Gate control method and device, equipment, storage medium and program product

Abstract

The invention provides a gate control method, device and equipment, a storage medium and a program product, which can be applied to the field of traffic. The method comprises the steps that an object image of a target object going to pass through a gate device is acquired, and the gate device supports at least two passing modes; image recognition is carried out on the object image to obtain a recognition result, a target passing mode to be used by the target object is determined based on the recognition result, and the at least two passing modes comprise the target passing mode; activating a target passing processing unit of the target passing mode, and verifying the target passing mode based on the target passing processing unit; when verification of the target passing mode based on the target passing processing unit is passed, the gate equipment is controlled to be in a passable state, so that the target object passes through the gate equipment; according to the invention, the hardware processing efficiency and the gate passing efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a gate control method, device, equipment, storage medium and program product. Background Art

[0002] A gate often supports multiple modes of access, such as access cards (such as one-way cards, one-way tickets), biometrics (such as face recognition, palm recognition), graphic codes (such as mini-program boarding codes), etc. In related technologies, in order to satisfy each user's desired access mode through the gate, the gate will keep the processing device of each access mode in an activated state. However, when users pass through the gate, they only use one access mode. In this way, not only will the hardware processing resources be excessively occupied and wasted, but the hardware processing efficiency will also be reduced, thereby affecting the gate's access efficiency. Summary of the invention

[0003] The embodiments of the present application provide a gate control method, device, electronic device, computer-readable storage medium and computer program product, which can reduce the occupation and waste of hardware processing resources and improve hardware processing efficiency and gate passage efficiency.

[0004] The technical solution of the embodiment of the present application is implemented as follows:

[0005] The present application provides a gate control method, including:

[0006] Acquire an object image of a target object to be passed through a gate device, wherein the gate device supports at least two passing modes;

[0007] Performing image recognition on the object image to obtain a recognition result, and determining a target passage mode to be used by the target object based on the recognition result, wherein the at least two passage modes include the target passage mode;

[0008] activating a target passage processing unit of the target passage mode, and verifying the target passage mode based on the target passage processing unit;

[0009] When the target passage mode is verified to be passed based on the target passage processing unit, the gate device is controlled to be in a passable state so that the target object can pass through the gate device.

[0010] The present application also provides a gate control device, including:

[0011] An acquisition module, used to acquire an object image of a target object to be passed through a gate device, wherein the gate device supports at least two passing modes;

[0012] a determination module, configured to perform image recognition on the object image to obtain a recognition result, and determine a target passage mode to be used by the target object based on the recognition result, wherein the at least two passage modes include the target passage mode;

[0013] an activation module, used for activating a target passage processing unit of the target passage mode, and verifying the target passage mode based on the target passage processing unit;

[0014] The control module is used to control the gate device to be in a passable state when the target passage mode is verified by the target passage processing unit, so as to allow the target object to pass through the gate device.

[0015] In the above solution, the gate device is equipped with an image acquisition unit; the acquisition module is also used to acquire an image of the target object passing through the gate device through the image acquisition unit to obtain an object image of the target object.

[0016] In the above scheme, the determination module is also used to perform object posture recognition on the object image to obtain the object posture of the target object; when the object posture is not a hand-held object posture, the hand movement of the target object is recognized to obtain a first recognition result, and the first recognition result indicates whether the hand movement is one of the following actions: a palm extension action for a hand information recognition unit, a password input action for a password input unit; when the object posture is a hand-held object posture, the hand-held object of the target object is recognized to obtain a second recognition result, and the second recognition result indicates whether the hand-held object is one of the following objects: a pass card, a mobile terminal with a display screen on, and a mobile terminal with a display screen off; wherein the recognition result includes the first recognition result and the second recognition result.

[0017] In the above scheme, when the recognition result is the first recognition result, the determination module is also used to determine that the target access mode to be used by the target object is the hand information recognition access mode when the first recognition result represents that the hand movement is the palm extension movement; when the first recognition result represents that the hand movement is the password input movement, the target access mode to be used by the target object is determined to be the password access mode; when the first recognition result represents that the hand movement is neither the palm extension movement nor the password input movement, the target access mode to be used by the target object is determined to be the facial information recognition access mode.

[0018] In the above scheme, when the recognition result is the second recognition result, the determination module is also used to determine that the target pass mode to be used by the target object is the first pass mode when the second recognition result represents that the handheld object is a pass card; when the second recognition result represents that the handheld object is a mobile terminal with a display screen on, determine that the target pass mode to be used by the target object is a graphic code pass; when the second recognition result represents that the handheld object is a mobile terminal with a display screen on, determine that the target pass mode to be used by the target object is the second pass mode.

[0019] In the above scheme, the determination module is also used to perform hand detection on the object image to obtain the hand area in the object image; perform object detection on the hand area to obtain the object area of ​​the hand area; perform object classification prediction on the object area to obtain the object classification result, and use the object classification result as the second recognition result.

[0020] In the above scheme, the determination module is also used to perform hand detection on the object image to obtain the hand area in the object image; perform pass card detection on the hand area to obtain a first detection result; when the first detection result indicates that the handheld object in the hand area is a pass card, use the first detection result as the second recognition result; when the first detection result indicates that the handheld object in the hand area is not a pass card, perform mobile terminal detection on the hand area to obtain a second detection result; when the second detection result indicates that the handheld object in the hand area is a mobile terminal, perform screen status detection on the mobile terminal in the hand area to obtain a third detection result; use the third detection result as the second recognition result, and the third detection result indicates that the display screen of the mobile terminal is in one of the following states: on state, off state.

[0021] In the above scheme, the acquisition module is also used to perform human biometric information detection on the object image after acquiring the object image of the target object to pass through the gate device, so as to obtain the biometric information image area in the object image; blur the biometric information image area to obtain the target object image; the determination module is also used to perform image recognition on the target object image to obtain a recognition result.

[0022] In the above scheme, the activation module is further used to output activation prompt information after activating the target passage processing unit of the target passage mode, and the activation prompt information is used to prompt that the target passage processing unit of the target passage mode has been activated.

[0023] In the above scheme, the activation module is also used to detect whether the silence condition of the target passage processing unit is met; when it is detected that the silence condition of the target passage processing unit is met, the target passage processing unit of the target passage mode is controlled to be in a silent state.

[0024] In the above scheme, the activation module is also used to obtain the passage mode data of the target object in the first time period when the passage frequency of the target object passing through the gate device in the first time period reaches a frequency threshold; based on the passage mode data, determine the first passage mode of the target object in the second time period, the at least two passage modes include the first passage mode, and the second time period is located after the first time period; when the target object passes through the gate device in the second time period, activate the first passage processing unit of the first passage mode.

[0025] The present application also provides an electronic device, including:

[0026] A memory for storing computer executable instructions;

[0027] The processor is used to implement the gate control method provided in the embodiment of the present application when executing the computer executable instructions stored in the memory.

[0028] The embodiment of the present application also provides a computer-readable storage medium storing computer-executable instructions or a computer program. When the computer-executable instructions or the computer program are executed by a processor, the gate control method provided by the embodiment of the present application is implemented.

[0029] The embodiment of the present application also provides a computer program product, including computer executable instructions or a computer program, which, when executed by a processor, implements the gate control method provided in the embodiment of the present application.

[0030] The embodiments of the present application have the following beneficial effects:

[0031] Applying the above-mentioned embodiment of the present application, first, an object image of the target object to be passed through the gate device is obtained, and the gate device supports at least two passage modes; then, image recognition is performed on the object image to obtain a recognition result, and based on the recognition result, the target passage mode to be used by the target object is determined; then, the target passage processing unit of the target passage mode is activated, and the target passage mode is verified based on the target passage processing unit; thereby, when the target passage mode is verified based on the target passage processing unit, the gate device is controlled to be in a passable state to allow the target object to pass through the gate device.

[0032] Here, when the target object passes through the gate device, the target passage mode to be used by the target object will be determined in advance based on the object image of the target object, so that only the target passage processing unit of the target passage mode is activated to support the target object to pass through the gate device based on the target passage mode; in this way, the occupation and waste of hardware processing resources are reduced, the hardware processing efficiency is improved, and the gate passage efficiency is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the architecture of the gate control system provided in the embodiment of the present application;

[0034] Figure 2 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0035] Figure 3 It is a flow chart of the gate control method provided in the embodiment of the present application;

[0036] Figure 4 It is a flow chart of the gate control method provided in the embodiment of the present application;

[0037] Figure 5 It is a flow chart of the gate control method provided in the embodiment of the present application;

[0038] Figure 6 It is a flow chart of the gate control method provided in the embodiment of the present application;

[0039] Figure 7 It is a flow chart of the gate control method provided in the embodiment of the present application;

[0040] Figure 8 It is a schematic diagram of the architecture of the gate control system provided in the embodiment of the present application;

[0041] Fig. 9 It is a flow chart of the gate control method provided in the embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0043] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0044] In the following description, the terms "first\second\third" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described here can be implemented in an order other than that illustrated or described here.

[0045] Unless otherwise defined, all technical and scientific terms used in the embodiments of the present application have the same meanings as those commonly understood by those skilled in the art. The terms used in the embodiments of the present application are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0046] Before further describing the embodiments of the present application in detail, the nouns and terms involved in the embodiments of the present application are explained. The nouns and terms involved in the embodiments of the present application are subject to the following interpretations.

[0047] 1) Client: An application running in the terminal to provide various services, such as a gate client.

[0048] 2) In response to: used to indicate the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more operations executed may be in real time or have a set delay. Unless otherwise specified, there is no restriction on the order in which the multiple operations executed are executed.

[0049] 3) Palm recognition: It identifies the identity by identifying the palm image from the fingertips to the wrist. It has the characteristics of simple sampling, rich image information, high user acceptance, not easy to forge, and little interference from noise.

[0050] 4) 3D cameras, including depth cameras and infrared cameras, can ensure information security. 3D cameras can detect the distance information of the shooting space and have dozens of functions such as 3D face recognition, gesture recognition, human skeleton recognition, 3D measurement, environmental perception, 3D map reconstruction, etc. They can be widely used in TV, mobile phones, robots, drones, logistics, VR / AR, smart homes, security, and car driving assistance.

[0051] 5) Near Field Communication (NFC), devices using NFC technology (such as mobile phones and gates) can exchange data when they are close to each other. It is an integration of contactless radio frequency identification technology (RFID) and interconnection technology. By integrating the functions of inductive card readers, inductive cards and point-to-point communication on a single chip, mobile terminals can be used to realize mobile payment, electronic ticketing, access control, mobile identity recognition, anti-counterfeiting and other applications.

[0052] 6) Gate machine is a channel blocking device (channel management equipment) that can be used in urban rail transit management, toll gate machine system, access control system, etc. to manage the flow of people and regulate the entry and exit of pedestrians. Therefore, paid areas and non-paid areas are formed. Its most basic and core function is to achieve only one person passing through at a time, and it can be used at the entrance of various toll and access control occasions.

[0053] Based on the above description of the nouns and terms involved in the embodiments of the present application, the embodiments of the present application are described in detail below. The embodiments of the present application provide a gate control method, device, electronic device, computer-readable storage medium and computer program product, which can reduce the occupation and waste of hardware processing resources and improve hardware processing efficiency and gate passage efficiency.

[0054] It should be noted that the collection and processing of relevant data in this application should be strictly in accordance with the requirements of relevant laws and regulations when applied in examples, and the informed consent or separate consent of the personal information subject should be obtained, and subsequent data use and processing should be carried out within the scope of authorization of laws and regulations and the personal information subject.

[0055] The gate control system provided by the embodiment of the present application is described below. Figure 1 , Figure 1 1 is a schematic diagram of the architecture of the gate control system provided in the embodiment of the present application. To support an exemplary application, the gate control system 100 includes: a server 200, a network 300, and a gate device 400. The gate device 400 is connected to the server 200 via the network 300. The network 300 can be a wide area network or a local area network, or a combination of the two, and a wireless or wired link is used to realize data transmission.

[0056] Here, the gate device 400 (for example, a gate client is running) detects a target object to be passed through the gate device 400, obtains an object image of the target object, and the gate device supports at least two passage modes; sends the object image of the target object to the server 200; the server 200 receives the object image of the target object sent by the gate device; performs image recognition on the object image to obtain a recognition result, and based on the recognition result, determines a target passage mode to be used by the target object, and the at least two passage modes include the target passage mode; returns the target passage mode to the gate device 400; the gate device 400 receives the target passage mode returned by the server; activates the target passage processing unit of the target passage mode, and verifies the target passage mode based on the target passage processing unit; when the target passage mode is verified based on the target passage processing unit, the gate device is controlled to be in a passable state to allow the target object to pass through the gate device.

[0057] In some embodiments, the gate control method provided in the embodiments of the present application is implemented by an electronic device. For example, it can be implemented by a gate device alone, or by a server alone, or by a gate device and a server in collaboration. Among them, the server (such as server 200) can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. The gate device and the server can be directly or indirectly connected via wired or wireless communication, and the embodiments of the present application do not limit this. The embodiments of the present application can be applied to various scenarios, including but not limited to cloud technology, artificial intelligence, access control, smart transportation, smart home, assisted driving, etc.

[0058] In some embodiments, the gate control method provided in the embodiments of the present application can be implemented with the aid of cloud technology (Cloud Technology). Cloud technology refers to a hosting technology that unifies a series of resources such as hardware, software, and network in a wide area network or a local area network to realize the calculation, storage, processing, and sharing of data. Cloud technology is a general term for network technology, information technology, integration technology, management platform technology, and application technology based on the cloud computing business model application, which can form a resource pool, which is used on demand and is flexible and convenient. Cloud computing technology will become an important support. The background services of the technical network system require a large amount of computing resources and storage resources. As an example, a server (such as server 200) can also be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.

[0059] In some embodiments, the terminal or server can implement the gate control method provided in the embodiments of the present application by running various computer executable instructions or computer programs. For example, the computer executable instructions can be microprogram-level commands, machine instructions or software instructions. The computer program can be a native program or software module in the operating system; it can be a native application (Application, APP), that is, a program that needs to be installed in the operating system to run, such as a gate client; it can also be a small program that can be embedded in any APP, that is, a program that only needs to be downloaded to the browser environment to run. In short, the above-mentioned computer executable instructions can be instructions in any form, and the above-mentioned computer program can be an application, module or plug-in in any form.

[0060] The electronic device for implementing the gate control method provided by the embodiment of the present application is described below. Figure 2 , Figure 2 Schematic diagram of the structure of the electronic device provided in the embodiment of the present application. The electronic device 500 provided in the embodiment of the present application can be a gate device or a server. Figure 2 As shown, the electronic device 500 includes: at least one processor 510, a memory 550, at least one network interface 520 and a user interface 530. The various components in the electronic device 500 are coupled together through a bus system 540. It can be understood that the bus system 540 is used to realize the connection and communication between these components. In addition to the data bus, the bus system 540 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, the bus system 540 is not used in the following examples. Figure 2 Various buses are labeled as bus system 540 .

[0061] The processor 510 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., where the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0062] The user interface 530 includes one or more output devices 531 that enable presentation of media content, including one or more speakers and / or one or more visual display screens. The user interface 530 also includes one or more input devices 532, including user interface components that facilitate user input, such as a keyboard, mouse, microphone, touch screen display, camera, other input buttons and controls.

[0063] The memory 550 may be removable, non-removable, or a combination thereof. The memory 550 may include one or more storage devices physically located away from the processor 510. The memory 550 includes a volatile memory or a non-volatile memory, and may also include both a volatile memory and a non-volatile memory. The non-volatile memory may be a read-only memory (ROM), and the volatile memory may be a random access memory (RAM). The memory 550 described in the embodiments of the present application is intended to include any suitable type of memory.

[0064] In some embodiments, the memory 550 can store data to support various operations, examples of which include programs, modules, and data structures, or a subset or superset thereof, as exemplarily described below.

[0065] Operating system 551, including system programs for processing various basic system services and performing hardware-related tasks, such as framework layer, core library layer, driver layer, etc., for implementing various basic services and processing hardware-based tasks;

[0066] A network communication module 552, used to reach other electronic devices via one or more (wired or wireless) network interfaces 520, exemplary network interfaces 520 include: Bluetooth, Wireless Compatibility Certification (WiFi), and Universal Serial Bus (USB), etc.;

[0067] a presentation module 553 for enabling presentation of information via one or more output devices 531 (e.g., display screen, speaker, etc.) associated with the user interface 530 (e.g., a user interface for operating peripherals and displaying content and information);

[0068] The input processing module 554 is used to detect one or more user inputs or interactions from one of the one or more input devices 532 and translate the detected inputs or interactions.

[0069] In some embodiments, the gate control device provided in the embodiments of the present application can be implemented in a software manner. Figure 2 A gate control device 555 stored in a memory 550 is shown, which may be software in the form of a program and a plug-in, including the following software modules: an acquisition module 5551, a determination module 5552, an activation module 5553 and a control module 5554. These modules are logical, and therefore may be arbitrarily combined or further split according to the functions implemented. The functions of each module will be described below.

[0070] The gate control method provided by the embodiment of the present application is described below. As mentioned above, the gate control method provided by the embodiment of the present application can be implemented by the server or the gate device alone, or by the server and the gate device in collaboration. Therefore, the execution subject of each step will not be repeatedly described below. Figure 3 , Figure 3 : is a flow chart of a gate control method provided in an embodiment of the present application. The gate control method provided in an embodiment of the present application includes:

[0071] Step 101: Acquire an object image of a target object to be passed through a gate device.

[0072] Among them, the gate equipment supports at least two modes of passage.

[0073] Here, a client may be provided in the electronic device, such as a gate client. When the electronic device runs the client, the gate device may be controlled to be in a passable state or an impassable state. In practical applications, the gate device supports at least two access modes, such as a first pass card access mode (such as a one-card pass, a one-way ticket, NFC), a second pass card access mode (such as a virtual pass card supported by the terminal, a simulated pass card, etc.), a biometric access mode (such as facial information recognition (such as face and iris), hand information recognition (such as palm prints and fingerprints)), a graphic code access mode (such as a gate pass client boarding code, a mini program boarding code), a password access mode, etc. The target object may be a user to be passed through the gate device.

[0074] In the embodiment of the present application, when the gate device detects a target object to be passed through the gate device, an object image of the target object is acquired in advance, thereby determining the target passage mode to be used by the target object in advance according to the object image, so as to control the activation and silencing of the passage processing units of various passage modes. Therefore, in step 101, an object image of the target object to be passed through the gate device can be acquired.

[0075] In some embodiments, the gate device is equipped with an image acquisition unit; based on this, the object image of the target object to be passed through the gate device can be obtained in the following manner: through the image acquisition unit, the image of the target object to be passed through the gate device is acquired to obtain the object image of the target object.

[0076] Here, when acquiring the object image of the target object, it can be acquired through an image acquisition unit. In practical applications, the image acquisition unit can be installed on the gate device, so that the image acquisition unit can be used to acquire the image of the target object passing through the gate device to obtain the object image of the target object. In practical implementation, the image acquisition unit can be a 2D camera or a 3D camera, such as a depth camera, an infrared camera, etc.

[0077] In some embodiments, see Figure 4 , after obtaining the object image of the target object to be passed through the gate device, the following steps can also be performed: Step 201: Perform human biometric information detection on the object image to obtain the biometric information image area in the object image; Step 202: Blur the biometric information image area to obtain the target object image; Based on this, image recognition can be performed on the object image in the following manner to obtain a recognition result: Image recognition is performed on the target object image to obtain a recognition result.

[0078] Here, after acquiring the object image of the target object, the human biometric information of the object image can also be desensitized to protect the security and privacy of the human biometric information (such as facial information, iris information, fingerprint information, palm print information, etc.) of the object in the object image. Specifically, firstly, human biometric information detection is performed on the object image to obtain the biometric information image area in the object image, and then the biometric information image area is blurred to obtain the target object image. In this way, when image recognition is performed on the object image later, image recognition is performed on the target object image.

[0079] In step 201, when performing human biometric information detection, hand detection can be performed on the object image. When a hand region is detected in the object image, the hand region is used as the biometric information image region. Face detection can also be performed on the object image. When a face region is detected in the object image, the face region is used as the biometric information image region. Both hand detection and face detection can be implemented using corresponding pre-trained machine learning models.

[0080] Taking face detection as an example, an initial face detection model can be pre-built, and the initial face detection model has initial model parameters. The initial face detection model can be built based on convolutional neural networks, deep neural networks, recurrent neural networks, etc. Then, image samples for training the initial face detection model are obtained, and each image sample is annotated with a corresponding sample label; the image sample is then input into the initial face detection model, and the detection result is output by the initial face detection model, so that the initial face detection model is trained based on the difference between the detection result and the sample label to update the model parameters of the initial face detection model and obtain the face detection model. In this way, face detection can be performed on the object image based on the face detection model to obtain the face area in the object image.

[0081] Taking hand detection as an example, an initial hand detection model can be pre-constructed, and the initial hand detection model has initial model parameters. The initial hand detection model can be constructed based on convolutional neural networks, deep neural networks, recurrent neural networks, etc. Then, image samples for training the initial hand detection model are obtained, and each image sample is annotated with a corresponding sample label; the image sample is then input into the initial hand detection model, and the detection result is output through the initial hand detection model, so that the initial hand detection model is trained based on the difference between the detection result and the sample label to update the model parameters of the initial hand detection model and obtain the hand detection model. In this way, hand detection can be performed on the object image based on the hand detection model to obtain the hand area in the object image.

[0082] In step 202, when blurring the biological information image region, the biological information image region may be filtered (such as edge-preserving filtering) to reduce the clarity of the filtered biological information image region; the biological information image region may also be smoothed to reduce the clarity of the filtered biological information image region.

[0083] In practical applications, the collected object image of the target object can be used only to determine the target's mode of passage, and may not be retained or uploaded to the cloud to ensure the privacy and security of the object image of the target object.

[0084] Step 102: Perform image recognition on the object image to obtain a recognition result, and determine the target passage mode to be used by the target object based on the recognition result.

[0085] Among them, at least two passage modes include a target passage mode.

[0086] In step 102, with respect to the acquired object image of the target object, image recognition may be performed on the object image to obtain a recognition result, and based on the recognition result, a target passage mode to be used when the target object passes through the gate device is determined.

[0087] In some embodiments, see Figure 5, the object image can be subjected to image recognition through the following steps 401-403 to obtain a recognition result: step 401, performing object posture recognition on the object image to obtain the object posture of the target object; step 402, when the object posture is not a hand-held object posture, the hand movement of the target object is recognized to obtain a first recognition result, and the first recognition result indicates whether the hand movement is one of the following movements: a palm extension movement for the hand information recognition unit, a password input movement for the password input unit; step 403, when the object posture is a hand-held object posture, the hand-held object of the target object is recognized to obtain a second recognition result, and the second recognition result indicates whether the hand-held object is one of the following objects: a pass card, a mobile terminal with a display screen in an on state, and a mobile terminal with a display screen in an off state; wherein the recognition result includes the first recognition result and the second recognition result.

[0088] In step 401, the posture recognition model can be called to perform object posture recognition on the object image to obtain the object posture of the target object. In practical applications, the posture recognition model can be called to 1) perform human body detection processing on the object image to find the human body parts of the object image, including the head, body, arms, etc.; 2) perform key point detection on the human body parts to find the positions of the key points of the human body parts, such as shoulders, elbows, palms, fingers, face, etc.; 3) perform posture estimation processing on the positions of the key points to obtain the human body posture, such as whether the target object's arm is straight or bent, whether the arm is raised, etc.; 4) perform posture classification on the human body posture to obtain the object posture of the target object, such as the posture of a handheld object and the posture of a non-handheld object. The posture recognition model can be pre-trained. For example, an initial posture recognition model can be pre-constructed. The initial posture recognition model has initial model parameters. The initial posture recognition model can be constructed based on a convolutional neural network, a deep neural network, a recurrent neural network, etc.; then, image samples for training the initial posture recognition model are obtained, and each image sample is annotated with a corresponding sample label; the image sample is then input into the initial posture recognition model, and the posture recognition result is output by the initial posture recognition model, so that the initial posture recognition model is trained based on the difference between the posture recognition result and the sample label to update the model parameters of the initial posture recognition model and obtain the posture recognition model.

[0089] In step 402, if the object posture is not a hand-held object posture, the hand movement of the target object can be identified to obtain a first recognition result. Here, the first recognition result indicates whether the hand movement is one of the following movements: a palm extension movement for the hand information recognition unit, a password input movement for the password input unit. It should be noted that the hand information recognition unit is used to identify hand information (such as palm prints, fingerprints) to support the hand information recognition pass mode; the password input unit is used for the target object to input a pass password from the gate device to support the password pass mode. In actual applications, the recognition of the hand movement can also be implemented using a pre-trained hand movement recognition model. For example, an initial hand motion recognition model can be constructed in advance, which has initial model parameters. The initial hand motion recognition model can be constructed based on a convolutional neural network, a deep neural network, a recurrent neural network, etc.; then, image samples for training the initial hand motion recognition model are obtained, and each image sample is annotated with a corresponding sample label; the image sample is then input into the initial hand motion recognition model, and the hand motion recognition result is output through the initial hand motion recognition model, so that the initial hand motion recognition model is trained based on the difference between the hand motion recognition result and the sample label to update the model parameters of the initial hand motion recognition model and obtain the hand motion recognition model.

[0090] In step 403, if the object posture is a handheld object posture, the handheld object of the target object can be identified to obtain a second identification result. Here, the second identification result indicates whether the handheld object is one of the following objects: a pass card, a mobile terminal with a display screen turned on, or a mobile terminal with a display screen turned off. In practical applications, the identification of the handheld object can also be implemented using a pre-trained handheld object identification model.

[0091] In some embodiments, see Figure 6 , step 403 can be implemented through the following steps 601-603: step 601, performing hand detection on the object image to obtain the hand area in the object image; step 602, performing object detection on the hand area to obtain the object area of ​​the hand area; step 603, performing object classification prediction on the object area to obtain the object classification result, and using the object classification result as the second recognition result.

[0092] In step 601, hand detection can also be implemented using a pre-trained hand detection model. Of course, it can also be implemented in the following ways: extracting features from the object image to obtain object image features, and obtaining pre-stored hand features; matching the object image features with the pre-stored hand features to determine the hand region in the object image based on the matching results, that is, the region in the object image where the object image features and the hand features match consistently is taken as the hand region. In step 602, object detection is continued on the hand region to obtain the object region of the hand region, and object detection can be implemented using a pre-trained object detection model. In step 603, object classification prediction is performed on the object region to obtain an object classification result, which indicates whether the handheld object is one of the following objects: a pass card, a mobile terminal with a display screen on, or a mobile terminal with a display screen off. In this way, the object classification result is the second recognition result.

[0093] In some embodiments, step 403 can also be implemented by the following steps: performing hand detection on the object image to obtain a hand area in the object image; performing a pass card detection on the hand area to obtain a first detection result; when the first detection result indicates that the handheld object in the hand area is a pass card, using the first detection result as the second recognition result; when the first detection result indicates that the handheld object in the hand area is not a pass card, performing a mobile terminal detection on the hand area to obtain a second detection result; when the second detection result indicates that the handheld object in the hand area is a mobile terminal, performing a screen status detection on the mobile terminal in the hand area to obtain a third detection result; using the third detection result as the second recognition result, the third detection result indicates that the display screen of the mobile terminal is in one of the following states: on state, off state.

[0094] Here, hand detection can also be implemented using a pre-trained hand detection model. Of course, it can also be implemented in the following way: extracting features from the object image to obtain object image features, and obtaining pre-stored hand features; matching the object image features with the pre-stored hand features to determine the hand area in the object image based on the matching result, that is, the area in the object image where the object image features and the hand features match consistently is taken as the hand area. When performing pass card detection, if the first detection result indicates that the handheld object in the hand area is a pass card, then the first detection result is taken as the second recognition result, that is, the handheld object of the target object is detected to be a pass card. If the first detection result indicates that the handheld object in the hand area is not a pass card, it is necessary to continue to detect the hand area, and at this time, the mobile terminal detection is performed on the hand area to obtain the second detection result; further, since the states of the display screen of the mobile terminal are different, corresponding to different modes of passage, therefore, when the second detection result indicates that the handheld object in the hand area is a mobile terminal, the screen state detection is performed on the mobile terminal in the hand area to determine whether the display screen of the mobile terminal is in the on state or in the off state.

[0095] In some embodiments, when the recognition result is the first recognition result, based on the recognition result, the target access mode to be used by the target object can be determined in the following manner: when the first recognition result represents that the hand movement is a palm extension movement, the target access mode to be used by the target object is determined to be a hand information recognition access mode; when the first recognition result represents that the hand movement is a password input movement, the target access mode to be used by the target object is determined to be a password access mode; when the first recognition result represents that the hand movement is neither a palm extension movement nor a password input movement, the target access mode to be used by the target object is determined to be a facial information recognition access mode.

[0096] It should be noted that the hand information recognition access method is a access method for recognizing hand information, and the hand information may be palm prints, fingerprints, etc. The password access method is a access method for verifying whether the password entered by the user is correct. The facial information recognition access method is a access method for recognizing facial information, and the facial information may be face, iris, etc.

[0097] In some embodiments, when the recognition result is the second recognition result, based on the recognition result, the target pass mode to be used by the target object can be determined in the following manner: when the second recognition result characterizes that the handheld object is a pass card, the target pass mode to be used by the target object is determined to be the first pass card pass mode; when the second recognition result characterizes that the handheld object is a mobile terminal with a display screen on, the target pass mode to be used by the target object is determined to be the graphic code pass mode; when the second recognition result characterizes that the handheld object is a mobile terminal with a display screen on, the target pass mode to be used by the target object is determined to be the second pass card pass mode.

[0098] It should be noted that the pass card of the first pass card access mode may include a one-card pass, a one-way ticket, NFC, etc.; the graphic code of the graphic code access mode may include a gate access client boarding code, a mini program boarding code, etc. The pass card of the second pass card access mode may include a virtual pass card or a simulated pass card supported by a mobile terminal.

[0099] It should be noted that the above-mentioned mobile terminal may include but is not limited to smart phones, tablet computers, smart watches, etc.

[0100] Step 103: activating a target passage processing unit of the target passage mode, and verifying the target passage mode based on the target passage processing unit.

[0101] In step 103, when it is determined that the target object passes through the communication mode to be used from the gate device as the target pass mode, the target pass processing unit of the target pass mode is activated, so that the target pass mode is verified based on the target pass processing unit. Here, each pass mode has a corresponding pass processing unit, which is the hardware (device) and program that supports the corresponding pass mode. When a pass processing unit is activated, the pass processing unit is controlled to be in an activated state and operate, so that the corresponding pass mode is verified based on the pass processing unit. Here, when the target pass mode is verified based on the target pass processing unit, the pass verification strategy corresponding to the target pass mode can be used for verification. Each pass mode is provided with a corresponding communication verification strategy.

[0102] In actual applications, if the target access method is the first access card access method (such as a card, a one-way ticket, NFC), then it can be verified whether the access card held by the target object is a access card for passing through the gate device, whether the balance of the access card is sufficient, etc.; if the target access method is the second access card access method (such as a virtual access card, a simulated access card, etc. supported by the terminal), then it can be verified whether the mobile terminal held by the target object is the binding terminal of the virtual access card or the simulated access card, whether the binding relationship is valid, whether the balance of the virtual access card or the simulated access card is sufficient, etc.; if the target access method is the facial information recognition access method, then it can be verified whether the facial information of the target object (such as face or iris) and the target facial information bound to the target object (pre-set facial information for passing through the gate device) match, The validity of the target facial information, whether the balance of the payment method bound to the target facial information is sufficient, etc.; if the target access method is a hand information recognition access method, then it can be verified whether the target object's hand information (such as palm prints or fingerprints) and the target hand information bound to the target object (the hand information pre-set for passing the gate device) match, the validity of the target hand information, whether the balance of the payment method bound to the target hand information is sufficient, etc.; if the target access method is a graphic code access method (such as a mini program boarding code), then it can be verified whether the graphic code is a graphic code that passes the gate device, the validity of the graphic code, whether the balance of the payment method bound to the graphic code is sufficient, etc.; if the target access method is a password access method, then it can be verified whether the entered password is the pre-set target password, etc.

[0103] In some embodiments, after the target passage processing unit of the target passage mode is activated, the following processing may be further performed: outputting activation prompt information, where the activation prompt information is used to prompt that the target passage processing unit of the target passage mode has been activated.

[0104] In practical applications, the activation prompt information may be audio, text (e.g., displayed on a display screen of the gate device), indicator light, etc. For example, an indicator light for each pass processing unit (or pass mode) may be set on the gate device. When the pass processing unit is activated, the indicator light of the pass processing unit may light up, flash, etc., and when the pass processing unit is muted, the indicator light of the pass processing unit may remain dark. In this way, the user can clearly understand the activated pass mode.

[0105] Step 104: When the target passage mode is verified by the target passage processing unit, the gate device is controlled to be in a passable state to allow the target object to pass through the gate device.

[0106] In step 104, if the target passage mode is verified by the target passage processing unit, the gate device is controlled to be in a passable state, for example, the gate of the gate device can be opened. At this time, the target object can pass through the gate device.

[0107] In actual applications, if the target pass mode is the first pass mode (such as a card, a one-way ticket, NFC), then it can be verified whether the pass card held by the target object is a pass card that passes the gate device, whether the balance of the pass card is sufficient, etc.; when the verification result characterizes that the pass card held by the target object is a pass card that passes the gate device and the balance of the pass card is sufficient, then it is determined that the target pass mode verification is passed based on the target pass processing unit.

[0108] If the target pass mode is a second pass mode (such as a virtual pass card, simulated pass card, etc. supported by the terminal), then it can be verified whether the mobile terminal held by the target object is the binding terminal of the virtual pass card or the simulated pass card, whether the binding relationship is valid, whether the balance of the virtual pass card or the simulated pass card is sufficient, etc.; when the verification result characterizes that the mobile terminal held by the target object is the binding terminal of the virtual pass card or the simulated pass card, the binding relationship is valid, and the balance of the virtual pass card or the simulated pass card is sufficient, then it is determined that the target pass mode verification is passed based on the target pass processing unit.

[0109] If the target access method is a facial information recognition access method, then it can be verified whether the facial information of the target object (such as face or iris) and the target facial information bound to the target object (facial information pre-set for passing through the gate device) match, the validity of the target facial information, whether the balance of the payment method bound to the target facial information is sufficient, etc.; when the verification result characterizes that the facial information of the target object and the target facial information bound to the target object match, the target facial information is valid, and the balance of the payment method bound to the target facial information is sufficient, then it is determined that the target access method is verified based on the target access processing unit.

[0110] If the target access method is a hand information recognition access method, then it can be verified whether the target object's hand information (such as palm prints or fingerprints) and the target hand information bound to the target object (the hand information pre-set for passing through the gate device) match, the validity of the target hand information, whether the balance of the payment method bound to the target hand information is sufficient, etc.; when the verification result indicates that the target object's hand information and the target hand information bound to the target object match, the target hand information is valid, and the balance of the payment method bound to the target hand information is sufficient, then it is determined that the target access method is verified based on the target access processing unit.

[0111] If the target access method is a graphic code access method (such as a mini-program boarding code), then it can be verified whether the graphic code is a graphic code that passes the gate device, the validity of the graphic code, whether the balance of the payment method bound to the graphic code is sufficient, etc.; when the verification result characterizes that the graphic code is a graphic code that passes the gate device, the graphic code is valid, and the balance of the payment method bound to the graphic code is sufficient, then it is determined that the target access method is verified based on the target access processing unit.

[0112] If the target access mode is a password access mode, then it can be verified whether the input password is a preset target password, etc. When the verification result indicates that the input password is a preset target password, then it is determined that the target access mode verification based on the target access processing unit is passed.

[0113] In some embodiments, after controlling the gate device to be in a passable state, the following processing can also be performed: detecting whether the silence condition of the target passage processing unit is met; when it is detected that the silence condition of the target passage processing unit is met, the target passage processing unit that controls the target passage mode is in a silent state.

[0114] Here, after the target passage processing unit is activated, it is also possible to detect in real time or periodically whether the silence condition of the target passage processing unit is met; when it is detected that the silence condition of the target passage processing unit is met, the target passage processing unit of the target passage mode can be controlled to be in a silent state. In this way, the occupation and waste of hardware processing resources can be reduced, the hardware processing efficiency can be improved, and the gate passing efficiency can be improved, which is convenient for the next user to pass through the gate device.

[0115] In practical applications, the silent condition may be whether the target object has passed through the gate device, whether the activation time of the target pass processing unit has reached the activation time threshold, etc. Thus, when the target object has passed through the gate device, it is determined that the silent condition of the target pass processing unit is satisfied; when the activation time of the target pass processing unit reaches the activation time threshold, it is determined that the silent condition of the target pass processing unit is satisfied.

[0116] In some embodiments, see Figure 7After controlling the gate device to be in a passable state, the following steps 301-303 can also be executed: Step 301, when the frequency of the target object passing through the gate device in the first time period reaches the frequency threshold, the pass mode data of the target object in the first time period is obtained; Step 302, based on the pass mode data, determine the first pass mode of the target object in the second time period, the at least two pass modes include the first pass mode, and the second time period is located after the first time period; Step 303, when the target object passes through the gate device in the second time period, activate the first pass processing unit of the first pass mode.

[0117] Here, for the target object, the traffic mode data of the target object in the first time period can be recorded, that is, the relevant data of the traffic mode of the target object through the gate device, such as the usage frequency, usage time point, usage cycle, etc. of each traffic mode. The first time period can be preset. Based on this, in step 301, when the traffic frequency of the target object passing through the gate device in the first time period reaches the frequency threshold, the traffic mode data of the target object in the first time period can be obtained.

[0118] In step 302, the traffic mode data of the target object in the first time period can be predicted to obtain the first traffic mode of the target object in the second time period. For example, a pre-trained traffic mode prediction model can be called to predict the traffic mode data to obtain the first traffic mode of the target object in the second time period. In practical applications, an initial traffic mode prediction model can be pre-constructed, and the initial traffic mode prediction model has initial model parameters. The initial traffic mode prediction model can be constructed based on a convolutional neural network, a deep neural network, a recurrent neural network, etc.; then a training sample (i.e., a traffic mode data sample in a period of time) for training the initial hand detection model is obtained, and the training sample is annotated with a corresponding sample label; then the image sample is input into the initial traffic mode prediction model, and the prediction result is output by the initial traffic mode prediction model, so that the initial traffic mode prediction model is trained based on the difference between the prediction result and the sample label to update the model parameters of the initial traffic mode prediction model to obtain the traffic mode prediction model. In this way, the traffic mode data in the first time period can be predicted based on the traffic mode prediction model to obtain the first traffic mode of the target object in the second time period. It should be noted that the second time period is located after the first time period, and the second time period may be adjacent to the first time period or may not be adjacent to the first time period.

[0119] In step 303, when the target object passes through the gate device within the second time period, the first passage processing unit of the first passage mode is activated to support the target object to pass through the gate device based on the first passage mode through the first passage processing unit.

[0120] It should be noted that the gate equipment is a channel blocking device (channel management equipment) used to manage the flow of people and regulate the entry and exit of pedestrians; it can be applied to urban rail transit management, toll gate systems, access control systems, etc.

[0121] Applying the above-mentioned embodiment of the present application, first, an object image of the target object to be passed through the gate device is obtained, and the gate device supports at least two passage modes; then, image recognition is performed on the object image to obtain a recognition result, and based on the recognition result, the target passage mode to be used by the target object is determined; then, the target passage processing unit of the target passage mode is activated, and the target passage mode is verified based on the target passage processing unit; thereby, when the target passage mode is verified based on the target passage processing unit, the gate device is controlled to be in a passable state to allow the target object to pass through the gate device.

[0122] Here, when the target object passes through the gate device, the target passage mode to be used by the target object will be determined in advance based on the object image of the target object, so that only the target passage processing unit of the target passage mode is activated to support the target object to pass through the gate device based on the target passage mode; in this way, the occupation and waste of hardware processing resources are reduced, the hardware processing efficiency is improved, and the gate passage efficiency is improved.

[0123] The following uses the passage of subway gates as an example to illustrate the exemplary application of the embodiments of the present application in an actual application scenario. A subway gate often centrally supports multiple modes of passage, such as pass cards (such as one-card passes, one-way tickets, NFC), biometrics (such as face scanning, palm scanning), graphic codes (such as mini-program boarding codes), etc. In the related art, in order to satisfy that each user can use the desired mode of passage to pass through the subway gate, the subway gate will keep the passage processing unit of each mode of passage in an activated state. However, when users pass through the subway gate, they only use one mode of passage. In the related art, not only will the hardware processing resources be over-occupied and wasted, but the hardware processing efficiency will also be reduced. As a result, when a large number of users queue up to pass through the subway gate one by one, it is easy to cause traffic congestion, which seriously affects the passage efficiency of the subway gate.

[0124] Based on this, the embodiment of the present application provides a gate control method to at least solve the above-mentioned problems. In the embodiment of the present application, based on the scenario of users queuing up to pass through the subway gate one by one, a central control camera is added to the subway gate, and the user's body posture and handheld objects are obtained through the central control camera, and the user's target passage mode this time is judged according to the body posture and handheld objects, so as to control the operation of the devices of different passage processing units according to the target passage mode, so as to regulate the activation and silencing of various passage modes, effectively avoid users from swiping by mistake, and improve the passage efficiency of the subway gate. The embodiment of the present application is described in detail below.

[0125] See also Figure 8 , Figure 8 Schematic diagram of the gate control system provided by the embodiment of the present application. Here, the gate control system includes: a registration and ticket purchase device end, a subway gate end, and a background service end, which are described below.

[0126] (I) Registration and ticket purchase device. Figure 8 As shown, the registration and ticket purchase device includes a one-way ticket unit, a recharge unit, a biometric activation unit, and an authorization unit.

[0127] (1) One-way ticket unit. The one-way ticket unit is used for users to purchase one-way tickets for the subway. Users use the purchased one-way tickets to swipe the card at the subway gate.

[0128] (2) Recharge unit. The recharge unit is used for users to recharge the pass card. Users can follow the operation prompts displayed on the screen at the registration and ticket purchase device to recharge the pass card, such as selecting a language, entering a recharge amount, payment method, etc.

[0129] (3) Biometric activation unit. When a user activates palm or face recognition to pass through the gate, he or she needs to complete the collection of biometric information (such as palm information, facial information), confirmation and consent of relevant service agreements, real-name authorization, and contract and bind the contract through the biometric activation unit, so that the biometric information can be used for subsequent gate passing.

[0130] (4) Authorization unit. The authorization unit is used to prompt the user to authorize the following operations in the subway scene: collect information (such as user posture, objects held by the user, etc.) through the central control camera to determine the user's current mode of travel in advance. At the same time, the authorization unit will also prompt the user that no data is retained for the information collected by the central control camera, and the data will be processed after desensitization and destroyed on use.

[0131] (ii) Subway gate terminal. Figure 8As shown, the subway gate terminal includes a central control camera, a central control controller, and a plurality of passage processing units (such as a biometric passage method (including the above-mentioned hand information recognition passage method and facial information passage method, such as face swiping and palm swiping), a card swiping passage method (i.e. the above-mentioned first pass card passage method, including a card, a one-way ticket and NFC), a code swiping passage method (i.e. the above-mentioned graphic code passage method, etc.). Among them, the central control camera can be a 3D camera. By installing a 3D camera on the subway gate, images or videos of users before entering the subway gate can be continuously collected, so that the user's body posture and handheld objects can be determined according to the collected images or videos. The user's target passage method this time can be determined according to the body posture and the handheld object, so that according to the target passage method, the central control controller is called to control the operation of different passage processing units (i.e., devices that realize the corresponding passage methods) to regulate the activation and silencing of various passage methods.

[0132] In practical applications, obtaining human posture from collected images or videos can be achieved through posture recognition algorithms. Posture recognition algorithms are a technology that uses image or video data to recognize human posture, including the following steps:

[0133] 1) Data collection: Use a 3D camera to collect user's image or video data.

[0134] 2) Human body detection: Use human body detection algorithms to process image or video data to find human body parts, including head, body, arms, legs, etc.

[0135] 3) Key point detection: Use the key point detection algorithm to further process the human body parts and find the locations of various key points, such as shoulders, elbows, waist, knees, etc.

[0136] 4) Posture estimation: Use the posture estimation algorithm to analyze and calculate the position of key points to derive the human body posture, such as whether the user is sitting or standing, whether the arms are straight or bent, etc.

[0137] 5) Posture classification: Classify human body postures, such as upright posture, forward leaning posture, sideways posture, etc., for subsequent application and analysis.

[0138] In actual implementation, deep learning technology can be used to perform tasks such as human body detection, key point detection, posture estimation and posture classification. For example, for each task, a machine learning model can be built through network structures such as convolutional neural network (CNN) and recurrent neural network (RNN), and the model can be trained and optimized using sample data sets, so as to achieve the corresponding task based on the trained machine learning model.

[0139] Combine the following Fig. 9 The gate control method provided in the embodiment of the present application is described. Fig. 9 is a flow chart of the gate control method provided in the embodiment of the present application, such as Fig. 9 As shown, the gate control method provided in the embodiment of the present application includes:

[0140] 1. User authorization; 2. User approaches the gate; 3. The gate collects the user's image; 4. Perform posture recognition on the image to determine whether the user is holding an object based on the recognized human posture; 5. Determine whether the user is holding a card, if so, execute 6, if not, execute 7; 6. Activate the access processing unit of the access card access mode; 7. Determine whether the user is holding a mobile phone, if so, execute 8, if not, execute 10; 8. Determine whether the screen of the mobile phone is on, if so, execute 9, if not, execute 6; 9. Activate the access processing unit of the graphic code access mode; 10. Determine whether the user performs a palm-stretching action, if so, execute 11, if not, execute 12; 11. Activate the access processing unit of the hand information recognition access mode; 12. Activate the access processing unit of the facial information recognition access mode; 13. The process ends.

[0141] In some embodiments, on the subway gate, the passage processing unit of each passage mode is provided with an indicator light. When the passage mode is activated, the indicator light corresponding to the passage mode will be lit, and when the passage mode is muted, the indicator light corresponding to the passage mode will be off.

[0142] In some embodiments, in order to ensure data security, the data collected by the central control camera can be used only on the current subway gate, and will not be uploaded to the cloud. The subway gate terminal also does not perform any storage, and is only used to assist in determining the user's target mode of passage this time, and is not used for any retention. For the image or video data collected by the central control camera, the subway gate terminal can also desensitize sensitive biometric information in the data. For example, the desensitization of facial information in the data can be achieved through the following steps: a) Face detection: Perform face detection processing on each frame of the collected image to determine the face area in each frame of the image; b) Facial key point positioning: Perform face key point positioning processing on the face area to determine the face key points of the face area, such as eyes, nose, mouth, etc.; c) Face feature point extraction: Perform face feature point extraction processing on the face key points to obtain face feature points, such as corners of the eyes, corners of the lips, etc.; d) Face area blurring: According to the face feature points, blur the face area to obtain a desensitized image.

[0143] In some embodiments, whether the user is holding a card can be determined through the following steps: 1) Data acquisition: Collect the user's hand image through a 3D camera (it can be a single image or a video). 2) Hand detection: Perform hand detection on the image to determine the hand area in the image. 3) Card detection: Perform card detection on the hand area to determine the card area in the hand area. 4) Card recognition: Perform card recognition on the card area to obtain a recognition result, which is used to indicate whether the card area is a pass card. For example, image recognition is performed on the card area to determine the text or pattern in the card area, and then determine whether the card area is a pass card based on the recognized text or pattern. In practical applications, deep learning technology can be used to perform tasks such as hand detection, card detection, and card recognition. For example, for each task, a machine learning model can be constructed through network structures such as CNN and RNN, and the model can be trained and optimized using sample data sets, so as to achieve the corresponding task based on the trained machine learning model.

[0144] In some embodiments, the following steps can be used to determine whether the user is holding a mobile phone: 1) Data collection: Collect the user's hand image (which can be a single image or a video) through a 3D camera. 2) Hand detection: Perform hand detection on the image to determine the hand area in the image. 3) Object detection: Perform object detection on the hand area to determine the object area in the hand area, such as the mobile phone screen, mobile phone frame, etc. 4) Object recognition: Perform object recognition on the object area to determine whether the object in the object area is a mobile phone (including the mobile phone screen, mobile phone frame, etc.). 5) Mobile phone status judgment: According to the result of object recognition and the placement of the mobile phone, judge the status of the mobile phone, that is, whether the user is holding the mobile phone, whether the mobile phone screen is facing up, whether the screen is lit, etc. In practical applications, deep learning technology can be used to perform tasks such as hand detection, object detection, and object recognition. For example, for each task, a machine learning model can be constructed through a network structure such as CNN, target detection network (such as Faster R-CNN, YOLO), and the model can be trained and optimized using a sample data set, so as to achieve the corresponding task based on the trained machine learning model.

[0145] In some embodiments, the palm swiping method can be implemented through the following steps: 1) Collecting the palm streaming media data of the user through a 3D camera; 2) Filtering the palm streaming media data. For example, according to the palm size, angle, image contrast, image brightness and clarity coefficient indicators in the palm streaming media data, filter out the palm image with the highest image quality. 3) Perform identity recognition on the palm image to determine whether the user can pass the current subway gate.

[0146] By applying the above-mentioned embodiments of the present application, a central control camera is added to the subway gate. The user's posture and handheld objects are obtained through the central control camera, and the user's target passage mode is judged based on the user's posture and handheld objects. Therefore, according to the target passage mode, the devices of different passage processing units are controlled to operate, so as to regulate the activation and silencing of various passage modes, effectively avoid users from swiping by mistake, and improve the passage efficiency of the subway gate.

[0147] The following is a description of an exemplary structure of the gate control device 555 provided in the embodiment of the present application implemented as a software module. In some embodiments, Figure 2 As shown, the software modules stored in the gate control device 555 of the memory 550 may include: an acquisition module 5551, used to acquire an object image of a target object to be passed through the gate device, the gate device supports at least two passage modes; a determination module 5552, used to perform image recognition on the object image to obtain a recognition result, and based on the recognition result, determine the target passage mode to be used by the target object, the at least two passage modes include the target passage mode; an activation module 5553, used to activate a target passage processing unit of the target passage mode, and verify the target passage mode based on the target passage processing unit; a control module 5554, used to control the gate device to be in a passable state when the target passage mode is verified based on the target passage processing unit, so as to allow the target object to pass through the gate device.

[0148] In some embodiments, the gate device is equipped with an image acquisition unit; the acquisition module 5551 is also used to acquire an image of a target object passing through the gate device through the image acquisition unit to obtain an object image of the target object.

[0149] In some embodiments, the determination module 5552 is also used to perform object posture recognition on the object image to obtain the object posture of the target object; when the object posture is not a hand-held object posture, the hand movement of the target object is recognized to obtain a first recognition result, and the first recognition result indicates whether the hand movement is one of the following actions: a palm extension action for a hand information recognition unit, a password input action for a password input unit; when the object posture is a hand-held object posture, the hand-held object of the target object is recognized to obtain a second recognition result, and the second recognition result indicates whether the hand-held object is one of the following objects: a pass card, a mobile terminal with a display screen on, and a mobile terminal with a display screen off; wherein the recognition result includes the first recognition result and the second recognition result.

[0150] In some embodiments, when the recognition result is the first recognition result, the determination module 5552 is also used to determine that the target access mode to be used by the target object is the hand information recognition access mode when the first recognition result represents that the hand movement is the palm extension movement; when the first recognition result represents that the hand movement is the password input movement, determine that the target access mode to be used by the target object is the password access mode; when the first recognition result represents that the hand movement is neither the palm extension movement nor the password input movement, determine that the target access mode to be used by the target object is the facial information recognition access mode.

[0151] In some embodiments, when the recognition result is the second recognition result, the determination module 5552 is also used to determine that the target pass mode to be used by the target object is the first pass mode when the second recognition result represents that the handheld object is a pass card; when the second recognition result represents that the handheld object is a mobile terminal with a display screen on, determine that the target pass mode to be used by the target object is a graphic code pass; when the second recognition result represents that the handheld object is a mobile terminal with a display screen on, determine that the target pass mode to be used by the target object is the second pass mode.

[0152] In some embodiments, the determination module 5552 is also used to perform hand detection on the object image to obtain a hand region in the object image; perform object detection on the hand region to obtain an object region in the hand region; perform object classification prediction on the object region to obtain an object classification result, and use the object classification result as the second recognition result.

[0153] In some embodiments, the determination module 5552 is also used to perform hand detection on the object image to obtain a hand area in the object image; perform pass card detection on the hand area to obtain a first detection result; when the first detection result indicates that the handheld object in the hand area is a pass card, use the first detection result as the second recognition result; when the first detection result indicates that the handheld object in the hand area is not a pass card, perform mobile terminal detection on the hand area to obtain a second detection result; when the second detection result indicates that the handheld object in the hand area is a mobile terminal, perform screen status detection on the mobile terminal in the hand area to obtain a third detection result; use the third detection result as the second recognition result, and the third detection result indicates that the display screen of the mobile terminal is in one of the following states: on state, off state.

[0154] In some embodiments, the acquisition module 5551 is further used to, after acquiring the object image of the target object to be passed through the gate device, perform human biometric information detection on the object image to obtain a biometric information image area in the object image; blur the biometric information image area to obtain a target object image; the determination module 5552 is further used to perform image recognition on the target object image to obtain a recognition result.

[0155] In some embodiments, the activation module 5553 is further used to output activation prompt information after activating the target passage processing unit of the target passage mode, and the activation prompt information is used to prompt that the target passage processing unit of the target passage mode has been activated.

[0156] In some embodiments, the activation module 5553 is also used to detect whether the silence condition of the target passage processing unit is met; when it is detected that the silence condition of the target passage processing unit is met, the target passage processing unit of the target passage mode is controlled to be in a silent state.

[0157] In some embodiments, the activation module 5553 is also used to obtain the passage mode data of the target object in a first time period when the passage frequency of the target object passing through the gate device in the first time period reaches a frequency threshold; based on the passage mode data, determine the first passage mode of the target object in a second time period, the at least two passage modes include the first passage mode, and the second time period is located after the first time period; when the target object passes through the gate device in the second time period, activate the first passage processing unit of the first passage mode.

[0158] It should be noted that the description of the device embodiment in this application is similar to the description of the above method embodiment, and has similar beneficial effects as the method embodiment, which will not be described in detail here. The unfinished technical details in the gate control device provided in the embodiment of this application can be understood based on the description of the technical details in the above method embodiment.

[0159] The embodiment of the present application also provides a computer program product, which includes computer executable instructions or computer programs, and the computer executable instructions or computer programs are stored in a computer readable storage medium. The processor of the electronic device reads the computer executable instructions or computer program from the computer readable storage medium, and the processor executes the computer executable instructions or computer program, so that the electronic device executes the gate control method provided in the embodiment of the present application.

[0160] The embodiment of the present application also provides a computer-readable storage medium, which stores computer-executable instructions or computer programs. When the computer-executable instructions or computer programs are executed by a processor, the processor will execute the gate control method provided by the embodiment of the present application.

[0161] In some embodiments, the computer-readable storage medium may be a memory such as RAM, ROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or may be various devices including one or any combination of the above memories.

[0162] In some embodiments, computer executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0163] As an example, computer-executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0164] As an example, computer executable instructions may be deployed to be executed on one electronic device, or on multiple electronic devices located at one site, or on multiple electronic devices distributed at multiple sites and interconnected by a communication network.

[0165] The above is only an embodiment of the present application and is not intended to limit the protection scope of the present application. Any modifications, equivalent substitutions and improvements made within the spirit and scope of the present application are included in the protection scope of the present application.

Claims

1. A gate control method, characterized in that: The method comprises: Acquire an object image of a target object to be passed through a gate device, wherein the gate device supports at least two passing modes; Performing image recognition on the object image to obtain a recognition result, and determining a target passage mode to be used by the target object based on the recognition result, wherein the at least two passage modes include the target passage mode; activating a target passage processing unit of the target passage mode, and verifying the target passage mode based on the target passage processing unit; When the target passage mode is verified to be passed based on the target passage processing unit, the gate device is controlled to be in a passable state so that the target object can pass through the gate device.

2. The method according to claim 1, characterized in that The gate device is equipped with an image acquisition unit; the acquisition of an object image of a target object to be passed through the gate device includes: The image acquisition unit is used to acquire an image of a target object to be passed through the gate device, thereby obtaining an object image of the target object.

3. The method according to claim 1, characterized in that The performing image recognition on the object image to obtain a recognition result includes: Performing object posture recognition on the object image to obtain the object posture of the target object; When the object posture is not a hand-held object posture, the hand motion of the target object is recognized to obtain a first recognition result, wherein the first recognition result indicates whether the hand motion is one of the following motions: a palm-stretching motion for a hand information recognition unit, or a password input motion for a password input unit; When the object posture is a handheld object posture, the handheld object of the target object is identified to obtain a second identification result, wherein the second identification result indicates whether the handheld object is one of the following objects: a pass card, a mobile terminal with a display screen turned on, or a mobile terminal with a display screen turned off; The recognition result includes the first recognition result and the second recognition result.

4. The method according to claim 3, characterized in that When the recognition result is the first recognition result, determining the target passage mode to be used by the target object based on the recognition result includes: When the first recognition result indicates that the hand motion is the palm-stretching motion, determining that the target passage mode to be used by the target object is a hand information recognition passage mode; When the first recognition result indicates that the hand motion is the password input motion, determining that the target access mode to be used by the target object is a password access mode; When the first recognition result indicates that the hand motion is not the palm-stretching motion and the password inputting motion, it is determined that the target access mode to be used by the target object is the facial information recognition access mode.

5. The method according to claim 3, characterized in that When the recognition result is the second recognition result, determining the target passage mode to be used by the target object based on the recognition result includes: When the second recognition result indicates that the handheld object is a pass card, determining that the target pass mode to be used by the target object is the first pass card pass mode; When the second recognition result indicates that the handheld object is a mobile terminal whose display screen is turned on, determining that the target access mode to be used by the target object is a graphic code access mode; When the second recognition result indicates that the handheld object is a mobile terminal whose display screen is in a closed state, it is determined that the target passage mode to be used by the target object is the second passage card passage mode.

6. The method according to claim 3, characterized in that The step of identifying the handheld object of the target object to obtain a second identification result includes: Performing hand detection on the object image to obtain a hand region in the object image; Performing object detection on the hand region to obtain an object region of the hand region; Perform object classification prediction on the object area to obtain an object classification result, and use the object classification result as the second recognition result.

7. The method according to claim 3, characterized in that The step of identifying the handheld object of the target object to obtain a second identification result includes: Performing hand detection on the object image to obtain a hand region in the object image; Performing a pass card detection on the hand area to obtain a first detection result; When the first detection result indicates that the handheld object in the hand area is a pass card, using the first detection result as the second recognition result; When the first detection result indicates that the handheld object in the hand area is not a pass card, performing mobile terminal detection on the hand area to obtain a second detection result; When the second detection result indicates that the handheld object in the hand area is a mobile terminal, performing screen status detection on the mobile terminal in the hand area to obtain a third detection result; The third detection result is used as the second recognition result, and the third detection result indicates that the display screen of the mobile terminal is in one of the following states: an on state and a off state.

8. The method according to claim 1, characterized in that After acquiring the object image of the target object to be passed through the gate device, the method further includes: Performing human body biometric information detection on the object image to obtain a biometric information image region in the object image; Performing fuzzy processing on the biological information image region to obtain a target object image; The performing image recognition on the object image to obtain a recognition result includes: performing image recognition on the target object image to obtain a recognition result.

9. The method according to claim 1, characterized in that After activating the target passage processing unit of the target passage mode, the method further includes: Output activation prompt information, where the activation prompt information is used to prompt that the target passage processing unit of the target passage mode has been activated.

10. The method according to claim 1, characterized in that After controlling the gate device to be in a passable state, the method further includes: Detecting whether a silence condition of the target traffic processing unit is satisfied; When it is detected that the silent condition of the target passage processing unit is satisfied, the target passage processing unit controlling the target passage mode is in a silent state.

11. The method according to claim 1, characterized in that After controlling the gate device to be in a passable state, the method further includes: When the frequency of the target object passing through the gate device in the first time period reaches a frequency threshold, obtaining the passing mode data of the target object in the first time period; Determine, based on the traffic mode data, a first traffic mode of the target object within a second time period, wherein the at least two traffic modes include the first traffic mode, and the second time period is after the first time period; When the target object passes through the gate device within the second time period, the first passage processing unit of the first passage mode is activated.

12. A gate control device, characterized in that: The device comprises: An acquisition module, used to acquire an object image of a target object to be passed through a gate device, wherein the gate device supports at least two passing modes; a determination module, configured to perform image recognition on the object image to obtain a recognition result, and determine a target passage mode to be used by the target object based on the recognition result, wherein the at least two passage modes include the target passage mode; an activation module, used for activating a target passage processing unit of the target passage mode, and verifying the target passage mode based on the target passage processing unit; The control module is used to control the gate device to be in a passable state when the target passage mode is verified by the target passage processing unit, so as to allow the target object to pass through the gate device.

13. An electronic device, characterized in that: The electronic device comprises: A memory for storing computer executable instructions; The processor is used to implement the gate control method described in any one of claims 1 to 11 when executing the computer executable instructions stored in the memory.

14. A computer-readable storage medium storing computer-executable instructions or a computer program, characterized in that: When the computer executable instructions or computer programs are executed by a processor, the gate control method described in any one of claims 1 to 11 is implemented.

15. A computer program product comprising computer executable instructions or a computer program, characterized in that When the computer executable instructions or computer programs are executed by a processor, the gate control method described in any one of claims 1 to 11 is implemented.

Images