A gate machine and its control system

By setting up an adjustable face acquisition module and an infrared detection module inside the gate, the problems of uneven structure and insufficient flexibility of the existing gate are solved, and efficient face recognition and power consumption of the gate are achieved.

CN118762431BActive Publication Date: 2024-11-19SHENZHEN ZHONGZHENG INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411216718.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-02
Publication Date
2024-11-19
Estimated Expiration
2044-09-02

AI Technical Summary

Technical Problem

The existing gates are easily damaged and lack flexibility due to the external camera and uneven structure of the camera. Professionals need to manually adjust the camera angle to meet the face collection requirements.

Method used

Design a gate machine, whose face acquisition module is set inside the gate machine body, adjust the acquisition angle through the control module, and is equipped with an infrared detection module to reduce the working frequency and power consumption of the face recognition module, improve the flatness of the gate machine and flexibility.

Benefits of technology

It improves the flatness of the gate and the protection effect of the face acquisition module, avoids the debugging needs of the face recognition module, and reduces the working frequency and power consumption of the face recognition module.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118762431B_ABST
    Figure CN118762431B_ABST
Patent Text Reader

Abstract

The present invention discloses a gate machine and its control system, and relates to the field of gate machine control technology. Among them, an opening area is set on any one side of the two ends of the gate machine body, and the gate machine includes: a control module; a face acquisition module, the face acquisition module is electrically connected to the control module, the face acquisition module adjusts the face acquisition angle according to the adjustment control signal output by the control module, the face acquisition module is arranged inside the gate machine body, and performs face acquisition through the opening area, and outputs the face acquisition signal; an infrared detection module, the infrared detection module is arranged on the surface of both sides of the two ends of the gate machine body; the output end of the infrared detection module is electrically connected to the control module, and the infrared detection module is used to output the infrared detection signal to the control module; wherein the control module is used to control the face acquisition angle of the face acquisition module according to the infrared detection signal. The present invention aims to improve the versatility and flexibility of the gate machine.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of gate control, and in particular to a gate and a control system thereof. Background Art

[0002] The gate machine is a commonly used product in the channel equipment of the security industry. The equipment is widely used in rail transit, office buildings, factories, scenic spots and other places. With the continuous increase in the flow of people, higher requirements are put forward for the safety and performance of the equipment. With the advancement of social science and technology, the door opening and closing control of the gate machine has gradually changed from card swiping control to face recognition control, which reduces the steps of people taking cards and swiping cards, and at the same time eliminates the situation where people forget the card and cannot open the door. However, most of the existing gate machines use a camera outside the gate machine body to collect facial information of people, which results in an uneven structure on the surface of the gate machine body. In addition, the camera placed outside the gate machine body will also cause the camera to be easily damaged due to lack of protection. Furthermore, in actual applications, the camera also needs to be manually adjusted by professionals according to actual needs to ensure that the angle of the camera to collect images meets the requirements, which also leads to the lack of flexibility of the gate machine. Summary of the invention

[0003] The main purpose of the present invention is to provide a gate machine, aiming to improve the versatility and flexibility of the gate machine.

[0004] In order to achieve the above-mentioned object, the present invention proposes a gate, which includes a gate body, and an opening area is provided on either side of the two ends of the gate body, and the gate includes:

[0005] Control module;

[0006] A face acquisition module, the face acquisition module is electrically connected to the control module, the face acquisition module adjusts the face acquisition angle according to the adjustment control signal output by the control module, the face acquisition module is arranged inside the gate body, and performs face acquisition through the opening area and outputs a face acquisition signal;

[0007] An infrared detection module, which is arranged on the surfaces of both ends of the gate body; the output end of the infrared detection module is electrically connected to the control module, and the infrared detection module is used to output an infrared detection signal to the control module;

[0008] Wherein, the control module is used to control the face acquisition angle of the face acquisition module according to the infrared detection signal.

[0009] In one embodiment, the gate machine further includes a communication module, which is connected to the control module by wired communication, and is connected to the cloud by wireless communication, and the communication module is used for data exchange between the control module and the cloud.

[0010] In one embodiment, the control module is further used to set the face collection angle and face collection distance of the face collection module according to the control instructions sent from the cloud.

[0011] In one embodiment, the control module includes: a memory, a processor, and a face recognition program stored in the memory and executable on the processor, wherein the face recognition program is configured to determine the authority of the person being collected through the face information pre-stored in the memory and the face collection signal.

[0012] In one embodiment, the control module is also used to upload the face collection signal to the cloud through the communication module when the face information pre-stored in the memory fails to match the face collection signal, so that the cloud can further determine the authority of the person being collected, and send a control instruction to the control module to make the gate perform a corresponding action; when the face information pre-stored in the cloud successfully matches the face collection signal, an update request signal is output to the cloud through the communication module, so that the cloud can send the face information pre-stored in the cloud to the control module.

[0013] In one embodiment, the specific steps of configuring the face recognition program include:

[0014] Acquire face acquisition signals and perform feature extraction after face alignment processing;

[0015] A loss function is trained based on samples obtained by feature extraction, and a recognition model is trained based on the loss function and samples obtained by feature extraction;

[0016] Among them, the training loss function is specifically:

[0017]

[0018] For a fixed sample, is a positive sample, is a negative sample, the function F is to convert the input face image into its corresponding embedding vector, and the function D is the distance between the two embedding vectors.

[0019] In one embodiment, the infrared detection module includes:

[0020] A first infrared detection module, the first infrared detection module is arranged at a first height on a side surface of one end of the gate, the detection angle of the first infrared detection module is a first detection angle horizontally downward, the output end of the first infrared detection module is electrically connected to the control module, and the first infrared detection module is used to output a first infrared detection signal to the control module;

[0021] A second infrared detection module, the second infrared detection module is arranged at a second height of a side surface of one end of the gate, the detection angle of the second infrared detection module is a horizontal second detection angle, the output end of the second infrared detection module is electrically connected to the control module, and the second infrared detection module is used to output a second infrared detection signal to the control module;

[0022] A third infrared detection module, the third infrared detection module is arranged at a third height of a side surface of one end of the gate, the detection angle of the third infrared detection module is a third detection angle horizontally upward, the output end of the third infrared detection module is electrically connected to the control module, and the third infrared detection module is used to output a third infrared detection signal to the control module;

[0023] Wherein, the third detection angle is greater than the first detection angle.

[0024] In one embodiment, the detection area corresponding to the detection angle of the first infrared detection module and the detection angle of the second infrared detection module is the first detection area; the detection area corresponding to the detection angle of the second infrared detection module and the detection angle of the third infrared detection module is the second detection area;

[0025] Wherein, the control module is used to adjust the face acquisition angle of the face acquisition module to the first detection area or the second detection area according to the infrared detection signal.

[0026] In one embodiment, the gate machine further includes:

[0027] A light detection module, the output end of which is electrically connected to the control module and outputs a light detection signal;

[0028] A fill light module, the input end of which is electrically connected to the control module, and the fill light module is used to output a supplementary light source according to a fill light control signal output by the control module;

[0029] A display module, an input end of which is electrically connected to the control module, and the display module is used to display relevant information;

[0030] An identity reading module, the input end of which is electrically connected to the control module, and the identity reading module is used to obtain identity information.

[0031] The present invention further provides a gate control system, which includes a cloud and a gate as described in any one of the above items.

[0032] The technical solution of the present invention improves the flatness of the gate body and the protection effect of the face acquisition module by setting the face acquisition module inside the gate body. In addition, the face acquisition module with adjustable angle is adopted to effectively improve the acquisition range of the gate face recognition, and the angle and focal length are adjusted by the control module, thereby avoiding the need to debug the face recognition module when the gate is installed. By setting an infrared detection module to detect whether there are people who need to pass through the gate, the working frequency of the face recognition module is effectively reduced, and its loss is reduced, while also reducing power consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0034] Figure 1 It is a schematic diagram of the module of the gate machine of the present invention;

[0035] Figure 2 This is a schematic diagram of a module of an embodiment of a gate machine of the present invention;

[0036] Figure 3 A schematic diagram of a module of another embodiment of a gate machine of the present invention;

[0037] Figure 4 It is a detection schematic diagram of a gate machine according to an embodiment of the present invention.

[0038] Description of Figure Numbers:

[0039] 10. Control module; 20. Face acquisition module; 30. Infrared detection module; 31. First infrared detection module; 32. Second infrared detection module; 33. Third infrared detection module; 40. Communication module; 50. Light detection module; 60. Fill light module; 70. Display module; 80. Identity reading module.

[0040] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0043] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0044] The gate machine is a commonly used product in the channel equipment of the security industry. The equipment is widely used in rail transit, office buildings, factories, scenic spots and other places. With the continuous increase in the flow of people, higher requirements are put forward for the safety and performance of the equipment. With the advancement of social science and technology, the door opening and closing control of the gate machine has gradually changed from card swiping control to face recognition control, which reduces the steps of people taking cards and swiping cards, and at the same time eliminates the situation where people forget the card and cannot open the door. However, most of the existing gate machines use a camera outside the gate machine body to collect facial information of people, which results in an uneven structure on the surface of the gate machine body. In addition, the camera placed outside the gate machine body will also cause the camera to be easily damaged due to lack of protection. Furthermore, in actual applications, the camera also needs to be manually adjusted by professionals according to actual needs to ensure that the angle of the image collected by the camera meets the requirements, which also leads to the lack of flexibility of the gate machine.

[0045] Therefore, the present invention proposes a gate, which includes a gate body, and an opening area is provided on either side of the two ends of the gate body. The gate includes:

[0046] Control module 10;

[0047] A face acquisition module 20, the face acquisition module 20 is electrically connected to the control module 10, the face acquisition module 20 adjusts the face acquisition angle according to the adjustment control signal output by the control module 10, the face acquisition module 20 is arranged inside the gate body, and performs face acquisition through the opening area, and outputs a face acquisition signal;

[0048] Infrared detection module 30, the infrared detection module 30 is arranged on the surfaces of both ends of the gate body; the output end of the infrared detection module 30 is electrically connected to the control module 10, and the infrared detection module 30 is used to output infrared detection signals to the control module 10;

[0049] The control module 10 is used to control the face acquisition angle of the face acquisition module 20 according to the infrared detection signal.

[0050] In this embodiment, the control module 10 can be implemented by a main controller, such as DSP (Digital Signal Process), FPGA (Field Programmable Gate Array), MCU (Microcontroller Unit), SOC (System On Chip), etc. In the gate system, the control module 10 is not only responsible for processing the input information from various sensors and card readers, but also loads and controls the actions of components such as gate wings, indicator lights, and alarms to ensure the stability and response speed of the gate operation. Among them, DSP is suitable for complex tasks such as image processing and video stream analysis due to its high-speed computing power and powerful digital signal processing capabilities. For example, in a gate with integrated face recognition function, DSP can efficiently process the large amount of data calculation required by the face recognition algorithm to achieve fast and accurate identity authentication. FPGA provides a high degree of flexibility and customizability through programmable logic blocks and interconnection resources. In gate control, FPGA can flexibly configure logic according to specific needs to achieve efficient data processing and real-time control, which is particularly suitable for application scenarios that require fast response and low-latency control. MCU integrates CPU, memory and various peripheral interfaces, and is an economical and efficient choice for basic logic control and data processing. In conventional gate systems, MCU is sufficient to handle basic functions such as reading card information, controlling motor operation, and managing indicator light display. It is widely used for its high cost-effectiveness and short development cycle. SOC is a highly integrated chip that integrates multiple functions such as processor, memory, graphics processing unit, communication module 40 on the same chip. In gate control, SOC can provide a high-performance, low-power integrated solution, which is particularly suitable for design requirements that require high integration, multi-functions and compact size, and helps to simplify system design, reduce energy consumption and improve system reliability. Therefore, the selection of control module 10 in the gate can be based on the system design requirements, cost budget, performance requirements and future scalability. Different control chips are suitable for different application scenarios and functional requirements. Designers will choose the most suitable solution according to actual conditions to build an efficient and stable gate control system.

[0051] In this embodiment, the gate body mainly includes a fuselage and a gate wing. The fuselage and the gate wing are usually composed of at least two to form a coordinated blocking effect. Among them, the fuselage, as the main structure of the gate, is usually made of stainless steel or aluminum alloy materials to ensure the firmness, beauty and corrosion resistance of its structure. The fuselage design must meet the requirements of load-bearing, stability and protection, and also provide installation space for internal components. The gate wing is a wing with two or more retractable wing doors, which are driven by a motor to open and close quickly to control the passage of pedestrians. In conventional fuselage designs, no opening area is set on the fuselage, but in this application, a corresponding opening area is set to facilitate the face acquisition module 20 to collect faces through the opening area. It can be understood that the opening area can determine whether to set a transparent partition according to the actual selection of the face acquisition module 20, and the shape of the opening area can also be adjusted according to actual needs. It should be understood that the face acquisition angle of the face acquisition module 20 is adjustable. Therefore, the setting of the opening area needs to meet the space requirements of the face acquisition module 20 for adjusting the acquisition angle.

[0052] In this embodiment, the face acquisition module 20 can be implemented by a variable focus spherical camera, an optical zoom camera module, etc. The face acquisition module 20 is arranged inside the gate body, and the face acquisition is performed through the opening area on the body. It can be understood that in actual applications, the installation setting of the face acquisition module 20 and the angles collected by it need to be debugged through actual measurement to confirm the installation position and angle. For example, under a residential building, the setting of the gate needs to consider the entrance and exit angles of the residents, so that the face acquisition module 20 can quickly obtain face information, avoiding the user's need to adjust the angle by himself to be collected by the face acquisition module 20. Therefore, the installation setting of the face acquisition module 20 and the angles collected by it all need to be debugged by professionals, and need to be re-debugged when the subsequent environment changes. This setting method has the problem of poor flexibility. This problem can be well solved by setting an angle-adjustable face acquisition module 20, and the face acquisition module 20 can adjust the angle through the control signal output by the control module 10. Among them, the control module 10 can establish a communication connection with an external terminal and the cloud through a communication module 40 such as Bluetooth and 4G, and then obtain a debugging instruction to facilitate the control of the angle adjustment of the face acquisition module 20, thereby avoiding the disassembly and debugging of the face recognition module. In addition, the face acquisition module 20 can also change the focal length of its face acquisition through the control signal output by the control module 10, thereby changing the distance that the face acquisition module 20 can collect. Further, the face acquisition module 20 in the existing gate machine tends to be fixed, which also leads to a relatively fixed angle that can be collected. It can be understood that the relatively fixed angle includes a horizontal angle and a vertical angle. Among them, the distance of the face acquisition module 20 is generally between 0.5 meters and 2 meters. This is because the pixels of the face at a long distance are not clear enough to ensure the accuracy of recognition, and at a too close distance, it is easy to cause problems such as image distortion and light reflection, affecting the recognition effect. In addition, the vertical angle that the face acquisition module 20 can collect is generally between 15° and 45°. For people with tall or short stature, the face acquisition module 20 needs to quickly acquire face information when the person is between 0.5 meters and 2 meters away from them, otherwise the group of people will be out of the acquisition range of the face acquisition module 20.

[0053] In this embodiment, there are multiple infrared detection modules 30, and they are respectively arranged on the surfaces of both sides of the two ends of the gate body. The infrared detection module 30 can be implemented by a reflective infrared sensor, a pyroelectric infrared sensor, an infrared beam sensor, etc. Among them, a pyroelectric infrared sensor is taken as an example. The pyroelectric infrared sensor detects infrared radiation emitted by a heating object such as a human body. The pyroelectric infrared sensor contains a pyroelectric element inside. When receiving infrared radiation from an external heat source (such as a human body), a slight change in the surface temperature of the element will cause the charge to be released. After circuit processing, the approach of the human body can be detected, and then an infrared detection signal is output to the control module 10. In the application of the gate, the pyroelectric infrared sensor is mainly used to sense the approach of the human body, rather than directly measuring the distance. It is suitable for triggering the gate to prepare for the preprocessing stage of face recognition or other identity authentication. The control module 10 can determine whether there is a person approaching and wanting to pass through the gate through the pyroelectric infrared sensor, and then control the face acquisition module 20 to prepare, so as to avoid the face acquisition module 20 working for a long time, resulting in a reduction in the service life of the face acquisition module 20 and an increase in power consumption.

[0054] In this embodiment, by setting the face acquisition module 20 inside the gate body, the flatness of the gate body is improved, and the protection effect of the face acquisition module 20 is improved. In addition, the face acquisition module 20 with adjustable angle is adopted to effectively improve the acquisition range of the gate face recognition, and the angle and focal length are adjusted by the control module 10, thereby avoiding the need to debug the face recognition module when the gate is installed. By setting the infrared detection module 30 to detect whether there are people who need to pass through the gate, the working frequency of the face recognition module is effectively reduced, and its loss is reduced, and the power consumption is also reduced.

[0055] In one embodiment of the present invention, the gate machine also includes a communication module 40, which is connected to the control module 10 by wired communication, and is connected to the cloud by wireless communication. The communication module 40 is used for data interaction between the control module 10 and the cloud.

[0056] In this embodiment, the communication module 40 can be implemented by using a 4G / 5G communication chip, a WIFI communication chip, a Bluetooth communication chip, a local area network communication chip, etc. Among them, for example, the 4G / 5G communication chip is connected to the control module 10 through one or more physical interfaces such as Mini PCIe, USB, SPI, and UART (serial communication interface). After the communication module 40 is initialized and configured, such as setting the working mode, network parameters (such as APN, IP address) and authentication information (such as the PIN code of the SIM card), the communication module 40 is connected to the operator's 4G / 5G network through AT instructions or a dedicated SDK to complete the network registration process, thereby establishing a data transmission channel, and configuring the corresponding network protocol stack according to the requirements of the cloud platform, and setting the address, port number and other parameters of the cloud server. It can be understood that the gate establishes wireless communication with the cloud through the communication module 40. The cloud can issue various control instructions through the communication module 40 in the gate, and can also receive relevant data uploaded by the main control module through the communication module 40 in the gate. For example, the cloud can realize angle adjustment and distance adjustment of the face acquisition module 20 through the communication module 40 in the gate.

[0057] In one embodiment of the present invention, the control module 10 includes: a memory, a processor, and a face recognition program stored in the memory and executable on the processor, wherein the face recognition program is configured to determine the authority of the person being collected through the face information pre-stored in the memory and the face collection signal.

[0058] In this embodiment, at the initial stage of gate machine deployment or when personnel authority is set, facial images of authorized personnel are collected through high-definition cameras and quality checks are performed to ensure that the images are clear and complete. After these images are processed by face detection, key point extraction, feature encoding, etc., the facial feature data (usually a high-dimensional vector) is stored in the memory of the control module 10 and / or the cloud server. These feature data are compact representations of facial images for subsequent comparison. The face acquisition module 20 on the gate machine captures the facial images of people who want to pass in real time and undergoes a series of preprocessing, including graying, denoising, brightness and contrast adjustment, and facial posture correction, to ensure the accuracy of authority matching. Specifically, feature vectors can be extracted from real-time captured facial images through deep learning or machine learning algorithms. This process is similar to the input process, which aims to extract representative features that can distinguish different individuals from the image. The face recognition program in the control module 10 quickly compares the feature vectors extracted in real time with the facial information (feature data) pre-stored in the memory. When the similarity between the real-time feature and a facial feature in the database exceeds a preset threshold, the recognition is considered successful, indicating that the person has access rights. Based on the comparison results, the control module 10 determines whether the user has the corresponding rights, and makes the next decision: if the recognition is successful, it sends instructions to the mechanical or electronic components of the gate, such as opening the door; if the recognition fails, it may trigger a warning, deny access, or prompt to try again.

[0059] In one embodiment of the present invention, the control module 10 is also used to upload the face collection signal to the cloud through the communication module 40 when the face information pre-stored in the memory fails to match the face collection signal, so that the cloud can further determine the authority of the person being collected, and send a control instruction to the control module 10 to make the gate perform a corresponding action; when the face information pre-stored in the cloud successfully matches the face collection signal, an update request signal is output to the cloud through the communication module 40, so that the cloud can send the face information pre-stored in the cloud to the control module 10.

[0060] In this embodiment, the face recognition program stored in the control module 10 in the gate machine will process the face acquisition signal output by the face acquisition module 20 and perform recognition and judgment according to the face information pre-stored in the memory of the control module 10. Among them, when the face acquisition signal and the face information pre-stored in the control module 10 match incorrectly, the control module 10 will first prompt the user to change the face information entry posture through the display module 70 to re-acquire the face image. When the face acquisition signal acquired by the face acquisition module 20 again matches the face information pre-stored in the memory and still fails, the control module 10 will upload the acquired face acquisition signal to the cloud through the communication module 40, so that the cloud can perform further face matching through the face information pre-stored in its server. It is understandable that in the early stage of gate machine deployment or after the personnel authority is set, there may be subsequent face information supplements, and these face information supplements may not be collected by the face acquisition module 20 on the gate machine, but collected by other face acquisition devices and uploaded to the cloud. Therefore, in a short period of time, the face information pre-stored in the cloud server contains some face information that is not stored in the memory of the gate control module 10. Further, when there is a face recognition signal that cannot be recognized in the gate and the match is successful in the cloud, the control module 10 will output an update request signal to the cloud through the communication module 40, so that the cloud sends the face information pre-stored in the cloud to the control module 10, so that the face information pre-stored in the control module 10 is updated, so that the user who failed to match successfully last time can quickly match successfully the next time he passes through the gate.

[0061] In one embodiment of the present invention, the specific steps of configuring the face recognition program include:

[0062] Acquire face acquisition signals and perform feature extraction after face alignment processing;

[0063] A loss function is trained based on samples obtained by feature extraction, and a recognition model is trained based on the loss function and samples obtained by feature extraction;

[0064] Among them, the training loss function is specifically:

[0065]

[0066] For a fixed sample, is a positive sample, is a negative sample, the function F is to convert the input face image into its corresponding embedding vector, and the function D is the distance between the two embedding vectors.

[0067] In this embodiment, in order to achieve fast and accurate matching of the face acquisition signal input by the gate control module 10 to the face recognition module, it is necessary to establish a relevant recognition model in advance. Specifically, the establishment of the face recognition model can automatically detect and locate faces from images or videos through target detection algorithms such as R-CNN, SSD, YOLO, FaceNet, etc. These algorithms can directly learn and optimize the facial feature representation from the original image data. Specifically, the training of the model requires a large number of face image samples to obtain a recognition model that meets actual needs. The training of the model specifically includes the acquisition of face images, that is, obtaining face acquisition signals and preprocessing the face acquisition signals. This stage involves detecting and locating faces from the original image, and face detection is usually performed using methods such as Haar cascade and MTCNN. Subsequently, the detected face area is adjusted to the standard viewing angle and position through face alignment technology to eliminate the influence of factors such as posture and occlusion. The pre-processed face images are analyzed in depth using a deep learning network (such as ResNet, DenseNet, MobileNet, ViT, etc.) to extract high-dimensional feature vectors, which can capture the key information of the face and are the basis for distinguishing different individuals. Among them, in this embodiment, the method adopted is a ternary loss function. In the triplet loss function, the goal is to ensure that the image of the same person, that is, the distance between the fixed sample and the positive sample in the embedding space is as small as possible, and the distance between the images of different people, that is, the fixed sample and the negative sample is as large as possible, so that the model can learn to distinguish different individuals by minimizing the triplet loss, thereby improving the accuracy of face recognition. The triplet loss improves the model's ability to distinguish by minimizing the distance between faces of the same type and maximizing the distance between faces of different types. The pre-processed face images and their corresponding labels are input into the deep learning model, and the selected loss function is used for back propagation and gradient descent to optimize the model parameters to achieve effective learning of face features. After training, the model is used for reasoning of new face images, that is, facial features are extracted through the model, and then matched with the existing facial features in the database to achieve identity authentication or recognition. This process includes preprocessing the input image again, the model inferring feature vectors, and calculating the similarity between feature vectors.

[0068] In one embodiment of the present invention, the infrared detection module 30 includes:

[0069] A first infrared detection module 31, the first infrared detection module 31 is arranged at a first height on a side surface of one end of the gate, the detection angle of the first infrared detection module 31 is a first detection angle horizontally downward, the output end of the first infrared detection module 31 is electrically connected to the control module 10, and the first infrared detection module 31 is used to output a first infrared detection signal to the control module 10;

[0070] A second infrared detection module 32, the second infrared detection module 32 is arranged at a second height of a side surface of one end of the gate, the detection angle of the second infrared detection module 32 is a horizontal second detection angle, the output end of the second infrared detection module 32 is electrically connected to the control module 10, and the second infrared detection module 32 is used to output a second infrared detection signal to the control module 10;

[0071] A third infrared detection module 33, the third infrared detection module 33 is arranged at a third height of a side surface of one end of the gate, the detection angle of the third infrared detection module 33 is a third detection angle horizontally upward, the output end of the third infrared detection module 33 is electrically connected to the control module 10, and the third infrared detection module 33 is used to output a third infrared detection signal to the control module 10;

[0072] Wherein, the third detection angle is greater than the first detection angle.

[0073] In this embodiment, the first infrared detection module 31, the second infrared detection module 32, and the third infrared detection module 33 are all arranged on the surface of one side of one end of the gate, and are used to detect whether there is a person approaching one side of one end of the gate, so as to output an infrared detection signal to the control module 10, so that the control module 10 controls the face acquisition module 20 to work. It can be understood that the gate has two ends and two sides. Therefore, in actual applications, the first infrared detection module 31, the second infrared detection module 32, and the third infrared detection module 33 can be respectively arranged on both sides of the two ends of the gate, that is, four groups of the first infrared detection module 31, the second infrared detection module 32, and the third infrared detection module 33 are set. Further, the first infrared detection module 31, the second infrared detection module 32, and the third infrared detection module 33 are respectively arranged at different heights on one side of one end of the gate body. Taking the gate height of 1 meter as an example, the first infrared detection module 31 is set at a position of 0.9 meters from the bottom surface of the gate, and the detection angle of the first infrared detection module 31 is 10° upwards with the horizontal angle, that is, the first detection angle; the second infrared detection module 32 is set at a position of 0.95 meters from the bottom surface of the gate, and the detection angle of the second infrared detection module 32 is a horizontal angle, that is, the second detection angle; the third infrared detection module 33 is set at a position of 1 meter from the bottom surface of the gate, that is, the surface of the top of the gate, and the detection angle of the third infrared detection module 33 is 15° upwards with the horizontal angle, that is, the third detection angle. Further, the detection area corresponding to the detection angle of the first infrared detection module 31 and the detection angle of the second infrared detection module 32 is the first detection area; the detection area corresponding to the detection angle of the second infrared detection module 32 and the detection angle of the third infrared detection module 33 is the second detection area; wherein, the control module 10 is used to adjust the face acquisition angle of the face acquisition module 20 to the first detection area or the second detection area according to the infrared detection signal. It is understandable that in the detection of facial features of passers-by by the gate machine, face collection often fails due to the height of the detected person and the distance from the person collection module. Therefore, it is necessary to further adjust the angle and focal length collected by the face collection module 20 to accurately obtain face information at different heights and distances. It is understandable that by setting infrared detection modules 30 at different heights, people at different heights and distances can be effectively detected. For example, if the height of the passer-by is 0.9 meters, he will be detected by the first infrared detection module 31 and the second infrared detection module 32 when approaching the gate machine. At this time, the first infrared detection module 31 and the second infrared detection module 32 will output infrared detection signals to the control module 10, so that the control module 10 outputs a control signal to the face collection module 20, thereby causing the collection angle of the face collection module 20 to deflect to the first detection area, thereby obtaining a face collection signal.Furthermore, the gate can also detect the distance between the person and the gate by setting up distance detection modules such as ultrasonic sensors, laser radar (LiDAR) or ToF (Time of Flight) sensors, and then control the focal length of the face acquisition module 20 through the distance detection signal, thereby improving the accuracy of the face acquisition signal. For another example, if the height of the passerby is 1.9 meters, he will be detected by the first infrared detection module 31, the second infrared detection module 32, and the third infrared detection module 33 when approaching the gate. At this time, the first infrared detection module 31, the second infrared detection module 32, and the third infrared detection module 33 will output the infrared detection signal to the control module 10, so that the control module 10 outputs the control signal to the face acquisition module 20, thereby causing the acquisition angle of the face acquisition module 20 to deflect to the second detection area, thereby obtaining the face acquisition signal.

[0074] In one embodiment of the present invention, the gate machine further includes:

[0075] A light detection module 50, wherein an output end of the light detection module 50 is electrically connected to the control module 10 and outputs a light detection signal;

[0076] A fill light module 60, the input end of which is electrically connected to the control module 10, and the fill light module 60 is used by the control module 10 to control the fill light module 60 according to the light detection signal;

[0077] A display module 70, the input end of which is electrically connected to the control module 10, and the display module 70 is used to display relevant information;

[0078] The identity reading module 80 has an input end electrically connected to the control module 10 , and the identity reading module 80 is used to obtain identity information and output it.

[0079] In this embodiment, the light detection module 50 can be implemented by photodiodes, photoresistors, etc. Among them, the photodiode can convert the received light intensity into an electrical signal, while the photoresistor changes the resistance value with the change of light intensity, thereby changing the electrical signal. The control module 10 determines the light conditions outside the gate by receiving the changing light detection signal output by the light detection module 50. When the external light conditions are poor, the control module 10 controls the fill light module 60 to work so that the face acquisition module 20 can obtain the face image more clearly and improve the accuracy of face recognition. Among them, the fill light module 60 can be implemented by LED light board / ring light, infrared LED light, etc. The display module 70 can be implemented by LCD liquid crystal display screen, LED dot matrix display screen, OLED display screen, touch screen, etc. The gate displays the user with interactive information such as passage information, operation instructions, error prompts, etc. through the control module 10 and the display module 70. The identity reading module 80 can be implemented by a magnetic stripe reader, RFID / NFC reader, etc. By setting up the identity reading module 80, the permissions of some users who failed to undergo face registration or face recognition in time can be obtained, thereby providing diversified channels for passing through the gate and meeting the actual needs of users.

[0080] The present invention further proposes a gate control system, which includes a gate and a cloud as described in any one of the above items. It is worth noting that since the gate control system of the present invention is based on the above gate, the embodiments of the gate control system of the present invention include all technical solutions of all embodiments of the above gate, and the technical effects achieved are also exactly the same, which will not be repeated here.

[0081] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the inventive concept of the present invention, or directly / indirectly applied in other related technical fields are included in the patent protection scope of the present invention.

[0082] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals (including but not limited to signals transmitted between user terminals and other devices, etc.) involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions. For example, the "face images", "face information", "face collection signals" involved in this disclosure are all obtained with full authorization."

Claims

1. A gate machine, It is characterized in that The gate includes a gate body, and an opening area is provided on either side of the two ends of the gate body. The gate includes: Control module; A face acquisition module, the face acquisition module is electrically connected to the control module, the face acquisition module adjusts the face acquisition angle according to the adjustment control signal output by the control module, the face acquisition module is arranged inside the gate body, and performs face acquisition through the opening area and outputs a face acquisition signal; An infrared detection module, which is arranged on the surfaces of both ends of the gate body; the output end of the infrared detection module is electrically connected to the control module, and is used to output an infrared detection signal to the control module; The infrared detection module comprises: A first infrared detection module, the first infrared detection module is arranged at a first height on a side surface of one end of the gate, the detection angle of the first infrared detection module is a first detection angle horizontally downward, the output end of the first infrared detection module is electrically connected to the control module, and the first infrared detection module is used to output a first infrared detection signal to the control module; A second infrared detection module, the second infrared detection module is arranged at a second height of a side surface of one end of the gate, the detection angle of the second infrared detection module is a horizontal second detection angle, the output end of the second infrared detection module is electrically connected to the control module, and the second infrared detection module is used to output a second infrared detection signal to the control module; A third infrared detection module, the third infrared detection module is arranged at a third height of a side surface of one end of the gate, the detection angle of the third infrared detection module is a third detection angle horizontally upward, the output end of the third infrared detection module is electrically connected to the control module, and the third infrared detection module is used to output a third infrared detection signal to the control module; The third detection angle is greater than the first detection angle; the detection area corresponding to the detection angle of the first infrared detection module and the detection angle of the second infrared detection module is the first detection area; the detection area corresponding to the detection angle of the second infrared detection module and the detection angle of the third infrared detection module is the second detection area; When a passerby is detected by the first infrared detection module and the second infrared detection module when approaching the gate, the first infrared detection module and the second infrared detection module will output infrared detection signals to the control module, so that the control module outputs a control signal to the face acquisition module, thereby causing the acquisition angle of the face acquisition module to deflect to the first detection area to obtain the face acquisition signal; When a person passing by is detected by the first infrared detection module, the second infrared detection module, and the third infrared detection module when approaching the gate, the first infrared detection module, the second infrared detection module, and the third infrared detection module will output infrared detection signals to the control module, so that the control module outputs a control signal to the face acquisition module, thereby causing the acquisition angle of the face acquisition module to deflect to the second detection area to obtain the face acquisition signal; The gate machine also detects the distance between the person and the gate machine by setting a distance detection module, and then controls the focal length of the face acquisition module through the distance detection signal to improve the accuracy of the face acquisition signal; Furthermore, the gate machine further comprises a communication module, the communication module is connected to the control module by wired communication, the communication module is connected to the cloud by wireless communication, and the communication module is used for data interaction between the control module and the cloud; The control module is also used to set the face collection angle and face collection distance of the face collection module according to the control instructions sent from the cloud.

2. The gate machine according to claim 1, It is characterized in that The control module includes: a memory, a processor, and a face recognition program stored in the memory and executable on the processor. The face recognition program is configured to determine the authority of the person being collected through the face information pre-stored in the memory and the face collection signal.

3. The gate machine according to claim 2, It is characterized in that The control module is also used to upload the face collection signal to the cloud through the communication module when the face information pre-stored in the memory fails to match the face collection signal, so that the cloud can further determine the authority of the person being collected, and send a control instruction to the control module to make the gate perform a corresponding action; when the face information pre-stored in the cloud successfully matches the face collection signal, an update request signal is output to the cloud through the communication module, so that the cloud can send the face information pre-stored in the cloud to the control module.

4. The gate machine according to claim 1, It is characterized in that The gate machine also includes: A light detection module, the output end of which is electrically connected to the control module and is used to output a light detection signal; A fill light module, the input end of which is electrically connected to the control module, and the fill light module is used to output a supplementary light source according to a fill light control signal output by the control module; A display module, an input end of which is electrically connected to the control module, and the display module is used to display relevant information; An identity reading module, the input end of which is electrically connected to the control module, and the identity reading module is used to obtain identity information.

5. A gate control system, It is characterized in that The gate control system includes a cloud and a gate as described in any one of claims 1-4.