A method and electronic device for identifying a tailgate

Abstract

The present application provides a method and electronic device for identifying tailgating through a gate, the method comprising: obtaining data to be processed, the data to be processed comprising: point cloud data within the gate channel range; extracting feature point data of multiple target objects from the point cloud data; judging whether the feature point data of multiple target objects are successfully matched with a preset human body shape feature model; if so, determining the distance between the multiple target objects according to the point cloud data, and determining the tailgating identification result within the gate channel range according to the distance between the multiple target objects. In the implementation process of the above scheme, the feature point data of multiple target objects extracted from the point cloud data within the gate channel range and the matching result of the preset human body shape feature model are combined with the matching result and the actual distance in three-dimensional space between the multiple target objects, so that the tailgating behavior can be more accurately identified, and the accuracy of tailgating identification is effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of pattern recognition and computer vision, and in particular to a method and electronic device for identifying a tailgate. Background Art

[0002] At the current airport security checkpoints or boarding gates, passengers often pass through in rapid succession, especially during peak hours. Traditional gates usually use infrared sensors to detect whether each confirmed passenger has passed through the gate. However, in actual operation, sometimes two passengers are very close to each other (for example, less than 10 cm). At this time, traditional infrared sensors are difficult to distinguish whether they are two independent individuals or one person, resulting in the occurrence of tailgating. Therefore, the accuracy of tailgating identification using infrared sensors is currently low. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a method and electronic device for identifying a tailgate, so as to improve the problem of low accuracy of tailgate identification.

[0004] The embodiment of the present application provides a method for identifying a tailgating gate, comprising: obtaining data to be processed, the data to be processed comprising: point cloud data within the gate channel range; extracting feature point data of multiple target objects from the point cloud data; judging whether the feature point data of multiple target objects are successfully matched with a preset human body shape feature model; if so, determining the distance between the multiple target objects according to the point cloud data, and determining the tailgating identification result within the gate channel range according to the distance between the multiple target objects. In the implementation process of the above scheme, by extracting the feature point data of multiple target objects from the point cloud data within the gate channel range, key body contour information can be captured, which is essential for identifying human body shape. Therefore, through the matching results of the feature point data of multiple target objects and the preset human body shape feature model, the recognition error caused by factors such as viewing angle changes and lighting conditions can be reduced, and further combined with the matching results and the actual distance in three-dimensional space between the multiple target objects, the tailgating behavior can be more accurately identified, effectively improving the accuracy of tailgating identification.

[0005] Optionally, in an embodiment of the present application, the tailgating recognition result within the gate channel range is determined based on the distance between multiple target objects, including: if the distance between two target objects among the multiple target objects is less than a preset distance threshold, the tailgating recognition result is determined as the presence of a tailgating phenomenon. In the implementation process of the above scheme, by accurately measuring the distance between the target objects and setting a reasonable preset distance threshold, the judgment method based on the distance and the threshold effectively reduces the false alarms caused by environmental changes (such as lighting, occlusion, etc.), and can effectively distinguish between normal passage and tailgating behavior, thereby improving the accuracy of recognition. Furthermore, it is not limited to the case of two people, but can be expanded to tailgating detection of more people, and is suitable for more complex scenarios. By adjusting the preset distance threshold, it can adapt to human bodies of different body shapes and postures, and the preset distance threshold can be dynamically adjusted according to different application scenarios and needs to adapt to different environmental conditions and safety requirements, thereby improving the flexibility of the system.

[0006] Optionally, in an embodiment of the present application, multiple target objects include: a first target object and a second target object; determining the tailgating recognition result within the gate channel range according to the distance between the multiple target objects includes: determining the height of the first target object and the height of the second target object according to the point cloud data; if the distance between the first target object and the second target object is less than a preset distance threshold, and the height of the first target object and the height of the second target object are both greater than a preset height threshold, then the tailgating recognition result is determined as the presence of a tailgating phenomenon. In the implementation process of the above scheme, by combining the information of the two dimensions of distance and height for comprehensive judgment, the tailing behavior can be more accurately identified. The distance is used to detect the relative position between the target objects, and the height is used to filter out non-human targets, which effectively improves the reliability of the system. Furthermore, the method is not limited to the case of two people, but can be extended to tailgating detection of more people, and is suitable for more complex scenes. By adjusting the preset threshold, it can adapt to people of different body shapes and postures, thereby improving the flexibility and scope of application of the system.

[0007] Optionally, in an embodiment of the present application, after determining the height of the first target object and the height of the second target object according to the point cloud data, it also includes: judging whether the height of the first target object and the height of the second target object are both greater than a preset height threshold; if not, sending a prohibition closing command or a delay closing command to the gate, the prohibition closing command is used to prohibit the closing of the leaf door of the gate within a preset safety time, and the delay closing command is used to delay the closing of the leaf door of the gate after a preset waiting time based on the preset safety time. In the implementation process of the above scheme, by sending a prohibition closing command or a delay closing command to the gate according to the height of the target object, misjudgment caused by luggage and children can be avoided, thereby preventing the leaf door from closing in inappropriate circumstances and reducing potential damage to personnel and objects. The preset height threshold can effectively filter out targets below the threshold, ensuring that only objects that meet the height of the human body can trigger tailing recognition, protecting the safety of children, thereby improving the safety of children when passing through the gate. Further, through the dual verification of distance and height, underreporting can be reduced, and the point cloud data can cover the entire gate channel range to ensure that there are no blind spots, thereby improving the comprehensiveness and accuracy of detection.

[0008] Optionally, in an embodiment of the present application, the data to be processed also includes: video data within the range of the gate channel; after determining whether the feature point data of multiple target objects are successfully matched with the preset human shape feature model, it also includes: if the feature point data of each target object in the multiple target objects are not successfully matched with the preset human shape feature model, then human body recognition is performed on the frame image in the video data to obtain a human body recognition result, and the tailing recognition result within the range of the gate channel is determined according to the human body recognition result. In the implementation process of the above scheme, the tailing behavior can be more accurately identified through the dual verification of point cloud data and video data. Even if the point cloud data fails to match successfully, the image information in the video data can provide additional confirmation, and a comprehensive judgment is made in combination with multiple data sources. Since the point cloud data provides three-dimensional spatial information, and the video data provides two-dimensional visual information, the combination of the two can better describe the scene, reduce false alarms and omissions caused by a single data source, and improve the reliability of the system.

[0009] Optionally, in an embodiment of the present application, the tailgating recognition result within the gate channel range is determined based on the human body recognition result, including: judging whether the human body recognition result meets preset conditions, the preset conditions include: the number of human bodies in the human body recognition result is multiple human bodies, and the distance between two human bodies among the multiple human bodies is less than a preset distance threshold; if so, the tailgating recognition result is determined as the existence of a tailgating phenomenon. In the implementation process of the above scheme, through the dual verification of point cloud data and video data, even in complex environments (such as dense crowds, changes in lighting, etc.), real-time monitoring can be achieved, while maintaining high robustness and accuracy. In addition, through contactless recognition by lidar and camera, users do not need additional operations, which improves the travel efficiency and user experience, and combines point cloud data and video data to form a multimodal fusion system to further improve the accuracy and reliability of recognition.

[0010] Optionally, in an embodiment of the present application, after obtaining the human body recognition result, it also includes: if the number of human bodies in the human body recognition result is one human body, the body thickness or body width of the human body is determined based on the point cloud data, and when the body thickness is greater than a preset thickness threshold or the body width is greater than a preset width threshold, an abnormal warning is issued. In the implementation process of the above scheme, by combining video data and point cloud data for comprehensive judgment, not only can the tailing phenomenon be detected, but also abnormal body size can be detected. Even in the case of a single human body, the system can detect potential abnormalities through the detection of body thickness or width, reducing the possibility of missed reports, thereby improving the comprehensiveness and accuracy of the system. Furthermore, by detecting abnormal body thickness or width, attempts to evade detection by disguise (such as wearing heavy clothing, carrying large items, etc.) can be effectively identified, thereby improving the security of the system.

[0011] Optionally, in an embodiment of the present application, the data to be processed also includes: infrared sensor data within a preset area of ​​the gate channel range; after determining the tailgating recognition result within the gate channel range according to the distance between multiple target objects, it also includes: if the tailgating recognition result cannot be obtained, then when the infrared sensor data is blocked by two obstacles at the same time, an abnormal warning is issued. In the implementation process of the above scheme, by combining point cloud data, video data and infrared sensor data for comprehensive judgment, not only can the tailgating phenomenon be detected, but also other abnormal situations can be detected, which improves the comprehensiveness and accuracy of the system. Even when a clear tailgating recognition result cannot be obtained through point cloud data and video data, infrared sensor data can be used as an additional verification method to reduce the possibility of missed reports.

[0012] Optionally, in an embodiment of the present application, after determining the tailing identification result within the gate channel range according to the distances between multiple target objects, it also includes: if the tailing identification result is that there is a tailing phenomenon, a closing command is sent to the gate, and the closing command is used to close the leaf door of the gate immediately, or to keep the leaf door of the gate closed. In the implementation process of the above scheme, once the tailing phenomenon is detected, the system can take immediate action to quickly close the leaf door of the gate to prevent unauthorized personnel from entering the controlled area, and by closing the gate immediately, the tailgating person can be effectively prevented from entering, reducing potential safety risks and safety hazards.

[0013] The embodiment of the present application also provides a device for identifying tailgating through a gate, including: a detection module and a control module, the detection module and the control module communicate with each other; the detection module is used to obtain data to be processed, and the data to be processed includes: point cloud data within the gate channel range; the control module is used to extract feature point data of multiple target objects from the point cloud data; the control module is also used to determine whether the feature point data of multiple target objects are successfully matched with a preset human body shape feature model; the control module is also used to determine the distance between the multiple target objects according to the point cloud data if the feature point data of the multiple target objects are successfully matched with the preset human body shape feature model, and determine the tailgating identification result within the gate channel range according to the distance between the multiple target objects.

[0014] Optionally, in the embodiment of the present application, the control module is further configured to determine the tailing recognition result as the presence of a tailing phenomenon if the distance between two target objects among the multiple target objects is less than a preset distance threshold.

[0015] Optionally, in an embodiment of the present application, multiple target objects include: a first target object and a second target object; the control module is also used to determine the height of the first target object and the height of the second target object based on the point cloud data; if the distance between the first target object and the second target object is less than a preset distance threshold, and the height of the first target object and the height of the second target object are both greater than a preset height threshold, the tailing recognition result is determined as the presence of a tailing phenomenon.

[0016] Optionally, in an embodiment of the present application, the control module is also used to determine whether the height of the first target object and the height of the second target object are both greater than a preset height threshold; if not, a prohibit closing command or a delay closing command is sent to the gate machine, and the prohibit closing command is used to prohibit the fan door of the gate machine from closing within a preset safety time period, and the delay closing command is used to delay the closing of the fan door of the gate machine after a preset waiting time on the basis of a preset safety time period.

[0017] Optionally, in an embodiment of the present application, the data to be processed also includes: video data within the gate channel range; a control module, which is also used to perform human body recognition on the frame image in the video data if the feature point data of each target object among multiple target objects does not successfully match the preset human body shape feature model, to obtain the human body recognition result, and determine the trailing recognition result within the gate channel range based on the human body recognition result.

[0018] Optionally, in an embodiment of the present application, the control module is also used to determine whether the human body recognition result meets preset conditions, and the preset conditions include: the number of human bodies in the human body recognition result is multiple human bodies, and the distance between two human bodies among the multiple human bodies is less than a preset distance threshold; if the human body recognition result meets the preset conditions, the tailing recognition result is determined as the existence of a tailing phenomenon.

[0019] Optionally, in an embodiment of the present application, the control module is also used to determine the body thickness or body width of the human body based on the point cloud data if the number of human bodies in the human body recognition result is one human body, and to issue an abnormal warning when the body thickness is greater than a preset thickness threshold or the body width is greater than a preset width threshold.

[0020] Optionally, in an embodiment of the present application, the data to be processed also includes: infrared sensor data within a preset area of ​​the gate channel range; the control module is also used to issue an abnormal warning if the tailgating recognition result cannot be obtained and the infrared sensor data is blocked by two obstacles at the same time.

[0021] Optionally, in an embodiment of the present application, the control module is also used to send a closing command to the gate if the tailgating identification result shows that a tailgating phenomenon exists. The closing command is used to close the fan door of the gate immediately, or to keep the fan door of the gate in a closed state.

[0022] Optionally, in an embodiment of the present application, the detection module includes: a laser radar module and / or a visual camera module; the laser radar module is used to scan point cloud data within the gate channel range, and the visual camera module is used to collect video data within the gate channel range.

[0023] An embodiment of the present application further provides an electronic device, including: a processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the machine-readable instructions execute the method described above when executed by the processor.

[0024] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method described above is executed.

[0025] The embodiment of the present application also provides a computer program product, including: a computer program or a computer instruction, and the computer program or the computer instruction executes the method described above when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0027] Figure 1 A schematic diagram of an application scenario of a tailgate identification system provided by an embodiment of the present application is shown;

[0028] Figure 2 A schematic diagram of the process of a method for identifying a tailgate provided by an embodiment of the present application is shown;

[0029] Figure 3 A schematic diagram showing the zoning of the gate channel range provided in an embodiment of the present application is shown;

[0030] Figure 4 A schematic diagram of two tailgating recognition results provided by an embodiment of the present application is shown;

[0031] Figure 5 A schematic diagram showing the height measurement of a target object in point cloud data provided by an embodiment of the present application is shown;

[0032] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown.

[0033] Icons: 100 - tailgate identification system; 110 - detection module; 120 - control module; 130 - gate; 300 - electronic device; 310 - processor; 320 - memory; 330 - computer-readable storage medium. DETAILED DESCRIPTION

[0034] To make the purpose, technical scheme and advantages of the embodiment of the present application clearer, the technical scheme in the embodiment of the present application will be clearly and completely described below in conjunction with the drawings in the embodiment of the present application. It should be understood that the drawings in the embodiment of the present application only serve the purpose of explanation and description, and are not used to limit the protection scope of the embodiment of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in the embodiment of the present application shows the operation implemented according to some embodiments of the embodiment of the present application. It should be understood that the operation of the flowchart can be implemented out of order, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the embodiment of the present application, or remove one or more operations from the flowchart.

[0035] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all embodiments. The components of the embodiments of the present application described and shown in the drawings generally here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the embodiments of the present application claimed, but merely represents the selected embodiments of the embodiments of the present application.

[0036] It is understandable that the "first" and "second" in the embodiments of the present application are used to distinguish similar objects. Those skilled in the art will understand that the words "first" and "second" do not limit the quantity and execution order, and the words "first" and "second" do not necessarily limit the difference. In the description of the embodiments of the present application, the term "and / or" is merely a description of the association relationship of the associated objects, indicating that there may be three relationships, such as A and / or B, which can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the related objects before and after are in an "or" relationship. The term "multiple" refers to more than two (including two), and similarly, "multiple groups" refers to more than two groups (including two groups).

[0037] It should be noted that the tailgate identification method provided in the embodiment of the present application can be executed by an electronic device, where the electronic device refers to a device terminal having the function of executing a computer program or the above-mentioned server, such as a smart phone, a personal computer, a tablet computer, a personal digital assistant or a mobile Internet device. A server refers to a device that provides computing services through a network, such as an x86 server and a non-x86 server, and the non-x86 server includes a mainframe, a minicomputer and a UNIX server.

[0038] Before introducing the tailgate identification method provided by the embodiment of the present application, the tailgate identification method can be applied to the airport boarding inspection scenario. The requirement of this inspection scenario is "one person, one verification, one pass", that is, each time a passenger is verified, the passenger can enter the gate channel. The passengers in the airport boarding inspection scenario are diverse, including adults, children, and the elderly, and the safety of passengers needs to be guaranteed while passing quickly. However, the layout of traditional infrared counter-radiation sensors in the gate chassis is difficult to fully cover the gate channel range, and there is usually a gap of 10CM to 15CM between two groups of infrared counter-radiation sensors. Therefore, when the distance between two adults is less than 10CM and they are closely following each other through the gate channel, the traditional system will think that an obstacle greater than 10CM blocks the two groups of infrared counter-radiation sensors, and it is difficult to distinguish between the situation where two adults are closely following each other.

[0039] See also Figure 1 A schematic diagram of an application scenario of a tailgating gate recognition system provided by an embodiment of the present application is shown; in response to the above-mentioned problems, an embodiment of the present application provides a tailgating gate recognition system 100, which may include: a detection module 110, a control module 120 and a gate 130; wherein the above-mentioned control module 120 may be used to execute the following tailgating gate recognition method, and the control module 120 communicates with the detection module 110 and the gate 130 respectively, and may communicate through a wired network or a wireless network of a local area network, or may communicate through a wide area network.

[0040] The above-mentioned detection module 110 can be set on the top of the gate, also known as the top visual detection module, so as to collect point cloud data and / or video data within the gate channel range from the top to the bottom. Specifically, the above-mentioned detection module may include: a Lidar module and / or a visual camera module, wherein the Lidar module may adopt a solid-state Lidar (Flash Lidar) and / or a rotating Lidar, and the Lidar module is used to scan the three-dimensional point cloud data within the gate channel range, and may specifically include: a laser transmitter and a laser receiver, and the laser transmitter is used to emit laser pulses, and the laser receiver receives the reflected signal to measure the distance and position of the target. The above-mentioned visual camera module is used to collect video data within the gate channel range through an image sensor, and the image sensor may be a depth CMOS image chip (Depth CMOS Image Sensor), which is a special image sensor that can capture two-dimensional image data and three-dimensional depth information at the same time, and specifically measures the distance of each pixel by calculating the time of flight (Time ofFlight, ToF) of the laser pulse. In addition, the detection module can also include an embedded AI chip to process the raw data obtained from the deep CMOS image chip and perform advanced image processing and artificial intelligence algorithms to efficiently and accurately obtain point cloud data and video data within the gate channel range.

[0041] The control module 120 can be used to execute the various steps and control logic in the following tailgate identification method. The control module can be in many forms, such as a control chip, a server, or a personal computer host. The control module can be stored in many places. The server stored in the computer room can be used as the control module, the control chip set inside the gate can be used as the control module, and the personal computer between the detection module and the gate can be used as the control module.

[0042] The gate 130 can receive and execute commands sent by the control module, and control the leaf doors on the gate according to the commands. Specifically, the leaf doors on the gate can be controlled by the application software SafeAisle running on the gate. In addition, other modules can be set on the gate, such as a speaker module, a camera model (for pass authentication of face recognition), a multi-function document reader (for reading the authentication information of the second-generation ID card) and an information verification display screen.

[0043] It can be understood that since the detection module collects point cloud data and / or video data within the gate channel range from the top to the bottom, the control module can complete functions such as trailing gate identification through point cloud data and / or video data. In this case, the number of infrared counter-radiation sensors set in the gate and the overall length of the gate chassis can be reduced, thereby reducing the volume of the overall hardware equipment, reducing the number of hardware equipment that needs to be maintained, and improving the maintenance efficiency of the hardware equipment.

[0044] See also Figure 2 The flowchart of the method for identifying a tailgate provided by an embodiment of the present application is shown; the embodiment of the present application provides a method for identifying a tailgate, including:

[0045] Step S210: Obtaining data to be processed, where the data to be processed includes: point cloud data within the gate channel range.

[0046] The data to be processed refers to data used to identify the phenomenon of tailgating through a gate, and the data may include: point cloud data within the gate channel range, and optionally, may also include video data within the gate channel range.

[0047] See also Figure 3 The schematic diagram of the zoning of the gate channel range provided by the embodiment of the present application is shown; the gate channel range mentioned above may include: a detection area, a monitoring area, a safety area and a departure area. Among them, the detection area is used to detect the stay area for authentication identity, and an infrared sensor can be set in the detection area. If the infrared sensor in the detection area is blocked by a person, the leaf door of the gate machine will remain in its original state, for example, if it is originally in an open and closed state, it cannot be closed, and if it is originally in a closed state, it cannot be opened. The monitoring area is used for the area monitored by the infrared sensor, and the length of the monitoring area can be adjusted according to the specific situation. The safety area is used for the area where the leaf door is opened and closed, and the departure area is used for the area specifically for passengers to leave. The four areas divided by the gate channel range mentioned above can all be processed by the horizon coordinate layer in the underlying algorithm module. Specifically, each area can be divided into a preset number of independent coordinates according to 5×5 pixels (for example, the departure area can be divided into 12,000 independent coordinates), each independent coordinate has a unique identifier (ID), and all independent coordinates can be freely combined, adjusted boundaries, judged and decided by the executable program according to time. For each of the above four areas, data to be processed can be collected. Point cloud data or video data of the area can be collected, or point cloud data and video data of the area can be collected. There are three situations in total.

[0048] The implementation method of the above step S210 is, for example: collecting the data to be processed through the detection module in the above-mentioned trailing gate recognition system, and collecting point cloud data within the gate channel range, or collecting point cloud data and video data within the gate channel range.

[0049] Step S220: extracting feature point data of multiple target objects from the point cloud data.

[0050] The implementation method of the above step S220 is, for example: after collecting the point cloud data within the gate channel range, a dynamic 3D map within the gate channel range can also be constructed based on the point cloud data, and feature point data of multiple target objects can be extracted from the point cloud data of the 3D map. The feature point data of the above target objects include: appearance feature point data, behavior feature point data, height feature points, motion trajectory coordinate points, and carry-on baggage feature points, etc.

[0051] Step S230: determining whether the feature point data of the plurality of target objects are successfully matched with the preset human body shape feature model.

[0052] The preset human body shape feature model is a mathematical or geometric model used to describe and identify typical human body shapes, and is used to describe the geometric shape and structural features of a standard human body.

[0053] It is understandable that the above-mentioned preset human body shape feature model can be defined and constructed in a variety of ways, including: the first construction method is to construct the model based on geometric features, for example: first, define the basic outline of the human body, including the head, torso, limbs and other parts, then define the proportional relationship between the various parts of the body, such as the head-to-body ratio, shoulder width to height ratio, etc., and finally, define the position and relative distance of key joints, such as the head, neck, shoulders, elbows, wrists, hips, knees, ankles, etc. The second construction method is to build a model based on statistics. The system extracts the feature points of the target object from the point cloud data and projects them into the principal component space. For example: first, a large number of points of human body are collected, including data of different body shapes, postures and viewing angles. Then, a set of feature points is extracted from each set of point cloud data, such as head, shoulder, elbow, wrist, hip, knee, ankle, etc., to obtain multiple sets of feature points; then, principal component analysis is performed on all the extracted feature points to obtain principal component vectors. These principal component vectors represent the main change directions of the human body shape. Finally, the principal component vectors extracted above are used to construct a human body shape feature model.

[0054] The implementation method of the above step S230 is, for example: the control module determines whether the feature point data of multiple target objects are successfully matched with the preset human body shape feature model. If they are all successfully matched, it means that there are multiple human bodies. In this case, there is a high probability that there is a tailing phenomenon or an abnormal phenomenon. It can be understood that when the feature point data of the target object matches the preset human body shape feature model, it can be determined that the target object is a human body. Then, the human body height of the target object can be determined based on the feature point data of the target object, so as to effectively distinguish whether it is an adult or a child based on the human body height.

[0055] Step S240: If the feature point data of multiple target objects are successfully matched with the preset human body shape feature model, the distance between the multiple target objects is determined according to the point cloud data, and the tailing recognition result within the gate channel range is determined according to the distance between the multiple target objects.

[0056] See also Figure 4 A schematic diagram of two types of tailgating recognition results provided in an embodiment of the present application is shown; the above-mentioned tailgating recognition results may include: side-by-side tailgating and hugging tailgating, etc.; when the feature point data of the target object matches the preset human body shape feature model, it means that the multiple target objects are all human bodies. Therefore, the above-mentioned side-by-side tailgating refers to the first human body and the second human body among the multiple human bodies passing through the gate channel shoulder to shoulder (for example, two passengers walking side by side), and hugging tailgating refers to the first human body and the second human body among the multiple human bodies passing through the gate channel with their chests pressed against their backs (for example, two passengers walking closely).

[0057] In the implementation process of the above scheme, by extracting the feature point data of multiple target objects from the point cloud data within the gate channel range, key body contour information can be captured, which is essential for identifying the human body shape. Therefore, through the matching results of the feature point data of multiple target objects with the preset human body shape feature model, the recognition error caused by factors such as viewing angle changes and lighting conditions can be reduced. Further combining the matching results with the actual distance in three-dimensional space between multiple target objects, the tailing behavior can be identified more accurately, effectively improving the accuracy of tailing identification.

[0058] As an optional implementation of the above step S240, the above implementation of determining the tailgating recognition result within the gate channel range according to the distance between multiple target objects may include:

[0059] Step S241: if the distance between two target objects among the multiple target objects is less than a preset distance threshold, the tailing recognition result is determined as the existence of a tailing phenomenon.

[0060] An implementation example of the above-mentioned step S241 is: when the feature point data of the target object matches the preset human body shape feature model, it means that the target object is a human body. Furthermore, if the distance between any two human bodies among the multiple human bodies is less than a preset distance threshold (for example, 10 cm or 12 cm, etc.), the tailgating recognition result is determined as the existence of tailgating phenomena such as side by side tailgating or hugging tailgating.

[0061] As an optional implementation of the above step S240, the above multiple target objects may include: a first target object and a second target object; the above implementation of determining the tailgating recognition result within the gate channel range according to the distance between the multiple target objects may include:

[0062] Step S242: Determine the height of the first target object and the height of the second target object according to the point cloud data.

[0063] It can be understood that when the feature point data of the target object matches the preset human body shape feature model, it means that the target object is a human body. The height of the human body is further determined based on the point cloud data, so that adults and children can be accurately distinguished to ensure the safe passage of child passengers.

[0064] See also Figure 5 The schematic diagram of the height measurement of the target object in the point cloud data provided by the embodiment of the present application is shown; the implementation method of the above step S242 is, for example: the height of the first target object in the point cloud data and the height of the second target object are calculated by the time-of-flight (ToF) technology of the laser pulse. The height of the first target object can be obtained by directly measuring the first target object in the point cloud data, or by subtracting the distance between the detection module and the ground from the distance between the detection module and the first target object. Similarly, the height of the second target object in the point cloud data can be obtained by subtracting the distance between the detection module and the ground from the distance between the detection module and the second target object.

[0065] Step S243: If the distance between the first target object and the second target object is less than the preset distance threshold, and the height of the first target object and the height of the second target object are both greater than the preset height threshold, the tailing recognition result is determined as the presence of a tailing phenomenon.

[0066] For example, the implementation of the above step S243 is as follows: when the feature point data of the target object matches the preset human body shape feature model, it indicates that the target object is a human body. Furthermore, if the distance between the first human body and the second human body is less than the preset distance threshold (such as 10 cm or 12 cm), it indicates that the distance between the first human body and the second human body is too close, which may be a normal passage of an adult holding a child or a situation where two adults are trailing each other. In combination with the fact that the height of the first human body and the height of the second human body are both greater than the preset height threshold (such as 1 meter, 1.1 meters or 1.2 meters), it is determined that both human bodies are adults, and the trailing recognition result can be determined to be trailing side by side or trailing with hugging. Therefore, if the distance between the first human body and the second human body is less than the preset distance threshold (such as 10 cm or 12 cm), and the height of the first human body and the height of the second human body are both greater than the preset height threshold (such as both are greater than 1.2 meters, indicating that both human bodies are adults), the trailing recognition result is determined to be the existence of the trailing phenomenon.

[0067] As an optional implementation of the above step S240, after determining the height of the first target object and the height of the second target object according to the point cloud data, the following may also be included:

[0068] Step S244: Determine whether the height of the first target object and the height of the second target object are both greater than a preset height threshold.

[0069] Step S245: If the height of the first target object or the height of the second target object is not greater than the preset height threshold, a prohibit closing command or a delay closing command is sent to the gate. The prohibit closing command is used to prohibit the closing of the leaf door of the gate within a preset safety time period, and the delay closing command is used to delay the closing of the leaf door of the gate for a preset waiting time based on the preset safety time period.

[0070] The implementation method of the above steps S244 to S245 is, for example: when the feature point data of the human body matches the preset human body shape feature model, it is indicated that the human body is a human body, and it is determined whether the height of the first human body and the height of the second human body are both greater than the preset height threshold (such as 1 meter, 1.1 meters or 1.2 meters). If the height of the first human body or the height of the second human body is not greater than the preset height threshold (such as 1 meter, 1.1 meters or 1.2 meters), a prohibit closing command is sent to the gate machine so that the gate machine prohibits the fan door of the gate machine from closing within the preset safety time (such as 1 minute), or a delayed closing command is sent to the gate machine so that the gate machine delays the preset waiting time (such as 40 seconds) to close the fan door of the gate machine on the basis of the preset safety time (such as 1 minute), so that the fan door is finally closed after 100 seconds. It can be understood that by sending a prohibit closing command to the gate machine, so that children can pass through the gate machine channel within the preset safety time, the probability of children being pinched when passing through the gate machine channel is reduced. Alternatively, by further sending a delayed closing command to the gate, the preset waiting time for children can be extended based on the preset safety time, so that children can have sufficient time to walk out of the gate channel, greatly reducing the probability of children being pinched when passing through the gate channel.

[0071] In the process of implementing the above solution, when the feature point data of the target object matches the preset human body shape feature model, it is indicated that the target object is a human body, and the height of the human body is further determined in combination with the point cloud data, thereby improving the defect that traditional gates cannot recognize children, effectively identifying child passengers and ensuring their safe passage, and greatly reducing the probability of children being pinched when passing through the gate channel.

[0072] As an optional implementation of the above-mentioned tailgate identification method, the above-mentioned data to be processed also includes: video data within the gate channel range; after determining whether the feature point data of multiple target objects are successfully matched with the preset human body shape feature model, it can also include:

[0073] Step S250: If the feature point data of each target object in the multiple target objects does not successfully match the preset human body shape feature model, human body recognition is performed on the frame image in the video data to obtain a human body recognition result, and the trailing recognition result within the gate channel range is determined based on the human body recognition result.

[0074] Optionally, as an optional implementation of the above-mentioned method for identifying tailgating gates, when the feature point data of each target object among multiple target objects does not successfully match the preset human shape feature model, OpenCV technology or YOLO V5 model can also be used to perform target recognition on the frame image in the video data, and determine the tailgating recognition result within the gate channel range based on the target recognition result. For example: determine whether the target detection result meets the preset conditions, and the preset conditions include: the number of target objects in the target detection result is multiple, and the target object categories in the target detection result are all human. It can be understood that the above-mentioned target object categories may include: human body, pets, and carry-on luggage, etc. If the number of target objects in the target detection result is multiple, and the target object categories in the target detection result are all human bodies, the tailgating recognition result is determined as the existence of tailing phenomenon, so as to reduce the probability of unauthorized passengers following authorized passengers into the gate channel. If the number of target objects in the target detection result is one and the target object category in the target detection result is a human body, the body thickness or body width of the human body is determined based on the point cloud data. When the body thickness is greater than a preset thickness threshold or the body width is greater than a preset width threshold, it means that someone may be carrying large luggage in their hands or on their shoulders and needs help, or may be holding a baby or a child and needs special attention. At this time, an abnormal warning can be issued to attract the attention of airport staff.

[0075] As an optional implementation of the above step S250, the above implementation of determining the tailgating recognition result within the gate channel range according to the human body recognition result may include:

[0076] Step S251: Determine whether the human body recognition result meets preset conditions, where the preset conditions include: the number of human bodies in the human body recognition result is multiple human bodies, and the distance between two human bodies among the multiple human bodies is less than a preset distance threshold.

[0077] Step S252: If the human body recognition result meets the preset conditions, the tailing recognition result is determined as the existence of tailing phenomenon.

[0078] An implementation example of the above-mentioned steps S251 to S252 is: when the feature point data of the target object matches the preset human body shape feature model, it means that the target object is a human body. Furthermore, if the number of human bodies in the human body recognition result is multiple human bodies, and the distance between any two of the multiple human bodies is less than a preset distance threshold (for example, 10 cm or 12 cm, etc.), the tailgating recognition result is determined as the existence of tailing phenomena such as side-by-side tailgating or hugging tailgating.

[0079] As an optional implementation of the above step S250, after obtaining the human body recognition result, the following may also be included:

[0080] Step S253: If the number of human bodies in the human body recognition result is one human body, determine the body thickness or body width of the human body according to the point cloud data, and issue an abnormal warning when the body thickness is greater than a preset thickness threshold or the body width is greater than a preset width threshold.

[0081] An implementation example of the above step S253 is as follows: if the number of human bodies in the human body recognition result is one human body, the body thickness or body width of the human body is determined based on the point cloud data, and when the body thickness is greater than a preset thickness threshold or the body width is greater than a preset width threshold, it means that someone may be carrying large luggage in their hands or on their shoulders and needs help, or may be holding a baby or a child and needs special attention. At this time, an abnormal warning can be issued to attract the attention of airport staff.

[0082] As an optional implementation of the above-mentioned tailgate identification method, the above-mentioned data to be processed may also include: infrared sensor data within a preset area of ​​the gate channel range; after determining the tailgate identification result within the gate channel range according to the distances between multiple target objects, it may also include:

[0083] Step S260: If the tailgating recognition result cannot be obtained, an abnormal warning is issued when the infrared sensor data is blocked by two obstacles at the same time.

[0084] For example, the implementation method of the above step S260 is as follows: the above gate can also be provided with an infrared beam sensor module for obtaining infrared sensor data in a preset area within the gate channel range. The preset area can be a safety area, a monitoring area, or a combined area of ​​a safety area and a monitoring area, or a combined area of ​​a safety area, a monitoring area, and a detection area. In the case where the tailgating recognition result cannot be obtained or the control module fails (for example, the point cloud data and video data cannot be obtained), it is determined whether the infrared sensor data in the preset area within the gate channel range is blocked by two obstacles at the same time. If the infrared sensor data in the preset area is blocked by two obstacles at the same time, it indicates that there may be a tailgating phenomenon or a side-by-side walking phenomenon, and therefore an abnormal warning is issued.

[0085] As an optional implementation of the above-mentioned tailgating gate identification method, after determining the tailgating identification result within the gate channel range according to the distances between the multiple target objects, the following may also be included:

[0086] Step S270: If the tailing identification result indicates that a tailing phenomenon exists, a closing command is sent to the gate, and the closing command is used to close the leaf door of the gate immediately, or to keep the leaf door of the gate closed.

[0087] An implementation example of the above step S270 is as follows: if the tailing identification result shows that tailing occurs, a closing command is immediately sent to the gate to close the leaf door of the gate immediately, or the leaf door of the gate is kept closed at all times.

[0088] Optionally, as an optional implementation of the above-mentioned trailing gate identification method, if the first target object among multiple target objects has been authenticated and authorized, for example, the ID card authentication and face recognition authentication both indicate that the authorized target object is successful, then the feature point data of the first target object can be bound to the current authorized target object, and the passage trajectory of the bound current authorized target object can be tracked until the current authorized target object passes through the gate channel, and then the binding is released. The unbinding here means releasing the binding relationship between the feature point data of the first target object and the current authorized target object. It can be understood that if the binding relationship between the feature point data of the first target object and the current authorized target object has not been released, and there is a second target object that enters the preset area within the gate channel range, the feature point data of the first target object bound to the current authorized target object can be matched with the feature point data of the second target object to obtain a matching result. If the matching result is no match, a closing command is issued to the gate, which is used to close the fan door of the gate, or the control component directly issues an early warning message to prevent authorized passengers from boarding the plane without passing through the gate channel, while unauthorized passengers mistakenly break into or follow through the gate channel and cause incorrect boarding.

[0089] An embodiment of the present application provides a tailgate identification device, including: a detection module and a control module, and the detection module and the control module communicate with each other.

[0090] The detection module is used to obtain the data to be processed, and the data to be processed includes: point cloud data within the gate channel range.

[0091] The control module is used to extract feature point data of multiple target objects from the point cloud data.

[0092] The control module is also used to determine whether the feature point data of multiple target objects are successfully matched with the preset human body shape feature model.

[0093] The control module is also used to determine the distance between the multiple target objects according to the point cloud data if the feature point data of the multiple target objects are successfully matched with the preset human body shape feature model, and determine the tailing recognition result within the gate channel range according to the distance between the multiple target objects.

[0094] As an optional implementation manner of the above device, the control module is further configured to determine the tailing recognition result as the presence of a tailing phenomenon if the distance between two target objects among the multiple target objects is less than a preset distance threshold.

[0095] As an optional implementation of the above-mentioned device, multiple target objects include: a first target object and a second target object; a control module is also used to determine the height of the first target object and the height of the second target object based on the point cloud data; if the distance between the first target object and the second target object is less than a preset distance threshold, and the height of the first target object and the height of the second target object are both greater than a preset height threshold, then the tailing recognition result is determined as the presence of a tailing phenomenon.

[0096] As an optional implementation of the above-mentioned device, the control module is also used to determine whether the height of the first target object and the height of the second target object are both greater than a preset height threshold; if not, a prohibit closing command or a delayed closing command is sent to the gate machine, and the prohibit closing command is used to prohibit the fan door of the gate machine from closing within a preset safety time period, and the delayed closing command is used to delay the closing of the fan door of the gate machine after a preset waiting time on the basis of a preset safety time period.

[0097] As an optional implementation of the above-mentioned device, the data to be processed also includes: video data within the gate channel range; a control module, which is also used to perform human body recognition on the frame image in the video data if the feature point data of each target object among the multiple target objects does not successfully match the preset human body shape feature model, obtain the human body recognition result, and determine the trailing recognition result within the gate channel range based on the human body recognition result.

[0098] As an optional implementation of the above-mentioned device, the control module is also used to determine whether the human body recognition result meets preset conditions, and the preset conditions include: the number of human bodies in the human body recognition result is multiple human bodies, and the distance between two human bodies among the multiple human bodies is less than a preset distance threshold; if the human body recognition result meets the preset conditions, the tailing recognition result is determined as the existence of a tailing phenomenon.

[0099] As an optional implementation of the above-mentioned device, the control module is also used to determine the body thickness or body width of the human body based on the point cloud data if the number of human bodies in the human body recognition result is one human body, and to issue an abnormal warning when the body thickness is greater than a preset thickness threshold or the body width is greater than a preset width threshold.

[0100] As an optional implementation of the above-mentioned device, the data to be processed also includes: infrared sensor data within a preset area of ​​the gate channel range; the control module is also used to issue an abnormal warning if the tailgating recognition result cannot be obtained and the infrared sensor data is blocked by two obstacles at the same time.

[0101] As an optional implementation of the above-mentioned device, the control module is also used to send a closing command to the gate if the tailgating identification result shows that a tailgating phenomenon exists. The closing command is used to close the fan door of the gate immediately, or to keep the fan door of the gate in a closed state.

[0102] As an optional implementation of the above-mentioned device, the detection module includes: a laser radar module and / or a visual camera module; the laser radar module is used to scan the point cloud data within the gate channel range, and the visual camera module is used to collect video data within the gate channel range.

[0103] It should be understood that the device corresponds to the above-mentioned embodiment of the method for identifying a tailgate, and can execute the various steps involved in the above-mentioned method embodiment. The specific functions of the device can be found in the above description, and the detailed description is appropriately omitted here. The device includes at least one software function module that can be stored in a memory in the form of software or firmware or solidified in the operating system (OS) of the device.

[0104] See also Figure 6 The electronic device 300 provided in the embodiment of the present application includes: a processor 310 and a memory 320, wherein the memory 320 stores machine-readable instructions executable by the processor 310, and when the machine-readable instructions are executed by the processor 310, the above method is executed.

[0105] The embodiment of the present application also provides a computer-readable storage medium 330, on which a computer program is stored, and the computer program is executed by the processor 310 to execute the above method. The computer-readable storage medium 330 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, disk or optical disk.

[0106] The embodiment of the present application also provides a computer program product, including: a computer program or a computer instruction, and the computer program or the computer instruction executes the method described above when executed by a processor.

[0107] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0108] In several embodiments provided by the embodiments of the present application, it should be understood that the disclosed device and method can also be implemented in other ways. The device embodiments described above are only schematic, for example, the flowcharts and block diagrams in the accompanying drawings show the possible implementation architecture, functions and operations of the devices, methods and computer program products according to the multiple embodiments of the embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a part of a module, a program segment or a code, and a part of a module, a program segment or a code includes one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also be different from the order of occurrence marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which is mainly determined by the functions involved.

[0109] In addition, each functional module of each embodiment in the embodiment of the present application can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part. In addition, in the description of this specification, the description of reference terms "one embodiment", "some embodiments", "example", "specific example", "some examples", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of the present application. In this specification, the schematic representation of the above terms does not necessarily target the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples without contradiction.

[0110] The above description is only an optional implementation manner of the embodiments of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the embodiments of the present application, which should be covered within the protection scope of the embodiments of the present application.

Claims

1. A method for identifying a trailing gate, characterized in that: include: Acquire data to be processed, wherein the data to be processed includes: point cloud data within the gate channel range; Extracting feature point data of multiple target objects from the point cloud data; Determining whether the feature point data of the plurality of target objects are all successfully matched with a preset human body shape feature model, wherein the preset human body shape feature model is used to describe the geometric shape and structural features of the human body; If yes, determining the distances between the multiple target objects according to the point cloud data, and determining the tailgating recognition result within the gate channel range according to the distances between the multiple target objects; Wherein, extracting feature point data of multiple target objects from the point cloud data includes: constructing a 3D map within the gate channel range according to the point cloud data, and extracting feature point data of the multiple target objects from the point cloud data of the 3D map.

2. The method according to claim 1, characterized in that The determining of the tailgating recognition result within the gate channel range according to the distances between the multiple target objects includes: If the distance between two target objects among the multiple target objects is less than a preset distance threshold, the tailing recognition result is determined as the presence of a tailing phenomenon.

3. The method according to claim 1, characterized in that The multiple target objects include: a first target object and a second target object; and determining the tailgating recognition result within the gate channel range according to the distance between the multiple target objects includes: Determine the height of the first target object and the height of the second target object according to the point cloud data; If the distance between the first target object and the second target object is less than a preset distance threshold, and the height of the first target object and the height of the second target object are both greater than a preset height threshold, the tailing recognition result is determined as the presence of a tailing phenomenon.

4. The method according to claim 3, characterized in that After determining the height of the first target object and the height of the second target object according to the point cloud data, the method further includes: Determine whether the height of the first target object and the height of the second target object are both greater than the preset height threshold; If not, a prohibit closing command or a delayed closing command is sent to the gate machine. The prohibit closing command is used to prohibit the closing of the leaf door of the gate machine within a preset safety time period, and the delayed closing command is used to delay the closing of the leaf door of the gate machine for a preset waiting time based on the preset safety time period.

5. The method according to claim 1, characterized in that The data to be processed also includes: video data within the gate channel range; after determining whether the feature point data of the multiple target objects are successfully matched with the preset human body shape feature model, it also includes: If the feature point data of each target object among the multiple target objects does not successfully match the preset human body shape feature model, human body recognition is performed on the frame image in the video data to obtain a human body recognition result, and the trailing recognition result within the gate channel range is determined based on the human body recognition result.

6. The method according to claim 5, characterized in that The determining of the tailgating recognition result within the gate channel range according to the human body recognition result includes: Determining whether the human body recognition result meets a preset condition, wherein the preset condition includes: the number of human bodies in the human body recognition result is multiple human bodies, and the distance between two human bodies among the multiple human bodies is less than a preset distance threshold; If so, the tailing identification result is determined as the existence of the tailing phenomenon.

7. The method according to claim 5, characterized in that After obtaining the human body recognition result, the method further includes: If the number of human bodies in the human body recognition result is one human body, the body thickness or body width of the human body is determined according to the point cloud data, and when the body thickness is greater than a preset thickness threshold or the body width is greater than a preset width threshold, an abnormal warning is issued.

8. The method according to any one of claims 1 to 7, characterized in that: The data to be processed also includes: infrared sensor data within a preset area of ​​the gate channel range; after determining the tailgating recognition result within the gate channel range according to the distances between the multiple target objects, it also includes: If the tailgating recognition result cannot be obtained, an abnormal warning is issued when the infrared sensor data is blocked by two obstacles at the same time.

9. The method according to any one of claims 1 to 7, characterized in that: After determining the tailgating recognition result within the gate channel range according to the distances between the multiple target objects, the method further includes: If the tailing identification result shows that tailing occurs, a closing command is sent to the gate, and the closing command is used to close the leaf door of the gate immediately, or to keep the leaf door of the gate closed.

10. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the machine-readable instructions are executed by the processor to perform any method according to claims 1 to 9.

Images