Radar data processing method, program product, electronic equipment and storage medium

Abstract

The invention provides a radar data processing method, a program product, electronic equipment and a storage medium, and the method comprises the steps: comparing collected radar data with a base value, and obtaining a difference value; if the difference value is smaller than a threshold; if yes, reporting the radar data according to the first data reporting frequency; if the difference value is not smaller than the threshold value, reporting the radar data according to a second data reporting frequency; updating the base value according to the base value and the threshold value; the second data reporting frequency is higher than the first data reporting frequency. The appropriate reporting frequency is selected based on the comparison result of the difference value and the threshold value, automatic adjustment of the reporting frequency is achieved, unnecessary data transmission can be reduced for non-abnormal conditions, and network bandwidth and storage resources are saved. And for abnormal conditions, the second data reporting frequency with higher frequency is adopted for data reporting, so that the problem of data omission is improved, and the accuracy of reported data is improved. By dynamically adjusting the base value, the real state of the radar data is reflected more accurately, and false alarms caused by a fixed base value are reduced.

Description

Technical Field

[0001] The present application relates to the field of radar, and in particular to a radar data processing method, a program product, an electronic device and a storage medium. Background Art

[0002] Radar is a radio device that uses electromagnetic waves to send and receive signals to detect the position, speed and other characteristics of target objects. Radar is widely used in many fields such as aviation, navigation, and transportation, and plays an important role. Uploading and processing large amounts of data will consume more computing resources and time, which may cause the system to respond more slowly and affect application scenarios with high real-time requirements. The current processing method is to use a periodic transmission method for radar data, which may cause important events to be missed and the safety monitoring system cannot accurately assess the safety status. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a radar data processing method, a program product, an electronic device and a storage medium, which are used to improve the problem of being unable to accurately assess the safety status caused by periodically transmitting radar data.

[0004] In a first aspect, an embodiment of the present application provides a radar data processing method, including: comparing the collected radar data with a base value to obtain a difference; if the difference is less than a threshold value; reporting the radar data at a first data reporting frequency; if the difference is not less than the threshold value, reporting the radar data at a second data reporting frequency; and updating the base value based on the base value and the threshold value, so that the subsequently collected radar data obtains the difference based on the updated base value; the second data reporting frequency is higher than the first data reporting frequency.

[0005] In the above implementation process, the collected radar data is compared with the base value to obtain the difference. Based on the comparison result of the difference and the threshold, the appropriate reporting frequency is selected to achieve automatic adjustment of the reporting frequency. For non-abnormal situations, unnecessary data transmission can be reduced, saving network bandwidth and storage resources. For abnormal situations, a second data reporting frequency with a higher frequency is used for data reporting to reduce the situation of missing emergency reports and improve the safety and accuracy of detection. In addition, by dynamically adjusting the base value, the real state of the radar data can be more accurately reflected, reducing false alarms or omissions caused by fixed base values.

[0006] Optionally, in an embodiment of the present application, the base value is updated according to the base value and the threshold, including: after the radar data is reported at a second data reporting frequency for a preset period or a preset number, a preset weight parameter is obtained, and the base value and the threshold are weighted and summed based on the weight parameter to obtain an updated base value.

[0007] In the implementation process of the above embodiment: after the radar data is reported according to the second data reporting frequency for a preset period or a preset number, the base value is dynamically updated according to the base value and the threshold value, so as to ensure the accuracy of the difference value, reflect the latest state of the monitored object, and reduce false alarms caused by the fixed base value. In addition, in the process of updating the base value, the contribution of the base value and / or the threshold value in the process of updating the base value is considered to improve the accuracy of the base value determination.

[0008] Optionally, in an embodiment of the present application, after updating the base value, the method also includes: obtaining a proportional relationship between the threshold value and the base value before the update; updating the threshold value based on the proportional relationship and the base value after the update; or; obtaining historical fluctuation data of historical radar data, and current fluctuation data of radar data; obtaining the fluctuation ratio based on the historical fluctuation data and the current fluctuation data; updating the threshold value according to the fluctuation ratio; wherein the historical fluctuation data is used to describe the degree of data fluctuation of the historical radar data; and the current fluctuation data is used to describe the degree of data fluctuation of the current fluctuation data.

[0009] In the implementation process of the above embodiment: by dynamically adjusting the threshold, the adjusted threshold can adapt to changes in the volatility of radar data, more accurately reflect the true state of the radar data, and reduce false alarms or missed alarms caused by fixed thresholds.

[0010] Optionally, in an embodiment of the present application, radar data is collected by a radar device; after comparing the collected radar data with a base value to obtain a difference, the method further includes: obtaining position information of points on a scanning surface of the radar device during the process of collecting radar data; if the difference is less than a threshold value; then photographing the point at a first shooting frequency based on the position information of the point by a camera; when the difference is not less than a threshold value, photographing the point at a second shooting frequency based on the position information of the point by a camera; wherein the ratio of the first data reporting frequency to the second data reporting frequency is equal to the ratio of the first shooting frequency to the second shooting frequency.

[0011] In the implementation process of the above embodiment: in the process of reporting radar data, the synchronously taken photos or videos provide intuitive evidence for the monitoring results, so that the combination of radar data and visual images can provide more comprehensive monitoring scene information and ensure the integrity of the data. And based on the comparison result of the difference and the threshold, the appropriate shooting frequency is selected to achieve automatic adjustment of the shooting frequency, thereby improving the safety and accuracy of detection.

[0012] Optionally, in an embodiment of the present application, obtaining point information scanned by a radar device during the process of collecting radar data includes: obtaining radar position parameters of the radar device in a digital elevation model; the position parameters include at least one of longitude, latitude, elevation, and zero point orientation of the radar antenna; mapping the radar data to the digital elevation model to obtain point position parameters of the point on the scanning surface relative to the digital elevation model; and determining the point position information of the point relative to the radar device based on the radar position parameters and the point position parameters.

[0013] In the implementation process of the above embodiment: this process combines the remote monitoring capability of the radar and the image capture capability of the camera, and through precise calculation and automated control, it realizes effective monitoring of ground deformation and timely capture of abnormal areas. And the point position information relative to the radar device is determined by the digital elevation model, thereby improving the accuracy of the point position information.

[0014] Optionally, in an embodiment of the present application, radar position parameters and point position parameters are converted to obtain point position information of the point relative to the radar device, including: if the zero position orientations of the camera and the radar device are inconsistent, respectively obtaining the camera coordinates, radar coordinates, camera zero position orientation and radar zero position orientation; wherein the camera coordinates and radar coordinates are coordinate information in the same coordinate system; according to the camera coordinates, radar coordinates, camera zero position orientation and radar zero position orientation, a calibration algorithm is used to obtain transformation parameters; the transformation parameters are used to adjust the camera zero position orientation and the radar zero position orientation to be consistent; the camera zero position orientation and the radar zero position orientation are adjusted to be consistent using the transformation parameters; when the camera zero position orientation and the radar zero position orientation are consistent, the radar position parameters and the point position parameters are converted to obtain the point position information of the point relative to the radar device.

[0015] In the implementation process of the above embodiment: the camera zero position orientation and the radar zero position orientation are adjusted to be consistent through the calibration algorithm, so as to achieve consistency in spatial perception between the radar device and the camera, and provide a data basis for subsequent point shooting, point information calculation and other tasks. A variety of camera setting methods are provided, supporting the situation where the camera is external, and adopting split work to improve shooting flexibility.

[0016] Optionally, in an embodiment of the present application, the method also includes: obtaining the photograph to be detected and the scene of the photograph to be detected; inputting the photograph to be detected into an anomaly recognition model corresponding to the scene to obtain an anomaly recognition result; and reporting the photograph to be detected and the corresponding anomaly recognition result.

[0017] In the implementation process of the above embodiment: through the automated image recognition process, manual intervention can be greatly reduced and processing efficiency can be improved. Selecting a suitable anomaly recognition model according to the scene of the photo to be detected can ensure the professionalism and pertinence of the anomaly recognition model, thereby improving the accuracy of the recognition result.

[0018] In the second aspect, an embodiment of the present application also provides a radar data processing device, including: a difference calculation module, used to compare the collected radar data with a base value to obtain a difference; a data reporting module, used to report the radar data according to a first data reporting frequency if the difference is less than a threshold; if the difference is not less than the threshold, report the radar data according to a second data reporting frequency; and update the base value according to the base value and the threshold, so that the subsequently collected radar data obtains the difference based on the updated base value; the second data reporting frequency is higher than the first data reporting frequency.

[0019] In a third aspect, an embodiment of the present application further provides a computer program product, including computer program instructions, which, when executed by a processor, execute the method provided by the first aspect or any one of the implementations of the first aspect.

[0020] In a fourth aspect, an embodiment of the present application further provides an electronic device, comprising: a processor and a memory, the memory storing computer program instructions, and the computer program instructions, when executed by the processor, execute the method provided by the first aspect or any one of the implementations of the first aspect.

[0021] In a fifth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method provided by the first aspect or any one of the implementations of the first aspect is executed.

[0022] The radar data processing method, device, electronic device and storage medium provided by the present application are used to compare the collected radar data with the base value to obtain the difference, and select the appropriate reporting frequency based on the comparison result of the difference and the threshold value to achieve automatic adjustment of the reporting frequency. For non-abnormal situations, unnecessary data transmission can be reduced, saving network bandwidth and storage resources. For abnormal situations, a second data reporting frequency with a higher frequency is used for data reporting to reduce the situation of missing emergency reports and improve the safety and accuracy of detection. In addition, by dynamically adjusting the base value, the true state of the radar data can be more accurately reflected, reducing false alarms or omissions caused by fixed base values. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 A schematic diagram of a flow chart of a radar data processing method provided in an embodiment of the present application;

[0025] Figure 2 A schematic diagram of the structure of a radar data processing device provided in an embodiment of the present application;

[0026] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] The following embodiments of the technical solution of the present application are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application, and are therefore only used as examples, and cannot be used to limit the scope of protection of the present application.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0029] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0030] There is a close relationship between the frequency of radar data upload and safety monitoring. Safety monitoring often has high requirements for the real-time performance of data. As an important monitoring tool, radar data can reflect the status and changes of the target in real time. Therefore, the frequency of radar data upload directly affects the real-time performance of safety monitoring. A higher upload frequency can ensure that the safety monitoring system can obtain the latest radar data faster, so as to make responses and decisions more timely.

[0031] Existing technologies use a fixed frequency and period to report radar data. However, if the upload frequency of radar data is too low, important events that occur between two uploads may be missed, thus affecting the integrity of the data. This may prevent the safety monitoring system from fully and accurately assessing the current safety situation, which in turn affects the monitoring effect and the accuracy of decision-making. The selection of radar data upload frequency also needs to consider the stability of the system. Too high an upload frequency may increase the burden on the system, causing the system to fail or crash due to being unable to handle the data.

[0032] The embodiment of the present application provides a radar data processing method, which compares the collected radar data with the base value to obtain the difference, selects the appropriate reporting frequency based on the comparison result of the difference and the threshold, and realizes automatic adjustment of the reporting frequency. For non-abnormal situations, unnecessary data transmission can be reduced, saving network bandwidth and storage resources. For abnormal situations, a second data reporting frequency with a higher frequency is used for data reporting, which improves the problem of missing emergency events and improves the safety and accuracy of detection. In addition, by dynamically adjusting the base value, the real state of the radar data can be more accurately reflected, reducing false alarms or omissions caused by fixed base values.

[0033] See also Figure 1 A flow chart of a radar data processing method provided by an embodiment of the present application is shown. The radar data processing method provided by an embodiment of the present application can be applied to an electronic device, which may include a physical device such as a server, a PC, a tablet computer, or a smart phone, or a virtual device such as a virtual machine or a container. The electronic device may be a single device, or a combination of multiple devices or a cluster of a large number of devices. The radar data processing method may include:

[0034] Step S110: Compare the collected radar data with the base value to obtain a difference.

[0035] Step S121: If the difference is less than the threshold, the radar data is reported according to the first data reporting frequency.

[0036] Step S122: If the difference is not less than the threshold, the radar data is reported according to the second data reporting frequency; and the base value is updated according to the base value and the threshold, so that the subsequently collected radar data obtains the difference based on the updated base value; the second data reporting frequency is higher than the first data reporting frequency.

[0037] In step S110, radar data may include basic detection data, radar image data, point cloud data, etc. Among them, basic detection data includes distance data, angle data and speed data, etc.; radar image data such as deformation cloud map or scattering cloud map, etc.; point cloud data consists of a large number of points, each point contains its three-dimensional coordinates (X, Y, Z) and possible other attributes (such as intensity, reflectivity, etc.). The above radar data can be collected by radar equipment or GNSS (Global Navigation Satellite System).

[0038] The base value can be regarded as a reference point for radar data, which is used to compare radar data. For the initial base value, the base value can be the initial data collected by the monitoring equipment within a certain period of time, and these data are used as a comparison benchmark for subsequent changes. For example, in deformation monitoring, the base value may be the elevation data of the terrain or structure collected for the first time after the equipment is installed. It is understandable that each type of radar data can have its corresponding base value. For example, if the radar data is angle data, then the base value should be an angle base value. For another example, if the radar data is deformation data, then the base value can be an elevation base value.

[0039] Compare the collected radar data with the base value to obtain a difference. For example, the collected radar data can be subtracted from its corresponding base value to obtain a difference. The difference is used to represent the change in the radar data corresponding to the monitored object since the base value was determined. If the radar data is deformation data, the change represents the deformation of the terrain or structure. The cause of the deformation may be surface uplift or surface subsidence.

[0040] In step S121, the threshold is used to represent the fluctuation of radar data. When the difference between the monitored data and the base value exceeds the threshold, it can be considered that a change or abnormality that requires attention has occurred at the point. The threshold can be set based on historical data or specific safety standards, and the threshold can be used to distinguish between normal fluctuations and abnormal fluctuations.

[0041] The difference is compared with a threshold value. If the difference is smaller than the threshold value, it indicates that the change of the radar data is within a normal fluctuation range, and the radar data is reported at a first data reporting frequency.

[0042] In step S122, if the difference is not less than the threshold (that is, the difference is greater than or equal to the threshold), it indicates that the change in the radar data exceeds the normal fluctuation range, and an abnormality may occur, such as a large deformation, etc., then the radar data is reported at a second data reporting frequency; the second data reporting frequency is higher than the first data reporting frequency, that is, the frequency of data reporting is accelerated when the difference is not less than the threshold, so as to quickly respond to actual changes in the radar data and promptly notify relevant personnel or systems to take action.

[0043] Considering that the position information provided by the radar data acquisition device records the relative position change relative to a certain starting point, if the acquisition device is used to monitor deformation or other types of displacement, and these displacements occur continuously, then the collected values ​​may gradually increase over time. In this case, if a fixed base value is used, the subsequent difference calculation may be inaccurate, and the difference may always be greater than (or equal to) the threshold, causing false alarms and other situations.

[0044] Based on this discovery, in the embodiment of the present application, after the radar data is reported according to the second data reporting frequency, the base value is updated according to the base value and the threshold value. In this way, for the radar data that is not reported subsequently, the difference can be calculated according to the new base value, and the reporting frequency is determined according to the difference.

[0045] As an implementation manner, after reporting radar data of a preset period at a second data reporting frequency, the base value may be updated according to a threshold value and a base value, and the updated base value may be compared with new radar data to obtain a difference.

[0046] In the above implementation process, the collected radar data is compared with the base value to obtain the difference. Based on the comparison result of the difference and the threshold, the appropriate reporting frequency is selected to achieve automatic adjustment of the reporting frequency. For non-abnormal situations, unnecessary data transmission can be reduced, saving network bandwidth and storage resources. For abnormal situations, a second data reporting frequency with a higher frequency is used for data reporting to improve the problem of missing emergency events and improve the safety and accuracy of detection. In addition, by dynamically adjusting the base value, the real state of the radar data can be more accurately reflected, reducing false alarms or omissions caused by fixed base values.

[0047] Optionally, in an embodiment of the present application, the base value is updated according to the base value and the threshold, including: after the radar data is reported at a second data reporting frequency for a preset period or a preset number, a preset weight parameter is obtained, and the base value and the threshold are weighted and summed based on the weight parameter to obtain an updated base value.

[0048] The preset period or preset number can be pre-configured, for example, the reporting period is preset to 2 hours or the preset number is set to 20 groups, and the radar data is reported according to the second data reporting frequency. After the reporting time reaches 2 hours or the reporting number reaches 20 groups, the base value is updated.

[0049] The method of updating the base value is: obtaining a preset weight parameter, which includes a base value weight and / or a threshold weight. The weight parameter is used to determine the contribution of the base value and / or the threshold in the process of updating the base value. For example, if the base value weight and the threshold weight are both 1, then the value after the base value and the threshold are summed is the updated base value.

[0050] In the implementation process of the above embodiment: after the radar data is reported according to the second data reporting frequency for a preset period or a preset number, the base value is dynamically updated according to the base value and the threshold value, so as to ensure the accuracy of the difference value, reflect the latest state of the monitored object, and reduce false alarms caused by the fixed base value. In addition, in the process of updating the base value, the contribution of the base value and / or the threshold value in the process of updating the base value is considered to improve the accuracy of the base value determination.

[0051] Optionally, in the embodiment of the present application, after the base value is updated, the threshold value may be dynamically adjusted, and the adjustment method may be any of the following two methods:

[0052] The first method is to obtain the proportional relationship between the threshold and the base value before the update; based on the proportional relationship and the base value after the update, the threshold is updated. For example, if the proportional relationship between the initial threshold and the initial base value is 1 / 5, then after the base value is updated, in order to ensure that the proportional relationship remains unchanged, the updated threshold is determined based on the updated base value and the proportional relationship.

[0053] The second method is to obtain the historical fluctuation data of the historical radar data and the current fluctuation data of the radar data. For example, the historical radar data of the historical time period (for example, the past 30 days) is selected, and the standard deviation of the historical radar data during this period is calculated to obtain the historical fluctuation data. Among them, the historical fluctuation data is used to describe the degree of data fluctuation of the historical radar data. The formula for calculating the historical fluctuation data can be:

[0054]

[0055] Among them, σ1 represents the historical fluctuation data, N represents the number of historical radar data, i represents the label of historical radar data, and X i represents the i-th historical radar data, and μ1 represents the average of the historical radar data.

[0056] The current fluctuation data is used to describe the degree of fluctuation of the current fluctuation data. The formula for calculating the current fluctuation data of the radar data can be:

[0057]

[0058] Among them, σ2 represents the current fluctuation data, M represents the number of current radar data, j represents the label of the current radar data, and X j represents the jth current radar data, and μ2 represents the average of the current radar data.

[0059] Based on the historical volatility data and the current volatility data, the volatility ratio is obtained. The volatility ratio is the ratio of the current volatility data to the historical volatility data. This ratio is used to measure the degree of change of the historical volatility data relative to the historical volatility data.

[0060] The threshold is updated according to the fluctuation ratio. For example, the updated threshold = threshold * fluctuation ratio. Of course, an adjustment coefficient can also be introduced to correct the updated threshold.

[0061] In the implementation process of the above embodiment: by dynamically adjusting the threshold, the adjusted threshold can adapt to changes in the volatility of radar data, more accurately reflect the true state of the radar data, and reduce false alarms or missed alarms caused by fixed thresholds.

[0062] Optionally, in the embodiment of the present application, radar data is acquired by collecting radar equipment, and the radar can be installed on a pan-tilt platform, which is a mechanical platform on which a camera or other equipment can be installed and the radar or camera can be rotated and tilted in the horizontal and vertical directions. The pointing angle of these instruments is adjusted by the pan-tilt platform to observe or collect data in a specific area.

[0063] After comparing the collected radar data with the base value to obtain the difference, the method further includes:

[0064] The position information of points on the scanning surface of the radar device during the process of collecting radar data is obtained. The position information refers to the point information relative to the radar device. This process can be achieved using a digital elevation model, which will be described in detail later.

[0065] If the difference is less than the threshold, the camera takes a picture of the point according to the first shooting frequency based on the location information of the point. As an implementation method, the shooting work can be performed in the interval of radar data reporting; the shooting work and the uploading work can also be performed at the same time. The timestamp of the captured image or video is matched with the timestamp of the radar data, and the radar data and images of the same time are packaged together and uploaded.

[0066] When the difference is not less than the threshold, the point is photographed at the second shooting frequency based on the location information of the point through the camera; wherein the ratio of the first data reporting frequency to the second data reporting frequency is equal to the ratio of the first shooting frequency to the second shooting frequency. The second data reporting frequency is higher than the first data reporting frequency, therefore, the second shooting frequency is higher than the first shooting frequency.

[0067] As an implementation method, the first shooting frequency can be equal to the first data reporting frequency, and the second shooting frequency can be equal to the second data reporting frequency. Then, regardless of whether the difference is greater than a threshold, the frequencies of data reporting and shooting are equal, and synchronous reporting of radar data and collected images can be achieved.

[0068] In the implementation process of the above embodiment: in the process of reporting radar data, the synchronously taken photos or videos provide intuitive evidence for the monitoring results, so that the combination of radar data and visual images can provide more comprehensive monitoring scene information and ensure the integrity of the data. And based on the comparison result of the difference and the threshold, the appropriate shooting frequency is selected to achieve automatic adjustment of the shooting frequency, thereby improving the safety and accuracy of detection.

[0069] Optionally, in an embodiment of the present application, obtaining point information scanned by the radar device during the process of collecting radar data includes: obtaining radar position parameters of the radar device in the digital elevation model; the position parameters include longitude, latitude, elevation, and at least one of the zero point direction of the radar antenna.

[0070] Map the radar data to the digital elevation model to obtain the position parameters of the points on the scanned surface relative to the digital elevation model. For example, the radar scans the monitoring area, obtains the surface deformation data, and projects these data onto the DEM model. Each point on the scanned surface (i.e., each point in the DEM model, also known as a pixel) will obtain its longitude, latitude, elevation, and deformation value information. This step is used to associate radar data (such as deformation data) with geographic location information for subsequent analysis and processing.

[0071] The point information of the point relative to the radar device is determined based on the radar position parameters and the point position parameters. The point information includes rotation angle, pitch angle and / or distance, etc. For example, for the pitch angle, the pitch angle refers to the angle between the radar beam and the horizontal plane, which can be calculated by the difference between the pixel elevation and the radar elevation and the horizontal distance. For the straight-line distance between the radar and the pixel, it can be calculated by the distance formula between two points in three-dimensional space. After obtaining the point information of the point relative to the radar device, provide accurate angle and focal length adjustment parameters for camera capture.

[0072] As an implementation method, the camera can be installed on the same pan-tilt with the radar device, so the camera and the radar antenna can be horizontally and vertically coaxial, sharing a two-dimensional pan-tilt, that is, the zero position orientation of the camera and the radar device is consistent. It can be understood that since the camera and the radar device are installed on the same pan-tilt, the point position information relative to the radar device is also the point position information relative to the camera. Therefore, after obtaining the point position information, the camera can directly rotate and focus according to the calculated point position information (such as pitch angle, rotation angle and distance information), and perform a snapshot action. For example, the camera rotates horizontally or vertically according to the rotation angle in the above point position information, and can also adjust the focal length of the camera according to the distance information in the point position information to ensure that the captured image is clear.

[0073] In the implementation process of the above embodiment: this process combines the remote monitoring capability of the radar and the image capture capability of the camera, and through precise calculation and automated control, it realizes effective monitoring of ground deformation and timely capture of abnormal areas. And the point position information relative to the radar device is determined by the digital elevation model, thereby improving the accuracy of the point position information.

[0074] Optionally, in another embodiment of the present application, when the radar device is installed indoors, if the camera and the radar device are installed on the same pan / tilt, the operation of the camera indoors is affected by the walls and windows, and effective monitoring cannot be achieved. Therefore, the radar and the camera can be set to be split. That is, the radar uses a separate two-dimensional pan / tilt, which can be adjusted in the horizontal and vertical directions. The camera uses an outdoor rainproof ball camera, which can achieve separate horizontal and vertical adjustments.

[0075] Since the camera and radar device are not installed on the same PTZ, the zero position orientation of the camera and radar device may be inconsistent. At this time, the camera zero position orientation and the radar zero position orientation should be adjusted to be consistent. On this basis, the radar position parameters and point position parameters are converted to obtain the point position information relative to the radar device. The following describes the process of adjusting the camera zero position orientation and the radar zero position orientation to be consistent.

[0076] The radar position parameters and the point position parameters are converted to obtain the point position information relative to the radar device, including: if the zero point position orientation of the camera and the radar device are inconsistent, the camera coordinates, radar coordinates, camera zero point position orientation and radar zero point position orientation are obtained respectively; wherein the camera coordinates and radar coordinates are coordinate information in the same coordinate system.

[0077] Zero position orientation refers to the direction directly in front of the device, a specific angle (such as 30 degrees north by east), or the direction relative to a known reference object. For example, the zero position orientation of a camera can refer to the direction directly in front of the camera, or it can refer to the direction angle of the camera relative to due north. It should be noted that the standards for the zero position orientation of the camera and the zero position orientation of the radar should be consistent, for example, the direction directly in front of each is used as the zero position orientation.

[0078] You can use RTK or a total station to accurately measure the installation center points of the camera and radar equipment, record the geographic coordinates (longitude, latitude, elevation) of the camera as the camera coordinates, and record the geographic coordinates (longitude, latitude, elevation) of the radar equipment as the radar coordinates.

[0079] According to the camera coordinates, radar coordinates, camera zero position orientation and radar zero position orientation, a calibration algorithm is used to obtain transformation parameters; the transformation parameters are used to adjust the camera zero position orientation and the radar zero position orientation to be consistent. The calibration algorithm is used to align the perception space of the camera and radar equipment so that the camera zero position orientation and the radar zero position orientation are adjusted to be consistent. The calibration algorithm can perform steps such as spatial geometric transformation, coordinate conversion and error correction. The transformation parameters include parameters such as rotation matrix and / or translation vector.

[0080] Use transformation parameters to adjust the camera zero position orientation and the radar zero position orientation to be consistent. The adjustment here can be physical adjustment or software adjustment. Physical adjustment refers to direct physical adjustment of the camera or radar device in terms of posture. Software adjustment refers to achieving logical alignment through transformation parameters. For example, before the camera shoots, the transformation parameters can be used for data preprocessing to simulate the state of adjusting the camera zero position orientation and the radar zero position orientation to be consistent.

[0081] When the camera zero point position orientation is consistent with the radar zero point position orientation, the radar position parameters and the point position parameters are converted to obtain the point position information relative to the radar device.

[0082] In the implementation process of the above embodiment: the camera zero position orientation and the radar zero position orientation are adjusted to be consistent through the calibration algorithm, so as to achieve consistency in spatial perception between the radar device and the camera, and provide a data basis for subsequent point shooting, point information calculation and other tasks. A variety of camera setting methods are provided, supporting the situation where the camera is external, and adopting split work to improve shooting flexibility.

[0083] Optionally, in an embodiment of the present application, the method also includes: acquiring photographs to be detected and scenes of the photographs to be detected; scenes of the photographs to be detected include, for example, foreign body intrusion, on-site status recognition, etc.

[0084] Input the photo to be detected into the anomaly recognition model corresponding to the scene to obtain the anomaly recognition result; the anomaly recognition result includes whether an anomaly occurs, the location of the abnormal event in the picture, or the classification of the abnormal scene. After obtaining the anomaly recognition result, the photo to be detected and the corresponding anomaly recognition result can be reported.

[0085] The anomaly recognition model can be a model formed by training data in different scenarios, or a model formed by training in a single audit scenario.

[0086] If the anomaly recognition model is a model trained in a single scene, then each scene has a corresponding model. When identifying the photo to be detected, the anomaly recognition model corresponding to the scene should be input to obtain the anomaly recognition result. The advantage of this is that the structure of the anomaly recognition model for each scene is relatively simple, the amount of data required for training the model and the time spent are relatively small, and the real-time performance and accuracy of the model are high.

[0087] If the anomaly recognition model is a model formed by training data in different scenarios, then all the photos to be detected in each scenario are detected by the anomaly recognition model. The advantage of this is that the trained anomaly recognition model can learn the image features of different scenarios. When reviewing the photos to be detected, there is no need to distinguish the scenes of the photos to be detected, which makes the operation more convenient.

[0088] Taking the anomaly recognition model as a model formed by training in a single scene as an example, sample images in the target scene can be obtained or collected, and the sample images include positive samples (images without abnormalities) and negative samples (abnormal images). The samples are marked respectively, and then the samples and their corresponding labels are input into a pre-set neural network model, and the parameters of the neural network model are adjusted to obtain the anomaly recognition model in the target scene. The above training process is performed for each scene, thereby obtaining the anomaly recognition model corresponding to each scene. The neural network model can be a YOLO model or a CNN model.

[0089] As an implementation method, two deployment forms can be adopted: front-end AI algorithm and back-end cloud platform identification.

[0090] For the front-end AI automatic processing algorithm, the algorithm module is deployed inside the host. After the equipment is installed, a large amount of image data is collected using the camera module. First, basic scene recognition and accumulation of original observation data are performed. At the same time, various objects identified in the monitoring image are distinguished (such as corners, poles, vegetation contours in the scene, terrain, etc.) to form an abnormal recognition model. After that, the image to be detected can be denoised, for example, factors affecting light, temperature, shadows, etc. are removed, and then the processed image is input into the model for automatic discrimination, such as pole changes, corner movement, animal touch, etc. The abnormal recognition results are then sent to the detection system.

[0091] The back-end platform identification solution is: deploy the AI ​​automatic processing algorithm service on the cloud platform. After the equipment is installed, use the camera module to collect a large amount of image sample data, and transmit the image sample data to the cloud platform through the communication module, and perform model training on the cloud platform. Then send the photos to be detected to the cloud platform for identification, obtain the abnormal identification results, and receive the abnormal identification results returned by the cloud platform.

[0092] In the implementation process of the above embodiment: through the automated image recognition process, manual intervention can be greatly reduced and processing efficiency can be improved. Selecting a suitable anomaly recognition model according to the scene of the photo to be detected can ensure the professionalism and pertinence of the anomaly recognition model, thereby improving the accuracy of the recognition result.

[0093] See also Figure 2 The structure diagram of the radar data processing device provided in the embodiment of the present application is shown; the embodiment of the present application provides a radar data processing device 200, including:

[0094] A difference calculation module 210 is used to compare the collected radar data with the base value to obtain a difference;

[0095] The data reporting module 220 is used to report the radar data at a first data reporting frequency if the difference is less than a threshold value; report the radar data at a second data reporting frequency if the difference is not less than the threshold value; and update the base value according to the base value and the threshold value so that the subsequently collected radar data obtains the difference based on the updated base value; the second data reporting frequency is higher than the first data reporting frequency.

[0096] Optionally, in an embodiment of the present application, the radar data processing device 200, the data reporting module 220 also includes a base value updating unit, which is used to obtain a preset weight parameter after reporting the radar data at a second data reporting frequency for a preset period or a preset number, and weightedly sum the base value and the threshold based on the weight parameter to obtain the updated base value.

[0097] Optionally, in an embodiment of the present application, the radar data processing device 200, the data reporting module 220 also includes a threshold updating unit, which is used to obtain the proportional relationship between the threshold and the base value before the update; based on the proportional relationship and the base value after the update, update the threshold; or; obtain historical fluctuation data of historical radar data, and current fluctuation data of the radar data; obtain the fluctuation ratio based on the historical fluctuation data and the current fluctuation data; update the threshold according to the fluctuation ratio; wherein the historical fluctuation data is used to describe the data fluctuation degree of the historical radar data; and the current fluctuation data is used to describe the data fluctuation degree of the current fluctuation data.

[0098] Optionally, in an embodiment of the present application, the radar data processing device 200, the radar data is obtained by collecting the radar data through the radar equipment; it also includes a radar-vision fusion module, which is used to obtain the position information of the points on the scanning surface of the radar equipment during the process of collecting the radar data; if the difference is less than a threshold value; the point is photographed at a first shooting frequency based on the position information of the point by the camera; when the difference is not less than the threshold value, the point is photographed at a second shooting frequency based on the position information of the point by the camera; wherein the ratio of the first data reporting frequency to the second data reporting frequency is equal to the ratio of the first shooting frequency to the second shooting frequency.

[0099] Optionally, in an embodiment of the present application, the radar data processing device 200, the radar-vision fusion module, is specifically used to obtain radar position parameters of the radar device in the digital elevation model; the position parameters include at least one of longitude, latitude, elevation and the zero point direction of the radar antenna; the radar data is mapped to the digital elevation model to obtain the point position parameters of the point on the scanning surface relative to the digital elevation model; and the point position information of the point relative to the radar device is determined based on the radar position parameters and the point position parameters.

[0100] Optionally, in an embodiment of the present application, the radar data processing device 200, the radar-vision fusion module, is also used to convert the radar position parameters and the point position parameters to obtain the point position information of the point relative to the radar device, including: if the zero position orientations of the camera and the radar device are inconsistent, then respectively obtain the camera coordinates, radar coordinates, camera zero position orientation and radar zero position orientation; wherein the camera coordinates and radar coordinates are coordinate information in the same coordinate system; according to the camera coordinates, the radar coordinates, the camera zero position orientation and the radar zero position orientation, a calibration algorithm is used to obtain transformation parameters; the transformation parameters are used to adjust the camera zero position orientation and the radar zero position orientation to be consistent; using the transformation parameters, the camera zero position orientation and the radar zero position orientation are adjusted to be consistent; when the camera zero position orientation and the radar zero position orientation are consistent, the radar position parameters and the point position parameters are converted to obtain the point position information of the point relative to the radar device.

[0101] Optionally, in an embodiment of the present application, the radar data processing device 200 further includes an image detection module, which is used to obtain the photographed photo to be detected and the scene of the photo to be detected; input the photo to be detected into an abnormality recognition model corresponding to the scene to obtain an abnormality recognition result; and report the photo to be detected and the corresponding abnormality recognition result.

[0102] It should be understood that the device corresponds to the above-mentioned radar data processing method embodiment and can execute the various steps involved in the above-mentioned method embodiment. The specific functions of the device can be referred to in the above description. To avoid repetition, the detailed description is appropriately omitted here. The device includes at least one software function module that can be stored in a memory in the form of software or firmware or solidified in the operating system (OS) of the device.

[0103] See also Figure 3 The electronic device 300 provided in the embodiment of the present application includes: a processor 310 and a memory 320, wherein the memory 320 stores machine-readable instructions executable by the processor 310, and when the machine-readable instructions are executed by the processor 310, the above method is executed.

[0104] Figure 3 Each component shown in can be implemented by hardware, software or a combination thereof. The electronic device 300 may be a physical device, such as a server, a PC, etc., or a virtual device, such as a virtual machine, a virtualized container, etc. Moreover, the electronic device 300 is not limited to a single device, but may also be a combination of multiple devices or a cluster consisting of a large number of devices.

[0105] An embodiment of the present application further provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above method is executed.

[0106] Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0107] In several embodiments provided by the embodiments of the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to the multiple embodiments of the embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and a part of the module, program segment or code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0108] In addition, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part.

[0109] The above description is only an optional implementation manner of the embodiments of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the embodiments of the present application, which should be covered within the protection scope of the embodiments of the present application.

Claims

1. A radar data processing method, characterized in that: include: Compare the collected radar data with the base value to obtain the difference; If the difference is less than a threshold value, the radar data is reported according to a first data reporting frequency; If the difference is not less than the threshold, reporting the radar data according to a second data reporting frequency; The base value is updated according to the base value and the threshold value, so that the radar data collected subsequently obtains the difference based on the updated base value; and the second data reporting frequency is higher than the first data reporting frequency.

2. The method according to claim 1, characterized in that The updating of the base value according to the base value and the threshold value comprises: After the radar data is reported at the second data reporting frequency for a preset period or a preset number, a preset weight parameter is obtained, and the base value and the threshold are weighted and summed based on the weight parameter to obtain the updated base value.

3. The method according to claim 1, characterized in that After updating the base value, the method further includes: Acquire a proportional relationship between the threshold value and the base value before updating; update the threshold value based on the proportional relationship and the base value after updating; or; Obtain historical fluctuation data of historical radar data and current fluctuation data of the radar data; obtain a fluctuation ratio based on the historical fluctuation data and the current fluctuation data; update the threshold value according to the fluctuation ratio; wherein the historical fluctuation data is used to describe the data fluctuation degree of the historical radar data; and the current fluctuation data is used to describe the data fluctuation degree of the current fluctuation data.

4. The method according to claim 1, characterized in that: The radar data is collected by radar equipment; after comparing the collected radar data with the base value to obtain the difference, the method further includes: Obtaining position information of points on a scanning surface of the radar device during the process of collecting the radar data; If the difference is less than the threshold, the point is photographed at a first shooting frequency based on the position information of the point by a camera; When the difference is not less than the threshold, the point is photographed at a second shooting frequency based on the position information of the point through a camera; wherein the ratio of the first data reporting frequency to the second data reporting frequency is equal to the ratio of the first shooting frequency to the second shooting frequency.

5. The method according to claim 4, characterized in that The obtaining of point information scanned by the radar device during the process of collecting the radar data includes: Acquire radar position parameters of the radar device in the digital elevation model; the position parameters include at least one of longitude, latitude, elevation, and zero point orientation of the radar antenna; Mapping the radar data to the digital elevation model to obtain point position parameters of points on the scanning surface relative to the digital elevation model; The point position information of the point relative to the radar device is determined according to the radar position parameter and the point position parameter.

6. The method according to claim 5, characterized in that The converting the radar position parameter and the point position parameter to obtain the point position information of the point relative to the radar device includes: If the zero point position orientations of the camera and the radar device are inconsistent, respectively obtaining the camera coordinates, the radar coordinates, the camera zero point position orientation, and the radar zero point position orientation; wherein the camera coordinates and the radar coordinates are coordinate information in the same coordinate system; According to the camera coordinates, the radar coordinates, the camera zero point position orientation and the radar zero point position orientation, a calibration algorithm is used to obtain transformation parameters; the transformation parameters are used to adjust the camera zero point position orientation and the radar zero point position orientation to be consistent; Using the transformation parameters, the camera zero point position orientation and the radar zero point position orientation are adjusted to be consistent; When the camera zero point position orientation is consistent with the radar zero point position orientation, the radar position parameter and the point position parameter are converted to obtain the point position information of the point relative to the radar device.

7. The method according to claim 4, characterized in that The method further comprises: Obtain the captured photos to be detected and the scenes of the photos to be detected; Inputting the photo to be detected into an anomaly recognition model corresponding to the scene to obtain an anomaly recognition result; The photo to be detected and the corresponding abnormal recognition result are reported.

8. A computer program product, characterized in that The method comprises computer program instructions, and when the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is executed.

9. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method according to any one of claims 1 to 7 is executed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is executed.

Images