WO2024082215A1

WO2024082215A1 - Seismic signal monitoring method and apparatus

Info

Publication number: WO2024082215A1
Application number: PCT/CN2022/126433
Authority: WO
Inventors: 杨平
Original assignee: 杨平
Priority date: 2022-10-20
Filing date: 2022-10-20
Publication date: 2024-04-25
Also published as: CN117157557A

Abstract

A seismic signal monitoring method and apparatus. The method comprises: for a plurality of collection moments, collecting a plurality of change images of a vibration response object, the collection moments being associated with the change images (S310); using a displacement processing model to respectively process the plurality of change images, and obtaining first displacement data, second displacement data and third displacement data of a three-dimensional space respectively corresponding to each of the plurality of collection moments (S320); and according to the first displacement data, the second displacement data and the third displacement data respectively corresponding to each collection moment, determining an amplitude information set respectively corresponding to each collection moment (S330). The seismic signal monitoring method can simply, rapidly and accurately determine a displacement data set corresponding to a three-dimensional space merely according to change images comprising a vibration response object, without needing to use three vibration response objects and three corresponding camera collection apparatuses.

Description

A method and device for monitoring earthquake signals

Technical Field

The present application relates to the field of signal monitoring, and in particular to a method and device for monitoring earthquake signals.

Background technique

Earthquakes include natural earthquakes and artificially stimulated earthquakes. At present, amplitude information is often used to characterize the signal generated by the earthquake, or velocity information obtained based on the amplitude information is used to characterize the signal generated by the earthquake. When determining the amplitude information, it is first necessary to collect the vibration characteristics of the stratum, and then determine the corresponding amplitude information and velocity based on the vibration characteristics. The detectors currently used to collect the vibration characteristics of the stratum include moving coil detectors and digital detectors to determine the corresponding amplitude information. The moving coil detector, as shown in FIG1A, includes a coil for collecting physical signals, and the coil vibrates in response to the physical signal of the earthquake, thereby forming electrical signals such as current, voltage, and capacitance, and then determining the corresponding amplitude information based on the electrical signal. In addition, since one coil and one chip can only be used to determine the amplitude information of one direction, when determining the amplitude information of three directions in three-dimensional space, three coils for collecting physical signals or three corresponding chips are required. The digital detector, as shown in FIG1B, includes three chips, one chip is used to determine the amplitude information of one direction in three-dimensional space.

How to determine the amplitude information in three directions in three-dimensional space through an object that collects physical signals to reduce resource waste is an urgent problem that needs to be solved in the existing technology.

Summary of the invention

The purpose of this application is to provide a seismic signal monitoring method and device to solve the problem of how to determine the amplitude information in three directions in three-dimensional space through an object for collecting physical signals, thereby reducing resource waste.

The above-mentioned purpose of the present application can be achieved by adopting the following technical solutions:

The present application provides a seismic signal monitoring method, comprising: acquiring multiple change images of a vibration response object at multiple acquisition moments, wherein the acquisition moments are associated with the change images; using a displacement processing model, processing the multiple change images respectively to obtain first displacement data, second displacement data and third displacement data in a three-dimensional space respectively corresponding to each acquisition moment in the multiple acquisition moments; and determining an amplitude information set respectively corresponding to each acquisition moment based on the first displacement data, the second displacement data and the third displacement data respectively corresponding to each acquisition moment, wherein the change image includes an image of the vibration response object changing in response to a seismic signal.

In some embodiments, the use of a displacement processing model to process the multiple change images separately to obtain first displacement data, second displacement data, and third displacement data corresponding to each of the multiple acquisition moments includes: determining an initial image and multiple vibration images from the multiple change images; and determining the first displacement data, the second displacement data, and the third displacement data corresponding to each of the multiple change images based on an image difference between a first vibration responder region included in the multiple vibration images and a second vibration responder region included in the initial image.

In some embodiments, the use of the displacement processing model to process the multiple change images separately to obtain the first displacement data, the second displacement data and the third displacement data corresponding to each of the multiple acquisition moments respectively includes: inputting the multiple change images into the trained three-dimensional displacement determination model respectively to obtain the first displacement data, the second displacement data and the third displacement data corresponding to each acquisition moment associated with each change image.

In some embodiments, the training process of the trained three-dimensional displacement determination model includes: using a preset three-dimensional displacement determination model to process multiple training change images included in a training sample set to obtain multiple predicted sample three-dimensional displacement data sets; and training the preset three-dimensional displacement determination model according to the difference between the multiple predicted sample three-dimensional displacement data sets and the multiple labels included in the training sample set to obtain the trained three-dimensional displacement determination model, wherein the predicted sample three-dimensional displacement data set includes predicted sample first displacement data, predicted sample second displacement data and predicted sample third displacement data, and the labels include sample first displacement data, sample second displacement data and sample third displacement data.

In some embodiments, the construction process of the training sample set includes: obtaining multiple first sample amplitude information corresponding to a change direction; performing expansion processing on each first sample amplitude information of the multiple first sample amplitude information to obtain the multiple sample first displacement data, the multiple sample second displacement data and the multiple sample third displacement data; obtaining an initial training change image; and processing the initial training change image according to the multiple sample first displacement data, the multiple sample second displacement data and the multiple sample third displacement data to obtain the multiple training change images.

In some embodiments, the expansion processing is performed on each of the multiple first sample amplitude information to obtain the multiple sample first displacement data, the multiple sample second displacement data and the multiple sample third displacement data, including: according to the mapping relationship between the amplitude information in three change directions, each of the first sample amplitude information is processed separately to obtain multiple second sample amplitude information and multiple third sample amplitude information; and sample expansion and normalization processing is performed on the multiple first sample amplitude information, the multiple second sample amplitude information and the multiple third sample amplitude information to obtain the multiple sample first displacement data, the multiple sample second displacement data and the multiple sample third displacement data.

In some embodiments, the acquiring of the initial training variation image corresponding to the first sample amplitude information includes: acquiring the plurality of variation images; performing overall boundary recognition on each variation image in the plurality of variation images, and based on the overall boundary, cropping each variation image in the plurality of variation images to obtain a plurality of cropped variation images; and determining the initial training variation image from the plurality of cropped variation images according to the number of cropped variation images similar to each cropped variation image in the plurality of cropped variation images.

In some embodiments, the processing of the initial training variation image according to the multiple sample first displacement data, the multiple sample second displacement data, and the multiple sample third displacement data to obtain the multiple training variation images includes: identifying the boundary of the vibration responder object for the initial training variation image, and removing the sub-image included in the boundary of the vibration responder object from the initial training variation image to obtain a target initial training variation image; performing variation processing on the sub-image according to the multiple sample first displacement data, the multiple sample second displacement data, and the multiple sample third displacement data to obtain a plurality of target sub-images; and splicing each of the multiple target sub-images with the target initial training variation image to obtain the multiple training variation images.

The present application also provides a seismic signal monitoring device, comprising: a box; a vibration responder, which is arranged inside the box and moves under the action of the seismic signal transmitted by the box; and a camera module, which is arranged inside the box facing the vibration responder to capture a changing image generated by the movement of the vibration responder.

In some embodiments, the seismic signal monitoring device further comprises a scale frame, the scale frame is disposed inside the box and fixedly connected to the box, and the vibration responder is disposed inside the scale frame.

In some embodiments, the seismic signal monitoring device further comprises a ruler, and the ruler is closely connected to the vibration responder.

In some embodiments, the vibration responder includes a mass block and an elastic component, and the mass block is elastically connected to the box through the elastic component.

In some embodiments, the mass block includes a light source, and the light beam emitted by the light source always falls into the changing image captured by the camera module.

In some embodiments, the vibration responder includes a plurality of elastic objects.

In some embodiments, the plurality of elastic objects have different masses respectively.

In some embodiments, the vibration-responsive substance comprises a fluid.

In some embodiments, the seismic signal monitoring device further includes a tail vertebrae mechanism, which is disposed outside the box and fixedly connected to the box to increase the coupling degree between the box and the formation.

In some embodiments, the seismic signal monitoring device further includes a counterweight to adjust the center of gravity of the housing so that the center of gravity of the housing remains on the vertical center line of the housing.

In some embodiments, the seismic signal monitoring device also includes a power supply mechanism, which is arranged inside the box to supply power to the camera module.

In some embodiments, the seismic signal monitoring device also includes a power supply mechanism, which is arranged inside the box to supply power to the camera module and the light source.

The features and advantages of this application are:

By collecting multiple change images of a vibration response object for collecting physical signals at each collection moment, and performing data operations on the multiple change images, the first displacement data, the second displacement data and the third displacement data of the three-dimensional space corresponding to each change image are determined, and then the amplitude information set is determined. Therefore, the first displacement data, the second displacement data and the third displacement data of the three-dimensional space can be determined simply, quickly and accurately only based on the change image including the vibration response object, without using three vibration response objects and three corresponding camera acquisition devices to determine the first displacement data, the second displacement data and the third displacement data of the three-dimensional space, thereby reducing the waste of resources and improving the relative consistency between the first displacement data, the second displacement data and the third displacement data of the three-dimensional space.

For the seismic signal monitoring device, since the electromagnetic coil is abandoned, any vibration response object that can vibrate in response to the seismic signal can be selected, making the seismic signal monitoring device more selective. In addition, when the vibration response object is a fluid or an elastic object, a signal with a wider frequency band can be collected. When the vibration response object includes an elastic component and a mass block, the vibration response of the vibration response object is completely determined by the elastic coefficient of the elastic component and the weight of the mass block, and is not subject to any electromagnetic interference. The signal-to-noise ratio is higher, so that the accuracy of the determined amplitude signal set is higher. Furthermore, since the amplitude information is calculated directly based on the acquired change image, there is no need for various electrical signals as intermediaries, so the reflection of the seismic signal amplitude is more realistic, that is, the signal fidelity is higher, and the cost is reduced.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

FIG1A is a schematic diagram of the structure of a moving coil three-component detector in the prior art;

FIG1B is a schematic diagram of the structure of a digital three-component detector in the prior art;

FIG2 is a schematic diagram of an implementation system of a seismic signal monitoring method according to an embodiment of the present application;

FIG3 is a flow chart of a method for monitoring seismic signals according to an embodiment of the present application;

FIG4 is a flow chart of a method for determining displacement data according to an embodiment of the present application;

FIG5A is a flow chart of a method for determining displacement data according to another embodiment of the present application;

FIG5B is a schematic diagram of an initial training variation image according to an embodiment of the present application;

FIG6 is a schematic structural diagram of a seismic signal monitoring device according to an embodiment of the present application;

FIG7 is a schematic structural diagram of a seismic signal monitoring device according to another embodiment of the present application;

FIG8 is a front view of a seismic signal monitoring device according to an embodiment of the present application;

FIG9A is a schematic structural diagram of a seismic signal monitoring device according to another embodiment of the present application;

FIG9B is a schematic structural diagram of a seismic signal monitoring device according to another embodiment of the present application;

FIG. 10 is a schematic diagram of the structure of a computer device according to an embodiment of the present application.

【Description of reference numerals】：

111, X-axis coil;

112. Y-axis coil;

113. Z-axis coil;

121, X-axis chip;

122. Y-axis chip;

123. Z-axis chip;

210. Electronic equipment;

220. Database;

230. The trained three-dimensional displacement determination model;

240, collection terminal;

250, amplitude information set;

260, changing images;

501. Multiple changing images;

502. A plurality of displacement data sets;

511. A plurality of first sample amplitude information;

512. A plurality of sample displacement data sets;

520, training sample set;

531. Preset three-dimensional displacement determination model;

532. The three-dimensional displacement determination model after training;

5011, initial training change image;

5012, overall boundary;

5013, sub-image;

610, box;

620. Vibration response object;

630, camera module;

740, scale frame;

721, mass block;

722. Elastic components;

822. Elastic components;

823, ruler;

840, scale frame;

910, box;

921, mass block;

922. Elastic components;

930, camera module;

940, scale frame;

950, counterweight;

960, power supply mechanism;

970, coccygeal mechanism;

1002. Computer equipment;

1004. Processing equipment;

1006. Storage resources;

1008. Driving mechanism;

1010, input/output module;

1012. Input device;

1014. Output device;

1016. Presentation equipment;

1018. Graphical user interface;

1020. Network interface;

1022. Communication link;

1024. Communication bus.

Detailed ways

The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, 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 "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

FIG. 1A is a schematic diagram of the structure of a moving coil three-component detector in the prior art, and FIG. 1B is a schematic diagram of the structure of a digital three-component detector in the prior art.

At present, the seismic detector used may be, for example, a moving coil detector, which includes three coils for determining the amplitudes in three directions in three-dimensional space. Specifically, the three coils include an X-axis coil 111, a Y-axis coil 112, and a Z-axis coil 113. The X-axis coil 111 may be, for example, a coil for detecting the amplitude in the propagation direction of an earthquake SV wave. The Y-axis coil 112 may be, for example, a coil for detecting the amplitude in the propagation direction of an earthquake SH wave. The Z-axis coil 113 may be, for example, a coil for detecting the amplitude in the propagation direction of an earthquake P wave.

The seismic detector used may also be, for example, a digital detector, which includes three chips for determining the amplitudes in three directions in a three-dimensional space, specifically, the three chips include an X-axis chip 121, a Y-axis chip 122, and a Z-axis chip 123. The X-axis chip 121 may be, for example, a chip for detecting the amplitude of the propagation direction of a shear wave (SV wave) in a vertical plane of an earthquake. The Y-axis chip 122 may be, for example, a chip for detecting the amplitude of the propagation direction of a shear wave (SH wave) in a parallel plane of an earthquake. The Z-axis chip 123 may be, for example, a chip for detecting the amplitude of the propagation direction of a longitudinal wave (P wave) of an earthquake.

Since both the moving coil detector and the digital detector require three vibration responders to determine the amplitude information in three directions in three-dimensional space, it results in a waste of resources.

FIG2 is a schematic diagram of an implementation system of a seismic signal monitoring method according to an embodiment of the present application, which may include an electronic device 210 and a database 220 .

The electronic device 210 can access the database 220 through a network, for example. The database 220 can store a plurality of training variation images and a label corresponding to each training variation image. The label includes sample first displacement data, sample second displacement data, and sample third displacement data.

In one embodiment, the electronic device 210 can read a labeled training change image from the database 220, and use the read training change image as a sample to train the preset three-dimensional displacement determination model. The preset three-dimensional displacement determination model is used to process each training change image to obtain a set of predicted sample three-dimensional displacement data corresponding to each training change image, and the predicted sample three-dimensional displacement data set includes predicted sample first displacement data, predicted sample second displacement data, and predicted sample third displacement data. The electronic device 210 can compare the obtained multiple predicted sample first displacement data, predicted sample second displacement data, and predicted sample third displacement data with multiple sample first displacement data, sample second displacement data, and sample third displacement data, and train the preset three-dimensional displacement determination model according to the comparison result to obtain a trained three-dimensional displacement determination model.

In one embodiment, the application scenario may also include a trained three-dimensional displacement determination model 230, an acquisition terminal 240, an amplitude information set 250, and a change image 260. The acquisition terminal 240 is connected to the electronic device 210 through a network or a mobile storage device. The acquisition terminal includes, for example, a photographic module. The acquisition terminal may also be a processing device connected to the photographic module. The processing device may be, for example, an electronic device. For example, the acquisition terminal 240 may obtain the trained three-dimensional displacement determination model 230 from the electronic device 210, and process the change image 260 based on the obtained trained three-dimensional displacement determination model 230 to obtain the first displacement data, the second displacement data, and the third displacement data of the three-dimensional space corresponding to the change image 260. Further, the amplitude information set 250 is determined according to the first displacement data, the second displacement data, and the third displacement data of the three-dimensional space. The amplitude information set 250 includes the first amplitude data, the second amplitude data, and the third amplitude data of the three-dimensional space.

It should be noted that FIG2 describes a method for processing the change image using a machine learning model to obtain an amplitude information set, but it is also possible to perform image recognition and mathematical calculations only on a plurality of change images collected corresponding to a time period to obtain first displacement data, second displacement data, and third displacement data in the three-dimensional space corresponding to each change image, and then obtain first amplitude data, second amplitude data, and third amplitude data in the three-dimensional space.

It should also be noted that the seismic signal monitoring method provided in the present application can generally be executed by the electronic device 210, or can be executed by a server or the like that is communicatively connected to the electronic device 210. It should be understood that the number and type of electronic devices, databases, change images, and acquisition terminals in FIG. 2 are merely illustrative. Any number and type of electronic devices, databases, change images, and acquisition terminals may be provided as required for implementation.

FIG3 is a flowchart of a seismic signal monitoring method according to an embodiment of the present application. This figure describes the process of determining the amplitude information set, but more or fewer operation steps may be included based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the order of many steps and does not represent the only execution order. When the system or device product is executed in practice, it can be executed in the order or in parallel according to the method shown in the embodiment or the accompanying drawings. Specifically, as shown in FIG3, the method may include:

S310, collecting multiple change images of the vibration response object at multiple collection moments, wherein the collection moments are associated with the change images;

S320, using a displacement processing model, processing the multiple change images respectively to obtain first displacement data, second displacement data, and third displacement data in a three-dimensional space corresponding to each of the multiple acquisition moments;

S330, determining an amplitude information set corresponding to each acquisition moment according to the first displacement data, the second displacement data, and the third displacement data corresponding to each acquisition moment.

The change image includes an image of changes in the vibration responder in response to the seismic signal.

According to an embodiment of the present application, for a time period, multiple acquisition moments are determined based on a preset acquisition frequency. The acquisition frequency can be configured according to actual conditions. At each acquisition moment, the photography module respectively acquires a change image including a vibration response, so that multiple change images are acquired within the time period. It should be noted that the change image can be, for example, a change image after preprocessing. Each change image is associated with the acquisition moment corresponding to the target image and stored. It should be noted that the change image includes an image of a vibration responder changing in response to a seismic signal, and the seismic signal refers to a physical signal, that is, a physical quantity that characterizes the vibration of the vibration responder caused by an earthquake.

The displacement processing model can be, for example, any model that can process change images corresponding to multiple continuous acquisition moments in a time period to determine a displacement data set in a three-dimensional space corresponding to each change image. The displacement data set includes first displacement data, second displacement data, and third displacement data. The vibration directions corresponding to the first displacement data, the second displacement data, and the third displacement data are perpendicular to each other. For example, the movement direction corresponding to the first displacement data is consistent with the propagation direction of the earthquake SV wave, the movement direction corresponding to the second displacement data is consistent with the propagation direction of the earthquake SH wave, and the movement direction corresponding to the third displacement value is consistent with the propagation direction of the earthquake P wave. Each change image is associated with an acquisition moment, and the displacement processing model processes the change image to obtain a corresponding displacement data set in a three-dimensional space, and the displacement data set in the three-dimensional space corresponds to the acquisition moment.

A rule is pre-configured to determine the corresponding amplitude information based on the displacement data. The rule is determined based on the meaning represented by the amplitude information. The amplitude information may be, for example, velocity information and acceleration information. In the case where the amplitude information represents velocity, the rule is a formula for determining velocity information based on the displacement information. In the case where the amplitude information represents acceleration, the rule is a formula for determining acceleration information based on the displacement information.

A rule for determining amplitude information is determined, and the first displacement data, the second displacement data, and the third displacement data that have been determined are processed by data calculation using the rule to obtain first amplitude information, second amplitude information, and third amplitude information. Thus, the first amplitude information, the second amplitude information, and the third amplitude information are combined to obtain an amplitude information set corresponding to the acquisition time. When the moving direction corresponding to the first displacement data is consistent with the propagation direction of the earthquake SV wave, the moving direction corresponding to the second displacement data is consistent with the propagation direction of the earthquake SH wave, and the moving direction corresponding to the third displacement value is consistent with the propagation direction of the earthquake P wave, the first amplitude information, the second amplitude information, and the third amplitude information are the amplitude information of the SV wave, the SH wave, and the P wave, respectively.

By using the multiple amplitude information sets corresponding to each of the multiple geographic location information, the epicenter of a natural earthquake, the intensity of the earthquake, and other information can be determined to carry out earthquake rescue. In addition, the multiple amplitude information sets can also be used in artificial earthquakes to detect the geographic location information of oil, gas or mineral resources, detect the development of cracks in the formation, predict the properties of underground fluids, and distinguish between dry layers, water layers, and oil and gas layers.

As shown in Figure 4, it is a flow chart of a method for determining displacement data in an embodiment of the present application. In this figure, the process of determining three displacement data in three-dimensional space is described, but more or fewer operation steps may be included based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the order of many steps and does not represent the only execution order. When the system or device product is executed in practice, it can be executed in the order of the method shown in the embodiment or the accompanying drawings or in parallel. Specifically, as shown in Figure 4, the method may include:

S421, determining an initial image and a plurality of vibration images from a plurality of change images;

S422, determining first displacement data, second displacement data and third displacement data respectively corresponding to each of the multiple change images according to an image difference between a first vibration responder region included in the multiple vibration images and a second vibration responder region included in the initial image.

According to another embodiment of the present application, determining an initial image and multiple vibration images from multiple changing images specifically includes identifying each changing image in the multiple changing images, determining the initial image from the multiple changing images based on the number of changing images similar to each changing image in the multiple changing images, and the remaining changing images are all vibration images, thereby obtaining multiple vibration images.

According to the quantity value of the change images similar to each change image in the multiple change images, the initial image is determined from the multiple change images specifically by using the image similarity processing model to process any two change images to determine the corresponding similarity values, and when it is determined that the similarity values are greater than or equal to a preset threshold, the two change images are determined to be similar; for each change image, the quantity value of images similar to the change image is counted respectively to obtain multiple quantity values; and the change image corresponding to the largest quantity value among the multiple quantity values is determined as the initial image.

For example, the multiple change images include a first image, a second image, a third image, and a fourth image. The first image and the second image are similar, and are respectively dissimilar to the third image and the fourth image, the second image is also dissimilar to the third image and the fourth image, and the third image is dissimilar to the fourth image. Then, it is determined that the quantity values corresponding to the first image and the second image are 1, respectively, and the quantity values corresponding to the third image and the fourth image are 0, respectively. Then the first image and the second image are used as the initial images. If the preset threshold is, for example, 1, then the first image and the second image are exactly the same, and any one of the images can be used as the initial image.

It should be noted that the identification of each of the multiple change images may also include, for example, identifying a boundary for each of the multiple change images, and based on the boundary, cropping each of the multiple change images to obtain multiple cropped change images; and then performing similarity calculation on the multiple cropped change images. The boundary may be, for example, a target identification area. When the vibration response object includes a mass block, the target identification area includes an area of the boundary of the mass block, and when the scale frame includes the vibration response object, the target is the area of the boundary formed by the scale frame. Since irrelevant pixels are cropped, the speed and accuracy of determining the initial image and the vibration image are improved.

The first vibration responder region is a region image of the vibration responder included in the vibration image, and the second vibration responder region is a region image of the vibration responder included in the initial image. It should be noted that the first vibration responder region and the second vibration responder region are regions where information such as the location and size of the vibration responder can be determined.

After determining the initial image and multiple vibration images, the vibration response object region is identified for the initial image and the multiple vibration images, and the position feature vector corresponding to each vibration response object is determined. The position feature vector represents the position coordinates of each pixel point of the vibration response object, or represents the position coordinates of the pixel corresponding to the target position of the vibration response object. The target position may be, for example, the center position of the vibration response object. For the initial image, the preset initial displacement data set is used as the corresponding displacement data set, and the initial displacement data set includes initial first displacement data, initial second displacement data, and initial third displacement data. For example, the initial displacement data set is (0,0,0). For each vibration image, the position feature vector of the vibration image is calculated differently from the position feature vector of the initial image to determine the first displacement data, the second displacement data, and the third displacement data of the vibration image. The difference calculation may, for example, determine the calculation formula of the displacement data scaled in the X-axis direction, and the calculation formula of the displacement data translated in the Y-axis and Z-axis directions. The calculation formula of the scaled displacement data may, for example, be a lens magnification formula.

When the vibration responder includes a mass block or an elastic object, the difference in the image refers to the position coordinates of the mass block or the elastic object in the corresponding change image. When the vibration responder is a fluid, the difference in the image refers to the flooding angle or height information of different areas of the fluid.

FIG. 5A is a flow chart of a method for determining displacement data according to another embodiment of the present application; FIG. 5B is a schematic diagram of an initial training change image according to an embodiment of the present application.

According to another embodiment of the present application, a plurality of change images are processed separately using a displacement processing model to obtain first displacement data, second displacement data, and third displacement data corresponding to each of a plurality of acquisition moments, including: inputting the plurality of change images into a trained three-dimensional displacement determination model, respectively, to obtain first displacement data, second displacement data, and third displacement data corresponding to each acquisition moment associated with each change image.

The trained three-dimensional displacement determination model is a trained neural network model, and the trained neural network model can determine the first displacement data, the second displacement data and the third displacement data corresponding to the image for the image. The trained neural network model is obtained by training the preset neural network model with the training sample set. The preset neural network model can be, for example, a neural network model composed of preset parameters, and the preset parameters can be, for example, randomly generated numbers. The deep learning model is used to determine the displacement data set corresponding to the image in which the vibration responder changes in response to the seismic signal, so as to realize the application of artificial intelligence methods in the earthquake monitoring process, improve the degree of automation and intelligence in the earthquake monitoring process, and also improve the speed and accuracy of determining the displacement data set.

According to another embodiment of the present application, the training process of the trained three-dimensional displacement determination model includes: using a preset three-dimensional displacement determination model to process multiple training change images included in a training sample set to obtain multiple predicted sample three-dimensional displacement data sets; and training the preset three-dimensional displacement determination model according to the difference between the multiple predicted sample three-dimensional displacement data sets and the multiple labels included in the training sample set to obtain the trained three-dimensional displacement determination model, wherein the predicted sample three-dimensional displacement data set includes predicted sample first displacement data, predicted sample second displacement data and predicted sample third displacement data, and the labels include sample first displacement data, sample second displacement data and sample third displacement data.

The preset three-dimensional displacement determination model includes a neural network model composed of preset parameters, and the neural network model can determine the predicted sample three-dimensional displacement data set corresponding to the image for the image. The preset parameters can be, for example, randomly generated numbers. The preset three-dimensional displacement determination model can be, for example, a network whose output is a predicted sample three-dimensional displacement data set (3×1) determined based on a target detection network, such as a network obtained based on a YOLO network or a FCOS network.

The training sample set includes a plurality of training variation images and a label corresponding to each training variation image. The label may be determined based on a sample amplitude information set, for example, and the sample amplitude information set includes first sample amplitude information, second sample amplitude information, and third sample amplitude information. The sample amplitude information set may be obtained based on any existing detector, for example. A preset image is obtained, and for each label, a variation is performed on the basis of the preset image to obtain a training variation image corresponding to each label.

According to the difference between the multiple labels included in the three-dimensional displacement data sets of multiple prediction samples and the training sample sets, the preset three-dimensional displacement determination model is trained to obtain the trained three-dimensional displacement determination model. Specifically, the multiple prediction sample three-dimensional displacement data sets and multiple labels are processed using a loss function to determine the loss function value, and the preset three-dimensional displacement determination model is trained based on the loss function value. When the preset conditions are met, the training three-dimensional displacement determination model of the current round is determined to be the trained three-dimensional displacement determination model. The preset conditions can be, for example, that the loss function value sequence obtained by multiple rounds of training processes converges and the training round is the target training round, etc.

According to another embodiment of the present application, the construction process of the training sample set includes: obtaining multiple first sample amplitude information corresponding to a change direction; performing expansion processing on each of the multiple first sample amplitude information to obtain multiple sample first displacement data, multiple sample second displacement data and multiple sample third displacement data; obtaining an initial training change image; and processing the initial training change image according to the multiple sample first displacement data, the multiple sample second displacement data and the multiple sample third displacement data to obtain multiple training change images.

A change direction may be, for example, any change direction, such as a change direction consistent with the X-axis direction, a change direction consistent with the Y-axis direction, or a change direction consistent with the Z-axis direction. The first sample amplitude information corresponding to the change direction may be, for example, the first sample amplitude information corresponding to the change direction obtained by using any existing detector.

The extended processing may include, for example, any existing method for determining amplitude information in two other directions based on amplitude information in one change direction and a model or method for expanding sample data, where the sample data is first sample amplitude information, second sample amplitude information, and third sample amplitude information corresponding to the three change directions, respectively. Based on a preset method for determining displacement data from amplitude information, the first sample amplitude information, the second sample amplitude information, and the third sample amplitude information are processed to obtain a plurality of sample first displacement data, a plurality of sample second displacement data, and a plurality of sample third displacement data.

The initial training variation image may be, for example, an initial image among a plurality of variation images. Since this embodiment uses the initial image among the variation images of the amplitude information set to be determined as the initial training variation image instead of the preset image, the error caused by the inconsistency between the initial image and the preset image due to the aging of the vibration responder is avoided, thereby achieving the accuracy of the displacement data set in the determined three-dimensional space without correction.

Determine the position feature vector of the initial image, determine multiple training position feature vectors based on the difference between each sample displacement data set and the initial sample displacement data set corresponding to the initial training variation image, and construct a corresponding training variation image based on each training position feature vector to obtain multiple training variation images. The one sample displacement data set includes a sample first displacement data, a sample second displacement data and a sample third displacement data. It should be noted that the initial sample displacement data set corresponding to the initial training variation image is (0,0,0).

For example, as shown in FIG5A , multiple first sample amplitude information 511 is expanded to obtain multiple sample displacement data sets 512. An initial training variation image is determined based on the multiple variation images 501 acquired, and image processing is performed on the initial training variation image based on the multiple sample displacement data sets 512 to obtain multiple training variation images, and a training sample set 520 is obtained based on the multiple training variation images and the multiple sample displacement data sets 512 as labels. The preset three-dimensional displacement determination model 531 is trained using the training sample set 520 to obtain a trained three-dimensional displacement determination model 532.

The collected multiple change images 501 are input into the trained three-dimensional displacement determination model 532 to obtain multiple displacement data sets 502. Each displacement data set includes a first displacement data, a second displacement data and a third displacement data, and each displacement data set corresponds to a change image and a corresponding collection time.

According to another embodiment of the present application, performing expansion processing on each of the multiple first sample amplitude information to obtain multiple sample first displacement data, multiple sample second displacement data and multiple sample third displacement data includes: performing processing on each of the first sample amplitude information according to the mapping relationship between the amplitude information in three change directions to obtain multiple second sample amplitude information and multiple third sample amplitude information; and performing sample expansion and normalization processing on the multiple first sample amplitude information, the multiple second sample amplitude information and the multiple third sample amplitude information to obtain multiple sample first displacement data, multiple sample second displacement data and multiple sample third displacement data.

The mapping relationship is specifically the correspondence between the amplitude information corresponding to one change direction and the amplitude information intervals of the other two change directions. For example, if the one change direction is a change direction consistent with the X-axis direction, then each first sample amplitude information has an amplitude interval corresponding to the Y-axis direction and an amplitude interval corresponding to the Z-axis direction. Based on the mapping relationship, for each first sample amplitude information, multiple second sample amplitude information and multiple third sample amplitude information are determined respectively. Thus, for each first sample amplitude information, multiple sub-sample amplitude information sets are determined, and each sub-sample amplitude information set includes a first sample amplitude information, a second sample amplitude information and a third sample amplitude information set. Multiple sub-sample amplitude information sets constitute a sample amplitude information set.

The sample amplitude information set is processed using any method that can achieve training sample expansion to obtain an expanded sample set to increase the data used for training samples. For example, each sample amplitude information is randomly interfered. Based on a preset method of determining displacement data from amplitude information, the expanded sample set is processed to obtain a plurality of sample first displacement data, a plurality of sample second displacement data, and a plurality of sample third displacement data, which are labels. The preset method of determining displacement data from amplitude information can be, for example, a normalization method, that is, the original expanded sample set is normalized to the range of [-1,1], and the normalized data is determined as the corresponding sample displacement data set.

According to another embodiment of the present application, obtaining an initial training change image corresponding to the first sample amplitude information includes: obtaining a plurality of change images; performing overall boundary recognition on each change image in the plurality of change images, and based on the overall boundary, cropping each change image in the plurality of change images to obtain a plurality of cropped change images; and determining an initial training change image from the plurality of cropped change images according to a number value of cropped change images similar to each cropped change image in the plurality of cropped change images.

The overall boundary recognition may be, for example, processing the change image using any model that can determine the target recognition area to obtain the overall boundary of each change image. The areas outside the overall boundary are all areas composed of pixels that are irrelevant to the determination of the displacement data set. For multiple cropped change images, for example, the method of step S421 in FIG. 4 above may be used to determine the initial image, and the initial image may be used as the initial training change image.

According to another embodiment of the present application, an initial training variation image is processed according to a plurality of sample first displacement data, a plurality of sample second displacement data, and a plurality of sample third displacement data to obtain a plurality of training variation images, including: performing vibration response object boundary identification on the initial training variation image, and removing a sub-image included in the vibration response object boundary from the initial training variation image to obtain a target initial training variation image; performing variation processing on a sub-image according to a plurality of sample first displacement data, a plurality of sample second displacement data, and a plurality of sample third displacement data to obtain a plurality of target sub-images; and splicing each of the plurality of target sub-images with the target initial training variation image to obtain a plurality of training variation images.

The boundary of the vibration-responsive object is a boundary that may include the entire vibration-responsive object. The sub-image is a regional image composed of all pixel points included in the boundary of the vibration-responsive object.

In the case where the vibration responder includes a mass block or an elastic object, all pixels of the mass block or the elastic object as a whole in the initial image are removed to obtain a target initial training change image. Then, the position and size of the mass block or the position and size of the elastic object are adaptively processed using multiple sample first displacement data, multiple sample second displacement data, and multiple sample third displacement data to obtain a target sub-image. The target sub-image is spliced with the target initial training change image to obtain multiple training change images. It should be noted that in the case where the vibration responder is a fluid, the change processing of the sub-image is the adaptive processing of the flooding angle or height of different regions of the fluid.

For example, as shown in FIG5B , the initial training variation image 5011 is subjected to overall boundary recognition, the overall boundary 5012 in the initial training variation image 5011 is determined, and the initial training variation image 5011 is cropped based on the overall boundary 5012 to obtain a cropped variation image. Furthermore, the cropped variation image is subjected to vibration response object boundary recognition, the boundary including the vibration response object is determined, and the area included by the boundary is used as a sub-image 5013.

The sub-image is adaptively processed using a plurality of sample first displacement data, a plurality of sample second displacement data, and a plurality of sample third displacement data to obtain a plurality of training change images. In the case where the vibration response object includes a mass block, the adaptive processing is at least one of shifting the sub-image to the left, shifting the sub-image to the right, and scaling the sub-image to obtain a corresponding target sub-image.

FIG6 is a schematic diagram of the structure of a seismic signal monitoring device according to an embodiment of the present application.

The present application provides a seismic signal monitoring device, including a box 610; a vibration responder 620, which is arranged inside the box 610 and moves under the action of the seismic signal transmitted by the box 610; and a camera module 630, which is arranged inside the box 610 and faces the vibration responder 620 to collect changing images generated by the movement of the vibration responder.

Since the electromagnetic coil is abandoned, any vibration response object that can vibrate in response to the seismic signal can be selected, making the seismic signal monitoring device more selective. Since the amplitude information is calculated directly based on the acquired change image, there is no need for various electrical signals as intermediaries, so the reflection of the seismic signal amplitude is more realistic, that is, the signal fidelity is higher and the cost is reduced.

According to an embodiment of the present application, the camera module 630 is fixedly connected to the box 610 to ensure that when the box vibrates, the camera module 630 and the box 610 are relatively stationary. The vibration response object 620 can be an object that moves relative to the box 610 when the box moves in response to the earthquake signal. Since the vibration response object 620 generates relative movement with the box 610, the vibration response object 620 also moves relative to the camera module 630, and then the camera module 630 can collect the changing image of the vibration response object 620 to determine the amplitude information set corresponding to the collection time.

The movement of the vibration response object 620 can be used to determine the amplitude information set based on the inertia principle. Specifically, when an earthquake occurs, the camera module 630 is fixed on the box 610, and the two move synchronously and are relatively still. Due to the effect of inertia, the vibration response object 620 produces relative movement with the camera module 630, so that the amplitude information of the earthquake signal can be determined based on the size of the movement.

It should be noted that the camera module 630 and the vibration responder 620 in FIG. 6 are facing each other left and right, but the camera module 630 and the vibration responder 620 may also be facing each other up and down to capture the changing images generated by the movement of the vibration responder.

It should be noted that the vibration responder 620 is of a preset size or there is a preset distance between the vibration responder 620 and the camera module 630 to ensure that no matter how the vibration responder 620 changes, it will completely fall into the change image captured by the camera module.

The camera module 630 can also be an electronic device with timing positioning, shooting, storage and data transmission functions, which can specifically include a master control module, a timing positioning module, a camera module, a storage module, and a data interface module. The camera module can be modified based on a general USB camera. The modification process can be, for example, using a 1/4 inch (about 3.9×2.5mm) CMOS photosensitive film and a 420-frame black and white imaging mode. The storage module can be, for example, a 128Gb memory card. The timing positioning module is used to ensure that the determined amplitude information set has corresponding time information and geographic location information.

The frame rate of the camera module is determined according to the Nyquist sampling theorem and research requirements, and needs to be twice or higher than the maximum frequency of the required effective signal. In the case of ultra-high-speed video recording with a frame rate of more than 500 frames per second, high-sampling shooting can be achieved by staggered shooting with two or more low-speed camera modules and interlacing images in the later stage. For example, 500 frames per second shooting can be achieved by two camera modules with a frame rate of 250 frames per second, starting the shooting separately with a difference of 2ms, and inserting the pictures taken by the later started modules into the back of the first 2ms pictures taken by the first started modules in the later stage. At this time, the sampling rate of all images is 2ms, and the actual frame rate is 500 frames per second.

It should be noted that the above-mentioned seismic signal monitoring method can be executed by the general control module included in the camera module 630, and can also be executed by an electronic device connected to the data interface module of the camera module.

According to another embodiment of the present application, the vibration responder 620 includes a plurality of elastic objects.

When the vibration responding object is an elastic object, a signal with a wider frequency band can be collected without any electromagnetic interference, and the signal-to-noise ratio is higher, so that the accuracy of the determined amplitude signal set is higher.

The elastic object may be, for example, an elastic ball or other elastic object. When the box 610 moves in response to the earthquake signal, the elastic object may generate relative motion.

According to another embodiment of the present application, the plurality of elastic objects have different masses respectively.

According to another embodiment of the present application, the plurality of elastic objects respectively have elastic components with different masses and different parameters.

Elastic objects of different masses can collect signals of different frequency bands, thereby collecting signals with a wider frequency band compared to the prior art.

According to another embodiment of the present application, the vibration responder 620 includes a fluid.

When the vibration responder is a fluid, a signal with a wider frequency band can be collected without any electromagnetic interference, and the signal-to-noise ratio is higher, so that the accuracy of the determined amplitude signal set is higher.

When seismic information exists, the fluid may produce corresponding deformations, and the camera module 630 collects corresponding change images at each collection moment to determine the corresponding amplitude information set.

FIG. 7 is a schematic diagram of the structure of a seismic signal monitoring device according to another embodiment of the present application.

According to another embodiment of the present application, the vibration responder includes a mass block 721 and an elastic component 722 , and the mass block 721 is elastically connected to the box body through the elastic component 722 .

When the vibration responder includes an elastic component and a mass block, the vibration response of the vibration responder is completely determined by the elastic coefficient of the elastic component and the weight of the mass block, is not subject to any electromagnetic interference, and has a higher signal-to-noise ratio, making the determined amplitude signal set more accurate.

The mass block 721 is suspended in the box through the elastic component 722. Due to the dual effects of inertia and elasticity, it produces relative movement with the camera module, so that the amplitude information of the seismic signal can be determined based on the size of the movement.

The elastic component 722 may be, for example, a coil spring, a spring sheet, a rubber band, or the like.

It should be noted that FIG. 7 only shows a case of one mass block, but the vibration responder may include, for example, multiple mass blocks, and the masses of the multiple mass blocks may be the same or different.

According to another embodiment of the present application, the seismic signal monitoring device further includes a scale frame 740 , which is disposed inside the box and fixedly connected to the box, and the vibration responder is disposed inside the scale frame 740 .

It should be noted that the box body may include a plurality of scale frames 740, each of which is fixedly connected to the box body, and each of which has a vibration responder disposed therein, such as 2, 4, 8, etc.

According to another embodiment of the present application, the mass block includes a light source, and the light beam emitted by the light source always falls into the changing image captured by the camera module.

The light source may be, for example, a thin beam laser emitting mechanism. The light source may also include a corresponding pattern inside, and the pattern may be imaged in the camera module to increase the accuracy of calculating the displacement data set of the change image.

FIG. 8 is a front view of a vibration responder according to an embodiment of the present application.

According to another embodiment of the present application, the seismic signal monitoring device further includes a scale 823, and the scale 823 is closely connected to the vibration response object.

As shown in Fig. 8, the front view of the vibration responder includes a scale frame 840, an elastic component 822, a scale 823, and the vibration responder wrapped by the scale 823. Corresponding scale lines can be configured at the boundaries of the scale frame 840, for example.

Fig. 9A is a schematic diagram of the structure of a seismic signal monitoring device according to another embodiment of the present application. Fig. 9B is a schematic diagram of the structure of a seismic signal monitoring device according to another embodiment of the present application.

According to another embodiment of the present application, the seismic signal monitoring device further includes a counterweight 950 to adjust the center of gravity of the box 910 so that the center of gravity of the box 910 remains on the vertical center line of the box 910 .

According to another embodiment of the present application, the seismic signal monitoring device further includes a power supply mechanism 960 , which is disposed inside the box 910 to supply power to the camera module 930 .

According to another embodiment of the present application, the seismic signal monitoring device further includes a power supply mechanism 960 , which is disposed inside the box 910 to supply power to the camera module 930 and the light source.

According to another embodiment of the present application, the seismic signal monitoring device further includes a tail vertebrae mechanism 970, which is disposed outside the box 910 and is fixedly connected to the box 910 to increase the coupling degree between the box 910 and the formation.

As shown in FIG. 9A , the seismic signal monitoring device includes a box 910 , a mass block 921 , an elastic component 922 , a camera module 930 , a scale frame 940 , a counterweight block 950 , a power supply mechanism 960 and a coccyx mechanism 970 .

When the seismic signal monitoring device is used to determine the amplitude information set, the seismic signal monitoring device is buried at a preset depth underground, or the seismic signal monitoring device is placed on the surface of a device whose amplitude information set is to be measured, and an adhesive is used to increase the coupling degree.

Specifically, the seismic signal monitoring device can be that the side length of the mass block 921 is 2×2×1 (cm), and f=2.8mm fixed focus imaging is adopted. The scale size including the mass block 921 is set to 2×2 (cm). The scale range of the scale frame 940 is 6×6 (cm). The X-axis coordinate of the center of the camera module 930 lens is 11 cm. The CMOS photosensitive film is located outside the X-axis coordinate 11.2 cm. When the ruler is in a stationary state, the object distance u=80 mm, and the magnification can be calculated to be approximately 0.036 from the convex lens magnification formula. At this time, the scale lines of the scale frame 940 and the imaging size of the ruler are approximately 2.16×2.16 (mm) and 0.72×0.72 (mm), respectively, thereby ensuring that they can be fully imaged on the CMOS photosensitive film. When the scale vibrates to the maximum in the X direction, the object distance u = 70 mm, and the magnification is 0.042. At this time, the image size of the scale line remains unchanged, and the image size of the scale is about 0.84×0.84 (mm), which is 1.17 times larger than the image size of the scale in the static state. The entire node instrument is powered by a built-in 20000mA lithium battery.

It should be noted that, for example, the outside of the box 910 may also include a directional mark to ensure that the seismic signal monitoring device does not tilt when it is placed in a preset position, so as to determine an accurate set of amplitude information. The directional mark may include, for example, three sub-directional marks, each of which corresponds to a direction in a three-dimensional space. The directional mark may be, for example, an arrow mark or a level tube, etc. The outer surface of the box 910 may be increased with undulating stripes according to the needs of the coupling strata to increase the coupling degree between the seismic signal monitoring device and the bottom layer. The seismic signal monitoring device may, for example, be a combination of two hollow rectangular or trapezoidal bodies that are larger on the top and smaller on the bottom, which may be opened to install and detect the internal equipment status, and a transparent observation window may be reserved, and the appearance is shown in FIG9B. The size of the upper sub-box in the up-down direction of the box may, for example, be 12×8×8 (cm), and the size of the lower sub-box may, for example, be 8×8×6 (cm). The lower part of the seismic signal monitoring device in the up-down direction of FIG9B may also include a tail vertebrae mechanism, which is not shown in the figure.

FIG10 is a schematic diagram of the structure of a computer device in an embodiment of the present application. The apparatus in the present application may be a computer device in the present embodiment, executing the method of the present application described above. The computer device 1002 may include one or more processing devices 1004, such as one or more central processing units (CPUs), each of which may implement one or more hardware threads. The computer device 1002 may also include any storage resource 1006, which is used to store any kind of information such as code, settings, data, etc. Non-limitingly, for example, the storage resource 1006 may include any one or more combinations of the following: any type of RAM, any type of ROM, flash memory device, hard disk, optical disk, etc. More generally, any storage resource may use any technology to store information. Further, any storage resource may provide volatile or non-volatile retention of information. Further, any storage resource may represent a fixed or removable component of the computer device 1002. In one case, when the processing device 1004 executes an associated instruction stored in any storage resource or a combination of storage resources, the computer device 1002 may perform any operation of the associated instruction. The computer device 1002 also includes one or more drive mechanisms 1008 for interacting with any storage resources, such as a hard disk drive mechanism, an optical disk drive mechanism, and the like.

The computer device 1002 may also include an input/output module 1010 (I/O) for receiving various inputs (via input devices 1012) and for providing various outputs (via output devices 1014). A specific output mechanism may include a presentation device 1016 and an associated graphical user interface (GUI) 1018. In other embodiments, the input/output module 1010 (I/O), the input device 1012, and the output device 1014 may not be included, and the computer device 1002 may be used as a computer device in a network. The computer device 1002 may also include one or more network interfaces 1020 for exchanging data with other devices via one or more communication links 1022. One or more communication buses 1024 couple the components described above together.

The communication link 1022 may be implemented in any manner, for example, through a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication link 1022 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc. governed by any protocol or combination of protocols.

An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above method is implemented.

An embodiment of the present application also provides a computer program product, which includes a computer program, and the computer program implements the above method when executed by a processor.

Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

The present application is described with reference to the flowchart and/or block diagram of the method, device (system) and computer program product according to the embodiment of the present application. It should be understood that each process and/or box in the flowchart and/or block diagram, and the combination of the process and/or box in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for realizing the function specified in one process or multiple processes in the flowchart and/or one box or multiple boxes in the block diagram.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

The above specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above are only specific embodiments of the present application and are not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

A method for monitoring earthquake signals, comprising:

At a plurality of acquisition moments, a plurality of change images of the vibration response object are acquired, wherein the acquisition moments are associated with the change images;

Using a displacement processing model, the plurality of change images are processed respectively to obtain first displacement data, second displacement data and third displacement data in a three-dimensional space corresponding to each of the plurality of acquisition moments; and

determining an amplitude information set corresponding to each acquisition moment according to the first displacement data, the second displacement data, and the third displacement data corresponding to each acquisition moment,

The change image includes an image of the vibration responder changing in response to the seismic signal.
The method according to claim 1, characterized in that the use of the displacement processing model to process the multiple change images respectively to obtain the first displacement data, the second displacement data and the third displacement data corresponding to each of the multiple acquisition moments respectively comprises:

determining an initial image and a plurality of shaken images from among the plurality of changed images; and

The first displacement data, the second displacement data and the third displacement data respectively corresponding to each of the multiple change images are determined according to an image difference between a first vibration responder region included in the multiple vibration images and a second vibration responder region included in the initial image.
The method according to claim 1, characterized in that the use of the displacement processing model to process the multiple change images respectively to obtain the first displacement data, the second displacement data and the third displacement data corresponding to each of the multiple acquisition moments respectively comprises:

The multiple change images are respectively input into the trained three-dimensional displacement determination model to obtain the first displacement data, the second displacement data and the third displacement data respectively corresponding to each acquisition moment associated with each change image.
The method according to claim 3, characterized in that the training process of the trained three-dimensional displacement determination model comprises:

Using a preset three-dimensional displacement determination model, a plurality of training change images included in the training sample set are processed to obtain a plurality of prediction sample three-dimensional displacement data sets; and

According to the differences between the plurality of predicted sample three-dimensional displacement data sets and the plurality of labels included in the training sample set, the preset three-dimensional displacement determination model is trained to obtain the trained three-dimensional displacement determination model,

The predicted sample three-way displacement data set includes predicted sample first displacement data, predicted sample second displacement data and predicted sample third displacement data, and the label includes sample first displacement data, sample second displacement data and sample third displacement data.
The method according to claim 4, characterized in that the process of constructing the training sample set comprises:

Acquire a plurality of first sample amplitude information corresponding to a change direction;

Performing expansion processing on each of the plurality of first sample amplitude information respectively to obtain the plurality of sample first displacement data, the plurality of sample second displacement data and the plurality of sample third displacement data;

obtaining an initial training variation image; and

The initial training change image is processed according to the plurality of sample first displacement data, the plurality of sample second displacement data and the plurality of sample third displacement data to obtain the plurality of training change images.
The method according to claim 5, characterized in that the step of performing expansion processing on each of the plurality of first sample amplitude information to obtain the plurality of sample first displacement data, the plurality of sample second displacement data, and the plurality of sample third displacement data comprises:

According to the mapping relationship between the amplitude information in three change directions, each of the first sample amplitude information is processed respectively to obtain a plurality of second sample amplitude information and a plurality of third sample amplitude information; and

Sample expansion and normalization processing are performed on the plurality of first sample amplitude information, the plurality of second sample amplitude information, and the plurality of third sample amplitude information to obtain the plurality of sample first displacement data, the plurality of sample second displacement data, and the plurality of sample third displacement data.
The method according to claim 5, characterized in that the obtaining of the initial training change image corresponding to the first sample amplitude information comprises:

acquiring the plurality of changed images;

Performing overall boundary recognition on each of the multiple change images, and based on the overall boundary, cropping each of the multiple change images to obtain multiple cropped change images; and

The initial training variation image is determined from the plurality of cropped variation images according to a number value of cropped variation images similar to each of the plurality of cropped variation images.
The method according to claim 5, characterized in that the processing of the initial training change image according to the plurality of sample first displacement data, the plurality of sample second displacement data and the plurality of sample third displacement data to obtain the plurality of training change images comprises:

Performing vibration response object boundary recognition on the initial training variation image, and removing the sub-image included in the vibration response object boundary from the initial training variation image to obtain a target initial training variation image;

According to the plurality of sample first displacement data, the plurality of sample second displacement data and the plurality of sample third displacement data, the sub-image is subjected to change processing to obtain a plurality of target sub-images; and

Each target sub-image in the multiple target sub-images is spliced with the target initial training variation image to obtain the multiple training variation images.
A seismic signal monitoring device, characterized in that it comprises:

Box;

a vibration responder, the vibration responder being arranged inside the box and moving under the action of a seismic signal transmitted by the box; and

A camera module is arranged inside the box and faces the vibration responder to collect changing images generated by the movement of the vibration responder.
The device according to claim 9 is characterized in that it also includes a scale frame, the scale frame is arranged inside the box and fixedly connected to the box, and the vibration responder is arranged inside the scale frame.
The device according to claim 9 is characterized in that it also includes a ruler, and the ruler is closely connected to the vibration responder.
The device according to claim 9 is characterized in that the vibration responder includes a mass block and an elastic component, and the mass block is elastically connected to the box through the elastic component.
The device according to claim 12 is characterized in that the mass block includes a light source, and the light beam emitted by the light source always falls into the changing image captured by the camera module.
The device according to claim 9 is characterized in that the vibration responder includes a plurality of elastic objects.
The device according to claim 14, characterized in that the plurality of elastic objects have different masses.
The device according to claim 9, characterized in that the vibration responsive substance comprises a fluid.
The device according to claim 9 is characterized in that it also includes a tail vertebrae mechanism, which is arranged outside the box and fixedly connected to the box to increase the coupling degree between the box and the formation.
The device according to claim 9 is characterized in that it also includes a counterweight block to adjust the center of gravity of the box so that the center of gravity of the box is maintained on the vertical center line of the box.
The device according to claim 9 is characterized in that it also includes a power supply mechanism, which is arranged inside the box to supply power to the camera module.
The device according to claim 13 is characterized in that it also includes a power supply mechanism, which is arranged inside the box to supply power to the camera module and the light source.