Earphone security encryption method based on fingerprint identification

Abstract

The application relates to the technical field of fingerprint identification, in particular to an earphone security encryption method based on fingerprint identification. The method comprises the following steps: integrating a micro fingerprint sensor on an earphone shell; when a user wears the earphone, collecting a fingerprint image of the user through the sensor to obtain original fingerprint image data; carrying out denoising and image enhancement on the original fingerprint image data to obtain high-quality fingerprint image data; carrying out fingerprint feature extraction on the fingerprint image data, identifying bifurcation points and terminal points to obtain fingerprint feature data; and carrying out fingerprint feature matching on the fingerprint feature data and a preset legal user fingerprint template to obtain fingerprint matching result data. The application does not need to upload the complete fingerprint image of the user, only needs to match the extracted fingerprint features with the local template, the fingerprint data of the user is always saved on the equipment side, and the risk of privacy leakage is avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the fingerprint identification technical field, and especially relates to an earphone security encryption method based on fingerprint identification. BACKGROUND

[0002] With the popularity of Bluetooth earphones, TWS earphones and other wireless audio devices, users have higher requirements for the security and privacy protection of audio transmission. At the same time, the progress of biometric technology provides a new solution for device security. The traditional authentication method has the risk of being cracked, forgotten and poor user experience. The earphone security encryption method based on fingerprint identification uses fingerprints, a biological feature, for identity authentication. Fingerprint is unique and difficult to forge, which greatly improves the authentication security. At the same time, the action of wearing earphones can naturally trigger fingerprint collection, without the need for manual input, improving convenience. However, the traditional earphone security encryption method based on fingerprint identification often has the following problems: earphones are easy to lose or be stolen, and there is no effective anti-theft mechanism for unauthorized users; the traditional password authentication method is inconvenient to operate on small devices such as earphones; and the conventional encryption method is difficult to securely store and update keys on small devices. SUMMARY

[0003] Therefore, it is necessary to provide an earphone security encryption method based on fingerprint identification to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned purpose, an earphone security encryption method based on fingerprint identification includes the following steps:

[0005] Step S1: integrating a micro fingerprint sensor on the earphone shell, collecting the fingerprint image of the user through the sensor when the user wears the earphone, and obtaining the original fingerprint image data; denoising and image enhancement are performed on the original fingerprint image data, thereby obtaining high-quality fingerprint image data;

[0006] Step S2: fingerprint feature extraction is performed on the fingerprint image data, the bifurcation point and the end point are identified, and the fingerprint feature data is obtained; fingerprint feature matching is performed between the fingerprint feature data and the pre-set legal user fingerprint template, thereby obtaining fingerprint matching result data; when the fingerprint matching result data shows a successful match, a unique user identity data is generated, and step S3 is performed, otherwise step S5 is performed;

[0007] Step S3: generating unique key data according to the fingerprint feature data, binding the user identity data with the key data to generate the bound key data, encrypting the audio transmitted by the earphone using the generated bound key data to obtain encrypted audio data, sending the encrypted audio data to the earphone receiving end through wireless transmission technology, and decrypting the received encrypted audio data using the generated bound key data to obtain the original audio data at the earphone receiving end;

[0008] Step S4: updating the key data according to a preset period and updating the bound user identity data to obtain updated key data, wherein the updated key data is generated according to the latest fingerprint features;

[0009] Step S5: judging the fingerprint matching result data according to the maximum number of user authentication attempts set in advance, recording the time and place of fingerprint recognition to obtain context data, detecting abnormal behavior on the context data to obtain abnormal behavior detection data, triggering a dynamic password verification mechanism according to the abnormal behavior detection data to obtain multi-factor authentication result data.

[0010] The application can realize user identity authentication by integrating a miniature fingerprint sensor. Fingerprint is a unique biological feature, and through collecting and processing fingerprint image, the user's identity can be accurately identified. By denoising and image enhancement processing of the original fingerprint image data, the quality and clarity of the fingerprint image can be improved, thereby improving the accuracy and reliability of fingerprint identification. By feature extraction and matching of the fingerprint image data, the fingerprint can be represented as a set of feature data for identity verification. Fingerprint features are unique and stable, and can accurately distinguish different users. When the fingerprint matches successfully, a unique user identity data can be generated for subsequent security authentication and data encryption process. By generating unique key data and binding it with the user identity, the security of the data can be ensured, and only the correct key data and legal user identity can decrypt and access the audio data, improving the confidentiality and protection of the data. By encrypting the audio data transmitted by the earphone, unauthorized persons can be prevented from eavesdropping or tampering with the data during transmission, ensuring the confidentiality and integrity of the communication. Regularly updating the key data can enhance the security of the system. By generating updated key data based on the latest fingerprint features, the timeliness and effectiveness of the key can be ensured, and potential security vulnerabilities can be reduced. By setting the maximum number of user authentication attempts and locking the matching result data, malicious users can be prevented from performing brute force attacks or repeated attempts, improving the security of the system. Recording context data such as fingerprint recognition time and location, and detecting abnormal behavior, can identify abnormal authentication behavior such as multiple attempts to authenticate in a short period of time or authentication in an abnormal location, thereby providing additional security. According to the abnormal behavior detection data, a dynamic password verification mechanism can be triggered to introduce additional authentication factors, increasing the complexity and security of authentication. The dynamic password can be a one-time password, SMS verification code or other forms of authentication, further ensuring the legitimacy of the user's identity. In summary, the above steps include user identity authentication, high-quality fingerprint image data, fingerprint feature extraction and matching, user identity generation, data security, encrypted communication, key update, security enhancement, abnormal behavior detection and multi-factor authentication, etc. These steps can be combined to build a secure and reliable fingerprint identification and authentication system, providing high-level user identity protection and data security protection.

[0011] Preferably, step S1 comprises the following steps:

[0012] Step S11: Integrate a miniature fingerprint sensor on the earphone shell. When the user wears the earphone, the sensor collects the user's fingerprint image to obtain the original fingerprint image data.

[0013] Step S12: Perform grayscale processing on the original fingerprint image data to obtain grayscale fingerprint image data.

[0014] Step S13: applying a Gaussian filter algorithm to the grayscale fingerprint image data to remove image noise and obtain denoised fingerprint image data;

[0015] Step S14: performing histogram equalization processing on the denoised fingerprint image data and improving image contrast to obtain equalized fingerprint image data;

[0016] Step S15: applying an adaptive binarization algorithm to the equalized fingerprint image data to convert the image into a black-and-white binary image and obtain binarized fingerprint image data;

[0017] Step S16: performing thinning processing on the binarized fingerprint image data and extracting fingerprint ridge lines to obtain thinned fingerprint image data;

[0018] Step S17: performing image enhancement processing on the thinned fingerprint image data to obtain high-quality fingerprint image data.

[0019] The fingerprint is a unique biological feature of each person, and accurate identity authentication can be performed by collecting the fingerprint image data of the user. In this way, only the authorized user can unlock and use the earphone when wearing the earphone, thereby improving the security and privacy of the device. Grayscale can convert a color image into a single-channel grayscale image, simplifying the complexity and computational load of subsequent processing steps. The Gaussian filter algorithm can effectively remove noise in the image, improve the clarity and quality of the fingerprint image, and reduce errors and interference in subsequent processing steps. Histogram equalization can adjust the brightness distribution of the image and enhance the contrast of the image, making the details in the fingerprint image more clear and visible. Binarized images can highlight the outlines and details of the fingerprint pattern, facilitating subsequent fingerprint feature extraction and matching. Thinning processing can further reduce the ridge line width in the fingerprint image, enhance the details of the fingerprint pattern, and improve the robustness and reliability of the fingerprint features. Image enhancement processing can further improve the quality and clarity of the fingerprint image, reduce noise and interference in the image, and thereby improve the accuracy and reliability of fingerprint recognition. The above steps can optimize and enhance the original fingerprint image data, improve the performance and effect of the fingerprint recognition system. The effects of these steps include user identity authentication, image quality improvement, noise removal, contrast enhancement, feature extraction and enhancement, and improvement of fingerprint recognition accuracy.

[0020] Preferably, step S17 includes the following steps:

[0021] Step S171: performing morphological processing on the thinned fingerprint image data to obtain repaired fingerprint image data, wherein the morphological processing is an opening operation of erosion followed by dilation;

[0022] Step S172: performing clarity enhancement processing on the repaired fingerprint image data to obtain enhanced fingerprint image data.

[0023] Step S173: Perform orientation field estimation based on local orientation of fingerprint texture on the segmented fingerprint image data to obtain fingerprint orientation field data;

[0024] Step S174: Perform anisotropic filtering on the segmented fingerprint image data based on the orientation field data to obtain high-quality fingerprint image data.

[0025] This invention removes small breaks or noise from fingerprint images through erosion, making the fingerprint texture more continuous and complete. Through dilation, it fills holes or breaks in the fingerprint image, repairing missing parts and improving readability and continuity. Enhancing the texture details makes the ridges clearer, improving the accuracy and reliability of fingerprint recognition. Analyzing the texture features allows estimation of the local orientation field, i.e., the direction of the fingerprint ridges, which is crucial for subsequent fingerprint feature extraction and matching. Anisotropic filtering uses information from the fingerprint orientation field to perform nonlinear filtering, emphasizing ridge details and further enhancing the quality and texture features of the fingerprint image. In summary, the above fingerprint image processing steps further improve the quality and texture features of fingerprint images, including removing small breaks, filling holes, texture enhancement, orientation field estimation, and fingerprint enhancement. These steps improve fingerprint recognition accuracy, enhance fingerprint image readability and continuity, and emphasize ridge details, thereby improving the robustness and reliability of fingerprint features. Attached Figure Description

[0026] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0027] Figure 1 This is a flowchart illustrating the steps of the fingerprint recognition-based headphone security encryption method of the present invention.

[0028] Figure 2 for Figure 1 A detailed flowchart of step S1;

[0029] Figure 3 for Figure 1 A detailed flowchart of step S2. Detailed Implementation

[0030] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0031] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0032] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0033] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a fingerprint recognition-based headphone security encryption method, the method comprising the following steps:

[0034] Step S1: Integrate a miniature fingerprint sensor on the earphone shell. When the user wears the earphone, the sensor collects the user's fingerprint image to obtain the original fingerprint image data. The original fingerprint image data is then denoised and enhanced to obtain high-quality fingerprint image data.

[0035] Step S2: Extract fingerprint features from the fingerprint image data, identify bifurcation points and endpoints to obtain fingerprint feature data; match the fingerprint feature data with the preset legitimate user fingerprint template to obtain fingerprint matching result data. If the fingerprint matching result data shows a successful match, generate unique user identification data and proceed to step S3; otherwise, proceed to step S5.

[0036] Step S3: generating unique key data according to the fingerprint feature data, binding the user identity data with the key data to generate the bound key data, encrypting the audio transmitted by the earphone by using the bound key data to obtain encrypted audio data, sending the encrypted audio data to the earphone receiving end by using the wireless transmission technology, and decrypting the received encrypted audio data by using the bound key data to obtain the original audio data at the earphone receiving end;

[0037] Step S4: updating the key data according to a preset period, and updating the bound user identity data to obtain the updated key data, wherein the updated key data is generated according to the latest fingerprint feature;

[0038] Step S5: judging the fingerprint matching result data according to a preset maximum number of user authentication attempts, recording the time and place of fingerprint recognition to obtain context data, detecting abnormal behaviors according to the context data to obtain abnormal behavior detection data, triggering a dynamic password verification mechanism according to the abnormal behavior detection data to obtain multi-factor authentication result data.

[0039] In the embodiment of the application, reference is made to Figure 1 The application is a step flow diagram of an earphone security encryption method based on fingerprint recognition, which includes the following steps in the example:

[0040] Step S1: integrating a micro fingerprint sensor on the earphone shell, collecting the fingerprint image of the user by the sensor when the user wears the earphone to obtain original fingerprint image data, and performing denoising and image enhancement on the original fingerprint image data to obtain high-quality fingerprint image data.

[0041] In the embodiment of the application, a micro fingerprint sensor is integrated on a suitable position of the earphone shell, and the sensor starts to collect the fingerprint image data of the user when the user wears the earphone. The collected fingerprint image data may contain noise or have low quality, and therefore needs to be denoised to remove interference factors, and the denoised fingerprint image data needs to be image-enhanced to improve the clarity and recognizability of the fingerprint image. The processed high-quality fingerprint image data is used for subsequent fingerprint feature extraction and matching.

[0042] Step S2: extracting the fingerprint feature data from the fingerprint image data, identifying the bifurcation point and the terminal point to obtain the fingerprint feature data, performing fingerprint feature matching between the fingerprint feature data and the preset legal user fingerprint template to obtain the fingerprint matching result data, generating unique user identity data when the fingerprint matching result data shows a successful match, and performing step S3, or performing step S5 when the fingerprint matching result data does not show a successful match.

[0043] The fingerprint feature extraction is performed on the collected fingerprint image data, and the feature information in the fingerprint image is extracted by using image processing and pattern recognition algorithms, such as detail enhancement, direction calculation, etc. In the fingerprint feature extraction process, key features such as bifurcation points and ending points in the fingerprint image are identified, and these features can uniquely identify a fingerprint pattern. The extracted fingerprint feature data is matched with a preset legal user fingerprint template, and the similarity or matching degree is compared. If the fingerprint matching result data shows that the matching is successful, that is, the fingerprint feature matches the legal user fingerprint template, a unique user identity identification data is generated, indicating that the user identity verification is passed, and the next step S3 is entered. If the fingerprint matching result data shows that the matching fails, that is, the fingerprint feature does not match the legal user fingerprint template, the user identity verification fails, and the next step S5 is entered.

[0044] Step S3: According to the fingerprint feature data, a unique key data is generated, and the user identity identification data is bound with the key data to generate bound key data. The audio transmitted by the earphone is encrypted by using the generated bound key data, so that encrypted audio data is obtained. The encrypted audio data is sent to the earphone receiving end through wireless transmission technology. In the earphone receiving end, the received encrypted audio data is decrypted by using the generated bound key data, so that the original audio data is obtained.

[0045] According to the fingerprint feature data verified, a unique key data is generated, and the user identity identification data is bound with the generated key data to ensure the uniqueness of the key and the association with the user identity. The audio transmitted by the earphone is encrypted by using the generated bound key data, and the symmetric encryption algorithm such as AES (Advanced Encryption Standard) can be used to encrypt the audio data to ensure the confidentiality of the transmission process. The encrypted audio data is sent to the earphone receiving end through wireless transmission technology (such as Bluetooth or Wi-Fi). In the earphone receiving end, the received encrypted audio data is decrypted by using the generated bound key data, and the original audio data is restored.

[0046] Step S4: According to the preset period, the key data is updated, and the bound user identity identification data is updated to obtain updated key data, wherein the updated key data is generated according to the latest fingerprint feature;

[0047] According to the preset period, for example, the key data is updated regularly every day or every week to enhance security. In the updating process, the updated key data is generated according to the latest fingerprint feature data to ensure the timeliness of the key and the association with the user fingerprint. At the same time, the bound user identity identification data is updated to maintain consistency with the latest key data.

[0048] Step S5: lock judgment is made on the fingerprint matching result data according to the pre-set maximum number of user authentication attempts, and context data is obtained by recording the time and location of fingerprint recognition; abnormal behavior detection is performed on the context data to obtain abnormal behavior detection data; a dynamic password verification mechanism is triggered according to the abnormal behavior detection data to obtain multi-factor authentication result data.

[0049] In the embodiment of the application, the fingerprint matching result data is judged according to the pre-set maximum number of user authentication attempts. When the maximum number is reached, the authentication system can be locked to prevent further authentication attempts. Meanwhile, context data such as the time and location of fingerprint recognition during the authentication process is recorded for subsequent abnormal behavior detection. Abnormal behavior detection is performed on the context data, which can use machine learning or rule engine methods to identify abnormal authentication behaviors such as frequent authentication attempts, unusual times or locations, etc. A dynamic password verification mechanism is triggered according to the abnormal behavior detection data, which can require the user to provide additional authentication factors, such as sending a dynamic password through a mobile phone message for verification. Finally, multi-factor authentication result data is obtained, which represents the final result of the authentication and can be used to determine whether the authentication is successful and further control the access rights of the system.

[0050] The application can realize user identity authentication by integrating a miniature fingerprint sensor. Fingerprint is a unique biological feature, and through collecting and processing fingerprint image, the user's identity can be accurately identified. By denoising and image enhancement processing of the original fingerprint image data, the quality and clarity of the fingerprint image can be improved, thereby improving the accuracy and reliability of fingerprint identification. By feature extraction and matching of the fingerprint image data, the fingerprint can be represented as a set of feature data for identity verification. Fingerprint features are unique and stable, and can accurately distinguish different users. When the fingerprint matches successfully, a unique user identity data can be generated for subsequent security authentication and data encryption process. By generating unique key data and binding it with the user identity, the security of the data can be ensured, and only the correct key data and legal user identity can decrypt and access the audio data, improving the confidentiality and protection of the data. By encrypting the audio data transmitted by the earphone, unauthorized persons can be prevented from eavesdropping or tampering with the data during transmission, ensuring the confidentiality and integrity of the communication. Regularly updating the key data can enhance the security of the system. By generating updated key data based on the latest fingerprint features, the timeliness and effectiveness of the key can be ensured, and potential security vulnerabilities can be reduced. By setting the maximum number of user authentication attempts and locking the matching result data, malicious users can be prevented from performing brute force attacks or repeated attempts, improving the security of the system. Recording context data such as fingerprint recognition time and location, and detecting abnormal behavior, can identify abnormal authentication behavior such as multiple attempts to authenticate in a short period of time or authentication in an abnormal location, thereby providing additional security. According to the abnormal behavior detection data, a dynamic password verification mechanism can be triggered to introduce additional authentication factors, increasing the complexity and security of authentication. The dynamic password can be a one-time password, SMS verification code or other forms of authentication, further ensuring the legitimacy of the user's identity. In summary, the above steps include user identity authentication, high-quality fingerprint image data, fingerprint feature extraction and matching, user identity generation, data security, encrypted communication, key update, security enhancement, abnormal behavior detection and multi-factor authentication, etc. These steps can be combined to build a secure and reliable fingerprint identification and authentication system, providing high-level user identity protection and data security protection.

[0051] Preferably, step S1 comprises the following steps:

[0052] Step S11: integrating a miniature fingerprint sensor on the earphone shell, when the user wears the earphone, the fingerprint image of the user is collected by the sensor to obtain the original fingerprint image data;

[0053] Step S12: performing grayscale processing on the original fingerprint image data to obtain grayscale fingerprint image data;

[0054] Step S13: applying a Gaussian filter algorithm to the gray-scale fingerprint image data to remove image noise, to obtain denoised fingerprint image data;

[0055] Step S14: performing histogram equalization processing on the denoised fingerprint image data, and improving image contrast, to obtain equalized fingerprint image data;

[0056] Step S15: applying an adaptive binarization algorithm to the equalized fingerprint image data, to convert the image into a black-and-white binary image, to obtain binarized fingerprint image data;

[0057] Step S16: performing thinning processing on the binarized fingerprint image data, and performing fingerprint ridge line extraction, to obtain thinned fingerprint image data;

[0058] Step S17: performing image enhancement processing on the thinned fingerprint image data, to obtain high-quality fingerprint image data.

[0059] As an embodiment of the present application, referring to Figure 2 , a detailed step flowchart of step S1 in the present application is shown, and step S1 in the embodiment of the present application includes the following steps: Figure 1

[0060] Step S11: integrating a micro fingerprint sensor on the earphone shell, when a user wears the earphone, collecting the fingerprint image of the user through the sensor, to obtain original fingerprint image data;

[0061] In the embodiment of the present application, a suitable micro fingerprint sensor is selected and integrated into a proper position of the earphone shell; when a user wears the earphone, the sensor starts to collect the fingerprint image data of the user, and the collection is performed through an optical or capacitive method.

[0062] Step S12: performing gray-scale processing on the original fingerprint image data, to obtain gray-scale fingerprint image data;

[0063] In the embodiment of the present application, the original fingerprint image is converted from a color or multi-channel format into a gray-scale image; the gray-scale processing can use a weighted average method or a method of extracting a single channel, and the red, green and blue channel values of the image are weighted in proportion or only the value of one channel is selected as a gray-scale value.

[0064] Step S13: applying a Gaussian filter algorithm to the gray-scale fingerprint image data to remove image noise, to obtain denoised fingerprint image data;

[0065] ​The Gaussian filtering of the embodiment of the application is a commonly used image filtering method, which removes high-frequency noise in the image by weighted average processing of each pixel point and the surrounding pixel points thereof; the Gaussian filtering algorithm is applied to the fingerprint image to smooth the image, reduce the noise in the image, and obtain the denoised fingerprint image data.

[0066] Step S14: histogram equalization processing is performed on the denoised fingerprint image data, and the contrast of the image is improved to obtain the equalized fingerprint image data.

[0067] The histogram equalization of the embodiment of the application is a commonly used image enhancement method, which enhances the contrast and details of the image by redistributing the gray levels of the image pixels; the histogram equalization algorithm is applied to the denoised fingerprint image data to make the gray level distribution of the image more uniform, increase the dynamic range of the image, and improve the contrast of the image.

[0068] Step S15: the adaptive binarization algorithm is applied to the equalized fingerprint image data to convert the image into a black-and-white binary image to obtain the binarized fingerprint image data.

[0069] The adaptive binarization of the embodiment of the application is a threshold segmentation method based on local pixel gray levels, which converts the image into a black-and-white binary image according to the gray distribution of the surrounding area of the pixel; the adaptive binarization algorithm is applied to the equalized fingerprint image data to automatically determine a suitable threshold according to the gray characteristics of the local image, and convert the image into the binarized fingerprint image data.

[0070] Step S16: the binarized fingerprint image data is subjected to thinning processing, and the fingerprint ridge line is extracted to obtain the thinned fingerprint image data.

[0071] The thinning of the embodiment of the application is an image processing method, which iteratively deletes sharp edges at the pixel points in the image to make the ridge lines in the image more slender; the thinning algorithm is applied to the binarized fingerprint image data to extract the ridge line information in the fingerprint image to obtain the thinned fingerprint image data.

[0072] Step S17: image enhancement processing is performed on the thinned fingerprint image data to obtain high-quality fingerprint image data.

[0073] The image enhancement of the embodiment of the application is a processing method, which aims to improve the visual quality of the image, enhance the details, contrast and clarity of the image; the image enhancement algorithm is applied to the thinned fingerprint image data to further improve the quality of the image, so that the fingerprint image is clearer and easier to analyze and identify.

[0074] The fingerprint of the present application is a unique biological feature of each person. By collecting the fingerprint image data of the user, accurate identity authentication can be performed. In this way, only the authorized user can unlock and use the earphone when wearing the earphone, thereby improving the security and privacy of the device. Grayscale conversion can convert a color image into a single-channel grayscale image, simplifying the complexity and computational load of subsequent processing steps. The Gaussian filter algorithm can effectively remove noise in the image, improve the clarity and quality of the fingerprint image, and reduce errors and interference in subsequent processing steps. Histogram equalization can adjust the brightness distribution of the image and enhance the contrast of the image, making the details in the fingerprint image more clear and visible. The binary image can highlight the outline and details of the fingerprint pattern, facilitating subsequent fingerprint feature extraction and matching. The thinning process can further reduce the ridge line width in the fingerprint image, enhance the details of the fingerprint pattern, and improve the robustness and reliability of the fingerprint features. Image enhancement processing can further improve the quality and clarity of the fingerprint image, reduce noise and interference in the image, and thus improve the accuracy and reliability of fingerprint recognition. The above steps can optimize and enhance the original fingerprint image data, improve the performance and effect of the fingerprint recognition system. The effects of these steps include user identity authentication, image quality improvement, noise removal, contrast enhancement, feature extraction and enhancement, and improvement of fingerprint recognition accuracy.

[0075] Preferably, step S17 comprises the following steps:

[0076] Step S171: performing morphological processing on the thinned fingerprint image data to obtain repaired fingerprint image data, wherein the morphological processing is an opening operation composed of erosion followed by dilation;

[0077] The morphological processing of the embodiment of the present application is an image processing method based on image shape, and the opening operation is a common operation of morphological processing, which is composed of erosion followed by dilation. The opening operation is applied to the thinned fingerprint image data, first erosion operation is performed, and then dilation operation is performed. In this way, small isolated points and noise in the image can be removed, while the main fingerprint line structure is maintained, and the repaired fingerprint image data is obtained.

[0078] Step S172: performing clarity enhancement processing on the repaired fingerprint image data to obtain enhanced fingerprint image data;

[0079] The clarity enhancement of the fingerprint texture of the embodiment of the present application is to increase the details and clarity of the fingerprint image through image enhancement technology to improve the accuracy of fingerprint recognition. Image enhancement algorithms such as sharpening, detail enhancement or frequency domain filtering are applied to the repaired fingerprint image data to enhance the texture details and clarity of the fingerprint image, and the enhanced fingerprint image data is obtained.

[0080] Step S173: performing direction field estimation based on local direction of fingerprint texture on the segmented fingerprint image data to obtain fingerprint direction field data;

[0081] The fingerprint direction field of the embodiment of the present application refers to the estimation result of the local fingerprint line direction of each pixel point in the fingerprint image. The fingerprint direction estimation algorithm is applied to the enhanced fingerprint image data. The local fingerprint direction of each pixel point is estimated by analyzing the texture features in the image, and the fingerprint direction field data is constructed.

[0082] Step S174: performing anisotropic filtering on the segmented fingerprint image data according to the direction field data to obtain high-quality fingerprint image data.

[0083] The anisotropic filtering of the embodiment of the present application is an image enhancement technique. The image can be nonlinearly filtered according to the direction field data to enhance the details and texture of the fingerprint image. The anisotropic filtering algorithm is applied to the segmented fingerprint image data. The filtering process is performed according to the fingerprint direction field data to enhance the quality and recognizability of the fingerprint image.

[0084] The present application can remove small fractures or noise points in the fingerprint image through the erosion operation, so that the texture of the fingerprint image is more continuous and complete. The present application can fill the holes or broken parts in the fingerprint image through the dilation operation, repair the missing parts of the fingerprint image, and improve the readability and continuity of the fingerprint image. By enhancing the texture details of the fingerprint image, the fingerprint lines in the fingerprint image can be more clearly visible, and the accuracy and reliability of the fingerprint recognition can be improved. By analyzing the texture features in the fingerprint image, the local direction field of the fingerprint image, i.e., the direction of the fingerprint lines, can be estimated, which is very important for subsequent fingerprint feature extraction and matching. The anisotropic filtering can nonlinearly filter the fingerprint image according to the information of the fingerprint direction field, emphasize the fingerprint line details in the fingerprint image, and further enhance the quality and texture features of the fingerprint image. In summary, through the above fingerprint image processing process, the quality and texture features of the fingerprint image can be further improved, including removing small fractures, filling holes, texture enhancement, direction field estimation, and fingerprint enhancement. The effects of these steps include improving the accuracy of fingerprint recognition, enhancing the readability and continuity of the fingerprint image, and emphasizing the fingerprint line details in the fingerprint image, thereby improving the robustness and reliability of the fingerprint features.

[0085] Preferably, step S2 comprises the following steps:

[0086] Step S21: performing ridge line tracking on the high-quality fingerprint image data and performing single-pixel width-based thinning processing to obtain fingerprint skeleton data;

[0087] Step S22: identifying and extracting the bifurcation points and termination points of the fingerprint skeleton data by the intersection number algorithm to obtain initial feature point data;

[0088] Step S23: Filter false feature points from the initial feature point data and calculate the attribute information based on the feature points to obtain the feature point attribute data, where the attribute information includes type, position coordinates and orientation angle.

[0089] Step S24: Analyze the relative relationships between feature points based on the feature point attribute data, and combine the feature point attribute data and feature point relationships to form complete fingerprint feature data;

[0090] Step S25: Read the fingerprint template data of the legitimate user from the preset storage medium; match the fingerprint feature data and the fingerprint template data using the least squares point pattern matching algorithm to obtain feature point matching data;

[0091] Step S26: Calculate the matching score of the feature point matching data based on the feature point matching data, and compare it with the preset threshold to obtain the fingerprint matching result data;

[0092] Step S27: If the fingerprint matching result data shows a successful match, then generate unique user identification data based on the fingerprint feature data; at the same time, merge and update the fingerprint feature data matched this time with the stored fingerprint template data to obtain the updated fingerprint template data, and proceed to step S3;

[0093] Step S28: If the fingerprint matching result data shows a matching failure, record the number of failures, update the authentication attempt counter data, and proceed to step S5.

[0094] As an embodiment of the present invention, reference Figure 3 As shown, Figure 1 A detailed flowchart of step S2 is shown below. In this embodiment of the invention, step S2 includes the following steps:

[0095] Step S21: Perform ridge tracking on the high-quality fingerprint image data and perform thinning processing based on single pixel width to obtain fingerprint skeleton data;

[0096] This invention uses image processing algorithms, such as orientation-based filters or Gabor filters, to detect and extract ridge (fingerprint ridge) information in fingerprint images; the extracted ridges are thinned to a single pixel width to preserve the main structure of the ridges and remove redundant details to obtain fingerprint skeleton data.

[0097] Step S22: The cross-tabulation algorithm is used to identify and extract the bifurcation points and termination points of the fingerprint skeleton data to obtain the initial feature point data;

[0098] The embodiment of the present application identifies and extracts the bifurcation points (bifurcation points of the fingerprint ridge line) and the termination points (termination points of the fingerprint ridge line) by analyzing the connection relationship of the pixel points in the fingerprint skeleton data; and finds the bifurcation points and the termination points in the fingerprint skeleton data by applying the cross number algorithm, so as to obtain the initial feature point data.

[0099] Step S23: performing false feature point filtering on the initial feature point data, and performing attribute information calculation based on the feature points, so as to obtain feature point attribute data, wherein the attribute information includes type, position coordinate and direction angle;

[0100] The embodiment of the present application excludes possible false feature points and retains real fingerprint feature points by judging the local features of the image around the feature points; and calculates the attribute information of the feature points, such as type (bifurcation point or termination point), position coordinate and direction angle, according to the position of the feature points and the pixel values around the feature points.

[0101] Step S24: performing relative relationship analysis between the feature points according to the feature point attribute data, and combining the feature point attribute data and the relationship between the feature points to form complete fingerprint feature data;

[0102] The embodiment of the present application analyzes the relative positions and connection relationships between the feature points to determine the topological structure therebetween, and combines the feature point attribute data and the relationship between the feature points to form complete fingerprint feature data according to the attribute data and the relative relationship of the feature points.

[0103] Step S25: reading the fingerprint template data of a legal user from a preset storage medium; and performing matching between the fingerprint feature data and the fingerprint template data by using a point pattern matching algorithm of the least square method, so as to obtain feature point matching data;

[0104] The embodiment of the present application reads the fingerprint template data of a legal user from a preset storage medium (such as a database, a file or the like), uses the least square method or other appropriate point pattern matching algorithm to perform matching between the fingerprint feature data and the fingerprint template data, so as to determine the similarity or matching degree. According to the matching result, the feature point matching data, i.e. the corresponding relationship between the feature points in the fingerprint feature data and the feature points in the fingerprint template data, is obtained.

[0105] Step S26: calculating a matching score of the feature point matching data according to the feature point matching data, and comparing the matching score with a preset threshold, so as to obtain fingerprint matching result data;

[0106] The embodiment of the present application calculates a matching score or a similarity value according to the feature point matching data, which is used to represent the matching degree of the fingerprint feature data and the fingerprint template data; and compares the calculated matching score with a preset threshold, so as to determine the fingerprint matching result data, i.e. matching success or matching failure.

[0107] Step S27: If the fingerprint matching result data shows a successful match, generate unique user identity data based on the fingerprint feature data; at the same time, fuse and update the fingerprint feature data of this match with the stored fingerprint template data to obtain fingerprint template update data, and proceed to step S3;

[0108] The embodiment of the application generates unique user identity data based on the fingerprint feature data of a successful match, which is used to identify and identify the user identity, and fuses and updates the fingerprint feature data of this match with the stored fingerprint template data to maintain the accuracy and integrity of the fingerprint template. According to the system flow, continue to perform subsequent operations, which can include data storage, log recording, etc.

[0109] Step S28: If the fingerprint matching result data shows a failed match, record the number of failures, update the authentication attempt counter data, and proceed to step S5.

[0110] In the fingerprint matching process, if the match fails, i.e. the fingerprint image cannot be successfully matched with the stored fingerprint template, the number of failures is recorded and the authentication attempt counter data is updated. This counter can be a variable or a field in a database, which is used to track the number of authentication attempts. When the match fails, the number of failures is incremented by one, and the result is saved in the authentication attempt counter; the counter is associated with the user's identity information for use in subsequent authentication attempts. After the above steps are completed, step S5 is continued, i.e. a new round of fingerprint collection and matching process is continued, so that the user can reattempt authentication.

[0111] The present application can extract the main ridge structure of the fingerprint by ridge tracking of the fingerprint image, and obtain the ridge line information in the fingerprint image; by thinning processing, the ridge line in the fingerprint image can be thinned to single-pixel width, unnecessary additional pixels are removed, and the skeleton structure of the fingerprint is obtained. By the intersection number algorithm, the bifurcation points (fingerprint ridge intersection) and termination points (end points of the fingerprint ridge) in the fingerprint skeleton can be identified, so that the initial feature points are extracted. By analyzing and screening the initial feature point data, false feature points introduced by noise or other factors can be excluded, and the accuracy and reliability of the feature points are improved; based on the filtered feature points, the attribute information is calculated, including the type (bifurcation point or termination point), position coordinates and direction angle of the feature points, etc., to provide necessary attribute information for subsequent fingerprint feature extraction and matching. By analyzing the relative position and direction relationship between the feature points, the connection relationship between the feature points can be established, and the structure information of the fingerprint features is formed; the attribute data of the feature points and the relationship between the feature points are combined together to form complete fingerprint feature data, which contains the structure and morphological information of the fingerprint image. The fingerprint template data of the legal user is read from the storage medium, which is used for matching comparison with the current fingerprint feature data; by the point pattern matching algorithm of the least square method, the current fingerprint feature data is matched with the fingerprint template data, and the matching relationship between the feature points is obtained. According to the feature point matching data, the similarity score of the fingerprint matching is calculated, which reflects the similarity between the fingerprint features; the matching score calculated is compared with the preset threshold value, and it is judged whether the fingerprint matching is successful or failed. According to the matching successful fingerprint feature data, the unique user identity data is generated, which is used for identifying and verifying the user identity; the fingerprint feature data matched successfully this time is fused and updated with the stored fingerprint template data, so as to facilitate the subsequent fingerprint matching and identification. The number of times of fingerprint matching failure is recorded, which is used for counting the number of authentication failures and subsequent processing, updating the authentication attempt counter data, which is used for limiting the number of authentication attempts or taking other security measures. In summary, through the execution of the above steps, the feature extraction, matching and authentication process of the fingerprint image can be completed, and the feature point data, matching result data and user identity data are provided, which are used for the purpose of fingerprint identification and authentication.

[0112] Preferably, step S23 comprises the following steps:

[0113] Step S231: neighborhood analysis is performed on the initial feature point data, and the ridge density around each feature point is calculated to obtain feature point density data;

[0114] The embodiment of the present application loads the fingerprint image data using an image processing library (for example, OpenCV) and performs neighborhood analysis on each feature point. A fixed-size neighborhood window, for example, a square window centered on the feature point, can be selected, and ridge density is calculated within the neighborhood window. The density can be estimated by counting the number of ridge pixels within the neighborhood. Ridge pixels generally refer to the obvious ridge structure in the fingerprint image, and the calculated feature point density data is saved for subsequent steps.

[0115] Step S232: Based on the feature point density data, a threshold is set to filter out feature points with abnormal density, and normal feature point data after density filtering is obtained.

[0116] The embodiment of the present application sets an appropriate density threshold to distinguish between normal feature points and feature points with abnormal density. The threshold can be adjusted according to the specific application scenario and the characteristics of the fingerprint image. The threshold filtering is performed on the feature point density data, the feature points with density values below the threshold are identified as abnormal density, and are excluded, the feature points that pass the density filtering are retained, and normal feature point data after density filtering is obtained.

[0117] Step S233: Perform broken ridge detection on the normal feature point data, and identify false bifurcation points caused by image quality, and obtain broken corrected feature point data.

[0118] The embodiment of the present application performs broken ridge detection on the normal feature point data. Broken refers to the interruption or discontinuity of the fingerprint ridge near the feature point. A broken detection algorithm, for example, a method based on ridge connectivity, is used to detect feature points in a broken state and mark them as broken points. False bifurcation points caused by image quality are identified. In some cases, image quality problems (such as noise or blur) can cause feature points to be incorrectly identified as bifurcation points. A false bifurcation point identification algorithm, for example, a method based on local direction consistency, is used to identify and exclude false bifurcation points. The feature point data after broken correction is obtained, in which the broken points are corrected or excluded.

[0119] Step S234: Perform bridge processing on the broken corrected feature point data, and perform direction consistency check, and remove feature points inconsistent with the direction of the surrounding ridge, and obtain direction consistent feature point data.

[0120] The embodiment of the present application performs bridge processing on the broken feature point data. Bridge refers to connecting the ridge line of the broken point to restore the continuity of the ridge line. The broken point is processed by using a bridge algorithm, such as a method based on local direction consistency and distance measurement, and direction consistency check is performed. The direction of the feature point should be consistent with the direction of the surrounding ridge line to increase the recognition accuracy and stability. A direction consistency check algorithm, such as a method based on neighborhood direction consistency, is used to remove the feature points that are inconsistent with the direction of the surrounding ridge line, and the feature point data processed by the direction consistency is obtained, which contains feature points that are specifically bridged and direction consistent.

[0121] Step S235: Type of each feature point in the direction consistent feature point data is distinguished to obtain feature point type data, wherein the feature point type data includes bifurcation point data and termination point data.

[0122] The embodiment of the present application distinguishes the type of each feature point in the direction consistent feature point data. The type of the feature point can be divided into bifurcation point and termination point. A feature point type distinguishing algorithm, such as a method based on local direction field, is used to judge the type of the feature point according to the ridge line structure and direction information around the feature point, and the feature point is classified as bifurcation point or termination point, and they are saved as feature point type data.

[0123] Step S236: The coordinates of each feature point in the direction consistent feature point data are accurately positioned by a sub-pixel accuracy algorithm to obtain feature point coordinate data.

[0124] The embodiment of the present application accurately positions each feature point in the direction consistent feature point data by a sub-pixel accuracy algorithm. This is to improve the accuracy of the position of the feature point. A sub-pixel accuracy positioning algorithm, such as a method based on interpolation and gradient optimization, is used to accurately position the coordinates of the feature point, and the accurately positioned coordinates of the feature point are saved as feature point coordinate data.

[0125] Step S237: Local direction field analysis is performed on each feature point in the direction consistent feature point data to calculate the direction angle thereof to obtain feature point direction angle data.

[0126] The embodiment of the present application performs local direction field analysis on each feature point in the direction consistent feature point data. The local direction field refers to the direction information of the ridge line around the feature point. A local direction field analysis algorithm, such as a method based on gradient calculation and smoothing filtering, is used to calculate the direction angle of the feature point, and the direction angle of the feature point is saved as feature point direction angle data.

[0127] Step S238: The feature point type data, the feature point coordinate data and the feature point direction angle data are integrated to form complete feature point attribute data.

[0128] The embodiment of the present application integrates feature point type data, feature point coordinate data and feature point direction angle data to form complete feature point attribute data; data structures such as arrays, lists or dictionaries can be used to store attribute information of each feature point, ensuring the correspondence between attributes, and the feature point attribute data can be used in subsequent fingerprint recognition and analysis processes.

[0129] The neighborhood analysis of the present application can determine the area range around each feature point, and then calculate the number of surrounding ridge lines of the feature point; the feature point density data provides local structure information about the fingerprint area, which can be used in subsequent feature point filtering and correction steps. Density filtering can exclude feature points with abnormal density, such as isolated feature points or feature points with very high density; the filtered normal feature point data has higher reliability and accuracy, which helps to improve the effect of subsequent steps. Broken ridge line detection can identify the broken ridge line caused by poor image quality or image damage, etc.; false bifurcation point identification can exclude incorrect bifurcation points caused by image quality problems; broken correction feature point data provides more accurate and complete feature point information, which helps to improve the accuracy of subsequent feature extraction and matching. Bridge processing can fill the broken ridge line, making the connection of feature points more continuous and complete; direction consistency check can ensure that the direction of the feature point is consistent with the direction of the surrounding ridge line, and exclude feature points with inconsistent directions, improving the accuracy of feature matching. Feature point type differentiation divides feature points into two categories: bifurcation points and termination points, providing more detailed feature point information, which helps to accurately extract and match fingerprint features. Sub-pixel accurate positioning can improve the accuracy of feature point coordinates, making the position of feature points more accurate and reliable; feature point coordinate data provides accurate position information of feature points in the fingerprint image, which is used in subsequent feature extraction and matching processes. Local direction field analysis can determine the fingerprint ridge line direction around each feature point, which helps subsequent feature matching and recognition; feature point direction angle data provides direction information of feature points, which can be used in direction consistency check and matching process in fingerprint matching and recognition algorithm. Integrating feature point type, coordinate and direction data can establish a complete feature point attribute data set; complete feature point attribute data can be used in various fingerprint applications, such as fingerprint recognition, feature matching and template generation, etc. In summary, the effects of the above steps include improving the accuracy and reliability of feature points, excluding abnormal and incorrect feature points, correcting broken and bifurcation situations, extracting more complete feature information, and improving the accuracy and robustness of fingerprint matching and recognition. These steps complement and optimize each other, forming a complete fingerprint feature extraction and processing flow.

[0130] Preferably, step S3 comprises the following steps:

[0131] Step S31: Hashing the fingerprint feature data and combining it with the random seed data to generate initial key data through a key derivation function;

[0132] The embodiment of the application loads fingerprint feature data, which can be extracted features of a fingerprint image or other forms of data representing a fingerprint; the fingerprint feature data is hashed, and a common hashing algorithm such as SHA-256 can be used; this converts the fingerprint feature data into a fixed-length hash value; random seed data is obtained, and a secure random number generator can be used to generate random seed data; the hash value is combined with the random seed data, and a key derivation function such as HMAC-SHA256 can be used to generate initial key data by taking the hash value and the seed data as input.

[0133] Step S32: Key expansion of the initial key data and binding with the user identity data to generate binding key data;

[0134] The embodiment of the application uses a key expansion algorithm such as PBKDF2 or HKDF to expand the initial key data. The key expansion algorithm can enhance the security of the key through multiple rounds of iteration and a salt value; the user identity data is obtained, which can be information that uniquely identifies the user, such as a username or user ID; the expanded key data is bound to the user identity data, and a hash function such as HMAC-SHA256 can be used to generate binding key data by taking the key data and the user identity data as input.

[0135] Step S33: Segmenting the binding key data and using a multi-round encryption strategy to block-encrypt the audio transmitted by the earphone to obtain preliminary encrypted audio data;

[0136] The embodiment of the application divides the binding key data into fixed-size key blocks, and a symmetric encryption algorithm such as AES can be used to block-encrypt the key blocks and the audio data; the audio data is divided into data blocks of appropriate size for block-encryption, and an appropriate block size can be selected to balance encryption efficiency and transmission delay; a multi-round encryption strategy such as the CBC mode or CTR mode of AES is used to encrypt each audio data block, and the key used in the encryption process comes from the key blocks generated earlier, resulting in preliminary encrypted audio data, i.e., a sequence of encrypted audio data blocks.

[0137] Step S34: Integrity protection of the preliminary encrypted audio data and calculation of a message authentication code to obtain audio authentication data;

[0138] The embodiment of the present application performs integrity protection processing on the preliminary encrypted audio data, and can use a message authentication code (MAC) algorithm, such as HMAC-SHA256. The encrypted audio data is taken as input to generate an authentication code related to the data; the message authentication code of the audio data is calculated, and the HMAC-SHA256 algorithm or the like can be used, the encrypted data is taken as input, and the authentication code is generated using the binding key data.

[0139] Step S35: The preliminary encrypted audio data is combined with the audio authentication data to obtain encrypted audio data; the encrypted audio data is packetized and encoded to obtain a transmission-ready encrypted data packet;

[0140] The embodiment of the present application combines the preliminary encrypted audio data and the audio authentication data into one data packet. The authentication data can be appended to the encrypted data to form a complete encrypted audio data packet; the encrypted audio data packet is packetized to divide it into an appropriate size for transmission and reception, and the packetization can be performed according to the requirements of a transmission protocol. The packetized encrypted data packet is encoded to ensure the transmission reliability of the data. An encoding scheme, such as Base64, can be used to convert binary data into a text format.

[0141] Step S36: The encrypted data packet is sent to the earphone receiving end through a wireless transmission technology; and at the earphone receiving end, the received encrypted data packet is decrypted using the generated binding key data to obtain the original audio data.

[0142] The embodiment of the present application uses a wireless transmission technology, such as Bluetooth or Wi-Fi, to send the encrypted data packet to the earphone receiving end. At the earphone receiving end, the encrypted data packet is received, and the generated binding key data is used to decrypt the encrypted data packet at the receiving end. According to the encryption algorithm and the encryption mode, the encrypted data is decrypted using a corresponding key and decryption algorithm, and the original audio data is obtained after decryption.

[0143] The hash processing can convert the fingerprint feature data into a fixed-length hash value, protect the confidentiality of the original fingerprint features, and increase the randomness and security of the key in combination with the random seed data. The key derivation function can generate a strong initial key data from the hash value and the random seed, providing a basis for subsequent key expansion. The key expansion can expand the initial key data into a longer key, increasing the complexity and security of the key. Binding the user identity data with the key can ensure the uniqueness and identity correlation of the key, improving the security. Generating the bound key data provides reliable key material for subsequent encryption and decryption operations. The segmentation processing can divide the longer key data into smaller blocks, improving the efficiency and feasibility of the encryption process. The multi-round encryption strategy increases the complexity and strength of the encryption, making it more difficult for the audio data to be cracked and decrypted. The preliminary encryption of the audio data protects the confidentiality of the audio, and only the receiving end with the correct key can decrypt and restore the original audio data. The integrity protection processing can prevent the encrypted audio data from being tampered with or damaged during transmission. The calculation of the message authentication code can verify the integrity and authenticity of the encrypted audio data, ensuring that the receiving end can correctly identify and use the legal audio data. Combining the encrypted audio data with the audio authentication data ensures the integrity and credibility of the audio data. The data packet segmentation and encoding enable the encrypted data packet to be effectively managed and processed during transmission, improving the transmission efficiency and reliability. Through wireless transmission technology, the encrypted data packet can be safely transmitted to the earphone receiving end, protecting the confidentiality of the audio data. At the earphone receiving end, the decryption processing using the generated bound key data can restore the original audio data, achieving secure data transmission and protection. In summary, the effects of the above steps include protecting the confidentiality of the fingerprint feature data, generating secure initial key data, encrypting the audio data and ensuring its integrity and authenticity, and ensuring the security and reliability of data transmission. These steps comprehensively use hash processing, key derivation function, key expansion, multi-round encryption strategy, integrity protection processing, message authentication code, and data packet segmentation and encoding techniques to ensure the confidentiality, integrity, and credibility of the audio data during transmission, while preventing data from being tampered with, leaked, or damaged.

[0144] Preferably, step S36 comprises the following steps:

[0145] Step S361: sending the encrypted data packet to the earphone receiving end through wireless transmission technology to obtain the receiving end encrypted data;

[0146] In the sending end, the encrypted data packet is sent to the earphone receiving end using appropriate wireless transmission technology such as Bluetooth or Wi-Fi. In the earphone receiving end, the same wireless transmission technology is used to receive the encrypted data packet. Related libraries or APIs can be used to handle wireless transmission.

[0147] Step S362: integrity verification is performed on the received encrypted data to obtain verification result data; when the verification result data shows that the verification is passed, the received encrypted data is decrypted using the binding key data to obtain decrypted audio data; when the verification result data shows that the verification is not passed, the received encrypted data is discarded, and an error notification is sent to the sending end to request retransmission;

[0148] In the earphone receiving end, the integrity of the received encrypted data is verified, and the message authentication code (MAC) algorithm such as HMAC-SHA256 can be used to verify the integrity of the encrypted data. The authentication code of the received encrypted data is calculated, the authentication code is generated using the binding key data, and the authentication code is compared with the received authentication code. If the verification is passed, it indicates that the data integrity verification is successful, and the decryption process can be continued; if the verification is not passed, it indicates that the data may have been tampered with, the received encrypted data is discarded, and an error notification is sent to the sending end to request retransmission.

[0149] Step S363: audio quality evaluation is performed on the decrypted audio data to obtain audio quality data.

[0150] In the earphone receiving end, the decrypted audio data is evaluated for audio quality. This can include evaluating the signal-to-noise ratio, distortion level, frequency response, and other indicators of the audio; using appropriate audio quality evaluation methods and tools, such as audio analysis libraries or algorithms, to evaluate the decrypted audio data and calculate the audio quality data.

[0151] Step S364: audio quality adjustment is performed on the decrypted audio data according to the audio quality data, thereby obtaining the original audio data.

[0152] In the earphone receiving end, the decrypted audio data is adjusted for audio quality according to the audio quality data. This can include noise reduction, equalization, volume adjustment, and other operations to improve the auditory experience of the audio; using appropriate audio processing tools and algorithms, such as audio processing libraries, to adjust and process the decrypted audio data to obtain better audio quality; after adjustment, the original audio data is obtained, which can be used for audio output of the earphone or further audio processing.

[0153] The application sends encrypted data packets to the earphone receiving end through wireless transmission technology, ensuring the secure transmission of encrypted data and avoiding eavesdropping or tampering of data during transmission. Integrity verification can ensure that the received encrypted data packets have not been tampered with or damaged during transmission, ensuring the integrity and reliability of the data. After verification, the original audio data can be restored by using the binding key data for multi-round decryption, realizing secure data decryption and protection. If the verification fails, the incorrect encrypted data packets are discarded and retransmission is requested, ensuring the accuracy and reliability of the data. Audio quality assessment can analyze and evaluate the decrypted audio data to determine the clarity, noise level, distortion level, and other audio quality indicators. Audio quality data provides performance evaluation of the audio decryption and transmission process, which helps identify any problems that may cause audio quality degradation. Audio quality adjustment based on audio quality data can improve the clarity, volume balance, dynamic range, and other aspects of the decrypted audio data, enhancing the auditory experience of the audio. Audio quality adjustment can optimize the sound characteristics and presentation of the decrypted audio data according to specific audio quality requirements and user preferences, making it more in line with user expectations. In summary, the above steps ensure the secure transmission and integrity verification of encrypted data, realize reliable data decryption and protection, evaluate audio quality and adjust audio quality as needed. These steps comprehensively use wireless transmission technology, integrity verification, multi-round decryption, audio quality assessment, and audio quality adjustment, ensuring the security and integrity of the encrypted data at the receiving end, ensuring the accuracy and quality of the decrypted audio data, and providing better audio experience and user satisfaction.

[0154] Preferably, step S4 comprises the following steps:

[0155] Step S41: Trigger the key update process according to the preset time period, and generate a key update trigger signal;

[0156] The embodiment of the application sets a preset time period in the system to trigger the key update process, and the frequency of key update can be determined according to security requirements and strategies; according to the preset time period, a key update trigger signal is generated at the beginning or end of each update period, which can be a software signal or a hardware trigger signal, used to notify the system to perform key update operation.

[0157] Step S42: After receiving the key update trigger signal, obtain the fingerprint template update data, and perform feature extraction on the fingerprint template update data to obtain updated fingerprint feature data;

[0158] After receiving the key update trigger signal, the embodiment of the application acquires the fingerprint template update data, which can be new fingerprint data acquired from the fingerprint sensor or update data downloaded from the external secure storage; feature extraction is performed on the received fingerprint template update data, which involves using a fingerprint feature extraction algorithm, such as a Minutiae or feature point extraction algorithm, to extract key features from the original fingerprint data; after feature extraction, the update fingerprint feature data, that is, a set of feature vectors or descriptors describing the fingerprint template, is obtained.

[0159] Step S43: generating a new key seed based on the update fingerprint feature data, and generating update master key data from the key seed, a current timestamp, and a random number through a key derivation function;

[0160] The embodiment of the application generates a new key seed based on the update fingerprint feature data, which can be converted into a unique key seed using a suitable algorithm or method, such as a hash function. The key derivation function is used to combine the key seed with the current timestamp and the random number to generate the update master key data. The key derivation function can be a cryptographic function, such as HKDF (HMAC-based Extract-and-Expand Key Derivation Function) or PBKDF2 (Password-Based Key Derivation Function 2).

[0161] Step S44: performing binding update according to the update master key data and the user identity data, thereby obtaining the update key data.

[0162] The embodiment of the application performs a binding update operation according to the update master key data, which can involve combining or calculating the update master key data with the user identity data to ensure the uniqueness and security of the key update; after the binding update, the update key data is obtained, that is, the newly generated key is used for encryption or decryption operations, and the update key data can be used internally in the system, and the related functions or modules using the key are updated accordingly.

[0163] The periodic triggering key update process of the present application can enhance the security of the system, and the risk of key leakage can be reduced by periodically changing the key, and the anti-attack ability of the system can be improved; the preset time period can be determined according to the security requirements and risk assessment, so as to ensure that the system updates the key within a proper time interval, and balance the relationship between security and performance. The acquisition of the fingerprint template update data can obtain the latest user fingerprint information, so as to ensure that the data used in the key update process is synchronized with the user's identity information; the feature extraction can process the fingerprint template update data, and extract the feature information related to the user's identity, so as to provide accurate input data for subsequent key generation and binding. The generation of a new key seed based on the updated fingerprint feature data can ensure the uniqueness of the key and the association with the user's identity, and enhance the security of the key; the key derivation function generates the update master key data according to the key seed, the current timestamp and the random number, so as to ensure the randomness and timeliness of the key, and improve the strength and security of the key. The binding update can bind the update master key data with the user identity data, so as to ensure that the generated update key data is only applicable to a specific user, and enhance the personalization and security of the key; the update key data is generated through the binding update process, and has the characteristics related to the user's identity, so as to ensure the correctness and applicability of the key. In summary, the effects of the above steps include periodic triggering of key update, keeping user data synchronized, generating a unique key associated with the user's identity, improving the randomness and timeliness of the key, and enhancing the personalization and security of the key. These steps comprehensively use periodic triggering, feature extraction, key generation and binding update, etc. technical means, to ensure the security and effectiveness of the key update process, and protect the data security of the system and the user.

[0164] Preferably, step S5 comprises the following steps:

[0165] Step S51: reading a pre-set maximum number of user authentication attempts threshold to obtain authentication threshold data;

[0166] The embodiment of the present application pre-sets a maximum number of user authentication attempts threshold in the system, which can be configured according to the security policy and requirements, and is used to determine the upper limit of the number of authentication attempts; in the authentication process, the pre-set maximum number of user authentication attempts threshold is read and used as authentication threshold data.

[0167] Step S52: comparing the authentication attempt counter data in the fingerprint matching result data according to the authentication threshold data, performing a lock judgment to obtain lock judgment result data;

[0168] In the fingerprint authentication process, the result of each authentication attempt is recorded in the fingerprint matching result data, including authentication attempt counter data; according to the authentication threshold data, the authentication attempt counter data in the fingerprint matching result data is compared with the authentication threshold to determine whether the upper limit of the number of authentication failure attempts is reached or exceeded; according to the comparison result, the lock judgment result data is obtained, indicating whether the device needs to be locked or the use of the authentication function is limited.

[0169] Step S53: If the lock judgment result data shows that locking is needed, activate the device locking mechanism to obtain device locking state data;

[0170] If the lock judgment result data shows that the device needs to be locked, the device locking mechanism is activated. This can be disabling the fingerprint authentication function, prohibiting user login or other limiting operations on the device; after activating the device locking mechanism, device locking state data is generated to record the locking state of the device for subsequent processing and analysis.

[0171] Step S54: Record the time and place of fingerprint recognition according to the device locking state data to obtain context data;

[0172] According to the device locking state data, the time and place information of fingerprint recognition are recorded, which can be achieved by obtaining the timestamp and geographic location information of the device; the recorded time and place information are combined into context data for subsequent behavior pattern analysis and anomaly detection.

[0173] Step S55: Obtain historical authentication pattern data and compare and analyze it with the context data to obtain behavior pattern deviation data;

[0174] The historical authentication pattern data can be the previous fingerprint authentication record or the user's authentication behavior data. The historical authentication pattern data is compared and analyzed with the current context data to compare whether the current authentication behavior deviates from the historical pattern or is abnormal. According to the comparison and analysis result, the behavior pattern deviation data is obtained for subsequent anomaly detection and risk assessment.

[0175] Step S56: Evaluate the behavior pattern deviation data through a pre-trained anomaly detection model to obtain preliminary anomaly score data;

[0176] The pre-trained anomaly detection model is used to input the behavior pattern deviation data into the model for evaluation. The anomaly detection model can be a machine learning model, such as an outlier detection algorithm based on statistical methods or an anomaly detection model based on deep learning. After model evaluation, preliminary anomaly score data is obtained, indicating whether the authentication behavior is determined to be abnormal.

[0177] Step S57: comprehensive risk assessment is performed according to the device lock state data and the preliminary abnormal score data, and abnormal behavior detection data is obtained;

[0178] According to the device lock state data and the preliminary abnormal score data, the embodiment of the application performs comprehensive risk assessment, considers the severity of the device lock state and the level of the abnormal score, and comprehensively judges whether the authentication behavior is abnormal; according to the result of the comprehensive risk assessment, abnormal behavior detection data is generated, indicating whether the authentication behavior is determined to be abnormal.

[0179] Step S58: a dynamic password verification mechanism is triggered according to the abnormal behavior detection data, so as to obtain multi-factor authentication result data.

[0180] According to the abnormal behavior detection data, the embodiment of the application triggers the dynamic password verification mechanism, and the dynamic password verification can include an SMS verification code, a one-time password or other forms of additional authentication factors; through the dynamic password verification mechanism, multi-factor authentication result data is obtained, indicating whether the user has passed the additional authentication verification.

[0181] The application can limit the number of authentication attempts by setting a maximum number of authentication attempts threshold for the user, preventing malicious attackers from making unauthorized authentication attempts through brute force cracking and other methods. Reading the authentication threshold data can obtain the upper limit of the number of authentication attempts set in the system, providing a basis for subsequent lock judgment and security decision-making. By comparing the authentication attempt counter data, it can be determined whether the number of authentication attempts by the user has reached or exceeded the authentication threshold, thereby determining whether the device needs to be locked. The lock judgment result data can indicate whether the device lock mechanism needs to be activated, improving the system's response to excessive authentication attempts and enhancing the system's security. Activating the device lock mechanism can temporarily lock devices with excessive authentication attempts, preventing malicious attackers from continuing to attempt authentication and protecting the system and user's safety. The device lock state data records the lock state of the device, providing context data for subsequent context data recording and risk assessment. According to the device lock state data, record the context information such as the time and location of fingerprint recognition, provide context data support for subsequent behavior pattern analysis and anomaly detection; Context data records environmental information during the authentication process, which helps to analyze and identify abnormal behavior and strengthen the monitoring and security control of the authentication process. Obtaining historical authentication pattern data can obtain the normal authentication behavior pattern of the user, which can be used as a reference benchmark for comparison analysis with the current authentication behavior; By comparing and analyzing the context data, the differences between the current authentication behavior and the historical behavior pattern can be detected, and potential abnormal behavior can be found. Based on the pre-trained anomaly detection model, the behavior pattern deviation data is evaluated, which can automatically identify potential abnormal behavior and provide preliminary anomaly score data; Preliminary anomaly score data can help the system judge the abnormality of authentication behavior, providing a basis for subsequent comprehensive risk assessment and security decision-making. Considering the device lock state and preliminary anomaly score data, comprehensive risk assessment can more accurately determine the nature and risk level of the authentication behavior; Abnormal behavior detection data provides a judgment of the nature of the authentication behavior, providing an important reference for subsequent security control and decision-making. Dynamic password verification mechanism can increase the security of multi-factor authentication, trigger dynamic password verification by combining abnormal behavior detection data, provide an additional authentication level, and protect the system and user's safety; Multi-factor authentication result data is the authentication result based on dynamic password verification, providing a more reliable authentication decision result, enhancing the security of the system and the ability to resist authentication attacks. In summary, the effects of the above steps include limiting the number of authentication attempts, lock judgment and device locking, context data recording, behavior pattern analysis and anomaly detection, comprehensive risk assessment, multi-factor authentication, etc. These steps comprehensively use authentication threshold settings, behavior pattern comparison, anomaly detection, and multi-factor authentication techniques to effectively improve the security of the authentication system and protect the user's identity and sensitive data from unauthorized access.

[0182] Preferably, step S58 comprises the following steps:

[0183] Step S581: determining a risk level according to the abnormal behavior detection data, and performing a multi-factor authentication strategy selection to obtain authentication strategy selection data;

[0184] According to the evaluation result of the abnormal behavior detection data, the risk level of the authentication behavior is determined in the embodiment of the application. The risk level can be determined according to factors such as the level of the abnormal score and the severity of the device lock state. According to the determined risk level, an appropriate multi-factor authentication strategy is selected. The multi-factor authentication strategy can include a combination of various authentication factors such as SMS verification code, one-time password, hardware token, and biometric recognition. Authentication strategy selection data is generated to represent the selected multi-factor authentication strategy.

[0185] Step S582: generating a dynamic password according to the authentication strategy selection data, and sending it to the user through a preset secure channel to obtain dynamic password sending state data;

[0186] According to the authentication strategy selection data, a dynamic password is generated in the embodiment of the application. The dynamic password can be a one-time password (OTP) or other dynamically generated password based on time, events or algorithms. The generated dynamic password is sent to the user through a preset secure channel. The secure channel can be a protected communication channel such as SMS, email, mobile application, etc. After sending the dynamic password, dynamic password sending state data is obtained, which represents the result of sending the dynamic password, including whether the sending is successful or not and possible error information.

[0187] Step S583: performing user password verification according to the dynamic password sending state data, thereby obtaining multi-factor authentication result data.

[0188] According to the dynamic password sending state data, user password verification is performed in the embodiment of the application. This can be the user inputting the received dynamic password in the authentication interface. It is verified whether the password input by the user matches the previously generated dynamic password. If the match is successful, the authentication is passed, indicating that the multi-factor authentication result is passed; otherwise, the authentication fails, indicating that the multi-factor authentication result is not passed.

[0189] The application can evaluate the abnormality and risk degree of the authentication behavior according to the risk level determined by the abnormal behavior detection data, and provide a basis for subsequent authentication strategy selection; the multi-factor authentication strategy selection selects a suitable authentication strategy according to the different risk levels, for example, uses dynamic password, SMS verification code, fingerprint and other authentication methods, and improves the security and reliability of authentication; the dynamic password is generated according to the authentication strategy selection data, which can increase the security in the authentication process; the dynamic password is sent through the preset secure channel, which can ensure that the transmission process of the password is not tampered with or stolen, and ensure the confidentiality and integrity of the dynamic password; the dynamic password sending state data records the sending result of the dynamic password, which is used for subsequent verification and judgment; the user password verification is performed according to the dynamic password sending state data, which can verify whether the password input by the user is consistent with the sent dynamic password, and ensure the accuracy of the user's identity and authentication information; the multi-factor authentication result data is the authentication result based on the user password verification and other authentication factors, which provides a more reliable authentication decision result, and enhances the security of the system and the ability to resist authentication attacks. In summary, the effects of the above steps include determining the risk level according to the abnormal behavior detection data, multi-factor authentication strategy selection, generating and sending dynamic password, and user password verification. These steps combine risk assessment, authentication strategy selection and multi-factor authentication, and provide stronger security and reliability to ensure the effectiveness of the authentication process and the protection of the user's identity. The use of dynamic password and the combination of user password verification increase the complexity and security of authentication, and reduce the risk of authentication attack.

[0190] Therefore, embodiments should be considered in all respects as illustrative and non-restrictive, the scope of the application being defined by the appended claims rather than the above description, and all changes falling within the meaning and range of equivalence of the essential features of the application are therefore intended to be embraced therein.

[0191] The above description is merely one specific implementation of the application, which enables those skilled in the art to understand or implement the application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the application. Therefore, the application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A fingerprint recognition-based secure encryption method for headphones, characterized in that, Includes the following steps: Step S1: Integrate a miniature fingerprint sensor on the earphone shell. When the user wears the earphone, the sensor collects the user's fingerprint image to obtain the original fingerprint image data. The original fingerprint image data is then denoised and enhanced to obtain high-quality fingerprint image data. Step S2: Extract fingerprint features from the fingerprint image data, identify bifurcation points and endpoints, and obtain fingerprint feature data; The fingerprint feature data is matched with the preset legitimate user fingerprint template to obtain fingerprint matching result data. If the fingerprint matching result data shows a successful match, a unique user identity identifier is generated and step S3 is performed; otherwise, step S5 is performed. Step S3: Generate unique key data based on fingerprint feature data, and bind the user identification data with the key data to generate bound key data; use the generated bound key data to encrypt the audio transmitted through the earphone to obtain encrypted audio data; Encrypted audio data is sent to the headphone receiver via wireless transmission technology; and the headphone receiver uses the generated binding key data to decrypt the received encrypted audio data to obtain the original audio data. Step S4: Update the key data according to the preset period and update the bound user identity data to obtain the updated key data, wherein the updated key data is generated based on the latest fingerprint features; Step S5: Lock the fingerprint matching result data according to the preset maximum number of user authentication attempts, and record the time and location of fingerprint recognition to obtain context data; Anomaly detection is performed on the context data to obtain anomaly detection data; a dynamic password verification mechanism is triggered based on the anomaly detection data to obtain multi-factor authentication result data.

2. The fingerprint recognition-based headphone security encryption method according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Integrate a miniature fingerprint sensor on the earphone shell. When the user wears the earphone, the sensor collects the user's fingerprint image to obtain the original fingerprint image data. Step S12: Perform grayscale processing on the original fingerprint image data to obtain grayscale fingerprint image data; Step S13: Apply a Gaussian filtering algorithm to the grayscale fingerprint image data to remove image noise and obtain denoised fingerprint image data; Step S14: Perform histogram equalization on the denoised fingerprint image data and improve the image contrast to obtain equalized fingerprint image data; Step S15: Apply an adaptive binarization algorithm to the equalized fingerprint image data to convert the image into a black and white binary image, thereby obtaining the binarized fingerprint image data; Step S16: Thin the binarized fingerprint image data and extract the fingerprint ridges to obtain thinned fingerprint image data; Step S17: Perform image enhancement processing on the refined fingerprint image data to obtain high-quality fingerprint image data.

3. The fingerprint recognition-based headphone security encryption method according to claim 2, characterized in that, Step S17 includes the following steps: Step S171: Perform morphological processing on the refined fingerprint image data to obtain repaired fingerprint image data, wherein the morphological processing is an opening operation of erosion followed by dilation. Step S172: Perform fingerprint texture clarity enhancement processing on the repaired fingerprint image data to obtain enhanced fingerprint image data; Step S173: Perform orientation field estimation based on local orientation of fingerprint texture on the segmented fingerprint image data to obtain fingerprint orientation field data; Step S174: Perform anisotropic filtering on the segmented fingerprint image data based on the orientation field data to obtain high-quality fingerprint image data.

4. The fingerprint recognition-based headphone security encryption method according to claim 3, characterized in that, Step S2 includes the following steps: Step S21: Perform ridge tracking on the high-quality fingerprint image data and perform thinning processing based on single pixel width to obtain fingerprint skeleton data; Step S22: The cross-tabulation algorithm is used to identify and extract the bifurcation points and termination points of the fingerprint skeleton data to obtain the initial feature point data; Step S23: Filter false feature points from the initial feature point data and calculate the attribute information based on the feature points to obtain the feature point attribute data, where the attribute information includes type, position coordinates and orientation angle. Step S24: Analyze the relative relationships between feature points based on the feature point attribute data, and combine the feature point attribute data and feature point relationships to form complete fingerprint feature data; Step S25: Read the fingerprint template data of the legitimate user from the preset storage medium; match the fingerprint feature data and the fingerprint template data using the least squares point pattern matching algorithm to obtain feature point matching data; Step S26: Calculate the matching score of the feature point matching data based on the feature point matching data, and compare it with the preset threshold to obtain the fingerprint matching result data; Step S27: If the fingerprint matching result data shows a successful match, then generate unique user identification data based on the fingerprint feature data; at the same time, merge and update the fingerprint feature data matched this time with the stored fingerprint template data to obtain the updated fingerprint template data, and proceed to step S3; Step S28: If the fingerprint matching result data shows a matching failure, record the number of failures, update the authentication attempt counter data, and proceed to step S5.

5. The fingerprint recognition-based headphone security encryption method according to claim 4, characterized in that, Step S23 includes the following steps: Step S231: Perform neighborhood analysis on the initial feature point data and calculate the ridge density around each feature point to obtain feature point density data; Step S232: Based on the feature point density data, set a threshold to filter out feature points with abnormal density, and obtain normal feature point data after density filtering; Step S233: Detect broken ridges on normal feature point data and identify false bifurcation points caused by image quality to obtain broken corrected feature point data; Step S234: Bridge the fracture correction feature point data and perform orientation consistency check, remove feature points that are inconsistent with the orientation of the surrounding ridge lines, and obtain orientation consistent feature point data. Step S235: Distinguish the type of each feature point in the feature point data with consistent direction to obtain feature point type data, which includes bifurcation point data and termination point data; Step S236: Accurately locate the coordinates of each feature point in the oriented feature point data using a sub-pixel precision algorithm to obtain the feature point coordinate data; Step S237: Perform local orientation field analysis on each feature point in the consistent orientation feature point data, calculate its orientation angle, and obtain the feature point orientation angle data; Step S238: Integrate the feature point type data, feature point coordinate data, and feature point direction angle data to form complete feature point attribute data.

6. The fingerprint recognition-based headphone security encryption method according to claim 5, characterized in that, Step S3 includes the following steps: Step S31: Hash the fingerprint feature data and combine it with the random seed data to generate the initial key data through the key derivation function; Step S32: Expand the initial key data and bind it with the user identity data to generate bound key data; Step S33: Segment the binding key data and use a multi-round encryption strategy to encrypt the audio transmitted by the headphones in blocks to obtain preliminary encrypted audio data; Step S34: Perform integrity protection processing on the initially encrypted audio data and calculate the message authentication code to obtain the audio authentication data; Step S35: Combine the initial encrypted audio data with the audio authentication data to obtain encrypted audio data; perform packet segmentation and encoding on the encrypted audio data to obtain encrypted data packets ready for transmission; Step S36: Send the encrypted data packet to the headphone receiver via wireless transmission technology; and at the headphone receiver, decrypt the received encrypted data packet using the generated binding key data to obtain the original audio data.

7. The fingerprint recognition-based headphone security encryption method according to claim 6, characterized in that, Step S36 includes the following steps: Step S361: Send the encrypted data packet to the earphone receiver via wireless transmission technology to obtain the encrypted data from the receiver; Step S362: Perform integrity verification on the encrypted data received at the receiving end to obtain verification result data; if the verification result data shows that the verification passed, use the binding key data to perform multiple rounds of decryption on the encrypted data received at the receiving end to obtain decrypted audio data; if the verification result data shows that the verification failed, discard the received encrypted data received at the receiving end and send an error notification to the sending end to request retransmission; Step S363: Evaluate the audio quality of the decrypted audio data to obtain audio quality data; Step S364: Adjust the audio quality of the decrypted audio data based on the audio quality data to obtain the original audio data.

8. The fingerprint recognition-based headphone security encryption method according to claim 7, characterized in that, Step S4 includes the following steps: Step S41: Trigger the key update process according to the preset time period and generate a key update trigger signal; Step S42: After receiving the key update trigger signal, obtain the fingerprint template update data, and extract features from the fingerprint template update data to obtain the updated fingerprint feature data; Step S43: Generate a new key seed based on the updated fingerprint feature data, and generate updated master key data based on the key seed, the current timestamp, and a random number using the key derivation function; Step S44: Bind and update based on the updated master key data and user identity data to obtain the updated key data.

9. The fingerprint recognition-based headphone security encryption method according to claim 8, characterized in that, Step S5 includes the following steps: Step S51: Read the pre-set maximum number of user authentication attempts threshold to obtain authentication threshold data; Step S52: Compare the authentication attempt counter data in the fingerprint matching result data with the authentication threshold data, perform a lock judgment, and obtain the lock judgment result data; Step S53: If the lock determination result data shows that locking is required, then activate the device locking mechanism and obtain the device lock status data; Step S54: Record the time and location of fingerprint recognition based on the device lock status data to obtain context data; Step S55: Obtain historical authentication pattern data and compare it with context data to obtain behavioral pattern deviation data; Step S56: Evaluate the behavioral pattern deviation data using a pre-trained anomaly detection model to obtain preliminary anomaly score data; Step S57: Conduct a comprehensive risk assessment based on the device lock status data and preliminary anomaly score data to obtain abnormal behavior detection data; Step S58: Trigger the dynamic password verification mechanism based on the abnormal behavior detection data to obtain multi-factor authentication result data.

10. The fingerprint recognition-based headphone security encryption method according to claim 9, characterized in that, Step S58 includes the following steps: Step S581: Determine the risk level based on the abnormal behavior detection data, and select a multi-factor authentication strategy to obtain authentication strategy selection data; Step S582: Select data to generate a dynamic password according to the authentication policy, and send it to the user through a preset secure channel to obtain dynamic password sending status data; Step S583: Verify the user password by sending status data based on the dynamic password, thereby obtaining multi-factor authentication result data.

Images