User verification method and device, electronic equipment and computer program product

Abstract

The invention discloses a user verification method and device, electronic equipment and a computer program product. Relates to the field of artificial intelligence, and the method comprises the steps: determining a verification strategy according to a financial verification request when the financial verification request sent by a to-be-verified user is received, and sending verification information to the to-be-verified user according to the verification strategy; receiving to-be-verified data sent by the to-be-verified user, and extracting M verification sub-data from the to-be-verified data, the to-be-verified data being data generated by identity verification after the to-be-verified user receives verification information; and verifying the M verification sub-data to obtain M verification results, and generating an authentication failure prompt message when any one of the M verification results is abnormal in verification. Through application of the method and the device, the problems of low verification precision and low detection efficiency when safety verification is performed on the user in related technologies are solved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular, to a user verification method, device, electronic device, and computer program product. Background Art

[0002] With the increasing development of the Internet, in order to ensure the security of user transactions, major application platforms usually adopt identity authentication mechanisms to verify the authenticity of user identities in risky scenarios. The relevant identity authentication methods mainly include password verification, SMS verification code and face recognition authentication, among which face recognition technology is widely used due to its convenience and high recognition rate. The face recognition authentication process usually involves face positioning, feature extraction and comparison. In order to further enhance the security of authentication, many platforms have also introduced liveness detection technology, such as requiring users to perform specific dynamic actions (such as blinking, nodding) to prevent the use of photos or videos to deceive the system.

[0003] However, with the advancement of deep learning technology and the rise of video forgery technology, the above-mentioned identity authentication methods are facing unprecedented challenges. Video forgery technology can generate highly realistic dynamic videos of human faces, and realize actions such as opening the mouth, turning the head, and blinking through dynamic processing software. It can even replace data samples in the detection, greatly increasing the possibility of forged videos deceiving the identity authentication system, thus seriously threatening the effectiveness and security of the identity authentication mechanism based on vision and motion comparison.

[0004] Although face recognition and motion matching can provide preliminary identity authentication, they lack effective detection mechanisms for advanced attack methods such as video generation and voice synthesis. Similarly, although manual video authentication can confirm user information through customer service or agent observation and conversation, when faced with video forgery attacks, it also has the problem of low recognition accuracy when relying solely on visual and auditory intuitive judgment.

[0005] With regard to the problems of low verification accuracy and low detection efficiency when performing security verification on users in related technologies, no effective solution has been proposed so far. Summary of the invention

[0006] The main purpose of the present application is to provide a user verification method, device, electronic device and computer program product to solve the problems of low verification accuracy and low detection efficiency when performing security verification on users in related technologies.

[0007] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a user verification method is provided. The method includes: in the case of receiving a financial verification request sent by a user to be verified, determining a verification strategy according to the financial verification request, and sending verification information to the user to be verified according to the verification strategy, wherein the verification strategy refers to a strategy for identity authentication of the user to be verified, and the verification information refers to prompting the user to be verified to authenticate the identity according to the verification strategy; receiving the data to be verified sent by the user to be verified, and extracting M verification sub-data from the data to be verified, wherein the data to be verified refers to the data generated by the user to be verified after receiving the verification information and performing identity authentication, and the M verification sub-data at least include verification video data, verification audio data and verification image data, and M is a positive integer; respectively verifying the M verification sub-data to obtain M verification results, and generating an authentication failure prompt message when any verification result among the M verification results is a verification abnormality.

[0008] Furthermore, determining the verification strategy based on the financial verification request includes: extracting the financial behavior type from the financial verification request, wherein the financial behavior type includes a financial transaction type and a device detection type; when the financial behavior type is a financial transaction type, determining the verification strategy as a first verification strategy, wherein the first verification strategy includes performing video verification and audio verification on the user to be verified; when the financial behavior type is a device detection type, determining the verification strategy as a second verification strategy, wherein the second verification strategy includes performing audio verification and image verification on the user to be verified.

[0009] Further, in the case where the M verification sub-data are M verification video data, the M verification sub-data are verified respectively to obtain M verification results including: for one verification video data, the verification video data is segmented to obtain N video segments, where N is a positive integer; K video segments are selected from the N video segments, and video frames of each video segment in the K video segments are segmented to obtain K groups of video frames, where each group of video frames includes Y video frames, and Y is a positive integer; for a group of video frames, the Y video frames are subjected to a consistency check to obtain a consistency check result, and the verification result of the verification video data is determined by the K consistency check results of the K groups of video frames.

[0010] Furthermore, performing continuity check on Y video frames to obtain a continuity check result includes: using an encoder to extract frame information of each video frame to obtain Y frame information, wherein each frame information is used to characterize time information of each video frame; performing feature comparison on every two video frames in the Y video frames according to each frame information to obtain L comparison results, wherein feature comparison refers to performing optical flow difference check and motion continuity check on two adjacent video frames, and L is a positive integer; determining a difference value corresponding to each comparison result, and generating a continuity check result with a verification abnormality when any difference value among the L difference values ​​is greater than a preset threshold; generating a continuity check result with no verification abnormality when all the L difference values ​​are less than the preset threshold.

[0011] Furthermore, in the case where the M verification sub-data are M verification audio data, the M verification sub-data are verified respectively to obtain M verification results including: for one verification audio data, the verification audio data is vectorized to obtain a verification audio vector; the verification audio vector is signal segmented to obtain P audio sub-vectors, and each audio sub-vector is signal transformed to obtain P frequency domain audio signals, where P is a positive integer; a sound spectrum recognition algorithm is obtained, and the P frequency domain audio signals are processed using the sound spectrum recognition algorithm to output a feature recognition result, where the feature recognition result is used to characterize whether the verification audio data is real audio; in the case where the feature recognition result characterizes that the verification audio data is real audio, a verification result indicating that there is no verification anomaly is generated; in the case where the feature recognition result characterizes that the verification audio data is not real audio, a verification result indicating that there is a verification anomaly is generated.

[0012] Furthermore, the P frequency domain audio signals are processed using a sound spectrum recognition algorithm, and the output feature recognition results include: inputting the P frequency domain audio signals into a Mel filter and outputting P Mel spectra; performing logarithmic calculation on each Mel spectrum to obtain P processed spectra, and performing discrete cosine processing on each processed spectrum to obtain P audio recognition coefficients; performing differential calculation on each audio recognition coefficient to generate a feature recognition result.

[0013] Furthermore, before vectorizing the verification audio data, the method also includes: obtaining a preset time window, dividing the verification audio data according to the preset time window, and obtaining Q sub-audio data, where Q is a positive integer; calculating the variance of two adjacent sub-audio data in the Q sub-audio data to obtain multiple variance values; when each variance value is within a preset variance threshold interval, executing the step of vectorizing the verification audio data; if there is a variance value outside the preset variance threshold interval, generating a verification result of a verification abnormality.

[0014] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a user verification device is provided. The device includes: a determination unit, which is used to determine a verification strategy according to the financial verification request when receiving a financial verification request sent by a user to be verified, and send verification information to the user to be verified according to the verification strategy, wherein the verification strategy refers to a strategy for identity verification of the user to be verified, and the verification information refers to prompting the user to be verified to perform identity verification according to the verification strategy; a receiving unit, which is used to receive the data to be verified sent by the user to be verified, and extract M verification sub-data from the data to be verified, wherein the data to be verified refers to the data generated by the user to be verified after receiving the verification information and performing identity verification, and the M verification sub-data at least include verification video data, verification audio data and verification image data, and M is a positive integer; a verification unit, which is used to verify the M verification sub-data respectively to obtain M verification results, and generate an authentication failure prompt message when any verification result among the M verification results is a verification abnormality.

[0015] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein when the executable program runs, the device where the computer-readable storage medium is located is controlled to execute any of the above-mentioned user verification methods.

[0016] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, comprising one or more processors and a memory, wherein the memory stores an executable program, and the processor is used to run the program, wherein when the one or more programs are executed by one or more processors, the one or more processors implement any one of the above-mentioned user verification methods.

[0017] According to another aspect of an embodiment of the present invention, a computer program product is further provided. The computer program product includes a computer program, wherein when the computer program is executed by a processor, any one of the above-mentioned user verification methods is implemented.

[0018] In the embodiment of the present application, the user verification method is adopted. When receiving the financial verification request sent by the user to be verified, the verification strategy is determined according to the financial verification request, and the verification information is sent to the user to be verified according to the verification strategy, wherein the verification strategy refers to the strategy for identity authentication of the user to be verified, and the verification information refers to prompting the user to be verified to authenticate the identity according to the verification strategy; receiving the data to be verified sent by the user to be verified, and extracting M verification sub-data from the data to be verified, wherein the data to be verified refers to the data generated by the identity authentication of the user to be verified after receiving the verification information, and the M verification sub-data at least include verification video data, verification audio data and verification image data, and M is a positive integer; verifying the M verification sub-data respectively to obtain M verification results, and generating authentication failure prompt information when any verification result in the M verification results is a verification abnormality. By receiving the data to be verified sent by the user to be verified, extracting the verification sub-data from the data to be verified, verifying the verification sub-data, and obtaining the verification result, the purpose of improving the verification defense capability and optimizing the user experience is achieved, thereby achieving the technical effect of improving the accuracy and efficiency of user security verification, and thus solving the problem of low verification accuracy and low detection efficiency when performing security verification on users. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative 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:

[0020] Figure 1 It is a hardware structure block diagram of a computer terminal (or mobile device) for implementing a user authentication method;

[0021] Figure 2 is a flowchart of a user verification method provided according to an embodiment of the present application;

[0022] Figure 3 is a schematic diagram of a user verification device provided according to an embodiment of the present application;

[0023] Figure 4 It is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[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 sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. 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] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set up between this system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0027] It should be noted that the collected information used in this application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse to use it.

[0028] Example 1

[0029] According to an embodiment of the present application, a method embodiment of user verification is also provided. 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.

[0030] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 It is a hardware structure diagram of a computer terminal (or mobile device) for implementing a user authentication method, such as Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more ( Figure 1 102a, 102b, ..., 102n are used to illustrate) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU (Microcontroller Unit) or a programmable logic device FPGA (Field-Programmable Gate Array)), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a keyboard, a cursor control device, a power supply and / or a camera. 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 above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0031] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0032] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the user verification method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, realizing the above-mentioned user verification method. 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 instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 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 combinations thereof.

[0033] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC) and a network interface, 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 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0034] The display may be, for example, a touch screen liquid crystal display (LCD), which may enable a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0035] Under the above operating environment, this application provides Figure 2 The authentication method for the user indicated. Figure 2 is a flowchart of a user verification method provided in accordance with an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0036] Step S201, upon receiving a financial verification request sent by a user to be verified, determining a verification strategy according to the financial verification request, and sending verification information to the user to be verified according to the verification strategy, wherein the verification strategy refers to a strategy for identity verification of the user to be verified, and the verification information refers to prompting the user to be verified to perform identity verification according to the verification strategy.

[0037] Specifically, the user to be verified refers to the user who needs to be authenticated. The financial verification in the financial verification request can be the user identity verification associated when the user logs into the application, or the user identity verification associated with the financial transaction. In order to securely verify the identity of the user and balance the user experience and system security, the financial verification request of the above user can be received first, and then the nature and risk of the financial behavior can be determined based on the request, and then the appropriate verification strategy can be determined based on the information obtained from the evaluation and the identity information of the user to be verified. For example, for high-risk transactions, it is necessary to perform video recognition, voice recognition and SMS verification code verification strategies at the same time; for low-risk transactions, password recognition or SMS verification code verification strategies can be selected.

[0038] Furthermore, after determining different verification strategies, verification information can be constructed according to the verification strategies. For example, the verification information can include specific actions that the user needs to perform, sentences to be recited, SMS verification codes that need to be entered, etc. The verification information is then sent to the user through multiple channels (such as SMS, email) to ensure that the information can reach the user in a timely manner.

[0039] Step S202, receiving the data to be verified sent by the user to be verified, and extracting M verification sub-data from the data to be verified, wherein the data to be verified refers to the data generated by the user to be verified after receiving the verification information and performing identity authentication, and the M verification sub-data at least include verification video data, verification audio data and verification image data, and M is a positive integer.

[0040] Specifically, the data to be verified refers to a collection of information generated by the user to be verified according to the verification strategy and uploaded by the user to be verified through the device terminal. It contains multiple types of data and can fully verify the authenticity of the user's identity. For example, the data to be verified may include video data, audio data, and image data. The video data may be videos and images taken by a camera, and the audio data may be audio recorded by a microphone. When identity authentication is performed based on the data to be verified, the data to be verified must first be processed and parsed to extract multiple verification sub-data from it to enhance the comprehensiveness and security of the verification.

[0041] For example, key frames can be extracted from the verification video data submitted by the user to be verified, and the key frames can be used to realize face recognition and motion comparison analysis; features can be extracted from the verification audio data submitted by the user to be verified, and the features can be used for voiceprint recognition and speech content analysis; static images submitted by the user to be verified can be extracted, and the images can be used for detailed comparison of facial features and consistency check with the facial features in the video frame.

[0042] Step S203: verify the M verification sub-data respectively to obtain M verification results. If any verification result among the M verification results is abnormal, an authentication failure prompt message is generated.

[0043] Specifically, the verification sub-data may include biometric information (such as face recognition images, fingerprint templates, voiceprint samples), device information (such as the unique identifier of the user's device and the device model), random questions and answers (such as random tasks required of users), security questions, etc. When performing identity authentication, it is necessary to verify multiple verification sub-data separately to ensure the authenticity, integrity and security of the data. For example, the biometric information can be compared with the template to check the match; verify whether the device information is consistent with the user's registration information to determine whether the device is trustworthy; analyze the user's behavior pattern to confirm whether it is within the normal range of the user's historical behavior; verify the results of the user's completion of the random challenge, such as whether the synthesized audio is consistent with the user's real voice, whether the performed actions meet the requirements, and determine whether the user's answers to security questions are correct.

[0044] Furthermore, the verification results of the above-mentioned verification sub-data are summarized, and each verification result is represented as normal verification or abnormal verification. If any verification result is marked as abnormal verification, that is, a certain data verification fails, it indicates that the user to be verified has not passed the identity authentication, and an authentication failure prompt message is generated at this time. The prompt message can describe which verification sub-data failed to pass the verification, and possible reasons (such as biometric mismatch, device information abnormality, suspicious behavior pattern, etc.), and can also provide the next remedial measures or re-authentication guidance, for example, it is recommended that the user resubmit clear biometric information, check the device network connection or re-answer security questions to improve the security and transparency of the entire verification process.

[0045] The user verification method provided in the embodiment of the present application determines a verification strategy according to the financial verification request when receiving a financial verification request sent by the user to be verified, and sends verification information to the user to be verified according to the verification strategy, wherein the verification strategy refers to a strategy for identity authentication of the user to be verified, and the verification information refers to prompting the user to be verified to perform identity authentication according to the verification strategy; receives the data to be verified sent by the user to be verified, and extracts M verification sub-data from the data to be verified, wherein the data to be verified refers to data generated by the identity authentication of the user to be verified after receiving the verification information, and the M verification sub-data at least include verification video data, verification audio data and verification image data, and M is a positive integer; verifies the M verification sub-data respectively to obtain M verification results, and generates authentication failure prompt information when any verification result in the M verification results is a verification abnormality, thereby solving the problems of low verification accuracy and low detection efficiency when performing security verification on users in the related technology, and receives the data to be verified sent by the user to be verified, extracts the verification sub-data from the data to be verified, verifies the verification sub-data, and obtains the verification result, thereby achieving the technical effect of improving the accuracy and efficiency of user security verification.

[0046] In order to ensure the security and compliance of the financial behavior of the user to be verified, optionally, in the user verification method provided in the embodiment of the present application, determining the verification strategy according to the financial verification request includes: extracting the financial behavior type from the financial verification request, wherein the financial behavior type includes a financial transaction type and a device detection type; when the financial behavior type is a financial transaction type, determining the verification strategy as a first verification strategy, wherein the first verification strategy includes performing video verification and audio verification on the user to be verified; when the financial behavior type is a device detection type, determining the verification strategy as a second verification strategy, wherein the second verification strategy includes performing audio verification and image verification on the user to be verified.

[0047] Specifically, in order to conduct security verification of the financial behavior of the user to be verified, it is first necessary to conduct verification of the type of financial behavior, and then adopt corresponding verification strategies according to different types of financial behavior. Financial behavior types can include financial transaction types and device detection types. Financial transaction types include transfers, payments, account changes, etc. These operations usually involve high amounts or sensitive information, so a more stringent identity authentication process is required; device detection types are usually performed when the user logs in for the first time, changes the device, or makes an abnormal login attempt. The main purpose is to confirm the security status of the device and prevent the account from being accessed by unauthorized devices.

[0048] When the financial behavior type of the above-mentioned user is a financial transaction type, a more stringent and comprehensive first verification strategy is required for identity authentication. For example, it can include video verification and audio verification (also known as voiceprint verification), where video verification is used to confirm that the user is alive and perform biometric comparison, while audio verification is used to verify the authenticity of the user's voice characteristics and command response. The combination of the two can improve the accuracy of identity authentication. When performing video verification, the user's permission to use the camera device on the device terminal of the user to be verified can be obtained, and a video containing specific movements and expression changes, such as head rotation, blinking, etc., and reading random sentences recorded by the user through the camera can be collected to comprehensively evaluate the user's identity; when performing audio verification, the user's permission to use the microphone device on the device terminal of the user to be verified can be obtained, and a voice recorded by the user containing personal specific information can be collected.

[0049] When the user's financial behavior type is a device detection type, a milder second verification process can be collected to perform identity authentication, that is, to verify the security of the device and operating environment and the user's biometrics, such as audio verification and image verification. This embodiment dynamically adjusts the verification strategy by identifying the financial behavior type, ensuring that high-risk transactions receive stricter identity authentication, enhancing the security of identity authentication, and avoiding unnecessary verification steps for low-risk operations, thereby improving user experience and reducing the complexity and waiting time of user operations.

[0050] Optionally, in the user verification method provided in the embodiment of the present application, when M verification sub-data are M verification video data, the M verification sub-data are verified separately to obtain M verification results including: for one verification video data, segmenting the verification video data to obtain N video segments, where N is a positive integer; selecting K video segments from the N video segments, segmenting the video frames of each of the K video segments, and obtaining K groups of video frames, where each group of video frames includes Y video frames, and Y is a positive integer; for a group of video frames, performing a consistency check on the Y video frames to obtain a consistency check result, and determining the verification result of the verification video data according to the K consistency check results of the K groups of video frames.

[0051] Specifically, when performing video verification on the user to be verified, the verification video data collected by the user can be authenticated. First, the verification video data submitted by the user needs to be preprocessed, including video format conversion, standardized resolution and compression, etc., to ensure that the video data can be effectively processed. Then, the complete verification video data can be divided into multiple video clips by random or fixed interval selection, ensuring that the selected clips cover all important stages of the entire video, thereby improving the comprehensiveness of the detection.

[0052] Furthermore, multiple clips are randomly selected from the above video clips, and for each selected video clip, video frames are further intercepted to obtain multiple video frames, and then the coherence and authenticity of the video are checked from multiple angles through the analysis of each group of video frames, for example, by comparing the facial features of the previous and next frames, detecting small changes in the movement of the head or lips, analyzing the consistency of light and shadow and environmental background, etc., wherein the coherence verification algorithm can detect whether there are sudden changes in motion, flickering, shaking or other unnatural parts in the video. In addition, deep learning and image processing technology can also be used for artifact detection, such as after collecting motion changes, flickering, shaking or other unnatural parts, judging the context and time coherence of these parts.

[0053] Finally, by calculating the variance of specific features in the video frame and observing the consistency of the global variance, it is determined whether the video is within a relatively stable range of change, and then the consistency check results of the above video frames are summarized. If any group of display abnormalities are detected in the consistency check results of the video frames, such as incoherent movements, visual artifacts, or unnatural feature changes, the verification result of the verified video data is determined to be abnormal. This embodiment can effectively detect forged videos by randomly selecting video clips and performing consistency checks on key frames in the clips without the need for a comprehensive analysis of the entire video, which not only improves the verification efficiency, but also reduces the user waiting time, improves the security of identity authentication, and optimizes the user experience.

[0054] Optionally, in the user verification method provided in the embodiment of the present application, a consistency check is performed on Y video frames to obtain a consistency check result, including: using an encoder to extract frame information of each video frame to obtain Y frame information, wherein each frame information is used to characterize the time information of each video frame; according to each frame information, feature comparison is performed on every two video frames in the Y video frames to obtain L comparison results, wherein feature comparison refers to performing optical flow difference verification and motion consistency verification on two adjacent video frames, and L is a positive integer; determining the difference value corresponding to each comparison result, and when any difference value among the L difference values ​​is greater than a preset threshold, generating a consistency check result with a verification abnormality; when all L difference values ​​are less than the preset threshold, generating a consistency check result with no verification abnormality.

[0055] Specifically, when analyzing the coherence of video frames, an encoder (such as an encoder based on a convolutional neural network) can be used to first encode each video frame to extract temporal information and spatial features that can characterize the video frame, where the temporal information may include the capture timestamp of the frame, and the spatial features may include the content of the image in the video frame, such as color, texture, edges, etc.

[0056] Since optical flow can describe the vector field of the object's motion changes in the video frame, when comparing the features of every two video frames based on each frame information, the optical flow difference between every two adjacent video frames can be calculated to detect the continuity of the object's motion. If the optical flow difference exceeds the preset threshold, it may indicate a sudden change in the object's motion or unnatural behavior. In addition, it can also be verified by identifying the continuity of the character's movements in the video frame, such as comparing whether the movement changes in consecutive frames are smooth and natural, and whether there are sudden interruptions or incoordination of the movement. The continuity of this dimension in the intercepted key frame group can also be calculated by the frame difference method or by calculating the pixel difference, for example, by comparing the rotation of the head, the blinking of the eyes, the opening and closing of the mouth, etc.

[0057] Since the comparison results of each of the two video frames mentioned above include the comprehensive results of the optical flow and motion consistency checks, the results of the two checks can be compared for difference, that is, the degree of deviation from the preset normal behavior pattern can be determined. When any of the difference values ​​is greater than the preset threshold, a consistency check result of "verification abnormality" is generated, and the verification result of the video data obtained from the consistency check result can indicate that the video data may be abnormal; conversely, if the difference values ​​are all less than the preset threshold, a consistency check result of "verification without abnormality" is generated, and the verification result of the video data obtained from the consistency check result indicates that the video data has good consistency and is a real video. This embodiment verifies whether there is an abnormality in the video data through optical flow difference and motion consistency analysis, which helps to more accurately judge the authenticity of the video data, reduce misjudgment, and ensure the security of financial transactions and identity authentication.

[0058] Optionally, in the user verification method provided in the embodiment of the present application, when the M verification sub-data are M verification audio data, the M verification sub-data are verified separately to obtain M verification results, including: for one verification audio data, vectorizing the verification audio data to obtain a verification audio vector; performing signal segmentation on the verification audio vector to obtain P audio sub-vectors, and performing signal transformation processing on each audio sub-vector to obtain P frequency domain audio signals, where P is a positive integer; obtaining a sound spectrum recognition algorithm, using the sound spectrum recognition algorithm to process the P frequency domain audio signals, and outputting a feature recognition result, where the feature recognition result is used to characterize whether the verification audio data is real audio; when the feature recognition result characterizes that the verification audio data is real audio, generating a verification result with no verification abnormality; when the feature recognition result characterizes that the verification audio data is not real audio, generating a verification result with verification abnormality.

[0059] Specifically, when performing audio verification on the user to be verified, it is possible to effectively distinguish between real audio and synthesized audio. In order to enhance the high-frequency components and improve the recognition effect of the voice signal, the verification audio data uploaded by the user to be verified can first be vectorized, that is, the audio signal is converted into a series of numerical feature vectors to obtain a verification audio vector, and then the verification audio vector is further divided into multiple audio sub-vectors, each sub-vector covering a specific time interval of the audio to reduce the sidelobe effect between frames. Then, each audio sub-vector is subjected to signal transformation processing, such as Fourier transformation, to obtain the corresponding frequency domain audio signal, so as to more clearly reveal the frequency components and structure of the audio, which is convenient for subsequent sound spectrum recognition.

[0060] Furthermore, a sound spectrum recognition algorithm, such as a deep neural network, a long short-term memory network, or an attention mechanism, can be used to extract and identify specific sound features, such as fundamental frequency, resonance peak, noise level, etc., from the frequency domain audio signal, and then process the frequency domain audio signal to output the feature recognition result, and then determine whether the verified audio data is real audio. If the feature recognition result shows that the characteristics of the audio data are consistent with real human speech, a verification result of "no abnormality in verification" can be generated at this time, indicating that the audio data is real audio; conversely, if the feature recognition result indicates that the audio data has abnormal features, such as unnatural frequency components, poor voice coherence, and inconsistent with the voiceprint in the user registration information, a verification result of "verification abnormality" is generated. This embodiment identifies audio data by analyzing the frequency domain features of the audio, thereby improving the accuracy and reliability of audio verification and reducing user waiting time.

[0061] Optionally, in the user verification method provided in the embodiment of the present application, P frequency domain audio signals are processed using a sound spectrum recognition algorithm, and the output feature recognition results include: inputting P frequency domain audio signals into a Mel filter and outputting P Mel spectra; performing logarithmic calculation on each Mel spectrum to obtain P processed spectra, and performing discrete cosine processing on each processed spectrum to obtain P audio recognition coefficients; performing differential calculation on each audio recognition coefficient to generate a feature recognition result.

[0062] Specifically, when processing the frequency domain audio signal, the frequency domain audio signal obtained by Mel cepstral coefficient extraction, fundamental frequency calculation, Fourier transform, etc. is first input into the Mel filter, and the frequency is nonlinearly mapped, and the corresponding Mel spectrum is output. That is, the above signal is weighted and summed through a set of filters based on the Mel frequency scale to obtain the Mel spectrum, wherein the Mel spectrum is based on the human ear's perception characteristics of sounds of different frequencies, and converts the frequency domain signal to a frequency scale that is closer to human auditory perception, which helps to more accurately identify the acoustic characteristics of speech.

[0063] Furthermore, in order to enhance the information in the low-frequency area, reduce the noise in the high-frequency area, and make the spectral features more prominent, each Mel spectrum can be logarithmically calculated to obtain the corresponding processed spectrum, and then each processed spectrum is converted by discrete cosine transform to obtain the audio recognition coefficient, that is, the Mel frequency cepstral coefficient (MFCCs: MelFrequency Cepstral Coefficient), in this way, the spectral features are further compressed, the main components of the speech are extracted, and the data dimension is reduced at the same time, reducing the amount of calculation for subsequent processing.

[0064] Finally, in order to capture the dynamic information of the speech signal, each audio recognition coefficient can be differentially calculated, that is, the first-order difference and second-order difference of the MFCCs coefficient are calculated, and then the feature recognition result is generated. If the feature recognition result shows that the dynamic characteristics and stability of the audio signal match the real human voice, a verification result of "no abnormality in verification" can be generated, indicating that the audio data is real audio; if the feature recognition result shows that the audio signal has abnormal dynamic changes or inconsistent stability, a verification result of "verification abnormality" is generated, indicating that the audio data may be synthesized. This embodiment extracts highly characterized audio representations through steps such as Mel filters, logarithmic calculations, and differential calculations, and then implements audio verification based on the obtained feature recognition results, which can significantly improve the ability to recognize synthesized audio, avoid user inconvenience caused by the complexity of the audio verification process, and thus improve the overall user experience.

[0065] Optionally, in the user verification method provided in the embodiment of the present application, before the verification audio data is vectorized, the method also includes: obtaining a preset time window, dividing the verification audio data according to the preset time window, and obtaining Q sub-audio data, where Q is a positive integer; calculating the variance of two adjacent sub-audio data in the Q sub-audio data to obtain multiple variance values; when each variance value is within a preset variance threshold interval, executing the step of vectorizing the verification audio data; if there is a variance value outside the preset variance threshold interval, generating a verification result of a verification abnormality.

[0066] Specifically, in order to effectively detect the stability of audio, the audio data can first be segmented through a pre-set time window to obtain multiple sub-audio data, each sub-audio data represents a fixed time segment in the audio. In this way, the audio data is decomposed into a series of smaller, independently analyzable units, which is convenient for subsequent stability detection and feature extraction. Among them, since the choice of time window will affect the accuracy of stability analysis and the efficiency of subsequent processing, the preset time window needs to comprehensively consider various factors such as speech speed, dialect differences, background noise, and the possible length of audio.

[0067] Furthermore, the variance of each two adjacent sub-audio data is calculated to obtain multiple variance values, and the variance values ​​are used to measure the amplitude of the audio signal changing over time, thereby detecting whether there are unnatural fluctuations or mutations in the audio. The variance values ​​are then compared according to the variance threshold interval. If all variance values ​​are within the preset variance threshold interval, it indicates that the audio signal has good stability within the preset time window, and the step of vectorization processing can be further performed; conversely, if there is a variance value exceeding the preset interval, this may indicate that there are unnatural fluctuations in the audio signal, and a verification result of "verification abnormality" will be generated, indicating that the audio may be abnormal audio. This embodiment can effectively identify unnatural fluctuations by detecting and verifying the stability of audio data, thereby significantly improving the accuracy of audio verification, reducing false positives and negatives, and ensuring that only audio data with good stability enters the next step of vectorization processing and feature recognition, thereby improving the efficiency of the entire audio verification process.

[0068] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0069] Example 2

[0070] The embodiment of the present application also provides a user verification device. It should be noted that the user verification device of the embodiment of the present application can be used to execute the user verification method provided by the embodiment of the present application. The user verification device provided by the embodiment of the present application is introduced below.

[0071] According to an embodiment of the present application, a device for implementing the above-mentioned user verification method is also provided. Figure 3 is a schematic diagram of a user verification device provided according to an embodiment of the present application, such as Figure 3 As shown, the device includes: a determination unit 30, a receiving unit 31, and a verification unit 32.

[0072] The determination unit 30 is used to determine the verification strategy according to the financial verification request when receiving the financial verification request sent by the user to be verified, and send verification information to the user to be verified according to the verification strategy, wherein the verification strategy refers to the strategy for performing identity authentication on the user to be verified, and the verification information refers to prompting the user to be verified to perform identity authentication according to the verification strategy;

[0073] The receiving unit 31 is used to receive the data to be verified sent by the user to be verified, and extract M verification sub-data from the data to be verified, wherein the data to be verified refers to the data generated by the user to be verified after receiving the verification information and performing identity authentication, and the M verification sub-data at least include verification video data, verification audio data and verification image data, and M is a positive integer;

[0074] The verification unit 32 is used to verify the M verification sub-data respectively to obtain M verification results, and generate authentication failure prompt information when any verification result among the M verification results is abnormal.

[0075] The user verification device provided in the embodiment of the present application determines a verification strategy according to the financial verification request when a determination unit 30 receives a financial verification request sent by a user to be verified, and sends verification information to the user to be verified according to the verification strategy, wherein the verification strategy refers to a strategy for identity authentication of the user to be verified, and the verification information refers to prompting the user to be verified to perform identity authentication according to the verification strategy; the receiving unit 31 receives the data to be verified sent by the user to be verified, and extracts M verification sub-data from the data to be verified, wherein the data to be verified refers to data generated by the user to be verified after receiving the verification information and performing identity authentication, and the M verification sub-data The data at least includes verification video data, verification audio data and verification image data, and M is a positive integer; the verification unit 32 verifies the M verification sub-data respectively to obtain M verification results. When any verification result among the M verification results is a verification abnormality, an authentication failure prompt message is generated, which solves the problems of low verification accuracy and low detection efficiency when performing security verification on users in related technologies. By receiving the data to be verified sent by the user to be verified, the verification sub-data is extracted from the data to be verified, the verification sub-data is verified, and the verification result is obtained, thereby achieving the technical effect of improving the accuracy and efficiency of user security verification.

[0076] Optionally, in the user verification device provided in the embodiment of the present application, the determination unit 30 includes: a first extraction module, used to extract the financial behavior type from the financial verification request, wherein the financial behavior type includes a financial transaction type and a device detection type; a first determination module, used to determine the verification strategy as a first verification strategy when the financial behavior type is a financial transaction type, wherein the first verification strategy includes performing video verification and audio verification on the user to be verified; a second determination module, used to determine the verification strategy as a second verification strategy when the financial behavior type is a device detection type, wherein the second verification strategy includes performing audio verification and image verification on the user to be verified.

[0077] Optionally, in the user verification device provided in the embodiment of the present application, the verification unit 32 includes: a capture module, used to capture fragments of a verification video data to obtain N video fragments, where N is a positive integer; a selection module, used to select K video fragments from the N video fragments, and capture video frames of each video fragment in the K video fragments to obtain K groups of video frames, where each group of video frames includes Y video frames, and Y is a positive integer; a verification module, used to perform a consistency check on Y video frames for a group of video frames to obtain a consistency check result, and determine the verification result of the verification video data according to the K consistency check results of the K groups of video frames.

[0078] Optionally, in the user verification device provided in the embodiment of the present application, the verification unit 32 includes: a second extraction module, used to use an encoder to extract frame information of each video frame to obtain Y frame information, wherein each frame information is used to characterize the time information of each video frame; a comparison module, used to perform feature comparison on every two video frames in the Y video frames according to each frame information, to obtain L comparison results, wherein feature comparison refers to performing optical flow difference verification and motion continuity verification on two adjacent video frames, and L is a positive integer; a third determination module, used to determine the difference value corresponding to each comparison result, and generate a continuity verification result with verification abnormality when any difference value among the L difference values ​​is greater than a preset threshold; a first generation module, used to generate a continuity verification result with no verification abnormality when all L difference values ​​are less than a preset threshold.

[0079] Optionally, in the user verification device provided in the embodiment of the present application, the verification unit 32 includes: a processing module, which is used to perform vectorization processing on a verification audio data to obtain a verification audio vector; a segmentation module, which is used to perform signal segmentation on the verification audio vector to obtain P audio sub-vectors, and perform signal transformation processing on each audio sub-vector to obtain P frequency domain audio signals, wherein P is a positive integer; an acquisition module, which is used to obtain a sound spectrum recognition algorithm, use the sound spectrum recognition algorithm to process the P frequency domain audio signals, and output a feature recognition result, wherein the feature recognition result is used to characterize whether the verification audio data is real audio; a second generation module, which is used to generate a verification result with no verification abnormality when the feature recognition result characterizes that the verification audio data is real audio; and a third generation module, which is used to generate a verification result with verification abnormality when the feature recognition result characterizes that the verification audio data is not real audio.

[0080] Optionally, in the user verification device provided in the embodiment of the present application, the verification unit 32 includes: an input module, used to input P frequency domain audio signals into a Mel filter and output P Mel spectra; a first calculation module, used to perform logarithmic calculation on each Mel spectrum to obtain P processed spectra, and perform discrete cosine processing on each processed spectrum to obtain P audio recognition coefficients; a second calculation module, used to perform differential calculation on each audio recognition coefficient to generate a feature recognition result.

[0081] Optionally, in the user verification device provided in the embodiment of the present application, the device also includes: an acquisition unit, used to acquire a preset time window before vectorizing the verification audio data, and divide the verification audio data according to the preset time window to obtain Q sub-audio data, where Q is a positive integer; a calculation unit, used to calculate the variance of two adjacent sub-audio data in the Q sub-audio data to obtain multiple variance values; an execution unit, used to execute the step of vectorizing the verification audio data when each variance value is within a preset variance threshold interval; a generation unit, used to generate a verification result of a verification abnormality if there is a variance value outside the preset variance threshold interval.

[0082] It should be noted that the above-mentioned determination unit 30, receiving unit 31, and verification unit 32 correspond to steps S201 to S203 in Example 1, and the above-mentioned units and corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules or units may be hardware components or software components stored in a memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n), and the above-mentioned units may also be part of the device and may be run in the computer terminal 10 provided in Example 1.

[0083] Example 3

[0084] The embodiment of the present application may provide a computer terminal, which may be any computer terminal device in a computer terminal group. Optionally, in this embodiment, the computer terminal may also be replaced by a terminal device such as a mobile terminal or an electronic device.

[0085] Optionally, in this embodiment, the computer terminal may be located in at least one network device among a plurality of network devices of the computer network.

[0086] In this embodiment, the above-mentioned computer terminal can execute the program code of the following steps in the user's verification method: when receiving a financial verification request sent by the user to be verified, determining the verification strategy according to the financial verification request, and sending verification information to the user to be verified according to the verification strategy, wherein the verification strategy refers to the strategy for identity authentication of the user to be verified, and the verification information refers to prompting the user to be verified to perform identity authentication according to the verification strategy; receiving the data to be verified sent by the user to be verified, and extracting M verification sub-data from the data to be verified, wherein the data to be verified refers to the data generated by the user to be verified after receiving the verification information and performing identity authentication, the M verification sub-data at least include verification video data, verification audio data and verification image data, and M is a positive integer; verifying the M verification sub-data respectively to obtain M verification results, and generating an authentication failure prompt message when any verification result among the M verification results is a verification abnormality.

[0087] Optionally, the above-mentioned computer terminal can execute the program code of the following steps in the user's verification method: extracting the financial behavior type from the financial verification request, wherein the financial behavior type includes a financial transaction type and a device detection type; when the financial behavior type is a financial transaction type, determining the verification strategy as a first verification strategy, wherein the first verification strategy includes performing video verification and audio verification on the user to be verified; when the financial behavior type is a device detection type, determining the verification strategy as a second verification strategy, wherein the second verification strategy includes performing audio verification and image verification on the user to be verified.

[0088] Optionally, the computer terminal may execute the program code of the following steps in the user verification method: for a verification video data, segment the verification video data to obtain N video segments, where N is a positive integer; select K video segments from the N video segments, segment the video frames of each of the K video segments to obtain K groups of video frames, where each group of video frames includes Y video frames, where Y is a positive integer; for a group of video frames, perform a consistency check on the Y video frames to obtain a consistency check result, and determine the verification result of the verification video data according to the K consistency check results of the K groups of video frames.

[0089] Optionally, the computer terminal may execute the program code of the following steps in the user verification method: extracting frame information of each video frame using an encoder to obtain Y frame information, wherein each frame information is used to characterize the time information of each video frame; performing feature comparison on every two video frames in the Y video frames according to each frame information to obtain L comparison results, wherein feature comparison refers to performing optical flow difference verification and motion continuity verification on two adjacent video frames, and L is a positive integer; determining the difference value corresponding to each comparison result, and generating a continuity verification result with a verification abnormality when any difference value among the L difference values ​​is greater than a preset threshold; generating a continuity verification result with no verification abnormality when all the L difference values ​​are less than the preset threshold.

[0090] Optionally, the above-mentioned computer terminal can execute the program code of the following steps in the user's verification method: for a verification audio data, vectorize the verification audio data to obtain a verification audio vector; perform signal segmentation on the verification audio vector to obtain P audio sub-vectors, and perform signal transformation processing on each audio sub-vector to obtain P frequency domain audio signals, where P is a positive integer; obtain a sound spectrum recognition algorithm, use the sound spectrum recognition algorithm to process the P frequency domain audio signals, and output a feature recognition result, where the feature recognition result is used to characterize whether the verification audio data is real audio; when the feature recognition result characterizes that the verification audio data is real audio, generate a verification result with no verification anomalies; when the feature recognition result characterizes that the verification audio data is not real audio, generate a verification result with verification anomalies.

[0091] Optionally, the computer terminal may execute the program code of the following steps in the user verification method: inputting P frequency domain audio signals into a Mel filter and outputting P Mel spectra; performing logarithmic calculation on each Mel spectrum to obtain P processed spectra, and performing discrete cosine processing on each processed spectrum to obtain P audio recognition coefficients; performing differential calculation on each audio recognition coefficient to generate a feature recognition result.

[0092] Optionally, the above-mentioned computer terminal can execute the program code of the following steps in the user's verification method: obtain a preset time window, divide the verification audio data according to the preset time window, and obtain Q sub-audio data, where Q is a positive integer; calculate the variance of two adjacent sub-audio data in the Q sub-audio data to obtain multiple variance values; when each variance value is within a preset variance threshold interval, execute the step of vectorizing the verification audio data; if there is a variance value outside the preset variance threshold interval, generate a verification result of a verification abnormality.

[0093] Optionally, Figure 4 is a structural block diagram of an electronic device according to an embodiment of the present application. Figure 4 As shown, the electronic device may include: one or more ( Figure 4 (only one is shown) processor 402, memory 404, storage controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0094] Among them, the memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the user's verification method and device in the embodiment of the present application, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the above-mentioned user's verification method. The memory 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 instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the 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 combinations thereof.

[0095] The processor can call the information and application programs stored in the memory through the transmission device to execute the above steps in the above user verification method.

[0096] By adopting the embodiment of the present application, a user verification scheme is provided. By receiving a financial verification request sent by a user to be verified, determining a verification strategy according to the financial verification request, and sending verification information to the user to be verified according to the verification strategy, wherein the verification strategy refers to a strategy for identity authentication of the user to be verified, and the verification information refers to prompting the user to be verified to authenticate the identity according to the verification strategy; receiving the data to be verified sent by the user to be verified, and extracting M verification sub-data from the data to be verified, wherein the data to be verified refers to the data generated by the identity authentication of the user to be verified after receiving the verification information, and the M verification sub-data at least include verification video data, verification audio data and verification image data, and M is a positive integer; respectively verifying the M verification sub-data to obtain M verification results, and generating an authentication failure prompt message when any verification result in the M verification results is a verification abnormality, thereby achieving the purpose of improving the verification defense capability and optimizing the user experience, and further solving the problem of low verification accuracy and low detection efficiency when performing security verification on users.

[0097] It can be understood by those skilled in the art that Figure 4 The structure shown is for illustration only, and the electronic device may also be a smart phone, a tablet computer, a PDA, a mobile Internet device (MID), a PAD or other terminal device. Figure 4The structure of the electronic device is not limited. Figure 4 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 4 Different configurations are shown.

[0098] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0099] Example 4

[0100] The embodiment of the present application further provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the user verification method provided in the first embodiment.

[0101] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0102] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: upon receiving a financial verification request sent by the user to be verified, determining a verification strategy based on the financial verification request, and sending verification information to the user to be verified based on the verification strategy, wherein the verification strategy refers to a strategy for identity authentication of the user to be verified, and the verification information refers to prompting the user to be verified to authenticate the user according to the verification strategy; receiving data to be verified sent by the user to be verified, and extracting M verification sub-data from the data to be verified, wherein the data to be verified refers to data generated by the user to be verified after receiving the verification information and performing identity authentication, the M verification sub-data at least include verification video data, verification audio data, and verification image data, and M is a positive integer; respectively verifying the M verification sub-data to obtain M verification results, and generating an authentication failure prompt message if any of the M verification results is a verification abnormality.

[0103] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing the steps of the user authentication method.

[0104] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0105] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0106] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0107] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0108] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0109] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk, etc., which can store program code.

[0110] The above are only preferred implementations of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A user verification method, characterized in that: include: Upon receiving a financial verification request sent by a user to be verified, determining a verification strategy according to the financial verification request, and sending verification information to the user to be verified according to the verification strategy, wherein the verification strategy refers to a strategy for identity verification of the user to be verified, and the verification information refers to prompting the user to be verified to perform identity verification according to the verification strategy; Receive the data to be verified sent by the user to be verified, and extract M verification sub-data from the data to be verified, wherein the data to be verified refers to data generated by the user to be verified after receiving the verification information and performing identity authentication, and the M verification sub-data at least include verification video data, verification audio data and verification image data, and M is a positive integer; The M verification sub-data are verified respectively to obtain M verification results. When any verification result among the M verification results is a verification abnormality, an authentication failure prompt message is generated.

2. The method according to claim 1, characterized in that Determining a verification strategy according to the financial verification request includes: Extracting a financial behavior type from the financial verification request, wherein the financial behavior type includes a financial transaction type and a device detection type; In a case where the financial behavior type is the financial transaction type, determining the verification strategy as a first verification strategy, wherein the first verification strategy includes performing video verification and audio verification on the user to be verified; In a case where the financial behavior type is the device detection type, the verification strategy is determined to be a second verification strategy, wherein the second verification strategy includes performing the audio verification and the image verification on the user to be verified.

3. The method according to claim 1, characterized in that: In the case where the M verification sub-data are M verification video data, the M verification sub-data are verified respectively to obtain M verification results including: For a verification video data, segment the verification video data to obtain N video segments, where N is a positive integer; Selecting K video segments from the N video segments, and intercepting video frames of each video segment from the K video segments to obtain K groups of video frames, wherein each group of video frames includes Y video frames, where Y is a positive integer; For a group of video frames, a consistency check is performed on the Y video frames to obtain a consistency check result, and the check result of the verification video data is determined by the K consistency check results of the K groups of video frames.

4. The method according to claim 3, characterized in that: The consistency check is performed on the Y video frames to obtain a consistency check result including: Extracting frame information of each video frame using an encoder to obtain Y pieces of frame information, wherein each piece of frame information is used to represent time information of each video frame; Performing feature comparison on every two video frames in the Y video frames according to each frame information to obtain L comparison results, wherein the feature comparison refers to performing optical flow difference verification and action continuity verification on two adjacent video frames, and L is a positive integer; Determine the difference value corresponding to each comparison result, and generate a consistency check result of a check exception when any difference value among the L difference values ​​is greater than a preset threshold; When the L difference values ​​are all smaller than the preset threshold, a consistency check result indicating no abnormality is generated.

5. The method according to claim 1, characterized in that In the case where the M verification sub-data are M verification audio data, the M verification sub-data are verified respectively, and the M verification results obtained include: For a verification audio data, vectorization processing is performed on the verification audio data to obtain a verification audio vector; Performing signal segmentation on the verification audio vector to obtain P audio sub-vectors, and performing signal transformation processing on each audio sub-vector to obtain P frequency domain audio signals, where P is a positive integer; Acquire a sound spectrum recognition algorithm, use the sound spectrum recognition algorithm to process the P frequency domain audio signals, and output a feature recognition result, wherein the feature recognition result is used to characterize whether the verification audio data is real audio; In the case where the feature recognition result indicates that the verification audio data is the real audio, generating a verification result indicating that there is no abnormality in the verification; When the feature recognition result indicates that the verification audio data is not the real audio, a verification result indicating a verification abnormality is generated.

6. The method according to claim 5, characterized in that The P frequency domain audio signals are processed using the sound spectrum recognition algorithm, and the output feature recognition results include: Input the P frequency domain audio signals into a Mel filter, and output P Mel spectra; Perform logarithmic calculation on each Mel spectrum to obtain P processed spectrums, and perform discrete cosine processing on each processed spectrum to obtain P audio recognition coefficients; A differential calculation is performed on each audio recognition coefficient to generate the feature recognition result.

7. The method according to claim 5, characterized in that Before vectorizing the verification audio data, the method further includes: Obtain a preset time window, and divide the verification audio data according to the preset time window to obtain Q sub-audio data, where Q is a positive integer; Calculating the variance of two adjacent sub-audio data among the Q sub-audio data to obtain multiple variance values; When each variance value is within a preset variance threshold interval, executing the step of performing vectorization processing on the verification audio data; If there is a variance value outside the preset variance threshold range, a verification result of verification abnormality is generated.

8. A user verification device, characterized in that: include: a determination unit, configured to, upon receiving a financial verification request sent by a user to be verified, determine a verification strategy according to the financial verification request, and send verification information to the user to be verified according to the verification strategy, wherein the verification strategy refers to a strategy for performing identity authentication on the user to be verified, and the verification information refers to prompting the user to be verified to perform identity authentication according to the verification strategy; a receiving unit, configured to receive the data to be verified sent by the user to be verified, and extract M verification sub-data from the verification data, wherein the data to be verified refers to data generated by the user to be verified after receiving the verification information and performing identity authentication, and the M verification sub-data at least include verification video data, verification audio data and verification image data, and M is a positive integer; The verification unit is used to verify the M verification sub-data respectively to obtain M verification results, and generate authentication failure prompt information when any verification result among the M verification results is a verification abnormality.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is executed, the device where the computer-readable storage medium is located is controlled to execute the user verification method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor is used to run the program, wherein the program executes the user verification method described in any one of claims 1 to 7 when running.

11. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the user verification method described in any one of claims 1 to 7 are implemented.

Images