A method and system for detecting crowd density based on millimeter wave base station
Through the detection method of the flow density of millimeter wave base station, radar detection data and spatial point cloud technology are used to solve the problems of low accuracy and privacy protection in complex environments, and high accuracy and privacy protection of flow density detection is achieved.
Patent Information
- Application Number
- CN202410905263.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-08
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-07-08
AI Technical Summary
Traditional abortion monitoring methods such as optical video analysis and infrared sensors are difficult to maintain high accuracy in complex environments and may invade personal privacy.
Using the millimeter wave base station's people's flow density detection method, by obtaining radar detection data, converting it into spatial point cloud data, extracting human body data and calculating the flow density, using millimeter wave radar technology to detect people's flow density, avoiding the influence of light and weather, and protecting individuals privacy.
It realizes high-accurate flow density detection in complex environments, and does not infringe on personal privacy, has high stability and accuracy, avoiding errors caused by light changes and temperature fluctuations.
Smart Images

Figure CN118509805B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of communication technology, and in particular to a method and system for detecting crowd density of a millimeter wave base station. Background Art
[0002] In today's ever-changing and rapidly developing society, with the surging tide of urbanization and the continuous growth of population, large gatherings, bustling shopping malls, transportation hubs and other crowded places are common. In these places, accurate perception of population density plays a vital role in improving the level of public safety management. Especially in emergencies such as fires and earthquakes, real-time and accurate crowd flow data can assist emergency response teams to quickly assess the situation on the scene and scientifically formulate evacuation plans, thereby minimizing casualties and property losses. In daily operations, understanding the density of people flow in the venue is also crucial. It can not only help managers optimize resource allocation and improve operational efficiency, but also ensure the improvement of customer experience and service quality to a large extent.
[0003] Traditional crowd monitoring methods, such as optical video analysis and infrared sensors, are widely used, but they each have their own limitations. For example, optical video analysis is limited by lighting conditions and line of sight obstructions, and it is difficult to maintain high accuracy at all times and in complex environments; infrared sensors may produce errors due to environmental factors such as temperature fluctuations. In addition, these methods usually involve the capture of personal images, which raises concerns about privacy protection. Summary of the invention
[0004] In order to solve the above-mentioned problems in the prior art, the present invention provides a method and system for detecting crowd density of a millimeter wave base station.
[0005] In a first aspect, an embodiment of the present application provides a method for detecting crowd density based on a millimeter wave base station, including:
[0006] Obtain radar detection data of millimeter wave base stations within the base station coverage area;
[0007] Converting the radar detection data into spatial point cloud data;
[0008] Extracting human body data from the spatial point cloud data, and counting the number of human bodies within the coverage area of the base station based on the human body data;
[0009] Based on the number of human bodies in the base station coverage area and the area of the base station coverage area, a first human flow density in the base station coverage area of the millimeter wave base station is determined.
[0010] Optionally, the acquiring radar detection data of the millimeter wave base station within the base station coverage area includes:
[0011] Acquiring initial radar detection data of the millimeter wave base station within the base station coverage area;
[0012] De-noising the initial radar detection data;
[0013] The radar detection data is obtained by performing intensity correction on the denoised initial radar detection data using a set distance compensation function.
[0014] Optionally, extracting human body data from the spatial point cloud data includes:
[0015] Extracting spatial features from the spatial point cloud data based on a convolutional neural network or a point cloud extraction algorithm;
[0016] Inputting the spatial features into a pre-built classifier to identify human body features;
[0017] Based on the human body features, human body data in the spatial point cloud data is determined.
[0018] Optionally, the method further includes: acquiring time series data of the millimeter wave base station within a time window; wherein each data element in the time series data includes the number of terminal accesses to the millimeter wave node corresponding to the current time node and the traffic corresponding to the current time node;
[0019] Based on the time series data of the millimeter wave base station in the time window, the second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station is determined.
[0020] Optionally, the method further includes: in response to a difference between a first crowd flow density and a second crowd flow density corresponding to the same time node being less than a preset density threshold, outputting the first crowd flow density or the second crowd flow density.
[0021] Optionally, the method further comprises:
[0022] In response to the difference between the first crowd flow density and the second crowd flow density corresponding to the same time node being greater than the preset density threshold, outputting a third crowd flow density; wherein the third crowd flow density is the average of the first crowd flow density and the second crowd flow density corresponding to the same time node;
[0023] Or in response to the difference between the first crowd flow density and the second crowd flow density corresponding to the same time node being greater than the preset density threshold, the radar detection data of the millimeter wave base station in the base station coverage area is re-acquired, and a new first crowd flow density is calculated based on the radar detection data, and the timing data of the millimeter wave base station in the time window is re-acquired, and a new second crowd flow density is calculated based on the timing data, and the difference between the new first crowd flow density and the new second crowd flow density is compared again; wherein, the new first crowd flow density and the new second crowd flow density correspond to the same time node.
[0024] Optionally, determining, based on the time series data of the millimeter wave base station in the time window, a second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station includes:
[0025] Embed each data element in the time series data based on the Informer layer to obtain an embedding value corresponding to each data element;
[0026] Based on the embedded value corresponding to each data element and the area of the base station coverage area, the second pedestrian flow density corresponding to different time nodes of the base station coverage area of the millimeter wave base station is determined.
[0027] Optionally, determining the second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station based on the embedded value corresponding to each data element and the area of the base station coverage area includes:
[0028] The embedded value corresponding to each data element is processed through the MSGNet layer to obtain the scaled representation of each data element;
[0029] The scaled representation of each data element is output through the Softmax layer, and combined with the area of the base station coverage area, the second pedestrian flow density corresponding to different time nodes of the base station coverage area of the millimeter wave base station is determined.
[0030] Optionally, the method further comprises:
[0031] Divide the base station coverage area of the millimeter wave base station into Q sub-areas; Q is a positive integer;
[0032] Acquire the arrival angle data of N millimeter wave nodes in the millimeter wave base station; determine the azimuth of each of the arrival angle data;
[0033] According to the azimuth of each of the arrival angle data, the corresponding terminal is divided into corresponding sub-areas;
[0034] Based on the number of terminal accesses in each sub-area and the area of each sub-area, the crowd density of each sub-area is determined.
[0035] In a second aspect, the present application also provides a crowd density detection system based on a millimeter wave base station, comprising:
[0036] An acquisition module is used to acquire radar detection data of a millimeter wave base station within a base station coverage area;
[0037] A spatial point cloud data generation module, used for converting the radar detection data into spatial point cloud data;
[0038] A human body detection module, used to extract human body data from the spatial point cloud data, and count the number of human bodies in the coverage area of the base station based on the human body data;
[0039] The density estimation module is used to determine the human flow density in the base station coverage area of the millimeter wave base station based on the number of human bodies in the base station coverage area and the area of the base station coverage area.
[0040] The beneficial effects of the present invention include: the embodiment of the present application first obtains the radar detection data of the millimeter wave base station in the base station coverage area, then converts the data into spatial point cloud data, and then extracts human body data from the spatial point cloud data, and then counts the number of human bodies in the base station coverage area based on the human body data. Finally, the first crowd density of the base station coverage area of the millimeter wave base station can be obtained by dividing the number of human bodies in the base station coverage area by the area of the base station coverage area. It can be seen that the embodiment of the present application provides a crowd density detection method using millimeter wave radar technology, which uses existing millimeter wave base station resources to detect crowd density within a spatial range. This method is not affected by light and weather, and can protect personal privacy. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] Figure 1 A flow chart of the steps of a method for detecting crowd density based on a millimeter wave base station provided in an embodiment of the present invention.
[0042] Figure 2 A flowchart of the steps of another method for detecting crowd density based on a millimeter wave base station provided in an embodiment of the present invention.
[0043] Figure 3 A flowchart of the steps of another method for detecting crowd density based on a millimeter wave base station provided in an embodiment of the present invention.
[0044] Figure 4 A module block diagram of a crowd density detection system based on millimeter wave base stations provided in an embodiment of the present invention.
[0045] Figure 5 A module block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0046] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.
[0047] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.
[0048] Traditional crowd monitoring methods, such as optical video analysis and infrared sensors, are widely used, but they each have their own limitations. For example, optical video analysis is limited by lighting conditions and line of sight obstructions, and it is difficult to maintain high accuracy at all times and in complex environments; infrared sensors may produce errors due to environmental factors such as temperature fluctuations. In addition, these methods usually involve the capture of personal images, which raises concerns about privacy protection.
[0049] In view of the above problems, the present application proposes the following embodiments to solve the above technical problems.
[0050] See also Figure 1 An embodiment of the present application provides a method for detecting crowd density based on a millimeter wave base station, including: steps 101 to 104.
[0051] Step 101: Acquire radar detection data of a millimeter wave base station within the base station coverage area.
[0052] It should be noted that due to its high penetration, low dependence on lighting and climate conditions, and non-intrusiveness to personal privacy, millimeter wave technology is used as a new option in the field of crowd detection. That is, first, the radar detection data of the millimeter wave base station in its base station coverage area is obtained.
[0053] Step 102: Convert the radar detection data into spatial point cloud data.
[0054] This step is used to convert the radar detection data into three-dimensional spatial point cloud data.
[0055] Step 103: extracting human body data from the spatial point cloud data, and counting the number of human bodies within the base station coverage area based on the human body data.
[0056] This step involves data classification, that is, filtering out human body data from the spatial point cloud data, and then using the human body data to count the number of human bodies in the entire base station coverage area.
[0057] Step 104: Determine a first human flow density in the base station coverage area of the millimeter wave base station based on the number of human bodies in the base station coverage area and the area of the base station coverage area.
[0058] The calculation of the first crowd density in the base station coverage area of the millimeter wave base station can refer to the following formula, including: , where D represents the first human flow density, N represents the number of human bodies in the base station coverage area, and A represents the area of the base station coverage area.
[0059] In summary, the method for detecting crowd density of a millimeter wave base station provided in the embodiment of the present application has the following beneficial effects, including:
[0060] The embodiment of the present application first obtains the radar detection data of the millimeter wave base station in the base station coverage area, then converts the data into spatial point cloud data, and then extracts human body data from the spatial point cloud data. Then, based on the human body data, the number of human bodies in the base station coverage area is counted. Finally, the number of human bodies in the base station coverage area is divided by the area of the base station coverage area to obtain the first crowd density of the base station coverage area of the millimeter wave base station. It can be seen that the embodiment of the present application provides a crowd density detection method using millimeter wave radar technology, which uses existing millimeter wave base station resources to detect crowd density within a spatial range. This method is not affected by light and weather, and can protect personal privacy.
[0061] In other words, compared with the detection through video in the prior art, this method will not be affected by lighting changes or line of sight obstructions, and is non-invasive to personal privacy and can protect personal privacy.
[0062] Compared with the detection by infrared sensors in the prior art, this method has higher stability and accuracy, and will not produce large errors due to environmental factors such as temperature fluctuations. In addition, the use of infrared sensors usually involves the capture of personal images and lacks privacy protection.
[0063] Optionally, the above steps of obtaining radar detection data of the millimeter wave base station within the base station coverage area may include: obtaining initial radar detection data of the millimeter wave base station within the base station coverage area; denoising the initial radar detection data; and performing intensity correction on the denoised initial radar detection data through a set distance compensation function to obtain radar detection data.
[0064] It should be noted that the initial radar detection data usually includes distance, angle and Doppler shift data. Here, the initial radar detection data can be denoised to improve data quality.
[0065] Specifically, a low-pass filter or other signal processing techniques may be used to reduce the noise in the data. The denoising formula may include: .in, represents the initial radar detection data after denoising; Represents initial radar detection data; Indicates filtering.
[0066] Then, the signal strength can be adjusted by distance to compensate for signal attenuation. The specific formula is as follows: .in, represents the distance compensation function, d is the distance from the radar to the target; Indicates the final radar detection data.
[0067] Optionally, the process of converting the radar detection data into spatial point cloud data in the above step may specifically include: converting the radar detection data into a point cloud format, where each point represents the position of an object in space, including coordinate conversion and data structuring. First, convert the polar coordinate data (distance and angle) of the radar into a Cartesian coordinate system. , the conversion is as follows:
[0068] ;
[0069] Where r is the straight-line distance of the target from the radar, is the azimuth, is the elevation angle. The transformed coordinate data is then organized into a point cloud data structure, which typically includes the location of the point and possible additional attributes (such as reflection intensity).
[0070] Optionally, the above steps extract human body data from spatial point cloud data, including: extracting spatial features in spatial point cloud data based on a convolutional neural network or a point cloud extraction algorithm; inputting the spatial features into a pre-built classifier to identify human body features; and determining human body data in the spatial point cloud data based on the human body features.
[0071] That is, detecting a human body from spatial point cloud data includes feature extraction and human body recognition. In an embodiment of the present application, a 3D convolutional neural network (CNN) or a point cloud feature extraction algorithm (such as PointNet or PointNet++) can be used to extract spatial features. The extracted features are input into a classifier (such as a support vector machine SVM or a deep learning model) to identify the human body, and then the human body data in the spatial point cloud data can be determined.
[0072] After obtaining the human body data, the number of human bodies can be counted.
[0073] Taking into account the problem of cross-overlap between human bodies, the embodiments of the present application can also detect the interaction and overlap between targets based on the JPDA (Joint Probabilistic Data Association) or MHT (Multiple Hypothesis Tracking) algorithm, thereby improving the accuracy of human body counting.
[0074] See also Figure 2 The method for detecting crowd density based on a millimeter wave base station provided in the embodiment of the present application also includes: step 201 to step 202.
[0075] Step 201: Obtain the timing data of the millimeter wave base station within the time window T.
[0076] Among them, each data element in the time series data includes the number of terminal accesses to N millimeter wave nodes corresponding to the current time node and the traffic corresponding to the current time node.
[0077] Here, the number of terminal accesses to the millimeter wave node can be counted through network detection software such as Tems Investigation and NetSpot. Assuming that each base station has N millimeter wave nodes, the number of accesses to each millimeter wave node at the current moment can be obtained. .
[0078] The downlink traffic data can be obtained by analyzing the base station traffic log to obtain the downlink traffic situation in the current time period T, or by measuring it through the operator management tool to obtain the total traffic F of the base station.
[0079] Specifically, taking time T as the observation window, the total traffic F and the number of accesses to each millimeter wave node are Time series data can be constructed ;
[0080] In this formula, It represents a data element, n=1, 2...
[0081] Step 202: Based on the time series data of the millimeter wave base station in the time window T, determine the second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station.
[0082] It can be seen that the embodiment of the present application also provides a method for measuring the time series data of the constructed millimeter wave base station within the time window T to obtain the second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station.
[0083] It should be noted that the embodiments of the present application provide two types of crowd density detection based on millimeter wave base stations, which can cooperate with each other to achieve more accurate crowd density detection.
[0084] Specifically, the first judgment method includes: in response to the difference between the first crowd flow density and the second crowd flow density corresponding to the same time node being less than a preset density threshold, outputting the first crowd flow density or the second crowd flow density.
[0085] That is, when the difference between the first crowd density and the second crowd density corresponding to the same time node is less than the preset density threshold, it means that the error of the crowd density measured by the two methods is very small. At this time, the first crowd density or the second crowd density can be output as the crowd density result corresponding to the current time node.
[0086] It can be seen that the embodiment of the present application compares the pedestrian density measured by the two methods to determine the error between the two, and ensures the reliability of the output result by comparing the error with the preset density threshold.
[0087] The above-mentioned preset density threshold can be set according to actual conditions, and there is no limitation on the specific value.
[0088] Specifically, the second judgment method includes: in response to the difference between the first crowd flow density and the second crowd flow density corresponding to the same time node being greater than a preset density threshold, outputting a third crowd flow density; wherein the third crowd flow density is the average of the first crowd flow density and the second crowd flow density corresponding to the same time node.
[0089] That is, when the difference between the first crowd density and the second crowd density corresponding to the same time node is greater than the preset density threshold, it indicates that there is a large error between the crowd densities measured by the two methods. It should be noted that both methods may be affected by different factors and cause a certain difference from the actual crowd density. Therefore, the processing method here is to output the average of the first crowd density and the second crowd density to reduce the error by taking the average value.
[0090] Specifically, the third judgment method includes: in response to the difference between the first crowd flow density and the second crowd flow density corresponding to the same time node being greater than a preset density threshold, re-acquiring the radar detection data of the millimeter wave base station in the base station coverage area, and calculating the new first crowd flow density based on the radar detection data, and re-acquiring the time series data of the millimeter wave base station in the time window T, and calculating the new second crowd flow density based on the time series data, and comparing the difference between the new first crowd flow density and the new second crowd flow density again; wherein, the new first crowd flow density and the new second crowd flow density correspond to the same time node.
[0091] That is, when the difference between the first crowd density and the second crowd density corresponding to the same time node is greater than the preset density threshold, it indicates that there is a large error between the crowd densities measured by the two methods. It should be noted that both methods may be affected by different factors and cause certain differences from the actual crowd density. Therefore, the processing method here is to recalculate using the two methods to re-compare and output the new first crowd density and the new second crowd density. This method can eliminate the situation where the error is large due to calculation reasons or abnormal collected data.
[0092] It should be noted that, when the difference between the new first crowd flow density and the new second crowd flow density is still greater than the preset density threshold, the second judgment method can be used to output the average of the two.
[0093] Optionally, the above steps determine the second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station based on the time series data of the millimeter wave base station within the time window T, and may also specifically include: embedding each data element in the time series data based on the Informer layer to obtain the embedded value corresponding to each data element; based on the embedded value corresponding to each data element and the area of the base station coverage area, determining the second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station.
[0094] Specifically, the Informer layer is used to embed the data elements in Data. Specifically, Normalize it and get , and then feed it into a one-dimensional convolution , so that it is projected into a multi-dimensional matrix, the formula is as follows:
[0095] ;
[0096] In this formula, PE represents the embedding position of each element in Data (the truncation point in time). A global timestamp embedding representing the science department of each element; Represents the embedded value. The embedded value can be used to represent the population (number of human bodies) corresponding to the time node.
[0097] After obtaining the embedded value, the second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station can be determined based on the embedded value corresponding to each data element and the area of the base station coverage area.
[0098] Optionally, the above steps determine the second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station based on the embedded value corresponding to each data element and the area of the base station coverage area, and may also specifically include: processing the embedded value corresponding to each data element through the MSGNet layer to obtain a scaled representation of each data element; outputting the scaled representation of each data element through the Softmax layer, and determining the second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station in combination with the area of the base station coverage area.
[0099] Specifically, after the embedding value is obtained, the embedding value corresponding to each data element is scaled and represented through the MSGNet layer. The formula is as follows:
[0100] ;
[0101] In the above formula, represents the calculated scaling representation; the purpose of scaling representation in this step is to minimize the pollution caused by some unstable access at the data level. Finally, combined with the coverage area of the base station Calculate the population density:
[0102] ;
[0103] In the above formula, the Softmax output is the population size in each time period corresponding to each scaling representation. By dividing it by the coverage area of the base station, the population density corresponding to different time nodes in the area can be obtained. .
[0104] See also Figure 3 Optionally, the above method also includes: step 301 to step 305.
[0105] Step 301: Divide the base station coverage area of the millimeter wave base station into Q sub-areas, where Q is a positive integer. In the embodiment of the present application, Q can be 4, that is, the base station coverage area of the millimeter wave base station is divided into four areas A, B, C, and D.
[0106] Step 302: Acquire arrival angle data of N millimeter wave nodes in the millimeter wave base station.
[0107] Step 303: Determine the azimuth of each arrival angle data.
[0108] Step 304: According to the azimuth of each arrival angle data, the corresponding terminal is divided into the corresponding sub-area.
[0109] Step 305: Determine the crowd density of each sub-area based on the number of terminals accessing each sub-area and the area of each sub-area.
[0110] Specifically, the above process analyzes the fingerprint of communication beamforming to perform regional population density segmentation. In the embodiment of the present application, the base station coverage is divided into four areas (A, B, C, D) and a world coordinate system is established. The specific population density of each area is obtained by calculating the number of people in each area.
[0111] Specifically, we first obtain the arrival angle data x of different millimeter wave nodes, and then divide the angle range into N parts according to the world coordinate system. , and generate basis vectors, and construct matrices.
[0112] Use the classic OMP (Orthogonal Matching Pursuit) algorithm to calculate the inner product of each basis vector in the matrix , take the basis vector with the largest inner product and add it to the reconstruction matrix . Reconstruct the signal into ,use Replace the data used in the inner product calculation and repeat the preset number of times. The final reconstructed matrix That is, it represents the direction corresponding to the arrival angle data x. The corresponding terminal is divided into the corresponding sub-area according to the direction. Finally, the crowd density of each sub-area can be determined based on the number of terminal accesses in each sub-area and the area of each sub-area.
[0113] In summary, the embodiment of the present invention also proposes a scene population density perception method based on millimeter wave base stations, which can also make up for the defects of existing infrared or video image sensors. By constructing time series data to calculate the total population of the base station area, and using MSGNet to scale the embedded results, the impact of unstable access on the measurement results is reduced, thereby improving the perception accuracy. In addition, the angle of arrival data is used to divide the area more finely to provide more detailed downstream task guidance information. The advantage of this method is that it reduces privacy concerns and uses existing millimeter wave base stations to measure population data without the need for additional dedicated equipment, thereby reducing costs.
[0114] See also Figure 4Based on the same inventive concept, the present application also provides a crowd density detection system 400 based on a millimeter wave base station, including: an acquisition module 401, used to acquire radar detection data of the millimeter wave base station in the base station coverage area; a spatial point cloud data generation module 402, used to convert the radar detection data into spatial point cloud data; a human body detection module 403, used to extract human body data from the spatial point cloud data, and count the number of human bodies in the base station coverage area based on the human body data; a density estimation module 404, used to determine the crowd density in the base station coverage area of the millimeter wave base station based on the number of human bodies in the base station coverage area and the area of the base station coverage area.
[0115] See also Figure 5 Based on the same inventive concept, the present application embodiment provides a module frame of an electronic device 500 that applies the above method. The electronic device 500 includes: at least one processor 501 ( Figure 5 Only one is shown in the figure), a memory 502, a computer program 503 stored in the memory 502 and executable on at least one processor 501, and when the processor 501 executes the computer program 503, the steps of the method in any of the foregoing embodiments are implemented.
[0116] The electronic device 500 may be a server, a personal computer, etc. Those skilled in the art will appreciate that Figure 5 The electronic device 500 is merely an example and does not constitute a limitation on the electronic device 500 , and may include more or less components than those shown in the figure, or may combine certain components, or may include different components.
[0117] The processor 501 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0118] In some embodiments, the memory 502 may be an internal storage unit of the electronic device 500, such as a hard disk or memory of the electronic device 500. In other embodiments, the memory 502 may also be an external storage device of the electronic device 500, 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 500. Furthermore, the memory 502 may also include both an internal storage unit of the electronic device 500 and an external storage device.
[0119] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.
[0120] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.
[0121] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.
[0122] An embodiment of the present application provides a computer program product. When the computer program product runs on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.
[0123] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / electronic device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electrical carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk.
[0124] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0125] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0126] In the embodiments provided in the present application, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0127] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0128] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A method for detecting crowd density based on a millimeter wave base station, characterized in that: include: Obtain radar detection data of millimeter wave base stations within the base station coverage area; Converting the radar detection data into spatial point cloud data; Extracting human body data from the spatial point cloud data, and counting the number of human bodies within the coverage area of the base station based on the human body data; Determine a first human flow density in the base station coverage area of the millimeter wave base station based on the number of human bodies in the base station coverage area and the area of the base station coverage area; The method further includes: obtaining time series data of the millimeter wave base station within a time window; wherein each data element in the time series data includes the number of terminal accesses to a millimeter wave node corresponding to a current time node and the flow corresponding to the current time node; based on the time series data of the millimeter wave base station within the time window, determining a second flow density of people corresponding to different time nodes in a base station coverage area of the millimeter wave base station; The determining, based on the time series data of the millimeter wave base station in the time window, the second crowd flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station includes: embedding each data element in the time series data based on the Informer layer to obtain the embedded value corresponding to each data element; determining the second crowd flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station based on the embedded value corresponding to each data element and the area of the base station coverage area; The method of determining the second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station based on the embedded value corresponding to each data element and the area of the base station coverage area includes: processing the embedded value corresponding to each data element through the MSGNet layer to obtain a scaled representation of each data element; outputting the scaled representation of each data element through the Softmax layer, and determining the second pedestrian flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station in combination with the area of the base station coverage area.
2. The method for detecting crowd density based on a millimeter wave base station according to claim 1, characterized in that: The obtaining of radar detection data of the millimeter wave base station within the base station coverage area includes: Acquiring initial radar detection data of the millimeter wave base station within the base station coverage area; De-noising the initial radar detection data; The radar detection data is obtained by performing intensity correction on the denoised initial radar detection data using a set distance compensation function.
3. The method for detecting crowd density based on a millimeter wave base station according to claim 1, characterized in that: The extracting human body data from the spatial point cloud data comprises: Extracting spatial features from the spatial point cloud data based on a convolutional neural network or a point cloud extraction algorithm; Inputting the spatial features into a pre-built classifier to identify human body features; Based on the human body features, human body data in the spatial point cloud data is determined.
4. The method for detecting crowd density based on a millimeter wave base station according to claim 1, characterized in that: The method further comprises: In response to the difference between the first crowd flow density and the second crowd flow density corresponding to the same time node being less than a preset density threshold, the first crowd flow density or the second crowd flow density is output.
5. The method for detecting crowd density based on a millimeter wave base station according to claim 4, characterized in that: The method further comprises: In response to the difference between the first crowd flow density and the second crowd flow density corresponding to the same time node being greater than the preset density threshold, outputting a third crowd flow density; wherein the third crowd flow density is the average of the first crowd flow density and the second crowd flow density corresponding to the same time node; or In response to the difference between the first crowd flow density and the second crowd flow density corresponding to the same time node being greater than the preset density threshold, the radar detection data of the millimeter wave base station in the base station coverage area is reacquired, and a new first crowd flow density is calculated based on the radar detection data, and the timing data of the millimeter wave base station in the time window is reacquired, and a new second crowd flow density is calculated based on the timing data, and the difference between the new first crowd flow density and the new second crowd flow density is compared again; wherein, the new first crowd flow density and the new second crowd flow density correspond to the same time node.
6. The method for detecting crowd density based on a millimeter wave base station according to claim 1, characterized in that: The method further comprises: Divide the base station coverage area of the millimeter wave base station into Q sub-areas; Q is a positive integer; Acquire arrival angle data of N millimeter wave nodes in the millimeter wave base station; determining the azimuth of each of the angle of arrival data; According to the azimuth of each of the arrival angle data, the corresponding terminal is divided into corresponding sub-areas; Based on the number of terminal accesses in each sub-area and the area of each sub-area, the crowd density of each sub-area is determined.
7. A crowd density detection system based on millimeter wave base stations, characterized in that: include: An acquisition module is used to acquire radar detection data of a millimeter wave base station within a base station coverage area; A spatial point cloud data generation module, used for converting the radar detection data into spatial point cloud data; A human body detection module, used to extract human body data from the spatial point cloud data, and count the number of human bodies in the coverage area of the base station based on the human body data; A density estimation module, used to determine the human flow density in the base station coverage area of the millimeter wave base station based on the number of human bodies in the base station coverage area and the area of the base station coverage area; The system is further used to obtain the time series data of the millimeter wave base station within the time window; wherein each data element in the time series data includes the number of terminal accesses to the millimeter wave node corresponding to the current time node and the flow corresponding to the current time node; based on the time series data of the millimeter wave base station within the time window, determine the second flow density corresponding to different time nodes in the base station coverage area of the millimeter wave base station; And specifically used to embed each data element in the time series data based on the Informer layer to obtain the embedded value corresponding to each data element; based on the embedded value corresponding to each data element and the area of the base station coverage area, determine the second pedestrian flow density corresponding to different time nodes of the base station coverage area of the millimeter wave base station; It is specifically used to process the embedded value corresponding to each data element through the MSGNet layer to obtain the scaled representation of each data element; output the scaled representation of each data element through the Softmax layer, and combine the area of the base station coverage area to determine the second pedestrian flow density corresponding to different time nodes of the base station coverage area of the millimeter wave base station.
