Mobile storage hardware protection method based on high-order biological fingerprint identification technology

Abstract

The invention provides a mobile storage hardware protection method based on a high-order biological fingerprint identification technology, and relates to the technical field of safety protection, and the method comprises the steps: collecting the data of the blood vessel texture and skin micro-surface morphology of a user, and carrying out the storage, preprocessing and data integration operation of the collected data, obtaining a high-order biological fingerprint data set; the method comprises the following steps: performing feature extraction operation on a high-order biological fingerprint data set to obtain high-order data features, and training a preset initial model to obtain a trained high-order biological fingerprint model; the high-order data features comprise a first high-order data feature and a second high-order data feature; and performing real-time authentication operation on the biological fingerprint of the user, and controlling the access state of the user according to a real-time authentication result. According to the method, the accuracy and the safety of the biological fingerprint identification system are improved.

Description

Technical Field

[0001] The present invention relates to the field of security protection technology, and in particular to a mobile storage hardware protection method based on high-order biometric fingerprint recognition technology. Background Art

[0002] Mobile storage hardware plays a vital role in the modern information society. However, with the development of technology, traditional protection methods are gradually becoming inadequate. Traditional mobile storage hardware protection methods mainly include password locks and encryption algorithms. However, these methods have a series of problems. First, passwords are easy to crack, especially when users set weak passwords; second, although encryption algorithms provide a certain degree of security, with the improvement of computer computing power, encryption cracking becomes easier. Among them, the biggest problem in traditional methods is that the authenticity verification of user identity is not strict enough and is vulnerable to malicious attacks; passwords are easy to guess, and encryption algorithms cannot solve the problem of identity forgery. In addition, although biometric technology has improved security to a certain extent, traditional fingerprint recognition systems are still likely to be counterfeited. Therefore, it is very necessary to design a mobile storage hardware protection method based on high-order biometric fingerprint recognition technology. Summary of the invention

[0003] The purpose of the present invention is to provide a mobile storage hardware protection method based on high-order biometric fingerprint recognition technology to improve the accuracy and security of the biometric fingerprint recognition system.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A mobile storage hardware protection method based on advanced biometric fingerprint recognition technology includes the following steps:

[0006] By collecting data on the user's blood vessel texture and skin micro-surface morphology, and storing, preprocessing and integrating the collected data, a high-level biometric fingerprint data set is obtained;

[0007] By performing a feature extraction operation on a high-order biometric fingerprint data set, a high-order data feature is obtained, and a preset initial model is trained to obtain a trained high-order biometric fingerprint model; the high-order data feature includes a first high-order data feature and a second high-order data feature;

[0008] Perform real-time authentication on the user's biometric fingerprint and control the user's access status based on the real-time authentication result.

[0009] Optionally, the blood vessel texture data includes:

[0010] Image data under multiple angles and lighting conditions, vascular position, vascular direction, image resolution, user ID, lighting conditions, acquisition time, image processing steps, mathematical model transformed from vascular texture features, user acquisition site, acquisition device information, acquisition protocol information and data security information.

[0011] Optionally, the skin microsurface data includes:

[0012] Skin surface image, user collection site, user ID, skin cleaning status record, lighting conditions, collection time, skin surface images under different collection conditions, image resolution, collection device information, image processing steps, data security information and time series identification.

[0013] Optionally, the collected data is stored, preprocessed and integrated, including:

[0014] The collected data is securely stored through encryption technology, denoised and contrast adjusted using algorithms, and image enhancement technology is used to improve the recognizability of the processed data. The enhanced vascular texture data and skin microsurface data are integrated into a high-order biometric fingerprint dataset.

[0015] Optionally, high-order data features are obtained by performing feature extraction operations on the high-order biometric fingerprint data set, including:

[0016] By performing data cleaning and standardization operations on the high-order biometric fingerprint data set, a high-order biometric fingerprint feature vector is established, feature selection is performed on the feature vector using a model-based method, and the principal component analysis method is used to perform feature transformation on the selected original features to obtain the first high-order feature, and the second high-order feature is obtained using a polynomial combination method.

[0017] Optionally, the preset initial model is trained to obtain a trained high-order biometric fingerprint model, including:

[0018] A linear regression model is established, and the linear regression model is trained using the gradient descent algorithm and the minimization loss function to obtain a trained high-order biological fingerprint model.

[0019] Optionally, the loss function is:

[0020]

[0021] Among them, MSE is the mean square error, n is the number of samples, and y i is the actual value, is the model's predicted value.

[0022] A mobile storage hardware protection system based on advanced biometric fingerprint recognition technology, comprising:

[0023] The acquisition and processing module is used to acquire data on the user's blood vessel texture and skin micro-surface morphology, and to store, pre-process and integrate the acquired data to obtain a high-order biometric fingerprint data set;

[0024] A data enhancement module is used to obtain high-order data features by performing feature extraction operations on high-order biometric fingerprint data sets;

[0025] A model training module is used to train a preset initial model to obtain a trained high-order biometric fingerprint model; the high-order data features include a first high-order data feature and a second high-order data feature;

[0026] The real-time authentication module is used to perform real-time identification and authentication operations on the user's biometric fingerprint, and control the user's access status based on the real-time authentication results and the user's real-time behavior operations.

[0027] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: the mobile storage hardware protection method based on high-order biometric fingerprint recognition technology provided by the present invention comprises: obtaining a high-order biometric fingerprint data set by collecting data on the user's vascular texture and skin micro-surface morphology, and storing, preprocessing and integrating the collected data; obtaining high-order data features by performing feature extraction operations on the high-order biometric fingerprint data set, and training a preset initial model to obtain a trained high-order biometric fingerprint model; the high-order data features include a first high-order data feature and a second high-order data feature; performing real-time authentication operations on the user's biometric fingerprint, and controlling the user's access status according to the real-time authentication results. This method ensures the convenience of the user and the practicality of the system, and effectively solves the limitations of traditional protection methods by introducing more complex biometric features and biometric authentication methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0029] Figure 1 This is a flow chart of a mobile storage hardware protection method based on advanced biometric fingerprint recognition technology according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The purpose of the present invention is to provide a mobile storage hardware protection method based on high-order biometric fingerprint recognition technology to improve the accuracy and security of the biometric fingerprint recognition system.

[0031] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] like Figure 1 As shown, the embodiment of the present invention provides a mobile storage hardware protection method based on high-level biometric fingerprint recognition technology, including the following steps:

[0033] Step 100 includes: collecting data on the user's blood vessel texture and skin micro-surface morphology.

[0034] Specifically, when collecting the user's vascular texture data, first select a biosensor device dedicated to vascular texture collection, which includes: a high-resolution camera and an infrared light source; secondly, under conditions of stable light intensity, ambient temperature and ambient humidity, the user places a designated body part on the biosensor device, and the biosensor device collects a high-resolution image of the user's body part that contains detailed vascular texture data, and the surface of the biosensor device is kept dry and clean during data collection; finally, collect multiple sets of high-resolution images under different lighting conditions and at different placement angles to improve the accuracy of subsequent identification and authentication.

[0035] Among them, the vascular texture data includes: high-resolution image data at multiple angles and under different lighting conditions; vascular location with markers or coordinates of the vascular area; vascular texture features with vascular direction, vascular branching pattern and vascular thickness; image quality indicators with image resolution, contrast and signal-to-noise ratio for evaluating the clarity and accuracy of the image; a user ID for identifying the relationship between the image and a specific user; light intensity, ambient temperature and ambient humidity during data collection; the specific time of each image collection; a record of image noise reduction, enhancement and standardization processing after the high-resolution image data is collected; the body part to which the high-resolution image data belongs; the model, manufacturer and software version of the biosensor device; and the collection steps, user instructions and cleaning procedures during the collection process.

[0036] More specifically, when collecting the user's skin micro-surface data, first select a high-resolution camera; secondly, under conditions of stable light intensity, ambient temperature and ambient humidity, the user places the designated body part under the camera, and the camera collects the tiny skin surface morphology detail images of the user's body part, and during the data collection period, the user's body part is ensured to be dry and clean; finally, multiple groups of high-resolution images are collected in different time periods and under different skin conditions to form dynamic change data of the skin surface.

[0037] Among them, the skin microsurface data includes: the body part to which the high-resolution image data belongs; the user ID used to identify the relationship between the image and a specific user; the skin cleanliness status of the specified body part before acquisition, the user's placement posture and placement duration of the part; the light intensity, ambient temperature and ambient humidity during data acquisition; the size of the high-resolution image and the dynamic change data of the skin surface; image quality indicators with image resolution, contrast and signal-to-noise ratio, which are used to evaluate the clarity and accuracy of the image; the model, manufacturer and software version of the high-resolution camera; the focal length and exposure time of the high-resolution camera; the image denoising, enhancement and standardization processing records performed after the high-resolution image data is acquired; and the time series identifier of each acquisition time node.

[0038] Step 100 also includes: storing, preprocessing and data integration operations on the collected data to obtain a high-level biometric fingerprint data set.

[0039] Specifically, firstly, encryption technology is used to store all the collected image data securely to ensure the confidentiality of the image data. The encryption technology complies with international data protection standards. Secondly, Gaussian filtering noise reduction technology is used to automatically repair the image data to eliminate the problems of noise and image blur, and adjust the contrast and brightness of the image data. Thirdly, Laplace operator sharpening technology and histogram equalization technology are used to improve the quality of image data and enhance the recognizability of image data. Finally, all image data are integrated into a high-order biometric fingerprint dataset. Among them, the high-order biometric fingerprint dataset includes: all vascular texture data and all skin micro-surface data.

[0040] More specifically, the encryption technology uses symmetric encryption (AES) technology, including: generating a strong key with a key length of 128, 192 or 256 bits to provide different levels of security; using AES technology and the generated key to encrypt the collected original image and related information; Gaussian filtering noise reduction technology is used for noise in the image; Laplace operator sharpening technology is used to enhance the edges and details of the image, improve the clarity of the image, and make the biometric features more obvious; histogram equalization technology is used to improve the overall visual effect of the image, making the dark and bright details of the image more clearly visible.

[0041] Furthermore, feature vectors are extracted using the high-order biometric fingerprint dataset; each data in the high-order biometric fingerprint dataset is converted into a quantitative value, and the quantitative value is adjusted to a specific range of 0 to 1 or converted into a value with unit variance, so that all data are at the same level, and a high-dimensional feature vector is constructed based on the converted data to obtain a numerical representation of each feature vector. The user's feature vector is:

[0042] feature vector =[f1 ,f 2 ,…f n ];

[0043] Where n is the number of selected feature vectors, f 1 ~f n It is the numerical representation of each selected feature vector, feature is the feature, and vector is the feature vector.

[0044] Step 200 includes: performing a feature extraction operation on a high-order biometric fingerprint data set to obtain high-order data features; the high-order data features include a first high-order data feature and a second high-order data feature.

[0045] Specifically, a statistical method is used to perform feature selection on the feature vectors of the high-order biometric fingerprint dataset, the selected features are used as original features, the data corresponding to the original features are the original data, and the principal component analysis (PCA) method is used to perform feature transformation on the original features to obtain new features.

[0046] More specifically, the statistical method implementation process is as follows:

[0047] 1) Perform data preprocessing through standard normalization to ensure that the data features have the same scale. The standard normalization is:

[0048]

[0049] Among them, x is the original data, μ is the mean, and σ is the standard deviation.

[0050] 2) The linear correlation between two continuous variables is calculated according to the Pearson correlation coefficient. The Pearson correlation coefficient is:

[0051]

[0052] Among them, r xy is the correlation coefficient, x i and i is the data point, and is the mean.

[0053] 3) Perform a hypothesis test on the means of two samples based on the t-test. The t-test is:

[0054]

[0055] in, and is the mean of the two samples, s p is the combined standard deviation, n 1 and n 2 is the sample size.

[0056] 4) Test whether there are significant differences in the means of three or more samples based on variance analysis. Variance analysis is:

[0057]

[0058] Among them, F is analysis of variance, SSB is the sum of squares between groups, SSE is the sum of squares within groups, df B is the degrees of freedom between groups, df E is the within-group degrees of freedom.

[0059] 5) The independence between categorical variables was tested according to the chi-square test, which is:

[0060]

[0061] Among them, O i is the observed frequency, E i is the expected frequency.

[0062] 6) Calculate the p-value. If the p-value is less than the significance level, it is considered that there is a significant association between the data feature and the target variable. The p-value is a probability value that measures statistical significance.

[0063] 7) Based on the results of the correlation coefficient, t-test, ANOVA, or chi-square test, select data features with p-values ​​less than the significance level.

[0064] The implementation process of the PCA method is as follows:

[0065] The original data is standardized so that the mean of each feature is 0 and the standard deviation is 1. The linear relationship between different features is obtained by calculating the covariance matrix of the standardized data, and the covariance matrix is ​​subjected to eigenvalue decomposition to obtain the eigenvalues ​​and the eigenvectors corresponding to the eigenvalues. The eigenvector describes the direction with the largest data variance, and the eigenvalue is the variance of the corresponding data in this direction. The eigenvalues ​​are sorted by size, and the eigenvectors corresponding to the largest k eigenvalues ​​are selected. The k eigenvectors form a projection matrix, and the new features are obtained by multiplying the original data with the projection matrix. k is the number of principal components that you want to retain.

[0066] Further, the original feature is polynomially expanded to obtain the first high-order data feature. The embodiment of the present invention uses the PolynomialFeatures class in Python to implement polynomial expansion, which is expressed as:

[0067] Python

[0068] from sklearn.preprocessing importPolynomialFeatures

[0069] poly=PolynomialFeatures(degree=2)

[0070] X_poly=poly.fit_transform(X);

[0071] Furthermore, a second high-order data feature is obtained according to a polynomial combination method to improve the performance of the training model. The polynomial combination method is: performing multiple power multiplication operations on an eigenvalue to obtain a second high-order data feature containing the square, cube and higher power of the corresponding eigenvalue; or performing multiplication operations on different eigenvalues ​​to obtain a second high-order data feature containing the product of different eigenvalues.

[0072] Step 200 also includes: training a preset initial model to obtain a trained high-order biometric fingerprint model.

[0073] Specifically, the high-order biometric fingerprint dataset is first divided into a training set and a test set, and an initial model is preset. The initial model adopts a linear regression model, which is:

[0074] Y=b 0 +b 1 x 1 +b 2 x 2 +…+b n x n ;

[0075] Among them, Y is the target variable, x 1 ~x n is the feature vector, b 0 ~b n are the parameters (coefficients) of the model.

[0076] Secondly, the linear regression model is trained using the gradient descent algorithm and the method of adjusting the model parameters to minimize the loss function.

[0077] Specifically, the loss function is the mean square error (MSE):

[0078]

[0079] Where n is the number of samples, y i is the actual value, is the model's predicted value.

[0080] The steps of the gradient descent algorithm include:

[0081] 1) The weight or bias is used as the initialization parameter of the model, and a loss function is constructed. The loss function of the embodiment of the present invention adopts the mean square error, and the mean square error is:

[0082]

[0083] Among them, L(θ) is the loss function, n is the number of samples, and h θ (x (i) ) is the model prediction value, y (i) is the actual value of the i-th sample.

[0084] 2) Calculate the gradient; the gradient is the partial derivative of the loss function with respect to each parameter, expressed as a vector, where the gradient of the linear regression model is:

[0085]

[0086] The gradient of the weight is:

[0087]

[0088] Among them, w is the weight.

[0089] The gradient of the bias is:

[0090]

[0091] Where b is the bias.

[0092] 3) Update the parameters using the gradient and learning rate until the update meets the stopping condition; the stopping condition is: the gradient is small enough, the predetermined number of iterations is reached, or the loss function value no longer decreases significantly; the learning rate is a hyperparameter that determines the step size of each update; the formula for updating the parameters is:

[0093]

[0094] Among them, θ (t) is the parameter value of the tth iteration, and α is the learning rate.

[0095] Finally, the linear regression model is evaluated using the test set, root mean square error (RMSE) and mean absolute error (MAE) to obtain a trained high-order biometric fingerprint model. The above feature extraction process and model training process are repeatedly iterated to optimize the model and ensure that the model can achieve good generalization performance on the data.

[0096] Specifically, the root mean square error is:

[0097]

[0098] Among them, h θ (x (i) ) is the value predicted by the trained model for the test data set, h θ (x(i) )-y (i) is the error between the predicted value and the actual value for each sample.

[0099] The present invention also provides a mobile storage hardware protection system based on high-level biometric fingerprint recognition technology, comprising:

[0100] The acquisition and processing module is used to acquire data on the user's vascular texture and skin microsurface morphology, and to store, preprocess and integrate the acquired data to obtain a high-order biometric fingerprint data set; both the acquisition and processing module and the data enhancement module use hardware devices that are compatible with the high-order biometric fingerprint model.

[0101] The data enhancement module is used to obtain high-order data features by performing feature extraction operations on high-order biometric fingerprint data sets; the model training module adopts a low-energy model architecture and algorithm to reduce energy consumption during model training; the model training module is used to achieve fast biometric recognition, reduce user waiting time, improve the overall user experience, and extend the battery life of mobile storage hardware.

[0102] Security protection module, including: security chip, which is used to provide encryption and protection functions for high-order biometric fingerprint data sets;

[0103] The real-time authentication module is used to perform real-time identification and authentication operations on the user's biometric fingerprint, and control the user's access status based on the real-time authentication results and the user's real-time behavior operations.

[0104] The mobile storage hardware protection system also includes: an intuitive user interface for simplifying user operation processes and improving user satisfaction; a system architecture for supporting future technology upgrades and new function additions; a remote update module for implementing remote update functions and having deployment model update and security patch functions.

[0105] Step 300 includes: performing a real-time authentication operation on a user's biometric fingerprint based on a mobile storage hardware protection system, and controlling the user's access status according to the real-time authentication result.

[0106] Specifically, the mobile storage hardware protection system conducts rapid biometric authentication on users, and identifies whether the user using the system is the user whose data is being collected based on the user's usage behavior and changes in the surrounding environment; if the user using the system behaves abnormally or fails to identify and authenticate, the system will issue an alarm. The mobile storage hardware protection system adopts a fine-grained permission management mode, setting different permission levels for different users; it has a detailed security log for post-audit and behavior analysis.

[0107] The beneficial effects of the present invention are as follows:

[0108] The present invention ensures user convenience and system practicability, and effectively solves the limitations of traditional protection methods by introducing more complex biometric features and biometric authentication methods.

[0109] The present invention uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only used to help understand the method and core ideas of the present invention. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A mobile storage hardware protection method based on advanced biometric fingerprint recognition technology, characterized in that: The steps include: By collecting data on the user's blood vessel texture and skin micro-surface morphology, and storing, preprocessing and integrating the collected data, a high-level biometric fingerprint data set is obtained; By performing a feature extraction operation on the high-order biometric fingerprint data set, a high-order data feature is obtained, and a preset initial model is trained to obtain a trained high-order biometric fingerprint model; the high-order data feature includes a first high-order data feature and a second high-order data feature; Perform real-time authentication on the user's biometric fingerprint and control the user's access status based on the real-time authentication results.

2. The mobile storage hardware protection method based on advanced biometric fingerprint recognition technology according to claim 1 is characterized in that: The blood vessel texture data includes: Image data under multiple angles and lighting conditions, vascular position, vascular direction, image resolution, user ID, lighting conditions, acquisition time, image processing steps, mathematical model transformed from vascular texture features, user acquisition site, acquisition device information, acquisition protocol information and data security information.

3. The mobile storage hardware protection method based on advanced biometric fingerprint recognition technology according to claim 2 is characterized in that: The skin micro-surface data includes: Skin surface image, user collection site, user ID, skin cleaning status record, lighting conditions, collection time, skin surface images under different collection conditions, image resolution, collection device information, image processing steps, data security information and time series identification.

4. The mobile storage hardware protection method based on advanced biometric fingerprint recognition technology according to claim 3 is characterized in that: The collected data is stored, preprocessed and integrated, including: The collected data is securely stored through encryption technology, denoised and contrast adjusted using algorithms, and image enhancement technology is used to improve the recognizability of the processed data. The enhanced vascular texture data and skin microsurface data are integrated into a high-order biometric fingerprint dataset.

5. The mobile storage hardware protection method based on advanced biometric fingerprint recognition technology according to claim 1 is characterized in that: By performing feature extraction operations on the high-order biometric fingerprint data set, high-order data features are obtained, including: By performing data cleaning and standardization operations on the high-order biometric fingerprint data set, a high-order biometric fingerprint feature vector is established, feature selection is performed on the feature vector using a model-based method, and feature transformation is performed on the selected original features using a principal component analysis method to obtain a first high-order feature, and a second high-order feature is obtained using a polynomial combination method.

6. The mobile storage hardware protection method based on advanced biometric fingerprint recognition technology according to claim 1 is characterized in that: The preset initial model is trained to obtain a trained high-order biometric fingerprint model, including: A linear regression model is established, and the linear regression model is trained using a gradient descent algorithm and a minimization loss function to obtain a trained high-order biometric fingerprint model.

7. The mobile storage hardware protection method based on advanced biometric fingerprint recognition technology according to claim 6 is characterized in that: The loss function is: Among them, MSE is the mean square error, n is the number of samples, and y i is the actual value, is the model's predicted value.

8. A mobile storage hardware protection system based on advanced biometric fingerprint recognition technology, characterized in that: include: The acquisition and processing module is used to acquire data on the user's blood vessel texture and skin micro-surface morphology, and to store, pre-process and integrate the acquired data to obtain a high-order biometric fingerprint data set; A data enhancement module, used to obtain high-order data features by performing feature extraction operations on the high-order biometric fingerprint data set; A model training module, used to train a preset initial model to obtain a trained high-order biometric fingerprint model; the high-order data features include a first high-order data feature and a second high-order data feature; The real-time authentication module is used to perform real-time identification and authentication operations on the user's biometric fingerprint, and control the user's access status based on the real-time authentication results and the user's real-time behavior operations.

Images