Intelligent access control remote authorization method and system applied to bank system

Abstract

The invention provides an intelligent access control remote authorization method and system applied to a bank system. Comprising the steps that related information of a target object is obtained, the target object is an object entering a bank through an access control system, the related information at least comprises one or more of face information, sound information, fingerprint information, talkback information and iris information, and the talkback information is dialogue information, obtained through talkback equipment, between the target object and an authorized object; performing identity verification on the target object according to the related information to obtain a plurality of verification results; and under the condition that all verification results represent that the verification is passed, controlling the entrance guard of the bank to be opened. According to the scheme, the problems that in the prior art, the traditional access control authorization capability is limited, real-time monitoring and remote authorization cannot be realized, an attacker can bypass the face recognition function of an access control system by using forged face information or embezzling face information of other people, and the safety is insufficient are solved.

Description

Technical Field

[0001] The present application relates to the field of access control security technology, and in particular to a method, device, computer program product, and system for remote authorization of intelligent access control applied to a banking system. Background Art

[0002] The financial industry is an industry where money is circulated in a centralized manner. As an important storage and circulation channel for money, banks should have stricter protection of security systems than ordinary units. However, the current bank security protection solutions are all manual monitoring, and the traditional access control authorization capabilities are limited, and real-time monitoring and remote authorization cannot be achieved. Attackers can use forged facial information or steal other people's facial information to bypass the facial recognition function of the access control system, resulting in insufficient security. Summary of the invention

[0003] The main purpose of this application is to provide a method, device, computer program product and system for intelligent access control remote authorization applied to a banking system, so as to at least solve the problems in the prior art that traditional access control authorization capabilities are limited, real-time monitoring and remote authorization cannot be achieved, and attackers can use forged facial information or stolen facial information of others to bypass the facial recognition function of the access control system, resulting in insufficient security.

[0004] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for remote authorization of intelligent access control applied to a bank system is provided, comprising: obtaining relevant information of a target object, wherein the target object is an object entering the bank through the access control system, and the relevant information includes at least one or more of facial information, sound information, fingerprint information, intercom information and iris information, and the intercom information is conversation information between the target object and the authorized object obtained through the intercom device, and the authorized object is an object that grants the target object the right to enter the bank or does not grant the target object the right to enter the bank; authenticating the target object according to the relevant information to obtain multiple verification results; and controlling the bank's access control to be opened when all of the verification results indicate that the verification is passed.

[0005] Optionally, the identity of the target object is authenticated according to the relevant information to obtain multiple verification results, including: extracting facial features according to the facial information, wherein the facial features include at least one or more of facial contour, expression and heat distribution; extracting sound features according to the sound information, wherein the sound features include at least one or more of pitch, volume, timbre and voiceprint; converting the sound information into text to obtain text features; fusing the facial features, the sound features and the text features to obtain comprehensive features, wherein the fusion method includes one or more of feature-level fusion, decision-level fusion, weighted average and deep learning fusion; performing identity recognition according to the comprehensive features to obtain a first verification result.

[0006] Optionally, identity identification is performed according to the comprehensive features to obtain a first verification result, including: obtaining credential information, wherein the credential information is information of the identity document of the target object; determining the first identity information of the target object according to the credential information; determining the second identity information of the target object according to the comprehensive features; in a case where the first identity information and the second identity information match successfully, determining the first verification result as identity authentication passed; in a case where the first identity information and the second identity information fail to match, determining the first verification result as identity authentication failed.

[0007] Optionally, the target object is authenticated according to the relevant information to obtain multiple verification results, including: obtaining a preset fingerprint library, wherein the preset fingerprint library includes multiple preset fingerprint information; matching the fingerprint information with multiple preset fingerprint information; if the fingerprint information matches any one of the preset fingerprint information successfully, determining that the identity authentication is passed and obtaining a second verification result; if the fingerprint information fails to match all of the preset fingerprint information, determining that the identity authentication is not passed and obtaining a second verification result.

[0008] Optionally, the identity of the target object is authenticated according to the relevant information to obtain multiple verification results, including: when the intercom information is obtained, obtaining remote verification information sent by the received remote terminal, wherein the remote verification information is decision information sent by the remote terminal where the authorized object is located as to whether to authorize the target object to open the door; when the remote verification information indicates that the door is authorized to be opened for the target object, determining that the identity authentication is passed and obtaining a third verification result; when the remote verification information indicates that the door is not authorized to be opened for the target object, determining that the identity authentication is not passed and obtaining a third verification result.

[0009] Optionally, in the process of controlling whether the bank's access control is opened, the method further includes: when the local authentication result indicates that the authentication is passed, obtaining a remote authentication result, wherein the remote authentication result is a result obtained by performing face recognition, manual review, and intercom confirmation through a remote server; when the remote authentication result indicates that the face recognition authentication is passed, the manual review authentication is passed, and the intercom confirmation authentication is passed, controlling the bank's access control to be opened; generating a door opening log, and storing the door opening log in a database, wherein the door opening log includes at least one or more of the door opening operation time, the bank's access control location, and the related information.

[0010] Optionally, after the identity of the target object is authenticated according to the relevant information to obtain multiple verification results, the method further includes: generating an alarm message when at least one of the verification results indicates that the identity authentication has failed; controlling the bank's alarm device to sound an alarm according to the alarm message; and sending the alarm message to a remote terminal.

[0011] According to another aspect of the present application, a device for remote authorization of intelligent access control applied to a bank system is provided, comprising: a first acquisition unit, used to acquire relevant information of a target object, wherein the target object is an object entering the bank through the access control system, and the relevant information includes at least one or more of facial information, sound information, fingerprint information, intercom information and iris information, the intercom information is conversation information between the target object and the authorized object obtained through the intercom device, and the authorized object is an object that grants the target object the right to enter the bank or does not grant the target object the right to enter the bank; a verification unit, used to authenticate the target object according to the relevant information to obtain multiple verification results; a first control unit, used to control the bank's access control to be opened when all of the verification results indicate that the verification is passed.

[0012] According to another aspect of the present application, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the computer program implements the steps of any one of the methods for remote authorization of smart access control applied to a banking system.

[0013] According to another aspect of the present application, a system for remote authorization of smart access control applied to a banking system is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include methods for executing any one of the methods for remote authorization of smart access control applied to a banking system.

[0014] By applying the technical solution of the present application, the identity of the target object can be verified through a variety of identity authentication methods, namely face recognition, voice recognition, fingerprint recognition and video intercom remote manual recognition. The bank door will only be opened if all verifications are passed, greatly improving the security of the bank's access control. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0016] Figure 1 A hardware structure block diagram of a mobile terminal for executing a method for remote authorization of intelligent access control applied to a banking system provided in an embodiment of the present application is shown;

[0017] Figure 2 A flow chart of a method for remote authorization of smart access control applied to a banking system provided in an embodiment of the present application is shown;

[0018] Figure 3 The overall flow chart of this scheme is shown;

[0019] Figure 4 A schematic diagram of the face recognition process of this solution is shown;

[0020] Figure 5 A structural block diagram of a device for remote authorization of intelligent access control applied to a banking system provided in accordance with an embodiment of the present application is shown.

[0021] The above drawings include the following reference numerals:

[0022] 102, processor; 104, memory; 106, transmission device; 108, input and output devices. DETAILED DESCRIPTION

[0023] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0024] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] For the convenience of description, some nouns or terms involved in the embodiments of the present application are explained below:

[0027] Remote access authorization: The authorized personnel of the access control system can authorize or refuse to open the door in the remote control center. The visitor's credentials and images, videos, and audio are synchronously transmitted to the remote authorization personnel, who will review and complete the authorization on their own terminals.

[0028] Passing the face: The "passing the face" technology uses technical means to turn illegally obtained facial information, ID card photos, etc. into dynamic videos, in order to crack the facial recognition verification program.

[0029] Local authentication: The access control front-end device performs authentication by reading the visitor's credentials and taking a photo of the person's face on site.

[0030] As introduced in the background technology, the bank security protection solutions in the prior art are all manually monitored, which is not secure enough. To solve the above problems, the embodiments of the present application provide a method, device, computer program product and system for remote authorization of smart access control applied to a bank system.

[0031] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0032] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 1 is a hardware structure block diagram of a mobile terminal for a method of remote authorization of intelligent access control applied to a banking system according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0033] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the display method of device information in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The transmission device 106 is used to receive or send data via a network. The above-mentioned specific examples of the network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0034] In this embodiment, a method for remote authorization of intelligent access control applied to a banking system is provided, which runs on a mobile terminal, a computer terminal or a similar computing device. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0035] Figure 2 FIG. 1 is a flow chart of a method for remote authorization of smart access control applied to a banking system according to an embodiment of the present application. Figure 2As shown, the method comprises the following steps:

[0036] Step S201, obtaining relevant information of the target object, wherein the target object is a person who enters the bank through the access control system, and the relevant information includes at least one or more of face information, voice information, fingerprint information, intercom information and iris information, the intercom information is the conversation information between the target object and the authorized object obtained through the intercom device, and the authorized object is the object who grants the target object the right to enter the bank or does not grant the target object the right to enter the bank;

[0037] Specifically, in the smart access control remote authorization system, when an object (i.e., the "target object") requests to enter the bank, the first step is to collect a series of information related to the object. This information includes at least one or more of the following types:

[0038] Face information: The face image of the target object is captured by the camera for subsequent face recognition and comparison to verify the identity of the object.

[0039] Sound information: The sound of the target object is collected through the microphone, including voice commands or spoken content, which can be used for voiceprint recognition or voice content analysis.

[0040] Fingerprint information: If the system is configured with fingerprint recognition function, the fingerprint data of the target object will be obtained through the fingerprint sensor to verify the unique identity.

[0041] Intercom information: This is the conversation between the target object and the authorized object, which is recorded by the intercom device integrated into the bank access control system. Intercom information is dynamic and real-time information that can be used to evaluate the visitor's intentions, behaviors, and on-site environmental conditions, and enhance security verification steps.

[0042] Iris information: refers to the biometric features extracted from the iris of the target object's eyes. The iris is the colored ring-shaped area around the pupil of the eye, which has rich textures and structures. These features are captured and analyzed through iris recognition technology. Iris recognition is characterized by high uniqueness and stability. Even in extreme cases, iris features rarely change, which makes iris information an important basis for identity authentication in high-security scenarios.

[0043] The "authorized object" mentioned here refers to the person who is given the authority to decide whether the target object can enter the bank. The authorized object can be the bank's security personnel, system administrators or specific managers. Their decision is based on the relevant information collected above to determine whether the target object has the legal right to enter the bank. If the authorized object believes that the target object's request is reasonable and the security verification is passed, they will grant the target object the right to enter the bank; otherwise, it will not be granted.

[0044] Step S202, performing identity authentication on the target object according to the relevant information to obtain multiple authentication results;

[0045] Specifically, the identity verification process is initiated, and the identity of the target object is carefully checked and confirmed using a combination of multimodal verification technology and manual review. This process will produce multiple verification results, each corresponding to a type of information:

[0046] Face information verification result: The system matches the face image of the target object with the face information of the authorized person registered in the database to determine whether the facial features of the target object are consistent with those of the authorized person.

[0047] Voice information verification results: Through voiceprint recognition, the system analyzes the voice characteristics of the target object and compares them with the voiceprints of known authorized personnel to confirm the legitimacy of the sound source.

[0048] Fingerprint information verification results: If fingerprint information is collected, the system compares the target object's fingerprint with the fingerprint records stored in the database to verify whether the fingerprints match.

[0049] Intercom information review results: After receiving the remote authorization request from the target object, the authorized object manually reviews and makes a decision on whether to pass by watching the video stream, listening to the intercom recording and directly talking to the target object based on the on-site situation and the target object's behavior.

[0050] Each verification result will be recorded by the system. If one or more verification results indicate a verification failure, the system will not proceed to the next step of door opening control.

[0051] Step S203, when all the above verification results indicate that the verification is passed, the bank's access control is controlled to be opened.

[0052] Specifically, in the smart access control system, the system will only perform the final door opening operation when all the identity authentication steps involved are passed, that is, when all the verification results indicate that the verification is passed. This means that if the face information, voice information, fingerprint information (if applicable) and intercom information review results of the target object are all confirmed to be legal and safe, the system will send a command to the access control execution device to control the bank's door to open automatically and allow the target object to enter.

[0053] Through this embodiment, the identity of the target object can be verified through a variety of identity authentication methods, namely face recognition, voice recognition, fingerprint recognition and video intercom remote manual recognition. The bank door will only be opened if all verifications are passed, greatly improving the security of the bank's access control.

[0054] Specifically, in the existing solutions, the access control authorization method has low timeliness. When the number of concurrent authorization requests at the outlets is high, the outlet staff will have to wait for a long time, resulting in a poor customer experience. Authorization personnel generally use the traditional manual authorization method, which cannot guarantee the correctness of the authorization personnel and the security of the process. This solution only relies on the face recognition technology of the first terminal, which cannot effectively prevent new "face-passing" technologies. Secondly, there is no record of the door opening operation log, making it difficult to track and trace dangerous events.

[0055] Specifically, according to bank security regulations, bank business premises should be equipped with a security access control system, including access card swiping, facial recognition and other technical means to ensure that only authorized personnel enter the premises to ensure the effectiveness of security management. The access control device for people entering and leaving the vault is an extremely important component of the security system, but the traditional access control authorization capabilities are limited, and it is impossible to achieve real-time monitoring and remote authorization, and it cannot adapt to the needs of modern management; and attackers can use forged facial information or steal other people's facial information to bypass the facial recognition function of the access control system. Therefore, this application proposes a solution for intelligent access control remote authorization applied to bank systems, such as Figure 3 As shown, local authentication is performed through the access control front-end device to identify whether the visitor's credentials and face are consistent. After the local authentication is passed, a remote authorization request is sent to the system (or server) to perform a multimodal face recognition module, video and photo comparison through the server, manual review, and intercom confirmation. The door can be opened only after these three items are passed. The operation log of remote authorized door opening operation is recorded and managed to help authorized personnel monitor and track the records of authorized door opening, ensuring the security and traceability of the access control system. The system records the door opening log.

[0056] In the specific implementation process, the identity of the target object is authenticated according to the above-mentioned relevant information to obtain multiple verification results, which can be achieved through the following steps: extracting facial features according to the above-mentioned facial information, wherein the above-mentioned facial features include at least one or more of facial contour, expression and heat distribution; extracting sound features according to the above-mentioned sound information, wherein the above-mentioned sound features include at least one or more of pitch, volume, timbre and voiceprint; converting the above-mentioned sound information into text to obtain text features; fusing the above-mentioned facial features, the above-mentioned sound features and the above-mentioned text features to obtain comprehensive features, wherein the fusion method includes one or more of feature-level fusion, decision-level fusion, weighted average and deep learning fusion; performing identity recognition according to the above-mentioned comprehensive features to obtain a first verification result.

[0057] This solution can significantly improve the security protection capabilities of financial places such as banks and effectively prevent unauthorized entry. At the same time, through comprehensive analysis of multimodal information, the accuracy and efficiency of identity authentication can be further improved.

[0058] Specifically, the system not only extracts features based on traditional facial images, but also uses thermal infrared images to obtain the heat distribution of the face, combined with facial contours and expression information captured by visible light, to form a more comprehensive and multi-dimensional facial feature set. This comprehensiveness helps the system accurately identify target objects under different environmental conditions (such as lighting changes, occlusion, etc.), and improves the robustness and accuracy of recognition.

[0059] In addition to facial features, the system also extracts rich sound features from the sound information, including pitch, volume, timbre and voiceprint, etc. Voiceprint features can identify individual speech patterns, while pitch, volume and timbre can help identify specific emotional states or the speaker's environmental background, providing additional dimensions and information for identity recognition.

[0060] By converting the collected audio information into text features, the system is able to understand and analyze the target object's speech content. This capability is particularly suitable for smart access control remote authorization systems, and can be used to determine the target object's intentions and instructions, and even detect potential threatening language. Text conversion helps enhance the system's understanding and response capabilities, enabling it to more intelligently assess security conditions.

[0061] The facial features, voice features and text features are fused to obtain a more comprehensive integrated feature. The system supports multiple fusion methods, including feature-level fusion, decision-level fusion, weighted average and deep learning fusion. The adoption of these fusion strategies enables the system to select the most appropriate fusion method according to different security scenarios and needs to improve the recognition accuracy and decision-making ability of the system. For example, feature-level fusion can merge information of different modalities in the early stage of feature extraction, while decision-level fusion performs decision fusion after each feature is independently judged. Weighted average can consider the importance of different features, and deep learning fusion automatically learns the association between features through neural networks to improve recognition performance.

[0062] Finally, the system uses comprehensive features for identity recognition and obtains the first verification result. Since comprehensive features contain information from multiple modalities, this recognition method can provide higher recognition accuracy and security, reducing the risk of misjudgment caused by single feature recognition. The first verification result is based on a comprehensive analysis of the target object's face, voice, and speech content, providing a key basis for subsequent remote authorization decisions.

[0063] Specifically, the access control front-end device first performs local authentication to identify whether the visitor's credentials and face are consistent. After the local authentication is passed, it enters the multimodal face recognition module, such as Figure 4 As shown in the figure, the main process of multimodal face recognition includes the following:

[0064] 1) Data collection: Use access control front-end devices, such as cameras, infrared cameras, sound sensors, etc., to collect different types of data such as facial images, thermal infrared images, audio, video, etc. Compared with the traditional single-modality facial image collection method, the multimodal face recognition system can simultaneously obtain multiple information, such as facial images, voiceprints, etc. The collection speed of this information is relatively fast, and a multimodal database can be quickly established to achieve a fast face recognition process;

[0065] 2) Data preprocessing: Preprocess the different types of data collected in 1), including denoising, standardization, feature extraction, etc. Since multimodal face recognition systems usually use parallel processing to preprocess multiple types of information at the same time, by making full use of computing resources, the data preprocessing process of face recognition can be accelerated;

[0066] 3) Feature extraction: Extract useful features from different types of data, such as facial contours and expression features from face images, voice features from sounds, and facial heat distribution features from thermal infrared images. Traditional single-modal face recognition may be affected by factors such as light and angle, resulting in a long authentication time. Multimodal face recognition systems use multiple types of information for feature extraction and matching, which can obtain more comprehensive and accurate feature information, reduce misidentification or missed recognition caused by a single factor, and thus improve recognition speed;

[0067] 4) Fusion algorithm: The different types of features extracted are used through a cross-modal learning model to learn the associations and connections between different types of features. Data of different modalities are input into different convolutional neural networks for feature extraction, the graph convolutional network is trained with facial image feature data, the hidden Markov acoustic model is trained with audio features, and the phonemes output by the acoustic model are input into the pre-trained text model Bert. By jointly training the above image model, acoustic model, and text model, different types of features can be used as input to jointly learn the relationships between them. At the same time, the multimodal database and model are continuously updated through the training process to adapt to new facial features and environmental changes, thereby maintaining the timeliness and accuracy of the system;

[0068] 5) Identification and verification: Use the fused features of the cross-modal learning model and the visitor’s credentials to verify whether they match.

[0069] Specifically, first, the system extracts useful features from data of different modalities. For example, it extracts visual features such as facial contour, expression, and texture from RGB face images; extracts temperature distribution features of the face from thermal infrared images; and extracts acoustic features such as fundamental frequency, timbre, and speech rate of speech from sound signals.

[0070] Use convolutional neural networks (CNNs) to extract and learn features from face images. CNNs are good at processing image data and can automatically learn patterns and structures in images to generate high-dimensional image feature vectors. Hidden Markov models (HMMs) or their variants, such as deep hidden Markov models (DHMMs), are used to model sound features. HMMs can capture the temporal characteristics of speech signals and model the dynamic characteristics of speech to identify specific sound patterns. The pre-trained text model Bert is used to process and understand the phoneme information extracted from the audio, converting sound features into possible textual meanings, for example, to assist in identity verification by identifying specific voice commands or language habits.

[0071] Through cross-modal learning models (such as multimodal convolutional neural networks, multimodal attention mechanism models, etc.), features extracted from different modalities are fused together to jointly learn and understand the associations and connections between these features. This process allows the model to understand individuals from multiple perspectives, for example, by considering facial expressions and voice features when speaking at the same time, to more accurately judge the identity and emotional state of an individual.

[0072] Specifically, the solution of the present application provides a two-line authentication strategy of access control front-end device plus multimodal face recognition. The multimodal face recognition is performed twice through local authentication of the access control front-end device and different types of data collected and sent to the server. While effectively preventing the "face-passing" technology, it also improves the security and timeliness of the access control system.

[0073] Specifically, deep learning technology is introduced into the multimodal face recognition module, especially for factors such as light, occlusion, and expression changes in different environments. Adaptive illumination compensation algorithms and occlusion detection algorithms, as well as expression recognition technology based on sentiment analysis, can also be used to enhance the system's recognition performance in complex scenarios. At the same time, by introducing 3D facial scanning technology, a three-dimensional facial model is constructed to improve the system's use of depth information, further preventing "face-passing" attacks using two-dimensional photos or videos.

[0074] Specifically, the "passing face" technique usually refers to an attack method in which an attacker bypasses a face recognition system based on a single modality (such as a 2D image) through technical means. An attacker can use a high-definition photo or video of the victim's face to display it in front of the camera of the access control front-end device through a device such as a smartphone or tablet, trying to deceive the image-based face recognition system. This attack method takes advantage of the system's ability to recognize 2D images without considering the authenticity and dynamics of the image. Deepfake technology, which is a more advanced "passing face" method, uses deep learning technology to synthesize a person's facial expressions and movements into another person's video in real time to generate a seemingly real video stream to deceive the access control system. This synthesized face video is highly deceptive and can simulate the victim's real reaction at different angles and expressions, increasing the concealment and complexity of the attack. 3D printing or mold manufacturing, attackers may use the victim's three-dimensional face model to make a realistic mask or head model through 3D printing or mold manufacturing, and then display it in front of the access control system to try to bypass face recognition based on depth information. This attack is ineffective against image-based face recognition systems, but may pose a threat to some systems based on thermal infrared or 3D scanning technology.

[0075] Specifically, the introduction of the multimodal face recognition module is intended to effectively prevent the above-mentioned "passing face" attack. By simultaneously analyzing face images, thermal infrared images, audio and video data, the system can detect and verify the authenticity and dynamic characteristics of the face. For example, thermal infrared images can detect the temperature distribution of real faces, audio data can verify whether the visitor's voice characteristics match the face, and video streams can analyze the dynamic facial expressions and movements, as well as whether the background environment is normal. These additional verification methods greatly increase the difficulty for attackers to bypass the access control system.

[0076] Specifically, in the field of voiceprint recognition and speech analysis, voice features refer to the characteristics that can uniquely or partially uniquely identify a person's voice. These characteristics can be extracted from voice signals and used in a variety of applications such as identity authentication, sentiment analysis, and health monitoring.

[0077] Specifically, after passing the face recognition, the target object's account can also be authenticated using multiple authentication methods, such as combining fingerprint recognition, card swiping, strong passwords and other authentication methods to strictly control the target object's identity.

[0078] Specifically, in the authentication process, in addition to traditional multiple authentication methods such as fingerprint recognition, card swiping, and strong passwords, iris scanning, palm print recognition, and vein recognition can also be included to increase the difficulty and security of authentication. These biometric features are highly unique and stable, difficult to forge, and can effectively prevent intrusion by unauthorized personnel.

[0079] In addition, abnormal access patterns can be identified, such as frequent access attempts, access requests during non-working hours, etc., and warnings can be automatically triggered. At the same time, the warning information is sent to authorized personnel and security teams in real time for immediate action.

[0080] In some embodiments, identity recognition is performed based on the above-mentioned comprehensive features to obtain a first verification result, which can be specifically achieved through the following steps: obtaining credential information, wherein the above-mentioned credential information is information of the identity document of the above-mentioned target object; determining the first identity information of the above-mentioned target object based on the above-mentioned credential information; determining the second identity information of the above-mentioned target object based on the above-mentioned comprehensive features; in a case where the above-mentioned first identity information and the above-mentioned second identity information match successfully, determining the above-mentioned first verification result as identity authentication passed; in a case where the above-mentioned first identity information and the above-mentioned second identity information fail to match, determining the above-mentioned first verification result as identity authentication failed.

[0081] In this solution, by combining the verification of certificate information with the analysis of multimodal features, it is possible to further effectively defend against "face-passing" attacks and other single-modal information forgery threats. Because the verification of certificate information provides static identity confirmation, and the analysis of multimodal features adds a dimension of dynamic verification, only when the two parties match successfully can the identity authentication be considered successful, which greatly improves the security barrier of the system.

[0082] Specifically, the system first obtains the information of the target object's identification document through the access control front-end device, such as an ID card, employee card or any other document containing identity information. This can be achieved through NFC, RFID card readers or optical character recognition (OCR) technology to ensure that the system can accurately read and parse the information on the document.

[0083] After the credential information is parsed, the system will extract and determine the target object’s primary identity information. This usually includes key data such as name, ID number, photo, etc., which is used for preliminary comparison with the authorized personnel information in the database to ensure that the target object has basic access rights.

[0084] In the previous process, the system has extracted and integrated the target object's facial features, voice features, and text features to form a comprehensive feature vector. Based on this comprehensive feature, the system uses advanced machine learning algorithms (such as deep neural networks) to determine the target object's second identity information. This process is not just a simple face recognition or voiceprint matching, but a deep analysis based on multimodal information, providing a more comprehensive and accurate basis for identity verification.

[0085] The system matches and verifies the first identity information (obtained from the certificate) with the second identity information (determined based on comprehensive features). If the two identity information are consistent, it indicates that the identity of the target object has been confirmed by multimodal information, and the system determines the first verification result as identity authentication passed. Conversely, if the match fails, whether an error occurs in the comparison of the certificate information and facial features, or inconsistencies are found in the analysis of voiceprints or other features, the system will determine the first verification result as identity authentication failed and reject the target object's entry request.

[0086] Visitor Credentials are information or documents used to verify the identity of visitors. They are an important part of the access control system to control the entry and exit of personnel. In a bank or similar high-security environment, visitor credentials usually include the following aspects: identity documents, electronic credentials, appointment information, passwords or PIN codes.

[0087] Identification documents: Visitors may need to present valid identification documents, such as ID cards, passports, work permits or membership cards, which usually contain personal photos, names, ID numbers and other information to initially confirm the visitor's identity. In some cases, the system may also require visitors to enter specific information on the document, such as the document number or birthday, for further verification.

[0088] Electronic credentials: In a system with a high degree of digitization, visitor credentials can exist in electronic form, such as a QR code, NFC tag, or an electronic pass in a specific application (App). These credentials can be generated by the system, and the visitor can show them through a device such as a mobile phone, and the system can authenticate the identity by reading the information in the electronic credentials.

[0089] Appointment information: For specific visits, such as visits by VIP customers of a bank, the system can require the visitor to make an appointment in advance and generate a temporary access credential based on the appointment information. The appointment information can include the visitor's name, appointment time, purpose of visit, etc., which is used to quickly verify the visitor's appointment status on the day of the visit to ensure the orderliness and compliance of the visit.

[0090] Password or PIN: Sometimes, visitor credentials can also include a password or PIN, and visitors need to enter the correct password to pass the identity verification. This dual verification method of combining physical credentials with passwords increases the security level and prevents physical credentials from being stolen.

[0091] In the specific implementation process, the identity of the target object is authenticated according to the relevant information to obtain multiple verification results, which can be achieved through the following steps: obtaining a preset fingerprint library, wherein the preset fingerprint library includes multiple preset fingerprint information; matching the fingerprint information with multiple preset fingerprint information; when the fingerprint information matches any one of the preset fingerprint information successfully, determining that the identity authentication is passed and obtaining a second verification result; when the fingerprint information fails to match all of the preset fingerprint information, determining that the identity authentication is not passed and obtaining a second verification result.

[0092] In this solution, fingerprint recognition is a highly personalized biometric verification method that is difficult to copy and forge. By comparing with the preset fingerprint library, the system can effectively prevent unauthorized personnel from impersonating authorized personnel to perform remote authorization operations, enhancing the system's security protection capabilities. Due to the extremely high uniqueness of fingerprints, the matching process can achieve a very high accuracy rate. Even under changes in environmental conditions such as light and humidity, the accuracy of fingerprint recognition can still be guaranteed, reducing misidentification caused by environmental factors.

[0093] Specifically, in the access control system, the preset fingerprint library is a series of fingerprint information of authorized personnel stored in advance by the system, which is entered into the database when the system is initialized or the authorized personnel are registered. The establishment of the preset fingerprint library ensures that the system can have a reference standard for subsequent real-time fingerprint matching and identity authentication.

[0094] When authorized personnel (i.e. bank staff or administrators who have the right to perform remote authorization operations) need to log in to the system for authorization operations, they need to place their fingers on the fingerprint sensor of the access control front-end device. The sensor captures real-time fingerprint information and sends it to the server. The server-side processing system compares the received real-time fingerprint information with multiple preset fingerprint information in the preset fingerprint library one by one to find out whether there is a matching fingerprint template.

[0095] In order to improve the accuracy of verification, the system may require the authorized person to press the fingerprint multiple times to ensure that the captured fingerprint information is clear and complete enough.

[0096] Once the real-time fingerprint information is captured, the access control front-end device uploads the information to the server. The server-side processing system compares the received real-time fingerprint information with multiple preset fingerprint information in the preset fingerprint library one by one. The comparison process can include matching based on feature points, matching based on grain direction, etc., to determine the similarity between the real-time fingerprint and the preset fingerprint.

[0097] If the real-time fingerprint information matches any preset fingerprint information in the preset fingerprint library successfully, this indicates that the identity of the authorized person has been confirmed, the system determines that the identity authentication has passed, and the second verification result is "passed", that is, the authorized person is allowed to perform remote authorization operations.

[0098] On the contrary, if the real-time fingerprint information cannot match all the preset fingerprint information, this means that the person attempting to log in to the system is not on the authorized list or there is an abnormality in his fingerprint information. The system determines that the identity authentication has not passed, and the second verification result is "failed", and the person's remote authorization operation request is rejected.

[0099] Specifically, the authorized person collects his fingerprint through a special fingerprint sensor on the remote authorization terminal device. The fingerprint sensor can be capacitive, optical or ultrasonic, among which capacitive sensors are widely used due to their high accuracy and reliability. The sensor converts the texture and details of the fingerprint surface into digital signals.

[0100] The original fingerprint image acquired by the sensor will be sent to the fingerprint recognition algorithm for feature extraction. The fingerprint recognition algorithm will identify and extract the key feature points of the fingerprint, such as the end point, bifurcation point, direction, spacing, etc. of the grain line. These feature points constitute the "feature template" of the fingerprint, which is the basis for subsequent comparison.

[0101] The extracted fingerprint feature template will be sent to the multimodal face recognition module on the server side for fingerprint comparison. The server stores the registered fingerprint templates of authorized personnel. By comparing the sent feature template with the stored template, it is determined whether the two match. If the matching degree reaches or exceeds the preset threshold, the fingerprint verification is considered to have passed, otherwise the verification fails.

[0102] In some embodiments, the identity of the target object is authenticated according to the relevant information to obtain multiple verification results, which can be specifically achieved through the following steps: when the intercom information is obtained, remote verification information sent by the received remote terminal is obtained, wherein the remote verification information is decision information sent by the remote terminal where the authorized object is located, indicating whether to authorize the target object to open the door; when the remote verification information indicates that the door is authorized to be opened for the target object, it is determined that the identity verification is passed and a third verification result is obtained; when the remote verification information indicates that the door is not authorized to be opened for the target object, it is determined that the identity verification is not passed and a third verification result is obtained.

[0103] In this solution, the intercom confirmation mechanism allows the authorized object to communicate directly with the target object, and can promptly detect and handle any abnormal situations, such as abnormal behavior of the target object or potential threats in the environment, thereby enhancing dynamic response capabilities and security.

[0104] Specifically, after the access control front-end device completes the initial local authentication (such as preliminary matching of ID information, face and fingerprint) and the remote verification of multimodal features, the system will start the intercom confirmation mechanism. The target object (i.e. the person requesting to enter the bank) will conduct a real-time intercom with the remote terminal (usually the desktop or mobile device of the authorized person) through the microphone and speaker of the access control front-end device. The intercom information includes but is not limited to the target object's voice characteristics, real-time conversation content, and any other voice information that helps to confirm the identity.

[0105] The authorized object (i.e., the bank staff or administrator who has the right to perform remote authorization operations) receives a remote verification request from the access control system on the remote terminal. The request contains the target object's credential information, facial features, voice features, and real-time status during intercom. The authorized object manually reviews the target object's identity by viewing and listening to this information, and decides whether to authorize the door to be opened. This decision information is sent back to the access control system through the remote terminal and becomes part of the remote verification information.

[0106] The remote authentication message is a key binary decision that clearly indicates whether the authorized party agrees to the target party's request to enter. This message can be a simple "authorized" or "denied" signal, or it can contain additional instructions or requirements from the authorized party to deal with specific situations.

[0107] After receiving the remote verification information, the access control system will make a final identity authentication judgment based on the decision in the information. If the remote verification information indicates that the authorized object agrees to authorize the door to be opened, the system will determine that the identity authentication has passed and obtain the third verification result as "passed". This means that the identity of the target object has been fully confirmed, including multimodal feature matching and manual review by the authorized object. The system will perform the door opening operation and allow the target object to enter the bank.

[0108] On the contrary, if the remote verification information indicates that the authorized object refuses to authorize the door to be opened, the system will determine that the identity verification has not passed, and the third verification result will be "failed". This may be because the authorized object has found an abnormality during the review, such as the target object does not match the certificate information, behaves abnormally during the intercom, etc., or makes a decision for higher security considerations. In this case, the access control system will remain closed and reject the target object's entry request.

[0109] Specifically, after the identity of the authorized personnel is verified, they can review the on-site video. If any abnormal behavior of the visitor is found, they can directly initiate an intercom through the intercom device associated with the video channel to communicate with the visitor (ie, the target object).

[0110] Specifically, this solution provides a safer and more effective authorization process. First, it adopts multiple identity authentication methods such as fingerprint recognition, card swiping, and password for authorized personnel; secondly, it uses video channel linkage intercom to communicate efficiently in real time about abnormal user behavior; finally, it records and manages the operation log of remote authorized door opening operations to help authorized personnel monitor and track the records of authorized door opening, ensuring the security and traceability of the access control system.

[0111] Specifically, the "intercom" function refers to a mechanism in which the authorized person can communicate with the person requesting access in real time through two-way audio or video calls after receiving the authorization request from the access control front-end device at the remote control center. The realization of this function usually requires the integration of a voice or video intercom module on the access control front-end device, and the corresponding intercom function on the control terminal of the authorized person.

[0112] When a visitor passes preliminary authentication (such as card swiping, face recognition) at the access control front-end device and sends a remote authorization request, the system automatically establishes a communication link between the authorized person and the visitor. This is usually achieved through microphones and speakers integrated in the access control front-end device and the remote authorization terminal.

[0113] Once the link is established, the authorized person and the visitor can have a real-time conversation. The authorized person can ask the visitor about his or her purpose, identity confirmation details, or observe the visitor's reaction. The visitor can also explain the situation to the authorized person or answer questions.

[0114] During the specific implementation process, in the process of controlling whether the bank's access control is opened, the above method also includes the following steps: when the local authentication result indicates that the authentication is passed, obtaining the remote authentication result, wherein the above remote authentication result is the result obtained by performing face recognition, manual review, and intercom confirmation through a remote server; when the above remote authentication result indicates that the face recognition authentication is passed, the manual review authentication is passed, and the intercom confirmation authentication is passed, controlling the bank's access control to be opened; generating a door opening log, and storing the above door opening log in a database, wherein the above door opening log at least includes one or more of the door opening operation time, the bank's access control location, and the above related information.

[0115] In this solution, the generation and storage of door opening logs provide the bank with detailed entry records, which facilitates security audits and event tracking. When a security incident occurs, the source of the problem can be quickly located by consulting the door opening logs, and the specific time and location of the incident, as well as the information of the personnel involved and the basis for decision-making, can be understood, which is helpful for the investigation and handling of the incident.

[0116] The local authentication result is the result obtained by local authentication through the access control front-end device.

[0117] Therefore, the existing access control remote authorization system has the following shortcomings: (1) It cannot prevent "face-passing" technology, that is, using face recognition to bypass the access control system, and its security is insufficient; (2) Most authorized personnel use traditional manual authorization methods, which cannot guarantee whether the target object has the authority to open the door and the rationality of the authorization process; (3) It cannot be linked to intercom to communicate efficiently in real time about abnormal user behavior; (4) It is difficult to track and locate dangerous door opening incidents afterwards.

[0118] The present application introduces a solution for remote authorization of intelligent access control applied to banking systems. Local authentication is performed through the access control front-end device. After the local authentication is passed, a remote authorization request is sent to the server. The door opening operation can be performed only after the server passes the multi-modal face recognition module, manual review, and intercom confirmation. The operation log of the remote authorized door opening operation is recorded and managed to help authorized personnel monitor and track the records of authorized door opening, thereby ensuring the security and traceability of the access control system.

[0119] Specifically, when the system completes all verification processes and determines that the remote authorization door opening operation can be performed, the system will generate a detailed door opening log. This log is a key document that records each access control operation and is used to track and manage the information of people entering the bank to ensure the traceability and security of the operation.

[0120] The generated door opening log will be stored in the database, and the database design needs to take into account data security and access efficiency. The stored door opening log should be encrypted to protect sensitive information, and there should be a corresponding backup mechanism to prevent data loss. The database should also support fast search and data extraction so that security managers can quickly obtain access control operation records at any point in time.

[0121] Specifically, when the authorized personnel agree to the remote door opening request, the system will record the door opening operation log. Through the access control remote authorization door opening log record, the authorized personnel can timely understand the remote authorization door opening situation, find abnormal operations or security risks, and take corresponding processing and preventive measures. At the same time, it also helps to monitor the system operation and performance evaluation, and improve the security and management efficiency of the access control system.

[0122] Specifically, the door opening log is an important component of the intelligent access control remote authorization system for monitoring, auditing and post-event tracing. It records in detail the relevant information of each authorized door opening operation, helping security managers track and analyze access control activities to ensure the transparency and security of operations.

[0123] Specifically, the door opening log record usually includes the following key information: operation time, operator information, visitor information, access point location, operation type, operation result, intercom record, environmental information, etc.

[0124] Operation time: Records the exact time when the authorized door opening operation occurred, including date and time stamp, accurate to the second, to facilitate tracking the timeline of events.

[0125] The operator information includes the authorized person ID, the authorized person name, and the authorized person role or department. The authorized person ID records the unique identifier of the authorized person who performs the remote authorization operation. The authorized person name is the name of the person who performs the authorization. The authorized person role or department is the authorized person's role or department in the bank or organization, which helps to understand the allocation and use of permissions.

[0126] Visitor information includes visitor ID or credential information, visitor name, and visitor biometrics. Visitor ID or credential information records the unique identification of the person trying to access or the access credential information they use. Visitor name is the name of the person trying to enter. Visitor biometrics records biometric information used for verification, such as fingerprints, facial images, etc.

[0127] Access control point location: specifically record the installation location of the access control equipment or the entrance it controls, such as "vault entrance 1", "west door of the bank hall", etc., to facilitate positioning and management.

[0128] Operation Type: The recorded operation is whether it is remote authorization to open the door, access denial, or other special commands such as emergency locking.

[0129] Operation Result: Records the result of the access control operation, whether the door was successfully opened, and the reasons for any failure, such as invalid documents, biometric mismatch, etc.

[0130] Intercom Recording: If the intercom function is used during the remote authorization process, recording the conversation between the authorized person and the visitor will help with auditing and security analysis.

[0131] Environmental information: Record environmental information during operation, such as on-site video or audio clips, lighting conditions, external environment status, etc., to provide additional verification basis and event context.

[0132] In some embodiments, after the identity of the target object is authenticated according to the relevant information and multiple verification results are obtained, the method further includes the following steps: generating an alarm message when at least one of the verification results indicates that the identity authentication has failed; controlling the bank's alarm device to sound an alarm according to the alarm message; and sending the alarm message to a remote terminal.

[0133] In this solution, when identity authentication fails, the immediate alarm response can quickly alert the surrounding personnel, prevent potential illegal intrusions, and protect the safety of bank assets and personnel. At the same time, the remote notification mechanism ensures that even security managers who are not on site can intervene in time, improving the speed and efficiency of emergency response.

[0134] Specifically, in a multimodal intelligent access control system, the identity authentication process may include multiple stages, such as credential information verification, biometric recognition (face, voice, fingerprint, etc.), authorized personnel review, etc. If at any stage of the verification, the system determines that the identity authentication of the target object has not passed, that is, the verification result indicates that the identity authentication has not passed, the system will immediately initiate a security emergency response.

[0135] When the system detects that at least one verification result is not passed, it generates an alarm message. The alarm message contains details about the failed verification, such as the type of failed verification (credential information does not match, biometric recognition error, etc.), basic information of the target object (if captured), timestamp of verification failure, access control location, etc. The design of the alarm message should ensure the completeness and clarity of the information so that security managers can quickly understand the situation.

[0136] After the alarm information is generated, the system will automatically control the alarm equipment in the bank to sound the alarm. This may include activating the sound and light alarm, triggering the alarm sound, sending the alarm signal to the monitoring center, etc. The response speed of the alarm equipment and the choice of alarm method are key, and should ensure that it can quickly attract the attention of surrounding personnel and security management personnel without causing excessive panic.

[0137] In addition to local alarms, the system also sends alarm information to remote terminals, such as the security manager's mobile phone, tablet computer or dedicated security monitoring system. The instant notification of the remote terminal enables the security manager who is not on site to respond quickly and conduct remote monitoring or direct on-site processing based on the alarm information.

[0138] Generally speaking, a major attack point of the access control system is that the attacker uses forged facial information or stolen facial information of others to bypass the facial recognition of the access control front-end device, which is the "face-passing" technology. The above-mentioned scheme of the present application, in addition to the facial recognition function of the access control front-end device itself, performs multimodal facial recognition again on the server side according to the different types of data sent by the front-end access control device, thereby improving the security and recognition accuracy of the access control system; ensuring the validity of the identity of the authorized personnel; through the video channel linkage intercom, efficient communication is carried out in real time for abnormal user behavior; recording the operation log of remote authorized door opening operation, for dangerous door opening events, it is convenient for subsequent tracking and positioning.

[0139] The embodiment of the present application also provides a device for remote authorization of smart access control applied to a banking system. It should be noted that the device for remote authorization of smart access control applied to a banking system in the embodiment of the present application can be used to execute the method for remote authorization of smart access control applied to a banking system provided in the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and those that have been described will not be repeated here. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0140] The following is an introduction to the device for remote authorization of smart access control applied to a banking system provided in an embodiment of the present application.

[0141] Figure 5 1 is a structural block diagram of a device for remote authorization of intelligent access control applied to a banking system according to an embodiment of the present application. Figure 5 As shown, the device comprises:

[0142] A first acquisition unit 10 is used to acquire relevant information of a target object, wherein the target object is an object that enters a bank through an access control system, and the relevant information includes at least one or more of face information, voice information, fingerprint information, intercom information, and iris information. The intercom information is conversation information between the target object and an authorized object obtained through an intercom device, and the authorized object is an object that grants the target object the right to enter the bank or does not grant the target object the right to enter the bank;

[0143] A verification unit 20, configured to authenticate the target object according to the relevant information to obtain multiple verification results;

[0144] The first control unit 30 is used to control the bank's access control to be opened when all the above verification results indicate that the verification is passed.

[0145] Through this embodiment, the identity of the target object can be verified through a variety of identity authentication methods, namely face recognition, voice recognition, fingerprint recognition and video intercom remote manual recognition. The bank door will only be opened if all verifications are passed, greatly improving the security of the bank's access control.

[0146] In the specific implementation process, the verification unit includes a first extraction module, a second extraction module, a conversion module, a fusion module and a recognition module. The first extraction module is used to extract facial features based on the above-mentioned facial information, wherein the above-mentioned facial features include at least one or more of facial contour, expression and heat distribution; the second extraction module is used to extract sound features based on the above-mentioned sound information, wherein the above-mentioned sound features include at least one or more of pitch, volume, timbre and voiceprint; the conversion module is used to convert the above-mentioned sound information into text to obtain text features; the fusion module is used to fuse the above-mentioned facial features, the above-mentioned sound features and the above-mentioned text features to obtain comprehensive features, wherein the fusion method includes one or more of feature-level fusion, decision-level fusion, weighted average and deep learning fusion; the recognition module is used to perform identity recognition based on the above-mentioned comprehensive features to obtain a first verification result.

[0147] This solution can significantly improve the security protection capabilities of financial places such as banks and effectively prevent unauthorized entry. At the same time, through comprehensive analysis of multimodal information, the accuracy and efficiency of identity authentication can be further improved.

[0148] In some embodiments, the identification module includes an acquisition submodule, a first determination submodule, a second determination submodule, a third determination submodule and a fourth determination submodule, the acquisition submodule is used to acquire credential information, wherein the credential information is information of the identity document of the target object; the first determination submodule is used to determine the first identity information of the target object based on the credential information; the second determination submodule is used to determine the second identity information of the target object based on the comprehensive characteristics; the third determination submodule is used to determine that the first verification result is identity authentication passed if the first identity information and the second identity information match successfully; the fourth determination submodule is used to determine that the first verification result is identity authentication failed if the first identity information and the second identity information fail to match.

[0149] In this solution, by combining the verification of certificate information with the analysis of multimodal features, it is possible to further effectively defend against "face-passing" attacks and other single-modal information forgery threats. Because the verification of certificate information provides static identity confirmation, and the analysis of multimodal features adds a dimension of dynamic verification, only when the two parties match successfully can the identity authentication be considered successful, which greatly improves the security barrier of the system.

[0150] During the specific implementation process, the verification unit includes a first acquisition module, a matching module, a first determination module and a second determination module. The first acquisition module is used to obtain a preset fingerprint library, wherein the preset fingerprint library includes multiple preset fingerprint information; the matching module is used to match the fingerprint information with multiple preset fingerprint information; the first determination module is used to determine that the identity authentication is passed when the fingerprint information matches any one of the preset fingerprint information successfully, and obtain a second verification result; the second determination module is used to determine that the identity authentication is not passed when the fingerprint information fails to match all of the preset fingerprint information, and obtain a second verification result.

[0151] In this solution, fingerprint recognition is a highly personalized biometric verification method that is difficult to copy and forge. By comparing with the preset fingerprint library, the system can effectively prevent unauthorized personnel from impersonating authorized personnel to perform remote authorization operations, enhancing the system's security protection capabilities. Due to the extremely high uniqueness of fingerprints, the matching process can achieve a very high accuracy rate. Even under changes in environmental conditions such as light and humidity, the accuracy of fingerprint recognition can still be guaranteed, reducing misidentification caused by environmental factors.

[0152] In some embodiments, the verification unit includes a second acquisition module, a third determination module and a fourth determination module, the second acquisition module is used to obtain the remote verification information sent by the received remote terminal when the above-mentioned intercom information is obtained, wherein the above-mentioned remote verification information is the decision information sent by the above-mentioned remote terminal where the above-mentioned authorized object is located whether to authorize the door to open the above-mentioned target object; the third determination module is used to determine that the identity authentication is passed and obtain a third verification result when the above-mentioned remote verification information indicates that the door is authorized to open the above-mentioned target object; the fourth determination module is used to determine that the identity authentication is not passed and obtain a third verification result when the above-mentioned remote verification information indicates that the door is not authorized to open the above-mentioned target object.

[0153] In this solution, the intercom confirmation mechanism allows the authorized object to communicate directly with the target object, and can promptly detect and handle any abnormal situations, such as abnormal behavior of the target object or potential threats in the environment, thereby enhancing dynamic response capabilities and security.

[0154] During the specific implementation process, the above-mentioned device also includes a second acquisition unit, a second control unit and a first generation unit. The second acquisition unit is used to obtain the remote authentication result when the local authentication result indicates that the authentication is passed during the process of controlling whether the bank's access control is opened, wherein the above-mentioned remote authentication result is a result obtained by performing face recognition, manual review and intercom confirmation through a remote server; the second control unit is used to control the bank's access control to be opened when the above-mentioned remote authentication result indicates that the face recognition authentication is passed, the manual review authentication is passed, and the intercom confirmation authentication is passed; the first generation unit is used to generate a door opening log and store the above-mentioned door opening log in a database, wherein the above-mentioned door opening log at least includes one or more of the door opening operation time, the bank's access control location and the above-mentioned related information.

[0155] In this solution, the generation and storage of door opening logs provide the bank with detailed entry records, which facilitates security audits and event tracking. When a security incident occurs, the source of the problem can be quickly located by consulting the door opening logs, and the specific time and location of the incident, as well as the information of the personnel involved and the basis for decision-making, can be understood, which is helpful for the investigation and handling of the incident.

[0156] In some embodiments, the above-mentioned device also includes a second generating unit, a third control unit and a sending unit. The second generating unit is used to generate an alarm message after authenticating the above-mentioned target object according to the above-mentioned relevant information and obtaining multiple verification results, when at least one of the above-mentioned verification results indicates that the identity authentication has failed; the third control unit is used to control the bank's alarm device to alarm according to the above-mentioned alarm information; and the sending unit is used to send the above-mentioned alarm information to a remote terminal.

[0157] In this solution, when identity authentication fails, the immediate alarm response can quickly alert the surrounding personnel, prevent potential illegal intrusions, and protect the safety of bank assets and personnel. At the same time, the remote notification mechanism ensures that even security managers who are not on site can intervene in time, improving the speed and efficiency of emergency response.

[0158] The above-mentioned device for remote authorization of intelligent access control applied to a banking system includes a processor and a memory. The above-mentioned first acquisition unit, verification unit and first control unit are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to implement corresponding functions. The above-mentioned modules are all located in the same processor; or, the above-mentioned modules are located in different processors in any combination.

[0159] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the problem of manual monitoring and insufficient security in the existing bank security protection solutions can be solved by adjusting kernel parameters.

[0160] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0161] An embodiment of the present invention provides a computer-readable storage medium, which includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the method of remote authorization of smart access control applied to a banking system.

[0162] An embodiment of the present invention provides a processor, which is used to run a program, wherein the program executes the method of remote authorization of smart access control applied to a banking system when running.

[0163] An embodiment of the present invention provides a device, which includes a processor, a memory, and a program stored in the memory and executable on the processor, and when the processor executes the program, at least the steps of a method for remote authorization of smart access control applied to a bank system are implemented. The device in this article may be a server, a PC, a PAD, a mobile phone, etc.

[0164] A computer program product includes a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for remote authorization of smart access control applied to a banking system in each embodiment of the present application are implemented.

[0165] The present application also provides a system for remote authorization of smart access control applied to a banking system, comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include methods for executing any one of the above-mentioned methods for remote authorization of smart access control applied to a banking system.

[0166] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0167] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0168] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0169] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0170] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0171] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0172] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0173] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0174] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0175] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0176] 1) The method of remote authorization of intelligent access control applied to the banking system of the present application can verify the identity of the target object through a variety of identity authentication methods, namely face recognition, voice recognition, fingerprint recognition and video intercom remote manual recognition. The door of the bank will only be opened if all the verifications are passed, which greatly improves the security of the bank access control.

[0177] 2) The intelligent access control remote authorization device applied to the banking system of the present application can verify the identity of the target object through a variety of identity authentication methods, namely face recognition, voice recognition, fingerprint recognition and video intercom remote manual recognition. The bank door will only be opened if all the verifications are passed, which greatly improves the security of the bank access control.

[0178] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for remote authorization of intelligent access control applied to a banking system, characterized in that: include: Acquire relevant information of a target object, wherein the target object is an object that enters a bank through an access control system, and the relevant information includes at least one or more of face information, voice information, fingerprint information, intercom information, and iris information, wherein the intercom information is conversation information between the target object and an authorized object acquired through an intercom device, and the authorized object is an object that grants the target object the right to enter the bank or does not grant the target object the right to enter the bank; Authentication of the target object is performed according to the relevant information to obtain multiple authentication results; When all the verification results indicate that the verification is passed, the bank's access control is controlled to be opened.

2. The method according to claim 1, characterized in that The target object is authenticated according to the relevant information to obtain multiple authentication results, including: Extracting facial features according to the facial information, wherein the facial features include at least one or more of facial contour, expression and heat distribution; Extracting sound features according to the sound information, wherein the sound features include at least one or more of pitch, volume, timbre and voiceprint; Convert the sound information into text to obtain text features; Performing feature fusion on the facial features, the voice features and the text features to obtain comprehensive features, wherein the fusion method includes one or more of feature-level fusion, decision-level fusion, weighted average and deep learning fusion; Identity recognition is performed based on the comprehensive features to obtain a first verification result.

3. The method according to claim 2, characterized in that Performing identity recognition according to the comprehensive features to obtain a first verification result includes: Acquire credential information, wherein the credential information is information of the identity document of the target object; Determine the first identity information of the target object according to the credential information; Determine the second identity information of the target object according to the comprehensive feature; If the first identity information and the second identity information match successfully, determining that the first verification result is identity verification passed; In the case where the first identity information and the second identity information fail to match, the first verification result is determined to be identity authentication failure.

4. The method according to claim 1, characterized in that: The target object is authenticated according to the relevant information to obtain multiple authentication results, including: Acquire a preset fingerprint library, wherein the preset fingerprint library includes a plurality of preset fingerprint information; Matching the fingerprint information with a plurality of preset fingerprint information; If the fingerprint information successfully matches any of the preset fingerprint information, it is determined that the identity authentication is passed, and a second verification result is obtained; In the case that the fingerprint information fails to match with all the preset fingerprint information, it is determined that the identity authentication has failed, and a second verification result is obtained.

5. The method according to claim 1, characterized in that The target object is authenticated according to the relevant information to obtain multiple authentication results, including: In the case of acquiring the intercom information, acquiring remote verification information sent by the received remote terminal, wherein the remote verification information is decision information sent by the remote terminal where the authorized object is located, indicating whether to authorize the target object to open the door; In the case where the remote verification information indicates that the target object is authorized to open the door, determining that the identity verification is passed, and obtaining a third verification result; In the case where the remote verification information indicates that the target object is not authorized to open the door, it is determined that the identity verification fails, and a third verification result is obtained.

6. The method according to any one of claims 1 to 5, characterized in that In the process of controlling whether the bank's access control is opened, the method further includes: When the local authentication result indicates that the authentication is passed, a remote authentication result is obtained, wherein the remote authentication result is a result obtained by performing face recognition, manual review, and intercom confirmation on a remote server; When the remote authentication result indicates that the face recognition authentication is passed, the manual review authentication is passed, and the intercom confirmation authentication is passed, the bank's access control is controlled to be opened; Generate a door opening log and store the door opening log in a database, wherein the door opening log includes at least one or more of the door opening operation time, the bank's access control location and the related information.

7. The method according to any one of claims 1 to 5, characterized in that After the target object is authenticated according to the relevant information to obtain multiple authentication results, the method further includes: If at least one of the verification results indicates that the identity authentication has not been passed, generating an alarm message; According to the alarm information, control the bank's alarm equipment to sound an alarm; The alarm information is sent to a remote terminal.

8. A device for remote authorization of intelligent access control applied to a banking system, characterized in that: include: A first acquisition unit is used to acquire relevant information of a target object, wherein the target object is an object that enters a bank through an access control system, and the relevant information includes at least one or more of face information, voice information, fingerprint information, intercom information, and iris information, wherein the intercom information is conversation information between the target object and an authorized object acquired through an intercom device, and the authorized object is an object that grants the target object the right to enter the bank or does not grant the target object the right to enter the bank; A verification unit, configured to authenticate the target object according to the relevant information to obtain a plurality of verification results; The first control unit is used to control the bank's access control to be opened when all the verification results indicate that the verification is passed.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for remote authorization of smart access control applied to a banking system as described in any one of claims 1 to 7 are implemented.

10. A system for remote authorization of intelligent access control applied to banking system, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the smart access control remote authorization method applied to a banking system as described in any one of claims 1 to 7.

Images