a second cutting window; According to the second cutting window, cutting the one external damage feature from the optical cable acoustic wave sequence diagram of the periodic event type to obtain a second feature segment as the cutting feature segment of the periodic event type.

4. The acoustic-based cable anti-physical damage method of claim 1, wherein, The multi-branch feature recognition model comprises an input unit, a feature splicing unit and a feature recognition unit connected in sequence, and the input unit contains a plurality of input branches; The method of processing the first feature segment by using the pre-trained multi-branch feature recognition model to obtain the external damage event information of the external damage event specifically comprises: Performing feature extraction on each first feature segment by using each input branch in the input unit to obtain corresponding first feature segments; The first segment features are weighted and spliced by the feature splicing unit to obtain weighted spliced features; The weighted spliced features are identified by the feature identification unit to obtain the external damage event information of the external damage event.

5. The acoustic-based cable anti-physical damage method of claim 1, wherein, The time sequence feature identification model comprises a feature extraction unit, a time sequence feature splicing unit and a time sequence feature identification unit connected in sequence; The second feature segment is processed by the pre-trained time sequence feature identification model to obtain the external damage event information of the external damage event, specifically including: A plurality of time sequence window segments are obtained by slidingly cutting the second feature segment according to a preset sliding window and a preset step size; The feature extraction unit processes each time sequence window segment in time sequence to obtain corresponding time sequence segment features; The time sequence feature splicing unit splices each time sequence segment feature in time sequence to obtain time sequence spliced features; The time sequence feature identification unit identifies the time sequence spliced features to obtain the external damage event information of the external damage event.

6. An acoustic-based optical cable external breakage prevention system characterized by, The system comprises: A data acquisition module is configured to acquire optical cable data of an optical cable, and obtain a corresponding optical cable acoustic wave sequence graph based on the optical cable data; A feature marking module is configured to mark external damage features in the optical cable acoustic wave sequence graph by using a pre-trained acoustic wave feature marking model; An event type identification module is configured to obtain a mean value of intervals between external damage features in the optical cable acoustic wave sequence graph, and the interval between the external damage features is a time interval between two continuous external damage features marked in the optical cable acoustic wave sequence graph; if the mean value is greater than or equal to a preset interval threshold, an external damage event type in the optical cable data is a transient event type; if the mean value is less than the preset interval threshold, the external damage event type in the optical cable data is a periodic event type; A feature cutting module is configured to cut the external damage features marked in the optical cable acoustic wave sequence graph according to the external damage event type to obtain a cutting feature segment corresponding to the external damage event type; An event information identification module is configured to process a first feature segment of the transient event type by using a pre-trained multi-branch feature identification model to obtain external damage event information of the external damage event, and process a second feature segment of the periodic event type by using a pre-trained time sequence feature identification model to obtain the external damage event information of the external damage event.

7. The acoustic-based cable external breach system of claim 6, wherein, The external damage features in the optical cable acoustic wave sequence graph are marked by using the pre-trained acoustic wave feature marking model, specifically including: The pre-trained acoustic wave feature marking model is used to obtain a region containing external damage features in the optical cable acoustic wave sequence graph; The region containing the external damage features in the optical cable acoustic wave sequence graph is framed out by using an adaptive window.

8. The acoustic-based cable anti-tamper system of claim 7, wherein, The external damage features marked in the optical cable acoustic wave sequence graph are cut according to the external damage event type to obtain a cutting feature segment corresponding to the external damage event type, specifically including: For the optical cable acoustic wave sequence graph of the transient event type: determining a size of a first cutting window according to a size of a maximum window marked by each external damage feature of the transient event type; cutting each external damage feature from the optical cable acoustic wave sequence graph of the transient event type according to the first cutting window to obtain a plurality of first feature segments as cutting feature segments of the transient event type; for the optical cable acoustic wave sequence graph of the periodic event type: determining a size of a second cutting window according to a size of a window marked by one external damage feature of the periodic event type; cutting the one external damage feature from the optical cable acoustic wave sequence graph of the periodic event type according to the second cutting window to obtain a second feature segment as a cutting feature segment of the periodic event type.

9. An electronic device, comprising: comprising: a memory for storing one or more computer programs; a processor, when the one or more computer programs are executed by the processor, implements the acoustic wave based optical cable external damage prevention method as claimed in any one of claims 1-5. 10.A computer readable storage medium, the computer readable storage medium storing computer instructions for causing a processor to implement the acoustic wave based optical cable external damage prevention method as claimed in any one of claims 1-5 when executed by the processor.

Images