Intelligent temporary platform non-contact image recognition mobile control method and system

Abstract

The present invention relates to the technical field of intelligent non-contact control of human flow, and in particular to a non-contact image recognition mobile control method and system for an intelligent temporary platform. It includes: generating a first area crowd density ρ1 corresponding to a target area according to a difference set between a first identification set and a second identification set; determining a second area crowd density ρ2 corresponding to the target area according to positioning information of a communication terminal by a communication base station; generating a crowd density ρ in the current target area according to ρ1, ρ2 and a probe compensation coefficient φ; generating a mobile control instruction according to ρ and a regional crowd density threshold Y. The final crowd density is obtained by fusing ρ1 and ρ2, thereby improving the calculation accuracy, guiding the flow of personnel more efficiently and accurately, and reducing the occurrence of public safety issues such as traffic congestion. In addition, the density of people in the area is counted in a non-contact and non-sensing manner, thereby reducing the requirements for the degree of cooperation of people entering the area.

Description

Background Art

[0002] Crowd movement control refers to measures to manage and control the flow of people in specific places or activities to ensure safe, orderly and efficient organization of people flow. This kind of control is particularly important in large-scale events, public transportation hubs, commercial centers, tourist attractions, emergency evacuations, etc. The goals of crowd movement control usually include preventing accidents caused by crowding, improving traffic efficiency, ensuring public safety, and quickly evacuating people in emergency situations.

[0003] In the prior art, a large number of manpower is usually organized to control the flow of people at some large gatherings. However, this method is usually inefficient and it is difficult to accurately grasp the crowd density in the corresponding area. As a result, it is very easy for the crowd density in some areas to be high, which in turn leads to public safety issues such as traffic congestion. Summary of the invention

[0004] In view of the above technical problems, the technical solution adopted by the present invention is:

[0005] According to one aspect of the present invention, a method for non-contact image recognition mobile control of an intelligent temporary platform is provided, the method comprising the following steps:

[0006] Acquire a first identification set A of communication devices collected by a plurality of WiFi probes disposed at the entrance of the target area;

[0007] Obtain a second identification set B of communication devices collected by multiple WiFi probes set at the exit of the target area; each device corresponds to a device identification in the identification set; the device identification is generated by the Mac address of the communication device;

[0008] According to the difference between the first identification set and the second identification set, the first area population density ρ1 corresponding to the target area is generated; ρ1 satisfies the following condition: ρ1=A−(A∩B);

[0009] Determine the second area crowd density ρ2 corresponding to the target area according to the positioning information of the communication terminal by the communication base station;

[0010] According to ρ1, ρ2 and the probe compensation coefficient φ, the crowd density ρ in the current target area is generated; φ∈(0,2), ρ satisfies the following conditions:

[0011] ρ = (φ × ρ1 + ρ2) / 2;

[0012] Generate mobile control instructions based on ρ and the regional population density threshold Y.

[0013] Furthermore, after determining the second area crowd density ρ2 corresponding to the target area according to the positioning information of the communication terminal by the communication base station, the method further includes:

[0014] Acquire image data corresponding to multiple angles in the target area;

[0015] Input multiple image data into the target density recognition model respectively to generate the crowd density sub-prediction value and sub-confidence corresponding to each image data; the target density recognition model includes the YOLO target detection model and the CNN density prediction model;

[0016] According to the crowd density sub-prediction value and sub-confidence corresponding to each image data, the third area crowd density ρ3 and confidence δ corresponding to the target area are generated; ρ3 and δ meet the following conditions:

[0017] ; ;

[0018] ρ i and δ i are the sub-prediction value and sub-confidence of the crowd density corresponding to the i-th image data; z is the total number of image data;

[0019] According to ρ1, ρ2, ρ3 and δ, the crowd density ρ in the current target area is generated; ρ satisfies the following conditions:

[0020] ρ=δ×ρ3+(1-δ)×[(φ×ρ1+ρ2) / 2];

[0021] Generate mobile control instructions based on ρ and the regional population density threshold Y.

[0022] Further, the first identification set A or the second identification set B is obtained according to the following steps:

[0023] Get the probe request frame captured by each WiFi probe;

[0024] Hash the device Mac address information included in each probe request frame to generate an address hash value corresponding to each device Mac address;

[0025] Match each address hash value with the existing address hash values ​​in the identification set;

[0026] If the match is successful, the current address hash value is discarded;

[0027] If the match fails, the current address hash value is added to the corresponding identification set.

[0028] Further, according to the positioning information of the communication terminal by the communication base station, determining the second area crowd density ρ2 corresponding to the target area includes:

[0029] Obtain a list of devices obtained from multiple target communication base stations during the collection period; the list of devices includes communication device identification, base station coordinates and communication distance; the number of target communication base stations is greater than or equal to 3; the target communication base station is a communication base station whose signal range completely covers the target area;

[0030] Based on the communication device identification, an intersection list of multiple device lists is obtained; each communication device identification in the intersection list corresponds to an information pair consisting of at least 3 base station coordinates and communication distances;

[0031] Determine the coordinate information corresponding to each communication device identifier according to the information pair consisting of the base station coordinates and the communication distance corresponding to each communication device identifier;

[0032] Compare the coordinate information corresponding to each communication device identifier with the coordinate domains [Jz, Jy] and [Ws, Wx] corresponding to the target area; Jz and Jy are respectively the longitude boundary values ​​corresponding to the target area; Ws and Wx are respectively the latitude boundary values ​​corresponding to the target area;

[0033] If J s n ∈[Jz,Jy] and W s n ∈[Ws, Wx], it is determined that the communication terminal corresponding to the nth communication device identifier is located in the target area; J s n and W s n The longitude and latitude of the communication terminal corresponding to the n-th communication device identifier are respectively.

[0034] Further, according to the information pair consisting of the base station coordinates and the communication distance corresponding to each communication device identifier, the coordinate information corresponding to each communication device identifier is determined, including:

[0035] The three-circle positioning method is used to determine the coordinate information corresponding to each communication device identifier based on the information pair consisting of the base station coordinates and the communication distance corresponding to each communication device identifier.

[0036] Furthermore, the communication distance is obtained according to the following steps:

[0037] Obtaining the sending timestamp and receiving timestamp of the information sent by the communication device received by the communication base station;

[0038] The communication distance is generated based on the time difference between the sending timestamp and the receiving timestamp and the speed of light.

[0039] Further, after obtaining the first identification set A of the communication devices collected by the multiple WiFi probes set at the entrance of the target area, the method further includes:

[0040] Obtaining a supplementary sub-identity set of communication devices collected by multiple supplementary WiFi probes in the target area;

[0041] The union of the multiple supplementary sub-identifier sets and the first identifier set is used as a new first identifier set.

[0042] Furthermore, multiple supplementary WiFi probes are arranged in some areas where people must pass.

[0043] Furthermore, the duration of the collection period is greater than or equal to 10 minutes.

[0044] According to a second aspect of the present invention, a non-contact image recognition mobile management and control system for an intelligent temporary platform is provided, the system comprising:

[0045] A first acquisition module is used to acquire a first identification set A of communication devices collected by a plurality of WiFi probes arranged at the entrance of the target area;

[0046] The second acquisition module is used to acquire a second identification set B of communication devices collected by multiple WiFi probes set at the exit of the target area; each device corresponds to a device identification in the identification set; the device identification is generated by the Mac address of the communication device;

[0047] The first generating module is used to generate a first area population density ρ1 corresponding to the target area according to the difference set between the first identification set and the second identification set; ρ1 satisfies the following condition: ρ1=A−(A∩B);

[0048] The second generating module is used to determine the second area crowd density ρ2 corresponding to the target area according to the positioning information of the communication terminal by the communication base station;

[0049] The fusion module is used to generate the crowd density ρ in the current target area according to ρ1, ρ2 and the probe compensation coefficient φ; φ∈(0,2), ρ satisfies the following conditions:

[0050] ρ = (φ × ρ1 + ρ2) / 2;

[0051] The command output module is used to generate mobile control commands based on ρ and the regional crowd density threshold Y.

[0052] The present invention has at least the following beneficial effects:

[0053] In this application, WiFi probes are set up to passively detect mobile communication devices with WiFi functions turned on by people entering and leaving the target area, so as to count the devices through the Mac addresses carried in the captured probe request frames. Usually, in most cases, a person will only carry one communication device (such as a smart phone). Therefore, the number of people in the target area can be roughly calculated by A-(A∩B), that is, ρ1. In addition, the positioning information of the communication terminal provided by multiple communication base stations around the target area is used to determine the number of people in the target area from another dimension, that is, ρ2. Then, ρ1 and ρ2 are added and averaged to obtain the final crowd density in the target area, so as to improve the calculation accuracy of the crowd density.

[0054] At the same time, since the WiFi probe is only effective when the WiFi function is turned on by the person, it will not be able to collect data from some devices that do not have the WiFi function turned on, resulting in collection errors. Therefore, when calculating, it is necessary to adjust ρ1 through the probe compensation coefficient φ to make the final crowd density more accurate. The probe compensation coefficient φ can be determined through historical compensation conditions.

[0055] In addition, due to the wide coverage of the communication base station, it is more suitable for detection and perception in a large area; while the effective range of the WiFi probe is usually from a few hundred meters to a few thousand square meters, which is more suitable for detection in some smaller areas. Therefore, this application uses WiFi probe technology to detect in the entrance and exit areas respectively, and uses the communication base station method to detect in the entire larger area. Then, the advantages of the two methods are used for detection to improve the final detection accuracy.

[0056] Moreover, since each detection scheme is a non-contact detection method, the density of people in the area can be counted in a non-contact and non-sensing way, thereby reducing the requirements for the cooperation of people entering the area. At the same time, control instructions can be generated according to the density of people in different areas, such as whether to close or open certain entrances and exits, or guide people from one area to another. In this way, the flow of people can be guided more efficiently and accurately, which can reduce the occurrence of public safety problems such as traffic congestion. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0058] Figure 1 A flow chart of a non-contact image recognition mobile control method for an intelligent temporary platform provided by an embodiment of the present invention;

[0059] Figure 2 A structural diagram of a non-contact image recognition mobile control system for an intelligent temporary platform provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0061] According to a possible embodiment of the present invention, Figure 1 As shown, a method for non-contact image recognition mobile control of an intelligent temporary platform is provided, and the method comprises the following steps:

[0062] S100: Acquire a first identification set A of communication devices collected by a plurality of WiFi probes disposed at the entrance of a target area.

[0063] WiFi probe devices passively monitor the surrounding radio signal bands (usually 2.4GHz and 5GHz ISM bands) through built-in antennas. These bands are standard bands for WiFi communication. The probe does not need to participate in the actual WiFi network communication, but "listens" to the wireless signals propagating in the surrounding air like a sniffer.

[0064] Usually in some target areas such as shopping malls, stations, stadiums, and some large parks with fixed entrances and exits, the areas corresponding to the entrances and exits are generally small. Therefore, by arranging WiFi probes in this smaller space, the Mac address of each person's communication device can be collected more accurately. In order to further improve the ability to identify the incoming personnel, in this embodiment, multiple WiFi probes can be used to locate the communication device to determine whether the person is in the target area. In addition, corresponding notifications or other interesting interactive methods can be set at the entrances and exits to improve the comprehensiveness of the WiFi probe's capture of the communication terminal. This will improve the accuracy of subsequent calculations.

[0065] In addition, after S100, the method further includes:

[0066] S101: Obtaining a set of supplementary sub-identifiers of communication devices collected by a plurality of supplementary WiFi probes in a target area. Further, a plurality of supplementary WiFi probes are arranged in some areas where people must pass, such as restaurants, certain intersections and bridges.

[0067] S102: taking a union of the plurality of supplementary sub-identifier sets and the first identifier set as a new first identifier set.

[0068] By setting up additional WiFi probes, the statistical comprehensiveness of the communication devices entering the target area can be further improved. Of course, the same method can also be used to arrange additional WiFi probes in other areas of the exit to improve the statistical comprehensiveness of the communication devices leaving the target area.

[0069] S200: Obtain a second identification set B of communication devices collected by multiple WiFi probes set at the exit of the target area. Each device corresponds to a device identification in the identification set. The device identification is generated by the Mac address of the communication device.

[0070] Specifically, the first identification set A or the second identification set B is obtained according to the following steps:

[0071] S201: Acquire a probe request frame captured by each WiFi probe.

[0072] The WiFi probe device will first be set to work in monitoring mode, which allows it to receive all surrounding WiFi signals without having to participate in any actual network communication. It will listen to the probe request frames sent by the device (when the device attempts to discover and connect to a known network) and other WiFi management frames, which usually contain the device's Mac address. The Mac address is the physical identifier of the device and is theoretically unique worldwide, so it can be used to distinguish different devices.

[0073] S202: Hash the device Mac address information included in each probe request frame to generate an address hash value corresponding to each device Mac address.

[0074] In order to improve the confidentiality of personnel and device information, the device Mac address needs to be encrypted while ensuring the uniqueness of a large amount of data. Therefore, hash processing can be used to generate address hash values, such as MD5 values.

[0075] S203: Match each address hash value with the existing address hash values ​​in the identification set.

[0076] S204: If the match is successful, the current address hash value is discarded.

[0077] S205: If the match is unsuccessful, the current address hash value is added to the corresponding identification set.

[0078] Every time a new address hash value is obtained, it is compared with the existing address hash value to avoid duplicate storage and thus reduce memory usage.

[0079] S300: Generate a first area population density ρ1 corresponding to the target area according to the difference between the first identification set and the second identification set. ρ1 satisfies the following condition: ρ1=A−(A∩B).

[0080] S400: Determine the second area crowd density ρ2 corresponding to the target area according to the positioning information of the communication terminal by the communication base station.

[0081] Specifically, S400 includes:

[0082] S401: Obtain a list of devices obtained from multiple target communication base stations during the collection period. The device list includes communication device identification, base station coordinates, and communication distance. The number of target communication base stations is greater than or equal to 3. The target communication base station is a communication base station whose signal range completely covers the target area. Furthermore, the duration of the collection period is greater than or equal to 10 minutes. The communication device identification can be a hash value corresponding to the device ID.

[0083] In daily use, communication terminals (i.e. mobile phones) usually establish frequent communication connections with corresponding communication base stations (such as mobile cellular base stations). These signals include: synchronization signals, channel allocation information, power control commands, call signals, traffic data signals, multimedia and high-speed data service signals, and heartbeat signals, etc. The distance and location between the communication terminal and the communication base station can be determined through these electromagnetic wave signals.

[0084] Furthermore, the communication distance is obtained according to the following steps:

[0085] S411: Obtaining a sending timestamp and a receiving timestamp of information sent by a communication device and received by the communication base station.

[0086] S421: Generate a communication distance based on the time difference formed by the sending timestamp and the receiving timestamp and the speed of light.

[0087] This method is based on the measurement of signal propagation speed and signal arrival time. The communication base station sends a signal to the mobile phone. After the mobile phone receives the signal, it records the time when the signal is received and sends this timestamp back to the communication base station. The communication base station can calculate the distance between the mobile phone and the base station based on the time difference of the round-trip signal and the propagation speed of electromagnetic waves in the air (approximately 3×10^8 meters / second). This method requires precise time synchronization between the mobile phone and the base station.

[0088] S402: Based on the communication device identification, an intersection list of multiple device lists is obtained. Each communication device identification in the intersection list corresponds to an information pair consisting of at least three base station coordinates and communication distances.

[0089] S403: Determine the coordinate information corresponding to each communication device identifier according to the information pair consisting of the base station coordinates and the communication distance corresponding to each communication device identifier. Specifically, the coordinate information corresponding to each communication device identifier is determined by using the three-circle positioning method according to the information pair consisting of the base station coordinates and the communication distance corresponding to each communication device identifier.

[0090] S404: Compare the coordinate information corresponding to each communication device identifier with the coordinate domains [Jz, Jy] and [Ws, Wx] corresponding to the target area. Jz and Jy are respectively the longitude boundary values ​​corresponding to the target area. Ws and Wx are respectively the latitude boundary values ​​corresponding to the target area.

[0091] S405: If J s n ∈[Jz,Jy] and W s n ∈[Ws, Wx], it is determined that the communication terminal corresponding to the nth communication device identifier is located in the target area. s n and W s n The longitude and latitude of the communication terminal corresponding to the n-th communication device identifier are respectively.

[0092] To determine the coordinate domains [Jz, Jy] and [Ws, Wx] corresponding to the target area, a communication terminal can be placed at the boundary of the target area in advance, and then the boundary of the target area can be outlined by using the positioning coordinates determined by the target communication base station. To determine the coordinate domain, a rectangular bounding box can be generated according to the area boundary to obtain [Jz, Jy] and [Ws, Wx].

[0093] S500: Generate the crowd density ρ in the current target area according to ρ1, ρ2 and the probe compensation coefficient φ. φ∈(0,2), ρ satisfies the following conditions:

[0094] ρ = (φ × ρ1 + ρ2) / 2;

[0095] Since WiFi probes are only effective when people turn on the WiFi function, they cannot collect data from some devices that do not have the WiFi function turned on, resulting in collection errors. Therefore, when calculating, it is necessary to adjust ρ1 through the probe compensation coefficient φ to make the final crowd density more accurate. The probe compensation coefficient φ can be determined based on historical compensation conditions. In other words, according to the errors that occurred in the recent WiFi probe statistics process, determine how large the corresponding φ should be.

[0096] S600: Generate a movement control instruction according to ρ and the regional crowd density threshold Y. The movement control instruction may be whether to close or open certain entrances and exits, or to guide people from one area to another.

[0097] In this application, WiFi probes are set up to passively detect mobile communication devices with WiFi functions turned on by people entering and leaving the target area, so as to count the devices through the Mac addresses carried in the captured probe request frames. Usually, in most cases, a person will only carry one communication device (such as a smart phone). Therefore, the number of people in the target area can be roughly calculated by A-(A∩B), that is, ρ1. In addition, the positioning information of the communication terminal provided by multiple communication base stations around the target area is used to determine the number of people in the target area from another dimension, that is, ρ2. Then, ρ1 and ρ2 are added and averaged to obtain the final crowd density in the target area, so as to improve the calculation accuracy of the crowd density.

[0098] Moreover, since each detection scheme is a non-contact detection method, the density of people in the area can be counted in a non-contact and non-sensing way, thereby reducing the requirements for the cooperation of people entering the area. At the same time, control instructions can be generated according to the density of people in different areas, such as whether to close or open certain entrances and exits, or guide people from one area to another. In this way, the flow of people can be guided more efficiently and accurately, which can reduce the occurrence of public safety problems such as traffic congestion.

[0099] As another possible embodiment of the present invention, after S400, the method further includes:

[0100] S700: Obtain image data corresponding to multiple angles in the target area.

[0101] The image data can be obtained by shooting with a surveillance camera fixed in the target area, or by using a drone to shoot in the corresponding area. By acquiring images at different angles in the same area, image distortion caused by occlusion and light problems can be avoided. In addition, the target area can be divided into different sub-areas for image shooting, and then the images of different areas are input into the target density recognition model respectively. The final crowd density can be summed to more accurately determine the total crowd density in the target area.

[0102] After the image is acquired, the original image needs to be preprocessed, including grayscale, noise reduction, correction of perspective distortion, etc., to optimize the effect of subsequent processing. This step may also include image enhancement to adapt to different lighting conditions and perspective changes.

[0103] S800: Input multiple image data into the target density recognition model respectively to generate a crowd density sub-prediction value and sub-confidence corresponding to each image data. The target density recognition model includes a YOLO target detection model and a CNN density prediction model.

[0104] In the target density recognition model of this embodiment, it is first necessary to identify and locate the pixels in the image that include the crowd. That is, crowd detection usually uses target detection algorithms such as YOLO (You Only Look Once) and SSD (Single Shot MultiBox Detector) to preliminarily identify the crowd area from the image. These algorithms are based on deep learning models and can quickly locate pedestrians or other targets of interest in the image.

[0105] Then, according to the crowd pixels in the image, the corresponding density map is predicted and generated. In order to estimate the number of people more accurately, convolutional neural networks (CNNs) are often used to generate crowd density maps. For example, CSRNet (Convolutional Spatial Pyramid Network) learns high-dimensional features and generates crowd density values ​​corresponding to each pixel, and then integrates to get the total number of people. There are also similar methods such as P2PNet and MCNN (Multi-column Convolutional Neural Network), which focus on improving the accuracy of density estimation.

[0106] Specifically, the training sample set used can be a collection of images or videos containing different crowd densities, lighting conditions, viewing angles, and occlusion conditions in the corresponding area as a training data set. To label these data, it is usually necessary to mark the location of each visible pedestrian, and sometimes a crowd density map is directly generated, in which the value of each pixel represents the number of pedestrians near the location, or an existing training set is used.

[0107] S900: Generate the third area crowd density ρ3 and confidence δ corresponding to the target area according to the crowd density sub-prediction value and sub-confidence corresponding to each image data. ρ3 and δ meet the following conditions: ; .

[0108] ρ i and δ i are the sub-prediction value and sub-confidence of the crowd density corresponding to the i-th image data. z is the total number of image data.

[0109] Since images may be blocked by light and obstacles during image capture, and the deep learning model ultimately obtains a job prediction value, the average crowd density and average confidence of multiple prediction results are obtained in this embodiment as the final usage value to improve the accuracy of the calculation.

[0110] S1000: Generate the crowd density ρ in the current target area based on ρ1, ρ2, ρ3 and δ. ρ satisfies the following conditions:

[0111] ρ=δ×ρ3+(1-δ)×[(φ×ρ1+ρ2) / 2].

[0112] S600: Generate a mobile control instruction based on ρ and the regional population density threshold Y.

[0113] In addition, during the calculation, confidence is used as the adjustment weight to fuse ρ1, ρ2, and ρ3. In some periods of weak light, such as in the evening or in scenes where there are many people holding umbrellas, the confidence of the results obtained by the model will be greatly reduced. Therefore, when fusing the results, the fusion ratio of each part of the results will be dynamically adjusted according to the size of δ, thereby improving the accuracy of the final result. In this embodiment, the crowd density result of the image recognition dimension is added to further adjust the accuracy of the prediction result to improve the final accuracy.

[0114] As another possible embodiment of the present invention, Figure 2 As shown, a non-contact image recognition mobile control system for an intelligent temporary platform is provided, and the system includes:

[0115] The first acquisition module is used to acquire a first identification set A of communication devices collected by multiple WiFi probes set at the entrance of the target area.

[0116] The second acquisition module is used to acquire a second identification set B of communication devices collected by multiple WiFi probes set at the exit of the target area. Each device corresponds to a device identification in the identification set. The device identification is generated by the Mac address of the communication device.

[0117] The first generating module is used to generate a first area population density ρ1 corresponding to the target area according to the difference between the first identification set and the second identification set. ρ1 satisfies the following condition: ρ1=A−(A∩B).

[0118] The second generating module is used to determine the second area crowd density ρ2 corresponding to the target area according to the positioning information of the communication terminal by the communication base station.

[0119] The fusion module is used to generate the crowd density ρ in the current target area based on ρ1, ρ2 and the probe compensation coefficient φ. φ∈(0,2), ρ satisfies the following conditions:

[0120] ρ=(φ×ρ1+ρ2) / 2.

[0121] The command output module is used to generate mobile control commands based on ρ and the regional crowd density threshold Y.

[0122] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0123] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by combining software with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0124] In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.

[0125] It will be appreciated by those skilled in the art that various aspects of the present invention may be implemented as a system, method or program product. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as a "circuit", "module" or "system".

[0126] The electronic device according to this embodiment of the present invention is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0127] The electronic device is presented in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: the at least one processor mentioned above, the at least one storage device mentioned above, and a bus connecting different system components (including storage devices and processors).

[0128] The storage stores program codes, which can be executed by the processor, so that the processor executes the steps according to various exemplary embodiments of the present invention described in the above “Exemplary Method” section of this specification.

[0129] The memory may include readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory, and may further include read only memory (ROM).

[0130] The storage may also include a program / utility having a set (at least one) of program modules, such program modules including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0131] The bus may represent one or more of several types of bus structures including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures.

[0132] The electronic device may also communicate with one or more external devices (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may communicate with one or more devices that enable a user to interact with the electronic device, and / or may communicate with any device that enables the electronic device to communicate with one or more other computing devices (e.g., routers, modems, etc.). This communication may be performed through an input / output (I / O) interface. In addition, the electronic device may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter. The network adapter communicates with other modules of the electronic device through a bus. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0133] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0134] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present invention may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary embodiments of the present invention described in the above "Exemplary Method" section of the present specification.

[0135] The program product may adopt any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0136] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0137] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0138] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0139] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to an exemplary embodiment of the present invention, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0140] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0141] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A non-contact image recognition mobile control method for an intelligent temporary platform, characterized in that: The method comprises the following steps: Acquire a first identification set A of communication devices collected by a plurality of WiFi probes disposed at the entrance of the target area; Obtain a second identification set B of communication devices collected by multiple WiFi probes set at the exit of the target area; each device corresponds to a device identification in the identification set; The device identification is generated by the Mac address of the communication device; Generate a first area population density ρ1 corresponding to the target area according to the difference between the first identification set and the second identification set; ρ1 satisfies the following condition: ρ1=A−(A∩B); Determine the second area crowd density ρ2 corresponding to the target area according to the positioning information of the communication terminal by the communication base station; Acquire image data corresponding to multiple angles in the target area; Inputting multiple image data into a target density recognition model respectively to generate a crowd density sub-prediction value and a sub-confidence corresponding to each image data; the target density recognition model includes a YOLO target detection model and a CNN density prediction model; According to the crowd density sub-prediction value and sub-confidence corresponding to each image data, the third area crowd density ρ3 and confidence δ corresponding to the target area are generated; ρ3 and δ satisfy the following conditions: ； ； ρ i and δ i are the sub-prediction value and sub-confidence of the crowd density corresponding to the i-th image data; z is the total number of image data; According to ρ1, ρ2, ρ3, δ and the probe compensation coefficient φ, the crowd density ρ in the current target area is generated; φ∈(0, 2), ρ satisfies the following conditions: ρ=δ×ρ3+(1-δ)×[(φ×ρ1+ρ2) / 2]; Generate mobile control instructions based on ρ and the regional population density threshold Y.

2. The method according to claim 1, characterized in that The first identification set A or the second identification set B is obtained according to the following steps: Get the probe request frame captured by each WiFi probe; Hash the device Mac address information included in each probe request frame to generate an address hash value corresponding to each device Mac address; Match each address hash value with the existing address hash values ​​in the identification set; If the match is successful, the current address hash value is discarded; If the match fails, the current address hash value is added to the corresponding identification set.

3. The method according to claim 1, characterized in that Determining the second area crowd density ρ2 corresponding to the target area according to the positioning information of the communication terminal by the communication base station includes: Obtain a list of devices respectively obtained from multiple target communication base stations during a collection period; the list of devices includes communication device identification, base station coordinates and communication distance; the number of the target communication base stations is greater than or equal to 3; the target communication base station is a communication base station whose signal range completely covers the target area; Based on the communication device identification, an intersection list of multiple device lists is obtained; each communication device identification in the intersection list corresponds to an information pair consisting of at least three base station coordinates and communication distances; Determine the coordinate information corresponding to each communication device identifier according to the information pair consisting of the base station coordinates and the communication distance corresponding to each communication device identifier; Compare the coordinate information corresponding to each communication device identifier with the coordinate domains [Jz, Jy] and [Ws, Wx] corresponding to the target area; Jz and Jy are respectively the longitude boundary values ​​corresponding to the target area; Ws and Wx are respectively the latitude boundary values ​​corresponding to the target area; If J s n ∈[Jz,Jy] and W s n ∈[Ws, Wx], it is determined that the communication terminal corresponding to the nth communication device identifier is located in the target area; J s n and W s n The longitude and latitude of the communication terminal corresponding to the n-th communication device identifier are respectively.

4. The method according to claim 3, characterized in that According to the information pair consisting of the base station coordinates and the communication distance corresponding to each communication device identifier, the coordinate information corresponding to each communication device identifier is determined, including: The three-circle positioning method is used to determine the coordinate information corresponding to each communication device identifier based on the information pair consisting of the base station coordinates and the communication distance corresponding to each communication device identifier.

5. The method according to claim 3, characterized in that: The communication distance is obtained according to the following steps: Obtaining the sending timestamp and receiving timestamp of the information sent by the communication device received by the communication base station; The communication distance is generated based on the time difference between the sending timestamp and the receiving timestamp and the speed of light.

6. The method according to claim 1, characterized in that After obtaining a first identification set A of communication devices collected by a plurality of WiFi probes arranged at the entrance of the target area, the method further includes: Obtaining a supplementary sub-identity set of communication devices collected by multiple supplementary WiFi probes in the target area; The union of the multiple supplementary sub-identifier sets and the first identifier set is used as a new first identifier set.

7. The method according to claim 6, characterized in that Multiple supplementary WiFi probes are placed in areas where people must pass.

8. The method according to claim 3, characterized in that The duration of the collection period is greater than or equal to 10 minutes.

9. An intelligent temporary platform non-contact image recognition mobile control system, characterized in that: The system comprises: A first acquisition module is used to acquire a first identification set A of communication devices collected by a plurality of WiFi probes arranged at the entrance of the target area; A second acquisition module is used to acquire a second identification set B of communication devices collected by multiple WiFi probes set at the exit of the target area; each device corresponds to a device identification in the identification set; the device identification is generated by the Mac address of the communication device; A first generating module is used to generate a first area population density ρ1 corresponding to the target area according to a difference set between the first identification set and the second identification set; ρ1 satisfies the following condition: ρ1=A−(A∩B); The second generating module is used to determine the second area crowd density ρ2 corresponding to the target area according to the positioning information of the communication terminal by the communication base station; A third acquisition module is used to acquire image data corresponding to multiple angles in the target area; The third generation module is used to input the multiple image data into the target density recognition model respectively to generate the crowd density sub-prediction value and sub-confidence corresponding to each image data; the target density recognition model includes the YOLO target detection model and the CNN density prediction model; The fourth generation module is used to generate the third area crowd density ρ3 and confidence δ corresponding to the target area according to the crowd density sub-prediction value and sub-confidence corresponding to each image data; ρ3 and δ meet the following conditions: ； ； ρ i and δ i are the sub-prediction value and sub-confidence of the crowd density corresponding to the i-th image data; z is the total number of image data; The fusion module is used to generate the crowd density ρ in the current target area according to ρ1, ρ2, ρ3, δ and the probe compensation coefficient φ; φ∈(0,2), ρ satisfies the following conditions: ρ=δ×ρ3+(1-δ)×[(φ×ρ1+ρ2) / 2]; The command output module is used to generate mobile control commands based on ρ and the regional crowd density threshold Y.

Images