Video-based seismic intensity calculation method, device and equipment

Abstract

The invention provides a video-based seismic intensity calculation method, device and equipment. The method comprises the following steps: acquiring a monitoring video of a target area where an earthquake occurs in a target time period before and after the earthquake moment, and calibrating a target object in the monitoring video; performing target tracking on the target object in the monitoring video, and outputting a tracking result; if the tracking result represents that the target object is lost in the monitoring video, recording an image when the target object is lost in the monitoring video as a target image, and performing full-image feature extraction on the target image to determine a suspected target object; recording an image before the target object is lost in the monitoring video as a historical image, and calculating a motion prediction area when the target object is lost based on the historical image; re-determining the target object based on the suspected target object and the motion prediction area; and calculating the seismic intensity of the target area according to the historical image and the target image after the target object is re-determined. According to the invention, the seismic intensity can be calculated timely and accurately.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical digital data processing, and particularly relates to a method, device, and equipment for calculating earthquake intensity based on video. Background Art

[0002] The rapid determination of earthquake intensity is of great significance for the disaster relief work of small and medium-sized earthquakes. However, according to the existing technical level, the magnitude assessment after an earthquake is a relatively complex, time-consuming, and laborious task with a huge workload.

[0003] Existing assessment methods are mostly qualitative assessments, lacking scientific objectivity. And existing earthquake intensity assessments mostly rely on microseismic instruments distributed in various regions to judge, but the signals of microseismic instruments depend on conventional communication means and cannot output signals in a timely manner. Moreover, for earthquakes with relatively mild intensities, the measurement data is not accurate enough. Summary of the Invention

[0004] Embodiments of the present invention provide a method, device, and equipment for calculating earthquake intensity based on video to solve the problem of inaccurate earthquake intensity calculation in the existing technology.

[0005] In a first aspect, embodiments of the present invention provide a method for calculating earthquake intensity based on video, including: Obtain the surveillance video of the target area where the earthquake occurred during the target time period before and after the earthquake moment, and calibrate the target object in the surveillance video.

[0006] Perform target tracking on the target object in the surveillance video and output the tracking result.

[0007] If the tracking result indicates that the target object is lost in the surveillance video, record the image when the target object is lost in the surveillance video as the target image, and perform feature extraction on the entire target image to determine the suspected target object.

[0008] Record the image before the target object is lost in the surveillance video as the historical image, and calculate the motion prediction area of the target object when it is lost based on the historical image.

[0009] Based on the suspected target object and the motion prediction area, re-determine the target object in the target image.

[0010] Calculate the earthquake intensity of the target area where the earthquake occurred based on the historical image and the target image after re-determining the target object.

[0011] In a possible implementation manner, performing feature extraction on the entire target image to determine the suspected target object includes: Perform feature extraction on the entire target image to obtain the features of each object in the target image, denoted as the first features.

[0012] Select the first feature that is most similar to the feature of the target object among each of the first features, and determine the object corresponding to the most similar first feature as the suspected target object.

[0013] In a possible implementation, calculating a motion prediction area of the target object when it is lost based on historical images includes: For each image in the historical images, determine a first area based on the target object on the image and a preset range.

[0014] Obtain the motion prediction area of the target object when it is lost according to all the first areas.

[0015] In a possible implementation, re-determining the target object in the target image based on the suspected target object and the motion prediction area includes: Determine the occurrence probability of each type of first area according to all the first areas.

[0016] Determine the target area based on each type of first area and the corresponding occurrence probability.

[0017] Re-determine the suspected target object in the target area as the target object in the target image.

[0018] In a possible implementation, calculating the seismic intensity of the target area where an earthquake occurs according to the historical images and the target image after re-determining the target object includes: Calculate the peak ground acceleration and peak ground velocity based on the target object in the historical images and the target object in the target image after re-determining the target object.

[0019] Calculate the seismic intensity of the target earthquake area based on the peak ground acceleration and peak ground velocity.

[0020] In a possible implementation, the method further includes: If the tracking result indicates that the target object is not lost in the surveillance video, calculate the peak ground acceleration and peak ground velocity based on the target object in the surveillance video.

[0021] Calculate the seismic intensity of the target earthquake area based on the peak ground acceleration and peak ground velocity.

[0022] In a possible implementation, performing target tracking on the target object in the surveillance video and outputting a tracking result includes: Perform target tracking on the target object in the surveillance video through a kernel correlation filter or a channel and spatial reliability tracker to obtain a tracking result.

[0023] If the tracking result indicates that the target object is lost in the surveillance video, output the moment when the target object is lost.

[0024] In a possible implementation, obtain the surveillance video of the target area where the earthquake occurred in the target time period before and after the earthquake moment, and calibrate the target object in the surveillance video, including: Obtain the surveillance video of the target area where the earthquake occurred in the target time period before and after the earthquake moment, and extract the first video frame image in the surveillance video, denoted as the starting video frame image.

[0025] Obtain the features of all objects in the starting video frame image.

[0026] For each object, calculate the feature similarity between this object and the other objects, and calculate the sum of all feature similarities between this object and the other objects to obtain the feature index corresponding to this object.

[0027] Calibrate the object corresponding to the minimum feature index as the target object in the video.

[0028] In a second aspect, an embodiment of the present invention provides a video-based earthquake intensity calculation device, including: A first processing module, configured to obtain the surveillance video of the target area where the earthquake occurred in the target time period before and after the earthquake moment, and calibrate the target object in the surveillance video.

[0029] A second processing module, configured to perform target tracking on the target object in the surveillance video and output the tracking result.

[0030] A third processing module, configured to, if the tracking result indicates that the target object is lost in the surveillance video, record the image when the target object is lost in the surveillance video as the target image, and perform feature extraction on the entire target image to determine the suspected target object.

[0031] A fourth processing module, configured to record the image before the target object is lost in the surveillance video as the historical image, and calculate the motion prediction area of the target object when it is lost based on the historical image.

[0032] A fifth processing module, configured to re-determine the target object in the target image based on the suspected target object and the motion prediction area.

[0033] A sixth processing module, configured to calculate the earthquake intensity of the target area where the earthquake occurred according to the historical image and the target image after re-determining the target object.

[0034] In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, it implements the method in the first aspect or any possible implementation manner of the first aspect as above.

[0035] In the embodiments of the present invention, a reference object is determined by calibrating the target object in the video of the target area during an earthquake. Then, the target object is tracked through a target tracking algorithm to obtain a tracking result, and an earthquake intensity calculation scheme is designed for different situations based on different tracking results. The principle is that when the earthquake intensity is relatively small, the target object will not be lost in the tracking result of the target tracking algorithm; when the earthquake intensity is relatively large, the target object may be lost in the tracking result of the target tracking algorithm. In addition, during an earthquake, due to the vibration of the ground itself, other objects similar to the target object may appear in the monitoring video. At this time, errors will occur in the results calculated by a simple target tracking algorithm. Similarly, due to the irregularity (vibration direction) of the earthquake vibration, the method of predicting the position where the target object appears based on historical data in the traditional method has too large a deviation and cannot meet the requirements of the accuracy of earthquake intensity calculation. Therefore, a method of combining suspected target objects and motion prediction areas to re-determine the target object is designed to ensure the accuracy of the re-determined target object, so as to ensure that the earthquake intensity of the target earthquake area can be accurately calculated when the earthquake intensity is relatively large. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a flowchart of the implementation of the method for calculating earthquake intensity based on video provided by the embodiments of the present invention; Figure 2 is a schematic structural diagram of the device for calculating earthquake intensity based on video provided by the embodiments of the present invention; [[ID=!2]] Figure 3 is a schematic diagram of the electronic device provided by the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] The embodiments of the present invention will be described in detail below with reference to the drawings.

[0038] See Figure 1 , which shows a flowchart of the implementation of the method for calculating earthquake intensity based on video provided by the embodiments of the present invention, and is described in detail as follows: Step 101, obtain the monitoring video of the target area where an earthquake occurs in the target time period before and after the earthquake moment, and calibrate the target object in the monitoring video.

[0039] In a possible implementation manner, step 101 may include: Obtain the monitoring video of the target area where an earthquake occurs in the target time period before and after the earthquake moment, and extract the first video frame image in the monitoring video, denoted as the starting video frame image.

[0040] Obtain the features of all objects in the starting video frame image.

[0041] For each object, calculate the feature similarity between this object and the remaining objects, and calculate the sum of all feature similarities between this object and the remaining objects to obtain the feature index corresponding to this object.

[0042] Calibrate the object corresponding to the minimum value of the feature index as the target object in the video.

[0043] Exemplarily, in order to ensure that the target object is an object that is easy to detect and to ensure the accuracy of target tracking, an object that is least similar to other objects is selected in the starting video frame image, that is, the object corresponding to the minimum value of the feature index.

[0044] Step 102, perform target tracking on the target object in the surveillance video and output the tracking result.

[0045] In a possible implementation manner, step 102 may include: Perform target tracking on the target object in the surveillance video through a Kernelized Correlation Filters (KCF) or a Channel and Spatial Reliability Tracking (CSRT) to obtain the tracking result.

[0046] If the tracking result indicates that the target object is lost in the surveillance video, output the moment when the target object is lost.

[0047] Exemplarily, the process of performing target tracking on the target object in the surveillance video through a kernelized correlation filter may specifically be: 1. Initialization: Extract the initial position and appearance features (such as grayscale, HOG features) of the target object.

[0048] Construct a target template and a background template, and train an initial classifier through a correlation filter.

[0049] 2. Tracking process: Search area: Set a search area centered on the target position in the next frame.

[0050] Feature extraction: Extract the same features (such as HOG) as those in the initialization for the search area.

[0051] Correlation operation: Use Fourier transform to calculate the correlation response between the target and the search area, and find the position with the maximum response as the target position.

[0052] Update the filter: Update the filter according to the features at the new position to adapt to the change of the target position.

[0053] 3. Optimization and robustness enhancement: Multi-scale estimation: Estimate the target scale change by scaling the search area.

[0054] Re-detection mechanism: If the tracking fails (such as the response value is too low), re-initialize and output the tracking result of tracking failure.

[0055] Although this scheme is simple (the algorithm structure is clear and easy to implement) and efficient (based on Fourier transform and GPU acceleration, suitable for real-time monitoring), it is sensitive to occlusion and fast movement (it is easy to fail when the target is occluded or the appearance changes drastically) and depends on the initial features (if the initial features are not selected properly, the tracking effect will decline under the condition of light change).

[0056] Exemplarily, the process of performing target tracking on a target object in a monitoring video and obtaining a tracking result through a channel and spatial reliability tracker can specifically be as follows: 1. Feature extraction: Extract multi-channel features of the target object (such as color histogram, HOG, depth map, etc.).

[0057] 2. Reliability calculation: Channel reliability: Calculate the confidence of each feature channel (such as variance, gradient energy).

[0058] Spatial reliability: Calculate the weight of each pixel according to the position of the target object and the background similarity (such as Gaussian distribution).

[0059] 3. Fusion and tracking: Perform weighted fusion on the features to construct a joint feature vector.

[0060] Use a classifier (such as SVM, random forest) or a regression model (such as Kalman filter) to predict the target object.

[0061] 4. Update and re-detection: Dynamically adjust the reliability weights according to the new observations.

[0062] If the tracking fails (such as the confidence is too low), trigger re-detection and output the tracking result of tracking failure.

[0063] This solution has good adaptability to occlusion, illumination changes and background interference, and can be combined with multiple feature channels to adapt to complex scenes. It can make up for the shortcomings of the kernel correlation filter. Therefore, in the solution of this application, the target can also be tracked simultaneously by two solutions. When the tracking of the two solutions fails at the same time, the output tracking result is defined as the target object being lost in the monitoring video. This combination can be more suitable for earthquake intensity calculation, because earthquakes are uncertain, and different earthquake intensities will also lead to obvious changes in the environment. Therefore, it is easy to fail to calculate the earthquake intensity by only using the same tracking detection method. In step 103 , if the tracking result indicates that the target object is lost in the surveillance video, the image in the surveillance video when the target object is lost is recorded as the target image, and full-image feature extraction is performed on the target image to determine a suspected target object.

[0064] In one possible implementation, performing full-image feature extraction on a target image to determine a suspected target object includes: Perform full-image feature extraction on the target image to obtain the features of each object in the target image, which are recorded as the first features.

[0065] A first feature that is most similar to a feature of a target object is selected from each first feature, and an object corresponding to the most similar first feature is determined as a suspected target object.

[0066] Exemplarily, the features of the target object may be recorded simultaneously after the target object is selected, and used as a label when determining the first feature.

[0067] Step 104 : Record the images before the target object is lost in the surveillance video as historical images, and calculate the motion prediction area of the target object when it is lost based on the historical images.

[0068] In one possible implementation, calculating the motion prediction area of the target object when it is lost based on historical images includes: For each image in the historical images, a first area is determined based on the target object in the image and a preset range.

[0069] According to all the first regions, a motion prediction region of the target object when it is lost is obtained.

[0070] For example, based on historical images, the approximate movement trajectory of the target object during the earthquake can be obtained. Combined with the preset range, it can ensure that a rough search range can be determined when the target object is lost, giving the range with the highest probability of successful search for the target object for re-determining the target object, thereby improving computing efficiency.

[0071] Step 105: Based on the suspected target object and the motion prediction area, re-determine the target object in the target image.

[0072] In a possible implementation, step 105 may include: Determine the occurrence probability of each type of first area according to all the first areas.

[0073] Based on each type of first area and the corresponding occurrence probability, determine the target area.

[0074] Re-determine the suspected target object in the target area as the target object in the target image.

[0075] Exemplarily, the type of the first area refers to the first areas at different positions formed with the target object at different positions as the center and a preset size as the radius. If the overlap degree of two first areas is higher than 95%, it can be considered that the two first areas are of the same type.

[0076] Exemplarily, the specific process of determining the occurrence probability of each type of first area may include: First, determine the total number N of all the first areas. Assume that there are three types of first areas, A, B, and C. The number of first areas of type A is a, the number of first areas of type B is b, and the number of first areas of type C is c. Then the occurrence probability of the first areas of type A is a / N, the occurrence probability of the first areas of type B is b / N, and the occurrence probability of the first areas of type C is c / N.

[0077] Exemplarily, the process of determining the target area may include: For each type of first area, form a second area with the center of the first area as the center and the product of the preset size and the occurrence probability of this type of first area as the radius; Take the union of all the second areas to obtain the target area.

[0078] Step 106: Calculate the seismic intensity of the target area where the earthquake occurred according to the historical image and the target image after re-determining the target object.

[0079] In a possible implementation, step 106 may include: Based on the target object in the historical image and the target object in the target image after re-determining the target object, calculate the peak ground acceleration and the peak ground velocity.

[0080] Based on the peak ground acceleration and the peak ground velocity, calculate the seismic intensity of the target earthquake area.

[0081] In a possible implementation, the method further includes: If the tracking result indicates that the target object is not lost in the surveillance video, then based on the target object in the surveillance video, the peak ground acceleration and the peak ground velocity are calculated.

[0082] Based on the peak ground acceleration and the peak ground velocity, the seismic intensity of the target earthquake area is calculated.

[0083] Exemplarily, by analyzing the position of the target object between multiple consecutive images, the direction and displacement of the target object can be obtained, and in combination with the time interval between multiple consecutive images, the velocity and acceleration of the target object can be calculated.

[0084] After calculating the peak ground acceleration and the peak ground velocity, in combination with the first formula and the second formula, the seismic intensity of the target earthquake area can be calculated.

[0085] The first formula can be:

[0086] The second formula can be:

[0087] Wherein, represents the seismic intensity of the target earthquake area.

[0088] In the above video-based seismic intensity calculation method, by calibrating the target object in the video of the target area during an earthquake, a reference object is determined. Then, the target object is tracked through a target tracking algorithm to obtain a tracking result, and different seismic intensity calculation schemes are designed based on different tracking results. The principle is that when the seismic intensity is small, the target object will not be lost in the tracking result of the target tracking algorithm; when the seismic intensity is large, the target object may be lost in the tracking result of the target tracking algorithm. In addition, during an earthquake, due to the vibration of the ground itself, other objects similar to the target object may appear in the surveillance video. At this time, the result calculated by a simple target tracking algorithm will have errors. Similarly, due to the vibration of the earthquake being irregular (in terms of vibration direction), the method of predicting the position where the target object appears based on historical data in the traditional way has too large a deviation and cannot meet the requirements of the accuracy of seismic intensity calculation. Therefore, a method of combining suspected target objects and motion prediction areas to re-determine the target object is designed to ensure the accuracy of the re-determined target object, so as to ensure that the seismic intensity of the target earthquake area can be accurately calculated when the seismic intensity is large.

[0089] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0090] The following is an embodiment of the device of the present invention. For details not described in detail, reference may be made to the corresponding method embodiment above.

[0091] Figure 2 The following shows a schematic structural diagram of a video-based earthquake intensity calculation device provided by an embodiment of the present invention. For ease of description, only parts related to the embodiment of the present invention are shown and are described in detail as follows: As Figure 2 shown, the video-based earthquake intensity calculation device includes: A first processing module 201, configured to obtain a monitoring video of a target area where an earthquake occurs in a target time period before and after the earthquake time, and calibrate a target object in the monitoring video.

[0092] A second processing module 202, configured to perform target tracking on the target object in the monitoring video and output a tracking result.

[0093] A third processing module 203, configured to, if the tracking result indicates that the target object is lost in the monitoring video, record the image when the target object is lost in the monitoring video as a target image, and perform feature extraction on the entire image of the target image to determine a suspected target object.

[0094] A fourth processing module 204, configured to record the image before the target object is lost in the monitoring video as a historical image, and calculate a motion prediction area of the target object when it is lost based on the historical image.

[0095] A fifth processing module 205, configured to re-determine the target object in the target image based on the suspected target object and the motion prediction area.

[0096] A sixth processing module 206, configured to calculate the earthquake intensity of the target area where the earthquake occurs according to the historical image and the target image after re-determining the target object.

[0097] In a possible implementation manner, the third processing module 203 may be configured to: Perform feature extraction on the entire image of the target image to obtain the features of each object in the target image, denoted as first features.

[0098] Select the first feature most similar to the feature of the target object from each first feature, and determine the object corresponding to the most similar first feature as the suspected target object.

[0099] In a possible implementation manner, the fourth processing module 204 may be configured to: For each image in the historical image, determine a first area based on the target object on the image and a preset range.

[0100] Based on all the first regions, obtain the motion prediction region of the target object at the time of loss.

[0101] In a possible implementation, the fifth processing module 205 can be used to: Based on all the first regions, determine the occurrence probability of each type of first region.

[0102] Based on each type of first region and the corresponding occurrence probability, determine the target region.

[0103] Redetermine the suspected target object in the target region as the target object in the target image.

[0104] In a possible implementation, the sixth processing module 206 can be used to: Based on the target object in the historical image and the target object in the target image after redetermining the target object, calculate the peak ground acceleration and peak ground velocity.

[0105] Based on the peak ground acceleration and peak ground velocity, calculate the seismic intensity of the target seismic region.

[0106] In a possible implementation, the third processing module 203 can also be used to: If the tracking result indicates that the target object is not lost in the surveillance video, calculate the peak ground acceleration and peak ground velocity based on the target object in the surveillance video.

[0107] Based on the peak ground acceleration and peak ground velocity, calculate the seismic intensity of the target seismic region.

[0108] In a possible implementation, the second processing module 202 can be used to: Perform target tracking on the target object in the surveillance video through a kernel correlation filter or a channel and spatial reliability tracker to obtain a tracking result.

[0109] If the tracking result indicates that the target object is lost in the surveillance video, output the moment when the target object is lost.

[0110] In a possible implementation, the first processing module 201 can be used to: Obtain the surveillance video of the target region where an earthquake occurs during the target time period before and after the earthquake moment, and extract the first video frame image in the surveillance video, denoted as the starting video frame image.

[0111] Obtain the features of all objects in the starting video frame image.

[0112] For each object, calculate the feature similarity between this object and the remaining objects, and calculate the sum of all feature similarities between this object and the remaining objects to obtain the feature index corresponding to this object.

[0113] Calibrate the object corresponding to the minimum value of the feature index as the target object in the video.

[0114] Figure 3 It is a schematic diagram of the electronic device provided by the embodiment of the present invention. As Figure 3 shown, the electronic device 5 of this embodiment includes: a processor 50 and a memory 51. The memory 51 stores a computer program 52. When the processor 50 executes the computer program 52, the steps in the above-mentioned method embodiments are implemented. Or, when the processor 50 executes the computer program 52, the functions of each module / unit in the above-mentioned device embodiments are implemented.

[0115] Exemplarily, the computer program 52 can be divided into one or more modules / units. The one or more modules / units are stored in the memory 51 and executed by the processor 50 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of completing specific functions, and these instruction segments are used to describe the execution process of the computer program 52 in the electronic device 5.

[0116] The electronic device 5 may include, but is not limited to, a processor 50 and a memory 51. Those skilled in the art can understand that Figure 3 this is only an example of the electronic device 5 and does not constitute a limitation on the electronic device 5. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, the electronic device 5 may also include input / output devices, network access devices, buses, etc.

[0117] The processor 50 may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or this processor may also be any conventional processor, etc.

[0118] The memory 51 can be an internal storage unit of the electronic device 5, such as the hard disk or memory of the electronic device 5. The memory 51 can also be an external storage device of the electronic device 5, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the electronic device 5. Further, the memory 51 can also include both the internal storage unit of the electronic device 5 and the external storage device. The memory 51 is used to store the computer program 52 and other programs and data required by the electronic device 5. The memory 51 can also be used to temporarily store the data that has been output or will be output.

[0119] For the convenience and simplicity of description, only the above-mentioned division of each functional module / unit is used as an example for illustration. In actual applications, the above functions can be assigned to different functional modules / units according to needs. The above-mentioned modules / units can be implemented in the form of hardware, or in the form of software, or in the form of a combination of hardware and software.

[0120] The embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the methods in the above-mentioned method embodiments are implemented.

[0121] >The embodiment of the present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, the methods in the above-mentioned method embodiments are implemented.

[0122] Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0123] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments. If there is no special explanation and logical conflict, the terms and / or descriptions between different embodiments are consistent and can be mutually referred to, and the technical features in different embodiments can be combined to form new embodiments according to their internal logical relationships.

[0124] The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for calculating seismic intensity based on video, characterized in that, Including: Obtain surveillance videos of a target area where an earthquake occurs in a target time period before and after the earthquake moment, and calibrate target objects in the surveillance videos; Perform target tracking on the target objects in the surveillance videos and output tracking results; If the tracking results indicate that the target objects are lost in the surveillance videos, record the images when the target objects are lost in the surveillance videos as target images, and perform feature extraction on the entire target images to determine suspected target objects; Record the images before the target objects are lost in the surveillance videos as historical images, and calculate the motion prediction area of the target objects when they are lost based on the historical images; Based on the suspected target objects and the motion prediction area, re-determine the target objects in the target images; Calculate the earthquake intensity of the target area where the earthquake occurs according to the historical images and the target images after re-determining the target objects; 2. The method for calculating seismic intensity based on video according to claim 1, wherein The performing feature extraction on the entire target images to determine suspected target objects includes: Perform feature extraction on the entire target images to obtain the features of each object in the target images, denoted as first features; Select the first feature most similar to the features of the target objects from each of the first features, and determine the object corresponding to the most similar first feature as the suspected target object; 3. The method for calculating seismic intensity based on video according to claim 1, characterized in that The calculating the motion prediction area of the target objects when they are lost based on the historical images includes: For each image in the historical images, determine a first area based on the target objects on the image and a preset range; According to all the first areas, obtain the motion prediction area of the target objects when they are lost; 4. The method for calculating seismic intensity based on video according to claim 3, wherein The re-determining the target objects in the target images based on the suspected target objects and the motion prediction area includes: Determine the appearance probability of each type of first area according to all the first areas; Based on each type of first area and the corresponding appearance probability, determine the target area; Re-determine the suspected target objects in the target area as the target objects in the target images; 5. The method for calculating seismic intensity based on video according to claim 1, characterized in that, The calculating the earthquake intensity of the target area where the earthquake occurs according to the historical images and the target images after re-determining the target objects includes: Calculate the peak ground acceleration and peak ground velocity based on the target objects in the historical images and the target objects in the target images after re-determining the target objects; Calculate the earthquake intensity of the target earthquake area based on the peak ground acceleration and peak ground velocity; 6. The method for calculating seismic intensity based on video according to claim 1, characterized in that The method further includes: If the tracking results indicate that the target objects are not lost in the surveillance videos, calculate the peak ground acceleration and peak ground velocity based on the target objects in the surveillance videos; Calculate the earthquake intensity of the target earthquake area based on the peak ground acceleration and peak ground velocity; 7. The method for calculating seismic intensity based on video according to claim 1, characterized in that, The performing target tracking on the target objects in the surveillance videos and outputting tracking results includes: Perform target tracking on the target objects in the surveillance videos through a kernel correlation filter or a channel and spatial reliability tracker to obtain tracking results; If the tracking result indicates that the target object is lost in the surveillance video, output the moment when the target object is lost.

8. The method for calculating seismic intensity based on video according to claim 1, wherein Obtaining surveillance videos of a target area where an earthquake occurred in a target time period before and after the earthquake time, and calibrating the target object in the surveillance video, includes: Obtaining surveillance videos of a target area where an earthquake occurred in a target time period before and after the earthquake time, and extracting the first video frame image in the surveillance video, denoted as the starting video frame image; Obtaining the features of all objects in the starting video frame image; For each object, calculating the feature similarity between this object and the remaining objects, and calculating the sum of all feature similarities between this object and the remaining objects to obtain the feature index corresponding to this object; Calibrating the object corresponding to the minimum value of the feature index as the target object in the video.

9. A video-based seismic intensity calculation device, characterized in that, Includes: A first processing module, configured to obtain surveillance videos of a target area where an earthquake occurred in a target time period before and after the earthquake time, and calibrate the target object in the surveillance video; A second processing module, configured to perform target tracking on the target object in the surveillance video and output a tracking result; A third processing module, configured to, if the tracking result indicates that the target object is lost in the surveillance video, record the image when the target object is lost in the surveillance video as a target image, and perform feature extraction on the entire target image to determine a suspected target object; A fourth processing module, configured to record the image before the target object is lost in the surveillance video as a historical image, and calculate the motion prediction area of the target object when it is lost based on the historical image; A fifth processing module, configured to re-determine the target object in the target image based on the suspected target object and the motion prediction area; A sixth processing module, configured to calculate the earthquake intensity of the target area where the earthquake occurred based on the historical image and the target image after re-determining the target object.

10. An electronic device, characterized in that, Includes a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the method described in any one of claims 1 to 8 is implemented.

Images