Face and smart card combined identity recognition method, system, device and storage medium

Abstract

The application provides a face and intelligent card combined identity recognition method, system, device and storage medium, wherein the method comprises: triggering a double recognition mechanism when a user approaches an access control system, acquiring a face image collected by a multi-source sensing device, and reading intelligent card information; inputting the face image into a trained neural network model to output face features, wherein a feature extraction module of the neural network model processes face images collected by each sensing device in parallel, comparing the face features with a feature database to obtain a first verification result of the identity; comparing the intelligent card information with an information database to obtain a second verification result of the identity; acquiring the first verification result and the second verification result, and judging whether to start the access control according to a preset security strategy. The application combines face and intelligent card double recognition to improve the security of the access control system, can select a corresponding preset security strategy for verification according to the current environment, and considers the adaptability in special environments.

Description

Technical Field

[0001] This invention relates to the field of identity verification technology, and in particular to a method, system, device, and storage medium for identity recognition using a combination of face and smart card. Background Technology

[0002] With the rapid development of science and technology, intelligent security technology has become an indispensable part of modern society. Identity recognition refers to the process of confirming an individual's identity through human biometrics or behavioral characteristics.

[0003] In related technologies, traditional access control systems, such as those based on physical keys, passwords, or single biometric technologies, are gradually failing to meet the security needs of modern society. Existing intelligent access control systems typically employ a single authentication method, such as facial recognition or smart card recognition. Facial recognition technology identifies facial features using computer vision, while smart card recognition verifies user identity by reading information stored in the smart card. However, these technologies have drawbacks. The accuracy of facial recognition may be affected by environmental conditions such as insufficient light or facial occlusion. Furthermore, related image technologies often acquire facial images using only a single sensor device, but different devices are affected by varying lighting conditions, leading to recognition errors to some extent. For example, infrared images are not sensitive to changes in lighting, while visible light images have rich texture details but are greatly affected by lighting conditions. Moreover, existing feature extraction networks have high computational complexity; for example, the VGG-16 network includes three fully connected layers with a huge number of parameters. When extracting facial features, it often extracts redundant features unrelated to identity, increasing the burden of subsequent matching calculations and leading to a higher false recognition rate. Using only smart cards also poses security risks due to the possibility of lost or stolen smart cards.

[0004] Based on the above analysis of the development status of this technology field, the existing technology lacks a multi-functional access control technology that combines facial recognition and smart card recognition. When using facial recognition technology, a scheme is needed that uses multi-source sensing devices to acquire facial images, uses an optimized neural network model to output the results, and determines whether to pass the identity verification based on a preset security policy. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, device, and storage medium for identity recognition using a combination of face and smart card, in order to solve the aforementioned problems in the prior art.

[0006] According to a first aspect of the present invention, a face-based smart card identification method is provided, comprising: When a user approaches the access control system, a dual recognition mechanism is triggered to acquire the facial image collected by multi-source sensors and read the smart card information. The face image is input into the trained neural network model and the face features are output. The feature extraction module of the neural network model processes the face images collected by each sensing device in parallel, compares the face features with the feature database, and obtains the first verification result of identity. The smart card information is compared with the information database to obtain a second verification result of the identity. Obtain the first verification result and the second verification result, and combine them with the preset security policy to determine whether to activate the access control.

[0007] According to a second aspect of the present invention, a face-based smart card-based identity recognition system is provided, comprising: A camera is used to capture facial images from multiple sensor devices. The face recognition module is used to input face images into a trained neural network model and output face features. The feature extraction module of the neural network model processes the face images collected by each sensing device in parallel, compares the face features with the feature database, and obtains the first verification result of identity. The smart card recognition module is used to read smart card information, compare the smart card information with the information database, and obtain a second verification result of the identity. The control module is used to obtain the first verification result and the second verification result, and combine them with the preset security policy to determine whether to activate the access control. Access control mechanisms are used to execute instructions from the control module; Databases are used to store feature databases and information databases; The display module is used to display the recognition results.

[0008] According to a third aspect of the present invention, an electronic device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the face-based smart card identification method provided in the first aspect of the present disclosure.

[0009] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, on which an information transmission implementation program is stored, which, when executed by a processor, implements the steps of the face-based smart card identity recognition method provided in the first aspect of the present disclosure.

[0010] The technical solution provided by the embodiments of the present invention has the following beneficial effects: combining facial recognition and smart card recognition technologies for dual identification improves the security and reliability of the access control system. It can also select corresponding preset security strategies for verification based on the current environmental conditions, taking into account environmental adaptability in special environments. It automatically determines the user's identity and controls the opening and closing of the access control mechanism according to the preset security strategy, thus possessing intelligent management functions. When using facial recognition technology, multi-source sensing devices are used to acquire facial images, maintaining a high recognition accuracy rate in complex environments such as insufficient light and facial occlusion. The use of smart card recognition has strong anti-interference capabilities, ensuring normal operation in various environments.

[0011] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in one or more embodiments of this specification or in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of the face-based smart card identity recognition method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the basic verification process according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a face-based smart card-based identity recognition system according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the system design architecture according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the system application architecture according to an embodiment of the present invention; Figure 6 This is a schematic diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0014] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0015] Method Implementation Examples According to embodiments of the present invention, a method for identity recognition combining face recognition and smart cards is provided. Figure 1 This is a flowchart of the face-based smart card identity recognition method according to an embodiment of the present invention, such as... Figure 1 As shown, the face recognition method combined with smart card identification according to an embodiment of the present invention specifically includes: In step S110, when a user approaches the access control system, a dual identification mechanism is triggered to acquire the facial image collected by the multi-source sensing device and read the smart card information, specifically including: When a user stands in front of the access control device, image capture and smart card reading are automatically triggered. Color facial images are acquired using a visible light image sensor, and infrared facial images are acquired using a ToF image sensor to ensure image clarity and accuracy. The captured facial images of the user are then used for subsequent facial feature extraction.

[0016] Smart card information, including the card number and cardholder's name, can be obtained by decoding using RFID or NFC technology.

[0017] In step S120, the face image is input into the trained neural network model, which outputs face features. The feature extraction module of the neural network model processes the face images collected by each sensing device in parallel, compares the face features with the feature database, and obtains the first identity verification result, specifically including: Preferably, the face region can be detected in the initial face image to eliminate background interference before being input into the neural network model; The face image is input into the input layer of the neural network model. The input image resolution supports 128×128 and 224×224 pixels. The face image is normalized by normalizing the pixel value of the input image to the range of [-1,1], which is achieved by subtracting 128 from the pixel value and then dividing by 128. At the same time, the image channel alignment is completed to ensure the consistency of subsequent module processing, and the color processed image and the infrared processed image are obtained respectively. The color-processed image is input into the VGG-16 model for feature extraction, yielding the first extraction result, which includes: This invention improves upon the original VGG-16 architecture by removing the three fully connected layers and replacing them with two 1×1 convolutional kernels to achieve full connectivity while retaining feature localization capabilities. It contains a total of six convolutional blocks, into which the color-processed image is input into the VGG-16 model consisting of six convolutional blocks. The VGG-16 model extracts facial contour features through the first convolutional block, which consists of two 5×5 convolutional kernels with a padding value of 1 and a convolutional stride of 1. It is then connected to a 2×2 max pooling layer with a pooling stride of 2 to initially extract facial contour features and filter image noise. The second to fifth convolutional blocks each extract detailed facial features. The second convolutional block contains two 4×4 convolutional kernels, and the third to fifth convolutional blocks each contain three 4×4 convolutional kernels. All convolutional kernels have a padding value of 1 and a convolutional stride of 1. Each convolutional block is followed by a 2×2 max pooling layer with a pooling stride of 2. This process gradually extracts detailed facial features such as the corners of the eyes, the tip of the nose, the corners of the mouth, and deep texture features. The original VGG-16 model replaces the fully connected layer with a sixth convolutional block to integrate facial contour and facial feature details. Two 1×1 convolutional kernels are used to integrate the features extracted by the preceding convolutional blocks, outputting a 4×4 feature vector. Simultaneously, it outputs the facial region detection box parameters (b1x, b1y, b1w, ​​b1h), where b1x and b1y are the coordinates of the detection box center point, and b1w and b1h are the width and height of the detection box. An adaptive feature filtering unit is added to the output of the sixth convolutional block, using a dynamic threshold to filter redundant texture features. The threshold is adaptively adjusted according to the sharpness of the input image; the lower the sharpness, the lower the threshold. Redundant texture features such as facial shadows and invalid features caused by slight occlusion are filtered out, retaining only key facial features and facial contour features, resulting in the first extraction result.

[0018] The infrared processed image is input into the YOLOv8 model for feature extraction, resulting in a second extraction result, which includes: The embodiments of the present invention are based on the original YOLOv8 architecture and are improved by inputting the infrared processed image into the YOLOv8 model composed of a backbone network and a neck network; Multi-level features are extracted through a backbone network consisting of sequentially arranged convolutional modules and C2f modules. Deformable convolution DCNv2 is used in the second convolutional block of each bottleneck block in the C2f module to improve the adaptability to tilted and slightly distorted faces. A channel attention pruning mechanism is introduced at the output of the C2f module to retain effective channels, that is, to dynamically filter the infrared image feature channels, retain the effective channels related to the face position, and remove redundant channels to reduce the number of model parameters by more than 30%, which is suitable for the lightweight deployment requirements of mobile terminals and solves the technical defects of large number of parameters and high energy consumption in the existing YOLOv8 improved model in infrared face detection. By employing a feature pyramid module in the neck network to fuse and enhance multi-level features, a second extraction result is obtained. A Separate Convolutional Attention (LSKA) mechanism is introduced into the pooling layer feature and lateral connection feature concatenation module and the second convolutional block of the neck network to enhance the capture capability of multi-scale facial features, adapting to the detection needs of small faces and dense scenes. After obtaining the feature map from the previous level, the first convolutional block of the neck network adjusts the feature weights through the adaptive weight adjustment factor in the LSKA mechanism to suppress background interference. Based on the brightness distribution of the face region in the infrared image (where the brightness of the face region is higher than the background), the attention weights are dynamically adjusted to strengthen the feature weights of the face region and suppress interference features in the background region, further improving the accuracy of infrared face location extraction in complex environments such as nighttime and low light.

[0019] Other parameters besides the above model settings can be adaptively adjusted according to actual needs.

[0020] The first extraction result and the second extraction result are adaptively fused to obtain facial features. This embodiment of the invention does not limit the specific fusion method. It is only necessary to combine the two results to obtain the final result. Facial features include the positions of eyes, nose, mouth and so on.

[0021] The facial features are compared with a pre-stored feature database to determine whether the user's identity is legitimate, thus obtaining the first verification result.

[0022] In step S130, the smart card information is compared with the information database to obtain a second verification result of the identity. Specifically, the access control device compares the smart card information it reads with the pre-stored smart card information database to further verify the user's identity and obtain a second verification result.

[0023] In step S140, the first verification result and the second verification result are obtained, and a determination is made whether to activate the access control system based on a preset security policy. Specifically, this includes: The system employs a dual-authentication strategy or a priority setting strategy to determine whether to activate the access control system and outputs control commands. The dual-verification strategy allows users to pass only when both the first and second verification results pass. This verification process is more stringent than priority-based systems, significantly improving the security of the access control system. By controlling access through the system, users are allowed to pass only when both verifications are successful, effectively preventing unauthorized intrusion and identity theft. The priority-based strategy determines whether to allow users to pass based on either the first or second verification result. If one verification result has a higher priority than the other, access can be granted only based on the higher-priority result. The optimized neural network model can improve the recognition accuracy to a certain extent. However, in situations such as severe lighting or severe facial occlusion, a priority setting strategy can be enabled, with the second verification result set as the highest priority, to ensure the stable operation of the access control system. The priority setting strategy improves high environmental adaptability. Under normal circumstances, a two-way verification strategy is used.

[0024] The access control mechanism is opened and closed according to control commands.

[0025] Preferably, the embodiments of the present invention can record user entry and exit information, providing data support for subsequent attendance management, data analysis, etc. The data can provide strong support for management decisions, further improving the operational efficiency and management level of enterprises. It also adopts a simple and clear operation process and a user-friendly interface design, enabling users to easily and quickly complete the identity verification process. In addition, it provides a variety of prompts and voice broadcast functions to help users understand the operation progress and results. These functions further enhance the user experience and improve user satisfaction.

[0026] The above technical solutions of the embodiments of the present invention will be illustrated with reference to the following accompanying drawings.

[0027] Figure 2 This is a schematic diagram of the basic verification process according to an embodiment of the present invention, as shown below. Figure 2 The diagram illustrates a facial recognition process combined with a smart card. When a user approaches the access control system, it triggers both the camera and smart card recognition. The camera captures a facial image, transmits it, and then performs facial recognition, reading the smart card information. The facial recognition and smart card recognition results are transmitted to the control module, which determines whether the user has passed authentication based on security policies. If authentication is successful, the control module outputs a command to open the access control mechanism; otherwise, the display module displays a failure message. This embodiment of the invention employs a dual recognition mechanism that activates simultaneously. The order of activation can be adjusted according to the user's actual operation without affecting the substance of the invention.

[0028] In summary, addressing the existing problems, this invention presents a face-smart card combined identity recognition method. By integrating face recognition and smart card recognition technologies, it enhances the security and reliability of access control systems. Furthermore, it can select appropriate preset security strategies for verification based on the current environment, considering environmental adaptability in special circumstances. It automatically determines user identity and controls the opening and closing of access mechanisms according to preset security strategies, providing intelligent management capabilities. The dual verification strategy effectively prevents unauthorized intrusion and identity theft, while the priority setting strategy improves verification flexibility and ensures stable operation of the access control system. When using face recognition technology, multi-source sensing devices are employed to acquire facial images, ensuring consistent performance in various environments. It boasts high recognition accuracy and strong anti-interference capabilities using smart card recognition, ensuring normal operation in various environments. An optimized neural network model is used for face recognition. Color-processed images are input into an optimized VGG-16 model for feature extraction, and convolutional kernels replace the original fully connected layers, reducing computational complexity while maintaining both detection and feature extraction functions. Dynamic threshold filtering preserves key features, effectively improving feature matching efficiency. Infrared-processed images are input into an optimized YOLOv8 model for feature extraction. Deformable convolution DCNv2 is used in the C2f module to improve adaptability to slightly distorted faces, and a large-scale separable convolutional attention mechanism LSKA is introduced in the neck network to enhance the capture of multi-scale facial features.

[0029] System Implementation Examples According to embodiments of the present invention, a face recognition system combined with a smart card is provided. Figure 3 This is a schematic diagram of a face-based smart card identity recognition system according to an embodiment of the present invention, as shown below. Figure 3 As shown, the face recognition system combined with a smart card according to an embodiment of the present invention specifically includes: Camera 30 is used to collect facial images from multi-source sensing devices and transmit them to the facial recognition module; The face recognition module 32 is used to input face images into a trained neural network model and output face features. The feature extraction module of the neural network model processes the face images collected by each sensing device in parallel, compares the face features with the feature database, and obtains the first verification result of identity. The smart card recognition module 34 is used to read the user's smart card information, compare the smart card information with the information database, and obtain a second verification result of the identity. The control module 36 is used to obtain the first verification result and the second verification result, and determine whether to activate the access control system based on the preset security policy. Access control mechanism 38 is used to execute instructions from the control module to control the opening and closing of the door; Database 310 is used to store feature databases and information databases for authentication. Display module 312 is used to display the recognition result and prompts such as "Verification successful, please pass" or "Verification failed, please try again".

[0030] Figure 4 This is a schematic diagram of the system design architecture of an embodiment of the present invention, as shown below. Figure 4 As shown, the design architecture of each functional module is illustrated. The camera, face recognition module, smart card recognition module, control module, access control mechanism, database, and display module are connected as a whole via cables or wirelessly. The camera and smart card recognition modules are responsible for data acquisition and preliminary processing, transmitting the results to the control module. The control module is the core of the system, responsible for data processing and decision output. The access control mechanism executes opening and closing operations according to the instructions of the control module. The database stores user information and security policies, providing data support for the system. The display module provides interactive information and prompts to the user. Figure 5 This is a schematic diagram of the system application architecture of an embodiment of the present invention, such as... Figure 5 The image shows the basic process of user identification using a combination of facial recognition and smart card verification.

[0031] In summary, addressing the existing problems, this invention presents a face recognition combined with smart card identification system. This dual-identification approach, combining face recognition and smart card recognition technologies, enhances the security and reliability of access control systems. Furthermore, it can select appropriate preset security strategies for verification based on the current environment, considering environmental adaptability in special circumstances. It automatically determines user identity and controls the opening and closing of access mechanisms according to preset security policies, providing intelligent management capabilities. The dual-authentication strategy effectively prevents unauthorized intrusion and identity theft, while the priority setting strategy improves verification flexibility and ensures stable operation of the access control system. When using face recognition technology, multi-source sensing devices are employed to acquire facial images, ensuring consistent performance in various environments. It boasts high recognition accuracy and strong anti-interference capabilities using smart card recognition, ensuring normal operation in various environments. An optimized neural network model is used for face recognition. Color-processed images are input into an optimized VGG-16 model for feature extraction, and convolutional kernels replace the original fully connected layers, reducing computational complexity while maintaining both detection and feature extraction functions. Dynamic threshold filtering preserves key features, effectively improving feature matching efficiency. Infrared-processed images are input into an optimized YOLOv8 model for feature extraction. Deformable convolution DCNv2 is used in the C2f module to improve adaptability to slightly distorted faces, and a large-scale separable convolutional attention mechanism LSKA is introduced in the neck network to enhance the capture of multi-scale facial features.

[0032] Electronic device examples Figure 6 This is a schematic diagram of an electronic device according to an embodiment of the present invention. The electronic device 600 may include at least one processor 610 and a memory 620. The processor 610 can execute instructions stored in the memory 620. The processor 610 is communicatively connected to the memory 620 via a data bus. In addition to the memory 620, the processor 610 can also be communicatively connected to an input device 630, an output device 640, and a communication device 650 via the data bus.

[0033] Processor 610 can be any conventional processor, such as a commercially available CPU. Processors may also include graphics processing units (GPUs), field-programmable gate arrays (FPGAs), systems-on-chips (SoCs), application-specific integrated circuits (ASICs), or combinations thereof.

[0034] The memory 620 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0035] In this embodiment of the present disclosure, the memory 620 stores executable instructions, and the processor 610 can read the executable instructions from the memory 620 and execute the instructions to implement all or part of the steps of the face-based smart card identification method in any of the exemplary embodiments described above.

[0036] Computer-readable storage medium embodiments In addition to the methods and systems described above, exemplary embodiments of this disclosure may also be a computer program product or a computer-readable storage medium storing the computer program product, the computer product including computer program instructions that can be executed by a processor to implement all or part of the steps described in any of the face-based smart card identification methods in the exemplary embodiments described above.

[0037] Computer program products can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. Programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages, and scripting languages ​​(e.g., Python). The program code can be executed entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0038] Computer-readable storage media may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example,, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media include: static random access memory (SRAM) having one or more electrically connected wires; electrically erasable programmable read-only memory (EEPROM); erasable programmable read-only memory (EPROM); programmable read-only memory (PROM); read-only memory (ROM); magnetic storage; flash memory; magnetic disk or optical disk; or any suitable combination thereof.

[0039] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identity recognition combining face recognition and smart cards, characterized in that, include: When a user approaches the access control system, a dual recognition mechanism is triggered to acquire the facial image collected by multi-source sensors and read the smart card information. The face image is input into a trained neural network model to output face features. The feature extraction module of the neural network model processes the face images collected by each sensing device in parallel and compares the face features with the feature database to obtain the first verification result of identity. The smart card information is compared with the information database to obtain a second identity verification result; Obtain the first verification result and the second verification result, and determine whether to activate the access control system based on the preset security policy.

2. The method according to claim 1, characterized in that, The acquisition of facial images from multi-source sensing devices specifically includes: acquiring color facial images through a visible light image sensor and acquiring infrared facial images through a ToF image sensor.

3. The method according to claim 1, characterized in that, The reading of smart card information specifically includes: obtaining smart card information, including the card number and cardholder's name, through RFID or NFC technology.

4. The method according to claim 1, characterized in that, The step of inputting the face image into the trained neural network model and outputting face features specifically includes: The face image is input into the input layer of the neural network model, and the face image is normalized to obtain a color-processed image and an infrared-processed image, respectively. The color-processed image is input into the VGG-16 model for feature extraction to obtain the first extraction result; The infrared processed image is input into the YOLOv8 model for feature extraction to obtain a second extraction result; The first extraction result and the second extraction result are adaptively fused to obtain facial features.

5. The method according to claim 4, characterized in that, The step of inputting the color-processed image into the VGG-16 model for feature extraction to obtain the first extraction result specifically includes: The color-processed image is input into a VGG-16 model consisting of 6 convolutional blocks; The VGG-16 model extracts facial contour features through its first convolutional block, and its second to fifth convolutional blocks each extract facial feature details. The sixth convolutional block replaces the fully connected layer in the original VGG-16 model to integrate the facial contour features and facial feature details. An adaptive feature filtering unit is added to the output of the sixth convolutional block to filter redundant texture features through dynamic thresholding, thus obtaining the first extraction result.

6. The method according to claim 4, characterized in that, The step of inputting the infrared processed image into the YOLOv8 model for feature extraction to obtain the second extraction result specifically includes: The infrared processed image is input into a YOLOv8 model consisting of a backbone network and a neck network. Multi-level features are extracted through a backbone network consisting of sequentially arranged convolutional modules and C2f modules. In each bottleneck block of the C2f module, the second convolutional block uses deformable convolution DCNv2. A channel attention pruning mechanism is introduced at the output of the C2f module to retain effective channels. By using the neck network of the feature pyramid module to fuse and enhance multi-level features, a second extraction result is obtained. A separate convolutional attention mechanism (LSKA) is introduced into the concatenation module of pooling layer features and lateral connection features and the second convolutional block of the neck network. The feature weights are adjusted by the adaptive weight adjustment factor in the LSKA mechanism to suppress background interference.

7. The method according to claim 1, characterized in that, The step of obtaining the first verification result and the second verification result, and determining whether to activate the access control system in conjunction with a preset security policy, specifically includes: The decision to activate access control is made using a dual authentication strategy or a priority setting strategy. The dual verification strategy allows the user to pass only when both the first and second verification results are passed; the priority setting strategy determines whether to allow the user to pass based on either the first or second verification result.

8. A face-based smart card identification system, used in the face-based smart card identification method according to any one of claims 1 to 7, characterized in that, include: A camera is used to capture facial images from multiple sensor devices. The face recognition module is used to input the face image into a trained neural network model and output face features. The feature extraction module of the neural network model processes the face images collected by each sensing device in parallel, compares the face features with the feature database, and obtains the first verification result of identity. The smart card recognition module is used to read smart card information, compare the smart card information with an information database, and obtain a second identity verification result. The control module is used to acquire the first verification result and the second verification result, and determine whether to activate the access control system based on the preset security policy; Access control mechanisms are used to execute instructions from the control module; A database is used to store the feature database and the information database; The display module is used to display the recognition results.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the face-based smart card identification method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an information transmission implementation program, which, when executed by a processor, implements the steps of the face-based smart card identification method as described in any one of claims 1 to 7.

Images