Multi-dimensional identification analysis method and device for dynamic floating population in region

By combining access control systems, RFID, GPS and video streaming information, the number of migrant populations is automatically identified and counted, and the problem of unregistered and high-risk personnel entering in the existing technology is solved, and the accuracy and security of migrant population management are improved.

CN118886567APending Publication Date: 2024-11-01BEIJING GZT NETWORK TECH +1
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202411155714.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-22
Publication Date
2024-11-01

AI Technical Summary

Technical Problem

When managing and monitoring migrant populations, it is difficult to prevent unregistered personnel and high-risk personnel from entering the area to be predicted through abnormal means, and it is impossible to effectively identify and manage migrant populations.

Method used

The multi-dimensional recognition and analysis method is adopted to automatically identify and count the number of migrant population through access control systems, RFID information, GPS position information, vibration sensors and video stream information, reduce the demand for active registration of migrant populations and improve monitoring accuracy.

Benefits of technology

Implicit identification and management of migrant populations is realized, avoiding missed and privacy disputes, and improving the security and management efficiency of the areas to be predicted.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118886567A_ABST
    Figure CN118886567A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-dimensional identification analysis method and device for dynamic floating population in a region, and relates to the technical field of population data statistics. The multi-dimensional identification and analysis method for the dynamic floating population in the area comprises the following steps: acquiring an access control movement period in a preset time period; obtaining the number of people entering the to-be-predicted area from the entrance position in the entrance guard motion period; acquiring identification quantity information of each RFID in an access control motion period; acquiring each piece of GPS position information in each access control motion period; and obtaining a first floating population number according to the RFID identification number information in each access control motion period, the GPS position information of each handheld terminal and the number of people entering the to-be-predicted area corresponding to the access control motion period. According to the method, the number of the non-floating persons entering the to-be-predicted area and the total number of the persons entering the to-be-predicted area are obtained through the RFID information and the GPS information to determine the number of the floating population, and the missing situation caused by the fact that the floating persons do not register can be prevented.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of population data statistics, and in particular to a multi-dimensional identification and analysis method for dynamic floating population in a region and a multi-dimensional identification and analysis device for dynamic floating population in a region. Background Art

[0002] In real life, pedestrians usually need to enter industrial parks, smart communities, etc. through access control systems. Generally, there are several types of people entering industrial parks and smart communities (hereinafter referred to as the predicted area):

[0003] (1) People who are inside the area to be predicted usually hold access cards corresponding to the access control system or built-in access cards in their mobile phones, and can enter and exit freely using access cards.

[0004] (2) People who are doing business or visiting others. These people are not considered as internal personnel of the area to be predicted, but they have specific business to do when entering the area to be predicted. When these people enter the area to be predicted, some will be brought into the area by internal personnel of the area to be predicted using their access control cards, some will enter the area by registering, and some will directly follow other internal personnel into the area if the management of the area to be predicted is relatively loose (generally, after the gate of the access control system is opened, if it detects that there are multiple people walking, in order to prevent the gate from hurting pedestrians who want to pass through when it closes, it is generally not closed forcibly, but waits until all pedestrians have left before closing).

[0005] (3) High-risk personnel, that is, they may intend to enter the predicted area for a special purpose. Such personnel generally enter by following other internal personnel, or by passing through the surrounding walls of the predicted area (e.g. Figure 3 Enter by climbing over the wall (as shown).

[0006] The existing technology for the management and monitoring of the area to be predicted usually adopts a registration method, that is, registering the floating population (i.e., people who are not permanent residents of the area to be predicted), registering their faces, and then allowing these floating population to enter the area to be predicted.

[0007] The disadvantages of this type of monitoring are as follows:

[0008] (1) Migrants are required to register, but many do not register at all, making it impossible to manage and identify them.

[0009] (2) It is impossible to prevent high-risk groups from entering through abnormal means (such as climbing over the firewall).

[0010] Therefore, it is desired to have a technical solution to overcome or at least alleviate at least one of the above-mentioned defects of the prior art.

[0011] Application Contents

[0012] The purpose of this application is to provide a multi-dimensional identification and analysis method for dynamic floating population in a region to overcome or at least alleviate at least one of the above-mentioned defects of the prior art.

[0013] To achieve the above objectives, the present application provides a multi-dimensional identification and analysis method for the dynamic floating population in a region, and the multi-dimensional identification and analysis method for the dynamic floating population in a region includes:

[0014] Obtaining the access control movement cycle of the access control system at the entrance position of the area to be predicted within a preset time period;

[0015] Get the number of people entering the predicted area from the entrance position in each access control movement cycle;

[0016] Obtain the RFID identification quantity information transmitted by the access control system during each access control movement cycle;

[0017] Obtain the GPS location information of each handheld terminal within each access control movement cycle;

[0018] The first floating population quantity of each access control movement cycle is obtained according to the RFID identification quantity information in each access control movement cycle, the GPS location information of each handheld terminal and the number of people entering the predicted area corresponding to the access control movement cycle.

[0019] Optionally, the multi-dimensional identification and analysis method for the dynamic floating population in the region further includes:

[0020] Respectively obtain vibration information transmitted by each vibration sensor set on the community fence within a preset time period;

[0021] Respectively obtain video stream information transmitted by the camera devices around the community fence within a preset time period;

[0022] The number of the second floating population within a preset time period is obtained based on the video stream information transmitted by the camera devices around the community fence and / or the vibration information transmitted by each vibration sensor.

[0023] Optionally, obtaining the number of people entering the area to be predicted from the entrance position in each access control movement cycle includes:

[0024] Obtain video stream information of the entrance position of each access control movement cycle respectively;

[0025] Each video stream information is identified, and the number of people entering the area to be predicted in each video stream information is identified.

[0026] Optionally, obtaining the first floating population number in each access control movement cycle according to the RFID identification number information in each access control movement cycle, the GPS location information of each handheld terminal, and the number of people entering the predicted area corresponding to the access control movement cycle includes:

[0027] Get the preset coordinate area;

[0028] Obtain the movement trajectory of each handheld terminal in each access control movement cycle within the preset coordinate area with GPS location information;

[0029] According to the movement trajectory of each handheld terminal, the number of people entering the predicted area in each access control movement cycle is obtained, and the number of people is the number of people entering with a handheld terminal in the corresponding access control movement cycle;

[0030] The RFID identification quantity information in each access control movement cycle is used as the number of RFID-entered personnel in the corresponding access control movement cycle;

[0031] Obtaining a correspondence table between RFID and handheld terminals, wherein the correspondence table between RFID and handheld terminals includes RFID identity codes and handheld terminal identity codes, and one RFID identity code corresponds to one handheld terminal identity code;

[0032] Deduplication is performed on the number of persons entering via the handheld terminal and the number of persons entering via the RFID according to the corresponding table of the RFID and the handheld terminal;

[0033] The sum of the number of people entering through handheld terminals and the number of people entering through RFID in the same access control movement cycle after deduplication is obtained as the first number of non-mobile people in the access control movement cycle;

[0034] The number of people who enter the predicted area from the entrance position in the same access control movement cycle is subtracted from the number of the first non-mobile people to obtain the first mobile population number.

[0035] Optionally, obtaining the access control movement cycle of the access control system at the entrance position in the area to be predicted within a preset time period includes:

[0036] Obtaining an opening signal of the access control system and recording a time point at which the opening signal is generated as a first time point;

[0037] Obtaining a closing signal from the access control system and recording the time point at which the closing signal is generated as a second time point;

[0038] in,

[0039] The time period between the first time point and the second time point is a door control movement cycle.

[0040] Optionally, the obtaining of the number of the second floating population within a preset time period according to the video stream information transmitted by the camera device around the community fence and / or the vibration information transmitted by each vibration sensor includes:

[0041] Obtain the camera device number and vibration sensor number associated with each community fence area respectively;

[0042] Proceed as follows for each community fenced area:

[0043] Obtaining image information transmitted by a camera device corresponding to a camera device number associated with the community fence area;

[0044] Obtaining vibration information transmitted by a vibration sensor corresponding to a vibration sensor number associated with the community fence area;

[0045] The number of floating population in the fenced area of ​​the community within a preset time period is obtained according to the image information and / or the vibration information; wherein,

[0046] The number of floating population in each of the community fenced areas constitutes the second number of floating population.

[0047] Optionally, the acquiring the number of floating population in the fenced area of ​​the community within a preset time period according to the image information and the vibration information includes:

[0048] Acquire vibration characteristics according to the vibration information;

[0049] Acquire image features according to the image information;

[0050] fusing the vibration feature with the image feature to obtain a fused feature;

[0051] Obtain a trained foreign body discrimination model;

[0052] The fused features are input into the foreign object discrimination model to obtain the type of the foreign object. When the type of the foreign object is a person, the foreign object is recorded as a floating population in the community fence area.

[0053] Optionally, obtaining the number of floating population in the fenced area of ​​the community within a preset time period according to the image information includes:

[0054] Each frame of image information is processed as follows:

[0055] Acquire a trained repair neural network, wherein the repair neural network includes an encoder and a decoder, and the decoder includes a multi-stage fusion module;

[0056] Inputting the image information into the encoder to extract image features;

[0057] Inputting the image features into the decoder to obtain a restored image;

[0058] Identify the repaired image and determine whether there is a human figure in the image. If so, determine that there is floating population; wherein the sum of the number of each floating population in the recognition results of the image information of each frame is the number of floating population in the fenced area of ​​the community.

[0059] The present application also provides a multi-dimensional identification and analysis device for dynamic floating population in a region, and the multi-dimensional identification and analysis device for dynamic floating population in a region comprises:

[0060] A door control movement cycle acquisition module, which is used to acquire the door control movement cycle of the door control system at the entrance position of the area to be predicted within a preset time period;

[0061] A personnel number acquisition module, which is used to obtain the number of personnel entering the area to be predicted from the entrance position in each access control movement cycle;

[0062] An RFID identification quantity acquisition module, the RFID identification quantity acquisition module is used to obtain the RFID identification quantity information transmitted by the access control system in each access control movement cycle;

[0063] A GPS location information acquisition module, which is used to acquire GPS location information of each handheld terminal within each access control movement cycle;

[0064] The first floating population number acquisition module is used to obtain the first floating population number in each access control movement cycle based on the RFID identification number information in each access control movement cycle, the GPS location information of each handheld terminal, and the number of people entering the predicted area corresponding to the access control movement cycle.

[0065] The multi-dimensional identification and analysis method for dynamic floating population in the region of the present application determines the number of floating population by obtaining the number of non-mobile personnel entering the area to be predicted and the total number of personnel entering the area to be predicted through RFID information and GPS information. With this method, the floating population does not need to actively register and report, but is recorded without their knowledge. On the one hand, it prevents the floating population from missing out due to their failure to register, and on the other hand, it prevents disputes caused by floating population believing that their privacy has been violated. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a flow chart of a multi-dimensional identification and analysis method for dynamic floating population in a region according to an embodiment of the present application.

[0067] Figure 2It is a schematic diagram of the structure of an electronic device in one embodiment of the present application.

[0068] Figure 3 It is a schematic diagram of a community fence in one embodiment of the present application. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical scheme and advantages of the implementation of this application clearer, the technical scheme in the embodiment of this application will be described in more detail below in conjunction with the drawings in the embodiment of this application. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of this application, not all of them. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain this application, and should not be construed as limitations on this application. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. The embodiments of this application are described in detail below in conjunction with the drawings.

[0070] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the scope of protection of the present application.

[0071] Figure 1 It is a flow chart of a multi-dimensional identification and analysis method for dynamic floating population in a region according to an embodiment of the present application.

[0072] like Figure 1 The multi-dimensional identification and analysis methods for the dynamic floating population in the region shown include:

[0073] Step 1: Obtain the access control movement cycle of the access control system at the entrance position in the area to be predicted within a preset time period.

[0074] Step 2: Obtain the number of people entering the predicted area from the entrance position in each access control movement cycle;

[0075] Step 3: Obtain the RFID identification quantity information transmitted by the access control system in each access control movement cycle;

[0076] Step 4: Obtain the GPS location information of each handheld terminal within each access control movement cycle;

[0077] Step 5: Obtain the first floating population quantity in each access control movement cycle according to the RFID identification quantity information in each access control movement cycle, the GPS location information of each handheld terminal, and the number of people entering the predicted area corresponding to the access control movement cycle.

[0078] The multi-dimensional identification and analysis method for dynamic floating population in the region of the present application determines the number of floating population by obtaining the number of non-mobile personnel entering the area to be predicted and the total number of personnel entering the area to be predicted through RFID information and GPS information. With this method, the floating population does not need to actively register and report, but is recorded without their knowledge. On the one hand, it prevents the floating population from missing out due to their failure to register, and on the other hand, it prevents disputes caused by floating population believing that their privacy has been violated.

[0079] In this embodiment, the preset time period is the monitoring period. For example, if you want to know the number of floating population within 24 hours, the preset time period is 24 hours. If you want to know the number of floating population within 1 hour, the preset time period is 1 hour.

[0080] In this embodiment, the access control system refers to an interactive system that requires a person to enter the predicted area by swiping a card or other verification methods. Generally, the access control system includes a gate and a controller that controls the gate. The controller and the gate communicate with each other, and the gate can be opened and closed by the control of the controller. It can be understood that the access control system belongs to the prior art and will not be described in detail here.

[0081] In this embodiment, obtaining the access control movement cycle of the access control system at the entrance position in the area to be predicted within a preset time period includes:

[0082] Obtaining an opening signal of the access control system and recording a time point at which the opening signal is generated as a first time point;

[0083] Obtaining a closing signal from the access control system and recording the time point at which the closing signal is generated as a second time point;

[0084] in,

[0085] The time period between the first time point and the second time point is a door control movement cycle.

[0086] For example, after the access control system receives an opening signal, the gate will open. At this time, pedestrians will walk from outside the predicted area into the predicted area (the gate of the present application is a one-way gate, that is, only entry but not exit, and other gates are required to leave the predicted area). Each time, only one person passes through the gate (the gates in the prior art are generally small, and there is no situation where two people enter at the same time). When the access control system detects that there is no one around, the gate will close and give the access control system a closing signal.

[0087] In order to prevent multiple people from queuing to enter, the gate will usually continue to detect whether there are other people entering after it is opened (for example, through obstacle detection methods or image recognition methods, which are prior art and will not be described here). Therefore, the time from when the access control system obtains the opening signal to when the access control system obtains the closing signal is the access control movement cycle.

[0088] For example, if an open signal is obtained at 12:01:10 on a certain day, and a close signal is obtained at 12:01:50, then the door movement cycle is 40 seconds. It is understandable that the open signal and the close signal in a door movement cycle of the present application are two adjacent signals, that is, there will be no other open signal or close signal between the open signal and the close signal. For example, the close signal is obtained at 12:01:50, that is, there is no other open signal or close signal between 12:01:50 and 12:01:10.

[0089] In this embodiment, an RFID receiving device is provided at the gate, and the RFID receiving device is used to receive RFID signals. The RFID sending device can be provided on a physical access control card.

[0090] In this embodiment, the handheld terminal may be a user's mobile phone. For example, the user's mobile phone has downloaded an APP for managing the area to be predicted. At this time, the GPS location information of the user's mobile phone may be obtained.

[0091] It is understandable that if it is a science and technology park, the science and technology park will generally distribute physical access cards. At this time, a GPS receiving device can also be set on the access card to obtain GPS location information.

[0092] In this embodiment, the dynamic floating population in the area of ​​the present application includes two parts. One part is the floating personnel entering from the gate of the access control system. Such floating personnel may enter by following other authorized persons or may be registered. The other part is crossing or climbing over the community fence. See Figure 3 Since the height of most community fences is not high, and in order to prevent injuries to people who climb over the fences, sharp objects are usually not set on the fences. Instead, some steel wire products are used for protection. However, steel wire products usually cannot completely prevent mobile people from climbing over.

[0093] Therefore, in this embodiment, the multi-dimensional identification and analysis method of the regional dynamic floating population of the present application further includes:

[0094] Respectively obtain vibration information transmitted by each vibration sensor set on the community fence within a preset time period;

[0095] Respectively obtain video stream information transmitted by the camera devices around the community fence within a preset time period;

[0096] The number of the second floating population within a preset time period is obtained based on the video stream information transmitted by the camera devices around the community fence and / or the vibration information transmitted by each vibration sensor.

[0097] By installing vibration sensors on the community fence, once the fence vibrates significantly, or when image recognition is performed on the video stream information to identify a human figure, it can be considered that there are migrant workers trying to climb over the community fence.

[0098] In this embodiment, obtaining the number of people entering the area to be predicted from the entrance position in each access control movement cycle includes:

[0099] Obtain video stream information of the entrance position of each access control movement cycle respectively;

[0100] Each video stream information is identified, and the number of people entering the area to be predicted in each video stream information is identified.

[0101] The following is an example of a method for identifying the number of people through a video stream. In this embodiment, there is no need to obtain the identity information of the entering person, and therefore, there is no need to use a frontal face shooting method, which can effectively protect the privacy of each entering person.

[0102] Therefore, the present application can use thermal imaging recognition technology to obtain thermal imaging technology for quantitative statistics, for example, it can be obtained by the following method:

[0103] Extract multiple image frames at every preset time interval from the video stream information. For example, if a door control motion cycle is 10 seconds, then it will be used for 10 seconds of continuous video stream information. However, it takes time for a person to walk, that is, the time it takes for a person to walk from one side of an image to the other side must be greater than the continuous multiple frames of image frames. For example, if there is one image frame every 1 second, then two adjacent frames of image frames may capture the same person twice. In this case, it is actually unnecessary to identify each image frame, which will waste recognition computing power. Therefore, a preset time interval can be set. For example, assuming that in a scene, the door control motion cycle is 10 seconds, and there are 20 image frames in the video stream information in 10 seconds, that is, one frame every 0.5 seconds. At this time, if these 20 frames are all image recognized, it is obviously a waste of computing power. Therefore, a preset time interval can be set, such as 2 seconds, that is, one frame can be taken at 0 seconds, one frame at 2 seconds, and one frame at 4 seconds, and so on. Or take a frame at 0.5 seconds, take a frame at 2.5 seconds, and so on.

[0104] It is understandable that the preset time interval can be set as needed.

[0105] After acquiring multiple image frames at a preset time interval, each image is processed as follows:

[0106] A multi-scale adaptive crowd detection model is used to identify and detect the tasks of each image to obtain the number of people in the image.

[0107] At this point, the number of people in each image can be obtained.

[0108] In this embodiment, in order to prevent each image frame from being unclear, each image frame may be repaired after being extracted and before being recognized.

[0109] Specifically, in this embodiment, each frame of image information is processed as follows:

[0110] Acquire a trained repair neural network, wherein the repair neural network includes an encoder and a decoder, and the decoder includes a multi-stage fusion module;

[0111] Inputting the image information into the encoder to extract image features;

[0112] Inputting the image features into the decoder to obtain a restored image;

[0113] Identify the repaired image and determine whether there is a human figure in the image. If so, determine that there is floating population; wherein the sum of the number of each floating population in the recognition results of the image information of each frame is the number of floating population in the fenced area of ​​the community.

[0114] In this embodiment, the multi-level fusion module includes:

[0115] A pooling and splicing module, wherein the pooling and splicing module is used to perform pooling and splicing on the image features provided by the encoder to form a first feature;

[0116] A first processing module, wherein the first processing module is used to perform a first processing on the first feature, so as to obtain a second feature;

[0117] a second processing module, the second processing module being used for performing a second processing on the first feature, thereby forming a third feature;

[0118] a third processing module, configured to add the second feature to the third feature and perform a convolution operation to obtain a fourth feature;

[0119] A linear transformation module, wherein the linear transformation module is used to perform a linear transformation on the fourth feature, thereby obtaining a linear transformation feature;

[0120] An activation feature module, wherein the activation feature module is used to perform Sigmoid activation on the fourth feature, thereby obtaining an activation feature;

[0121] A multiplication module, wherein the multiplication module is used to multiply the linear transformation feature and the activation feature to obtain a fifth feature;

[0122] An adding module, wherein the adding module is used to add the fifth feature to the fourth feature, so as to obtain a final fusion feature.

[0123] In this embodiment, the first processing module includes:

[0124] An average pooling module, wherein the average pooling module is used to perform average pooling on the first feature, thereby obtaining an average pooling feature;

[0125] A maximum pooling module, wherein the maximum pooling module is used to perform maximum pooling on the first feature, thereby obtaining a maximum pooling feature;

[0126] A first splicing module, wherein the first splicing module is used to splice the average pooling feature and the maximum pooling feature to obtain a first splicing feature;

[0127] A position attention mechanism module, wherein the position attention mechanism module is used to perform a position self-attention operation on the first splicing feature;

[0128] An upsampling module is used to upsample the first splicing feature that performs the position self-attention operation, so as to obtain a second feature.

[0129] In this embodiment, the second processing module includes:

[0130] an anti-overflow module, the anti-overflow module being used to perform an anti-overflow process on the first feature;

[0131] A channel attention mechanism module, wherein the channel attention mechanism module is used to perform a channel self-attention operation on the first feature that has been processed with anti-overflow, so as to obtain a third feature.

[0132] In this way, clearer image frames can be obtained.

[0133] In this embodiment, even after each image frame is acquired at a preset time interval, it is possible that a person may appear in two image frames at the same time. For example, suppose pedestrian A appears on the left side of the screen in an image frame at one time point (for example, an image frame at 0.5 seconds), and may appear on the right side of the screen in an image frame at the next time point (for example, an image frame at 2.5 seconds). At this time, if no elimination is performed, the number of pedestrians in the image frame at 0.5 seconds already includes pedestrian A, and the number of pedestrians in the image frame at 2.5 seconds also includes pedestrian A, which leads to the problem of duplicate counting.

[0134] In this embodiment, the extracted image frames are referred to as to-be-recognized image frames for the convenience of description.

[0135] Therefore, the present application further includes:

[0136] Each image frame to be identified is processed as follows:

[0137] Extracting the identification area of ​​each person in the image frame to be identified;

[0138] The similarity values ​​of each recognition area of ​​the image frame to be recognized are calculated respectively with the recognition areas of the image frame to be recognized that is closest to the image frame to be recognized in time, and it is determined whether the similarity value of one recognition area exceeds a preset threshold. If so, the two recognition areas whose similarity values ​​exceed the preset threshold are determined to be the same recognition areas.

[0139] For example, suppose there are three human figures on the image frame A to be identified (the image frame acquired at 0.5 seconds), namely human figure 1, human figure 2 and human figure 3, and the image frame B to be identified (the image frame acquired at 2.5 seconds) is the image frame to be identified that is closest in time to the image frame to be identified and has three human figures on it, namely human figure 3, human figure 4 and human figure 5.

[0140] Extract the identification area of ​​each person in the image frame to be identified. The identification area refers to the wearing area of ​​the human figure (i.e., the area of ​​the picture occupied by the clothes and pants of the person). The area can be obtained by a feature extraction method (such as a CNN network).

[0141] The recognition area of ​​human figure 1 is respectively calculated to have similarity with the recognition areas of human figure 3, human figure 4, and human figure 5 to obtain similarity values ​​respectively, the recognition area of ​​human figure 2 is respectively calculated to have similarity with the recognition areas of human figure 3, human figure 4, and human figure 5 to obtain similarity values ​​respectively, and the recognition area of ​​human figure 3 is respectively calculated to have similarity with the recognition areas of human figure 3, human figure 4, and human figure 5 to obtain similarity values ​​respectively. It can be understood that the similarity calculation of two images is a prior art and will not be described in detail here.

[0142] Since it is unlikely that most people who appear at the same time are wearing exactly the same clothes, by using image recognition to compare the similarity of each figure's clothing, it is possible to find out whether the two images are of the same person. In this way, when there is the same person in two images, the number of people can be counted only once, thus preventing duplicate counting.

[0143] Taking the above-mentioned image frame A to be identified and image frame B to be identified as an example, through the above-mentioned similarity calculation, it may appear that the similarity value between the recognition area of ​​the human figure 3 in the image frame A to be identified and the recognition area of ​​the human figure 3 in the image frame B to be identified exceeds the preset threshold, so only the human figure 3 in one of the images is selected for personnel counting. At this time, there are a total of 4 people in the image frame A to be identified and the image frame B to be identified, that is, the number of people in the area to be predicted is 4.

[0144] In this embodiment, obtaining the first floating population number according to the RFID identification number information in each access control movement cycle, the GPS location information of each handheld terminal, and the number of people entering the predicted area corresponding to the access control movement cycle includes:

[0145] Get the preset coordinate area;

[0146] Obtain the movement trajectory of each handheld terminal in each access control movement cycle within the preset coordinate area using GPS location information;

[0147] According to the movement trajectory of each handheld terminal, the number of people entering the predicted area in each access control movement cycle is obtained, and the number of people is the number of people entering with a handheld terminal in the corresponding access control movement cycle;

[0148] The RFID identification quantity information in each access control movement cycle is used as the number of RFID-entered personnel in the corresponding access control movement cycle;

[0149] Obtaining a correspondence table between RFID and handheld terminals, wherein the correspondence table between RFID and handheld terminals includes RFID identity codes and handheld terminal identity codes, and one RFID identity code corresponds to one handheld terminal identity code;

[0150] Deduplication is performed on the number of persons entering via the handheld terminal and the number of persons entering via the RFID according to the corresponding table of the RFID and the handheld terminal;

[0151] The sum of the number of people entering through handheld terminals and the number of people entering through RFID in the same access control movement cycle after deduplication is obtained as the first number of non-mobile people in the access control movement cycle;

[0152] The number of people who enter the predicted area from the entrance position in the same access control movement cycle is subtracted from the number of the first non-mobile people to obtain the first mobile population number.

[0153] Take an access control movement cycle as an example. Suppose that within the access control movement cycle, there are passers-by A, B, C and D. Among them, passer-by A has a handheld terminal, which can provide GPS location information, passer-by B has an RFID transmitter, and passer-by C has both a handheld terminal and an RFID transmitter. Passer-by D is not a mobile person, that is, he has nothing.

[0154] The tracks of passerby A and passerby C can be obtained by tracking the GPS location information. In this embodiment, the GPS location information can be obtained once every preset time, for example, once every 2 seconds, and the time can be set as needed.

[0155] In this embodiment, the location of the gate is preset in the system, that is, through the GPS location information and the location of the gate each time, it can be known whether the holder of the handheld terminal enters the area to be predicted.

[0156] It is understandable that it is very likely that a user carries a handheld terminal and an RFID transmitter at the same time. In this case, in order to prevent duplicate counting, deduplication processing is performed through the RFID and handheld terminal correspondence table.

[0157] For example, the RFID and handheld terminal correspondence table includes the RFID identity code and the handheld terminal identity code. The RFID identity code can be obtained when the RFID transmitting device sends it to the RFID receiving device, and the handheld terminal identity code can be obtained in the form of a short message. When it is found by searching the RFID and handheld terminal correspondence table that there is a person (such as the passerby C mentioned above) whose RFID identity code and handheld terminal identity code are in correspondence in the RFID and handheld terminal correspondence table, then at this time, the number of passerby C only needs to be counted once.

[0158] In the above embodiment, the number of people entering with handheld terminals is 2 (passerby A and passerby C), and the number of people entering with RFID is 2 (passerby B and passerby C). Through deduplication processing, the number of people entering with handheld terminals or the number of people entering with RFID are randomly removed, and the sum of the number of people entering with handheld terminals in the same access control movement cycle after deduplication (if the number of people entering with handheld terminals is removed, the number is 1) and the number of people entering with RFID (the number is 2) is taken as the first number of non-mobile people in the access control movement cycle, that is, the first number of non-mobile people is 3.

[0159] Through the above-mentioned image recognition method, it can be identified that the number of people entering the area to be predicted is 4 (passerby A, passerby B, passerby C and passerby D), and the difference between the two is 1, so the number of the first floating population is 1.

[0160] In this embodiment, the method of obtaining the number of the second floating population within a preset time period according to the video stream information transmitted by the camera device around the community fence and / or the vibration information transmitted by each vibration sensor includes:

[0161] Obtain the camera device number and vibration sensor number associated with each community fence area respectively;

[0162] Proceed as follows for each community fenced area:

[0163] Obtaining image information transmitted by a camera device corresponding to a camera device number associated with the community fence area;

[0164] Obtaining vibration information transmitted by a vibration sensor corresponding to a vibration sensor number associated with the community fence area;

[0165] Acquire the number of floating population in the fenced area of ​​the community within a preset time period according to the image information and / or the vibration information;

[0166] The number of floating population in each of the community fenced areas constitutes the second number of floating population.

[0167] In this embodiment, the number of floating population in the fenced area of ​​the community within a preset time period can be obtained according to the image information and / or the vibration information in the following selection methods:

[0168] (1) Only use image information to obtain the number of floating population in the community fence area. The disadvantage of this method is that if the image clarity is not enough or the camera device is blocked, the recognition accuracy will be relatively low.

[0169] (2) Only vibration information is used to obtain the number of floating population in the fenced area of ​​the community within a preset time period. The disadvantage of this method is that if there is strong wind or other weather conditions, the vibration frequency and amplitude may be large, resulting in low judgment accuracy.

[0170] (3) Using the image information and the vibration information simultaneously, the number of floating population in the fenced area of ​​the community within a preset time period is obtained.

[0171] In this embodiment, only the image information is used to obtain the number of floating population in the community fence area, and image recognition is used to identify whether there are people on the community fence.

[0172] In this embodiment, the method of obtaining the number of floating population in the fenced area of ​​the community within a preset time period using only vibration information only requires setting a vibration frequency threshold and a vibration amplitude threshold, because the vibration amplitude threshold and vibration frequency threshold transmitted when a person climbs over must be different from those in situations such as animal interference. Therefore, screening can be performed by setting a vibration amplitude threshold and a vibration frequency threshold.

[0173] In this embodiment, obtaining the number of floating population in the fenced area of ​​the community within a preset time period according to the image information and the vibration information includes:

[0174] Acquire vibration characteristics according to the vibration information;

[0175] Acquire image features according to the image information;

[0176] fusing the vibration feature with the image feature to obtain a fused feature;

[0177] Obtain a trained foreign body discrimination model;

[0178] The fusion feature is input into the foreign body discrimination model to obtain the type of foreign body. When the type of foreign body is human, the number of floating population in the community fence area is recorded.

[0179] For example, recognition can be achieved by fusing vibration features and image features, which can take into account both frequency domain changes and time domain changes, making recognition more accurate.

[0180] In this embodiment, both vibration features and image features can be extracted through a Vision Transformer (ViT) model.

[0181] In this embodiment, the foreign body identification model may select a random forest model for classification.

[0182] The present application also provides a multi-dimensional identification and analysis device for dynamic floating population in a region, wherein the multi-dimensional identification and analysis device for dynamic floating population in a region comprises a door control movement cycle acquisition module, a personnel quantity acquisition module, an RFID identification quantity acquisition module, a GPS location information acquisition module, and a first floating population quantity acquisition module, wherein:

[0183] The access control movement cycle acquisition module is used to obtain the access control movement cycle of the access control system at the entrance position of the area to be predicted within a preset time period;

[0184] The personnel quantity acquisition module is used to obtain the number of personnel entering the predicted area from the entrance position in each access control movement cycle;

[0185] The RFID identification quantity acquisition module is used to obtain the RFID identification quantity information transmitted by the access control system in each access control movement cycle;

[0186] The GPS location information acquisition module is used to obtain the GPS location information of each handheld terminal within each access control movement cycle;

[0187] The first floating population number acquisition module is used to obtain the first floating population number in each access control movement cycle based on the RFID identification number information in each access control movement cycle, the GPS location information of each handheld terminal, and the number of people entering the predicted area corresponding to the access control movement cycle.

[0188] It should be noted that the above explanations of the method embodiment are also applicable to the device of this embodiment and will not be repeated here.

[0189] The present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the computer program, it implements the above-mentioned multi-dimensional identification and analysis method for dynamic floating population in a region.

[0190] The present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it can implement the above-mentioned multi-dimensional identification and analysis method for dynamic floating population in a region.

[0191] Figure 2 It is an exemplary structural diagram of an electronic device capable of implementing a multi-dimensional identification and analysis method for dynamic floating population in a region provided according to an embodiment of the present application.

[0192] Figure 2 The electronic device shown can also be implemented as comprising: a memory storing computer executable instructions; and one or more processors, which can implement the combination when executing the computer executable instructions. Figure 1 A multi-dimensional identification and analysis method for the dynamic floating population in a region is described.

[0193] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0194] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0195] Computer readable media include permanent and non-permanent, removable and non-removable, and the media can be implemented by any method or technology to store information. The information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), data versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0196] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0197] In addition, it is obvious that the word "comprising" does not exclude other units or steps. Multiple units, modules or devices stated in the device claim can also be implemented by one unit or the overall device through software or hardware.

[0198] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and a module, a program segment or a part of a code includes one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes identified in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or the total flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0199] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0200] In addition, it is obvious that the word "comprising" does not exclude other units or steps. Multiple units, modules or devices stated in the device claim can also be implemented by one unit or the overall device through software or hardware.

[0201] Although the present application is disclosed as above in terms of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

[0202] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application is described in detail with reference to the above embodiments, a person skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A multi-dimensional identification and analysis method for dynamic floating population in a region, characterized in that: The multi-dimensional identification and analysis method of the dynamic floating population in the region includes: Obtaining the access control movement cycle of the access control system at the entrance position of the area to be predicted within a preset time period; Get the number of people entering the predicted area from the entrance position in each access control movement cycle; Obtain the RFID identification quantity information transmitted by the access control system during each access control movement cycle; Obtain the GPS location information of each handheld terminal within each access control movement cycle; The first floating population quantity of each access control movement cycle is obtained according to the RFID identification quantity information in each access control movement cycle, the GPS location information of each handheld terminal and the number of people entering the predicted area corresponding to the access control movement cycle.

2. The multi-dimensional identification and analysis method for dynamic floating population in a region as claimed in claim 1, characterized in that: The multi-dimensional identification and analysis method for the dynamic floating population in the region further includes: Respectively obtain vibration information transmitted by each vibration sensor set on the community fence within a preset time period; Respectively obtain video stream information transmitted by the camera devices around the community fence within a preset time period; The number of the second floating population within a preset time period is obtained based on the video stream information transmitted by the camera devices around the community fence and / or the vibration information transmitted by each vibration sensor.

3. The multi-dimensional identification and analysis method for dynamic floating population in a region as claimed in claim 2, characterized in that: The method of obtaining the number of people entering the predicted area from the entrance position in each access control movement cycle includes: Obtain video stream information of the entrance position of each access control movement cycle respectively; Each video stream information is identified, and the number of people entering the area to be predicted in each video stream information is identified.

4. The multi-dimensional identification and analysis method for dynamic floating population in a region as claimed in claim 3, characterized in that: The method of obtaining the first floating population number in each access control movement cycle according to the RFID identification number information in each access control movement cycle, the GPS location information of each handheld terminal, and the number of people entering the predicted area corresponding to the access control movement cycle includes: Get the preset coordinate area; Obtain the movement trajectory of each handheld terminal in each access control movement cycle within the preset coordinate area with GPS location information; According to the movement trajectory of each handheld terminal, the number of people entering the predicted area in each access control movement cycle is obtained, and the number of people is the number of people entering with a handheld terminal in the corresponding access control movement cycle; The RFID identification quantity information in each access control movement cycle is used as the number of RFID-entered personnel in the corresponding access control movement cycle; Obtaining a correspondence table between RFID and handheld terminals, wherein the correspondence table between RFID and handheld terminals includes RFID identity codes and handheld terminal identity codes, and one RFID identity code corresponds to one handheld terminal identity code; Deduplication is performed on the number of persons entering via the handheld terminal and the number of persons entering via the RFID according to the corresponding table of the RFID and the handheld terminal; The sum of the number of people entering through handheld terminals and the number of people entering through RFID in the same access control movement cycle after deduplication is obtained as the first number of non-mobile people in the access control movement cycle; The number of people who enter the predicted area from the entrance position in the same access control movement cycle is subtracted from the number of the first non-mobile people to obtain the first mobile population number.

5. The multi-dimensional identification and analysis method for dynamic floating population in a region as claimed in claim 4, characterized in that: The method of obtaining the access control movement cycle of the access control system at the entrance position of the area to be predicted within the preset time period includes: Obtaining an opening signal of the access control system and recording a time point at which the opening signal is generated as a first time point; Obtain a closing signal from the access control system and record the time point at which the closing signal is generated as a second time point; wherein, The time period between the first time point and the second time point is a door control movement cycle.

6. The multi-dimensional identification and analysis method for dynamic floating population in a region as claimed in claim 5, characterized in that: The method of obtaining the number of the second floating population within a preset time period according to the video stream information transmitted by the camera device around the community fence and / or the vibration information transmitted by each vibration sensor includes: Obtain the camera device number and vibration sensor number associated with each community fence area respectively; Proceed as follows for each community fenced area: Obtaining image information transmitted by a camera device corresponding to a camera device number associated with the community fence area; Obtaining vibration information transmitted by a vibration sensor corresponding to a vibration sensor number associated with the community fence area; The number of floating population in the fenced area of ​​the community within a preset time period is obtained according to the image information and / or the vibration information; wherein, The number of floating population in each of the community fenced areas constitutes the second number of floating population.

7. The multi-dimensional identification and analysis method for dynamic floating population in a region as claimed in claim 6, characterized in that: The step of obtaining the number of floating population in the fenced area of ​​the community within a preset time period according to the image information and the vibration information includes: Acquire vibration characteristics according to the vibration information; Acquire image features according to the image information; fusing the vibration feature with the image feature to obtain a fused feature; Obtain a trained foreign body discrimination model; The fused features are input into the foreign object discrimination model to obtain the type of the foreign object. When the type of the foreign object is a person, the foreign object is recorded as a floating population in the community fence area.

8. The multi-dimensional identification and analysis method for dynamic floating population in a region as claimed in claim 6, characterized in that: The step of obtaining the number of floating population in the fenced area of ​​the community within a preset time period according to the image information includes: Each frame of image information is processed as follows: Acquire a trained repair neural network, wherein the repair neural network includes an encoder and a decoder, and the decoder includes a multi-stage fusion module; Inputting the image information into the encoder to extract image features; Inputting the image features into the decoder to obtain a restored image; Identify the repaired image and determine whether there is a human figure in the image. If so, determine that there is floating population; wherein the sum of the number of each floating population in the recognition results of the image information of each frame is the number of floating population in the fenced area of ​​the community.

9. A multi-dimensional identification and analysis device for dynamic floating population in a region, characterized in that: The multi-dimensional identification and analysis device for the dynamic floating population in the region comprises: A door control movement cycle acquisition module, which is used to acquire the door control movement cycle of the door control system at the entrance position of the area to be predicted within a preset time period; A personnel number acquisition module, which is used to obtain the number of personnel entering the area to be predicted from the entrance position in each access control movement cycle; An RFID identification quantity acquisition module, the RFID identification quantity acquisition module is used to obtain the RFID identification quantity information transmitted by the access control system in each access control movement cycle; A GPS location information acquisition module, which is used to acquire GPS location information of each handheld terminal within each access control movement cycle; The first floating population number acquisition module is used to obtain the first floating population number in each access control movement cycle based on the RFID identification number information in each access control movement cycle, the GPS location information of each handheld terminal, and the number of people entering the predicted area corresponding to the access control movement cycle.

Citation Information

Patent Citations

  • System and method for quickly positioning newly-added floating population in field of community treatment

    CN111028118A

  • Specific area personnel access management method and system based on wireless radio frequency

    CN113988231A

  • Personnel multi-dimensional data management method

    CN116311452A