Intelligent archival repository identity data processing method and system

Abstract

The invention discloses an intelligent archival repository identity data processing method and system, and relates to the technical field of intelligent archival repository identity authentication, and the method comprises the steps: collecting real-time environment data, carrying out the preprocessing, and generating an environment parameter matrix; collecting biological characteristics of a user through a multi-mode biological recognition terminal, and performing environment compensation correction based on the environment parameter matrix to generate a temporary biological characteristic code; the method comprises the following steps: collecting a microscopic surface image of a target archive carrier, extracting frequency domain features through fast Fourier transform, and generating a physical unclonable feature code of the target archive carrier; based on the real-time environment data, dynamically adjusting the temporary biological feature code weight and the physical unclonable feature code weight, and calculating and evaluating a feature code matching score; dynamic compensation is carried out on the biological characteristics and the frequency domain fingerprints through the environment parameter matrix, a two-factor authentication mechanism based on the physical unclonable characteristic codes is achieved, the unauthorized access rate is reduced, and meanwhile the authentication efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart archive warehouse identity authentication, and in particular to a smart archive warehouse identity data processing method and system. Background Art

[0002] With the increasing demand for smart archive warehouse management, biometric technology has been widely used in the field of identity authentication. At present, mainstream solutions mostly use single-modality biometrics (such as face or fingerprint) for identity authentication. After the identity authentication is passed, the user's access to the archives is managed according to the preset static permission allocation mechanism. However, the archive warehouse environment has interference factors such as temperature and humidity fluctuations and light changes, which may lead to a decrease in the accuracy of biometric collection. In addition, existing technologies usually rely on fixed weight matching strategies, which are difficult to adapt to dynamic environments and changes in user behavior, and have problems such as insufficient security or high misrecognition rates.

[0003] The limitations of existing technologies are mainly reflected in their lack of environmental adaptability. Multimodal biometric technology has improved the robustness of biometric recognition in complex environments to a certain extent, but it still does not fully consider the interference caused by environmental parameters on the feature extraction process, resulting in reduced feature matching stability in cross-environmental scenarios. Summary of the invention

[0004] In view of the problems in the prior art such as poor stability of multimodal biometric recognition in complex environments and serious impact of environmental interference, the present invention provides a method and system for processing identity data in a smart archive warehouse, aiming to improve the robustness and security of identity authentication and enhance the adaptability to environmental changes and user behaviors.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for processing identity data of a smart archive warehouse, which includes collecting real-time environmental data, and performing preprocessing to generate an environmental parameter matrix; Through the multimodal biometric terminal, the user's biometric characteristics are collected, and environmental compensation correction is performed based on the environmental parameter matrix to generate a temporary biometric characteristic code; Collect the microscopic surface image of the target archive carrier, extract the frequency domain features through fast Fourier transform, and generate the physical unclonable feature code of the target archive carrier; Based on real-time environmental data, the temporary biometric signature weight and the physical unclonable signature weight are dynamically adjusted, and the signature matching score is calculated and evaluated; The UWB positioning base station is used to track the user's location in real time to generate behavior trajectory data, and the user's permissions are dynamically adjusted based on the entropy value of the residence time to generate a dynamically adjusted permission set and dynamically update the status of the smart archive warehouse.

[0006] As a preferred solution of the method for processing identity data of the smart archive warehouse of the present invention, the real-time environmental data is collected and pre-processed to generate an environmental parameter matrix. The specific steps are as follows: Real-time collection of temperature, humidity, light intensity and air velocity to obtain real-time environmental data, and perform outlier removal, missing value filling and median filtering to denoise; The preprocessed real-time environmental data is compensated by a dynamic compensation algorithm and organized into a structured matrix according to the space-time dimension to generate an environmental parameter matrix.

[0007] As a preferred solution of the method for processing identity data in the smart archive warehouse of the present invention, the specific steps of generating a temporary biometric code are as follows: The IEEE 1588 precision time protocol is used to synchronize the clock of the multimodal biometric terminal, and the FPGA is used to generate synchronous trigger pulses to synchronously collect three-dimensional facial features, palm vein features, and voiceprint features to obtain user biometric features; The environmental parameter matrix is ​​decomposed into compensation parameters, and the three-dimensional Euclidean distance is calculated using the Gaussian weighted average algorithm. The three-dimensional Euclidean distance is used to calculate the spatial weight, and the effective value is output, and the user's biometric characteristics are compensated by the effective value; The compensated user biometrics are fused, and Z-score normalization, timestamp splicing and position encoding are performed to generate a temporary biometric code using the SHA3-256 algorithm.

[0008] As a preferred solution of the identity data processing method of the smart archive warehouse of the present invention, the environmental parameter matrix is ​​decomposed into compensation parameters, and the three-dimensional Euclidean distance is calculated using the Gaussian weighted average algorithm, the three-dimensional Euclidean distance is used to calculate the spatial weight, the effective value is output, and the user's biometric characteristics are compensated by the effective value. The specific steps are as follows: Based on the environmental parameter matrix, the covariance matrix is ​​calculated through principal component analysis, and characteristic decomposition is performed to extract temperature compensation parameters, humidity compensation parameters, light compensation parameters and air flow rate compensation parameters; The Gaussian weighted average algorithm is used to calculate the spatial weight according to the three-dimensional Euclidean distance between each sensor and the biometric terminal, and the weighted environmental parameters of the same parameter type are normalized to calculate the weighted average value, and the effective value of temperature, humidity, light and wind speed is output; The illumination effective value and temperature effective value are used to perform illumination compensation and temperature compensation on the three-dimensional facial features, the humidity effective value is used to perform humidity compensation on the palm vein features, and the Doppler effect compensation on the voiceprint features is performed by the wind speed effective value.

[0009] As a preferred solution of the method for processing identity data of the smart archive warehouse of the present invention, the physical unclonable feature code of the target archive carrier is generated, and the specific steps are as follows: Capture the micro texture of the target archive carrier through multi-view scanning and obtain multi-view images; The multi-view images are denoised using non-local mean combined with adaptive median filtering, the CLAHE algorithm is used to enhance the local texture contrast, and the multi-view images are registered based on SIFT feature point matching to extract ROI images of fixed size. The ROI image is divided into overlapping sub-blocks and fast Fourier transform is performed to generate a complex spectrum matrix, and the frequency domain resolution is improved by zero padding; The optimized complex spectrum matrix is ​​filtered by a two-dimensional Gaussian bandpass filter, and the energy distribution, information entropy and spectrum slope of the filtered complex spectrum matrix are calculated to construct the frequency domain feature vector, and the dimension is reduced by principal component analysis to retain the previous principal components to generate frequency domain fingerprints; The local binary pattern histogram of the ROI image is extracted by the local binary pattern algorithm, the frequency domain fingerprint is concatenated with the local binary pattern histogram to generate a hybrid feature descriptor, and a SHA3-256 hash operation is performed to generate a physical unclonable feature code.

[0010] As a preferred solution of the method for processing identity data in the smart archive warehouse of the present invention, the following specific steps are taken to dynamically adjust the temporary biometric feature code weight and the physical unclonable feature code weight based on real-time environmental data, calculate the feature code matching score and evaluate it. Based on real-time environmental data, the influence coefficient of each environmental factor is calculated using the Gaussian weighted average method, and the weight of the temporary biometric signature code and the physical unclonable signature code are dynamically adjusted based on the influence coefficient of the environmental factor through a linear weighted adjustment method; The dynamically adjusted temporary biometric signature code is matched with the physical unclonable signature code, and the signature code matching score is calculated using the cosine similarity method, and the matching threshold is set. , evaluate the feature code matching degree.

[0011] As a preferred solution of the identity data processing method of the smart archive warehouse of the present invention, wherein: the real-time tracking of the user location by the UWB positioning base station generates behavior trajectory data, dynamically adjusts the user authority in combination with the residence time entropy value, generates a dynamically adjusted authority set, and dynamically updates the status of the smart archive warehouse. The specific steps are as follows: Based on the evaluation results, multiple UWB positioning base stations in the warehouse capture the positioning tag signals worn by users when accessing archives, and calculate the three-dimensional coordinates in real time based on the TDOA algorithm. Kalman filtering is used to eliminate positioning jump points and cubic spline interpolation is used to complete the occlusion data to generate a continuous trajectory. According to the user's continuous trajectory in front of the filing cabinet, the Shannon entropy within the time window is calculated, the abnormal state of user behavior is evaluated, and the permission downgrade is automatically triggered. The permission tag is sent to the electromagnetic lock controller through the MQTT protocol, and the real-time user location and dynamic permission set are synchronized through Redis to update the status of the smart archive warehouse.

[0012] In a second aspect, the present invention provides a smart archive warehouse identity data processing system, including a data collection module, a feature code generation module, an archive processing module, an evaluation module and a warehouse authority management module; The data acquisition module is used to collect real-time environmental data, perform preprocessing, and generate an environmental parameter matrix; A feature code generation module is used to collect user biometric features through a multimodal biometric terminal, and perform environmental compensation correction based on an environmental parameter matrix to generate a temporary biometric feature code; The file processing module is used to collect the microscopic surface image of the target file carrier, extract the frequency domain features through fast Fourier transform, and generate the physical non-clonable feature code of the target file carrier; An evaluation module, used to dynamically adjust the temporary biometric signature weight and the physical unclonable signature weight based on real-time environmental data, calculate signature matching scores and perform evaluation; The warehouse authority management module is used to track the user's location in real time through the UWB positioning base station to generate behavior trajectory data, dynamically adjust the user's authority based on the entropy value of the residence time, generate a dynamically adjusted authority set, and dynamically update the status of the smart archive warehouse.

[0013] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the smart archive warehouse identity data processing method as described in the first aspect of the present invention is implemented.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the smart archive warehouse identity data processing method as described in the first aspect of the present invention.

[0015] The beneficial effects of the present invention are as follows: dynamic compensation of biometrics and frequency domain fingerprints is performed through the environmental parameter matrix to realize a two-factor authentication mechanism based on a physically unclonable feature code; combined with the UWB behavior trajectory entropy analysis, high-precision identity authentication and dynamic authority control of the smart archive warehouse are realized, and the problems of biometric distortion, easy duplication of carrier features, and difficulty in tracing abnormal behaviors in complex environments in traditional methods are solved, thereby reducing the unauthorized access rate and improving the authentication efficiency, thereby forming a full-dimensional security protection system covering "user-environment-carrier-behavior". BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0017] Figure 1 This is a flow chart of the identity data processing method of the smart archive warehouse in Example 1.

[0018] Figure 2 This is a schematic diagram of the identity data processing system of the smart archive warehouse in Example 1.

[0019] Figure 3 This is a flow chart of generating a temporary biometric code in Example 1.

[0020] Figure 4 This is a sub-flow chart for generating a physical unclonable signature code in Example 1. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0022] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0023] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0024] Example 1, reference Figure 1~Figure 4This embodiment provides a method for processing identity data of a smart archive warehouse, comprising the following steps: S1. Collect real-time environmental data, perform preprocessing, and generate an environmental parameter matrix.

[0025] Temperature, humidity, light intensity and air velocity are collected in real time to obtain real-time environmental data, and outliers are eliminated, missing values ​​are filled in, and median filtering is used for denoising.

[0026] It should be explained that the specific steps for outlier removal are: the temperature, humidity, light intensity and air velocity in the real-time environmental data are processed in a sliding window manner, and the window size example is 5 consecutive sampling points. The mean and standard deviation of the real-time environmental data in each window are calculated, and the threshold example is set as the mean ± 3 times the standard deviation. The real-time environmental data points in the window are traversed. If a real-time environmental data point exceeds the threshold, it is marked as an outlier and removed, and the vacant position is retained to enter the missing value completion process; Missing value filling is specifically to detect the vacancy in the real-time environmental data. The vacancy may be caused by the removal of outliers or the temporary failure of the sensor. The missing values ​​are filled by linear interpolation for temperature, humidity, light intensity and air velocity. The interpolation basis is the nearest valid data point before and after. If the vacancy is at the beginning or end of the sequence and there is no valid neighboring point, the current sensor’s historical mean value of the same period is used to fill it. The historical period example is the mean value of the data at the same time in the past 24 hours. The specific steps of median filtering denoising are as follows: sliding window processing is performed on the completed temperature, humidity, light intensity and air velocity data respectively. The window size example is 5 consecutive sampling points. The values ​​are sorted in the window and the median is taken to replace the original value of the window center point. The sliding step is 1 point, and the filtered data is updated point by point. Finally, the smoothed real-time environmental data is output for subsequent dynamic compensation algorithm.

[0027] The preprocessed real-time environmental data is compensated by a dynamic compensation algorithm and organized into a structured matrix according to the space-time dimension to generate an environmental parameter matrix.

[0028] It should be explained that the deviation between the pre-processed temperature, humidity, light intensity and air velocity data and the standard environmental reference value is calculated, and the real-time environmental data is adjusted using a linear compensation formula (for example, a compensation coefficient of 0.8). The compensated real-time environmental data is spatially aligned according to a three-dimensional rectangular coordinate system, and time synchronization with 1μs accuracy is achieved through the IEEE 1588 protocol. Time alignment is performed at 1 second intervals, and then a four-dimensional matrix (X / Y / Z coordinates + timestamp) is constructed to store the four environmental parameters in 32-bit floating point format. The arithmetic average of multiple sensor data at the same spatiotemporal coordinates is taken, NaN is filled in the uncovered position, and re-collection is triggered when NaN exceeds 5%. The continuous NaN area is patched using a three-dimensional linear interpolation of 3×3×3 spatial units and a 5-second time window. The completed matrix is ​​Min-Max normalized (range [0,1]), converted to HDF5 format for storage and metadata is recorded, and finally the environmental parameter matrix is ​​output after verifying that the statistical characteristics (mean, variance, spatial gradient) conform to historical laws.

[0029] S2. Collect user biometrics through a multimodal biometric terminal, perform environmental compensation correction based on an environmental parameter matrix, and generate a temporary biometric code.

[0030] The IEEE 1588 precision time protocol is used to synchronize the clock of the multimodal biometric terminal, and FPGA is used to generate synchronous trigger pulses to synchronously collect three-dimensional facial features, palm vein features and voiceprint features to obtain user biometric features.

[0031] It should be explained that a PTP (Precision Time Protocol) communication link is established between the master clock and the multimodal biometric terminal, and the Ethernet physical layer is used to transmit time synchronization messages. The master clock periodically sends Sync (synchronization message) and Follow_Up (follow message), and the multimodal biometric terminal records the message arrival time and calculates the clock offset. The network transmission delay is measured through the Delay_Request (delay request) / Delay_Response (delay response) mechanism, and finally the clock synchronization accuracy of each terminal is better than 1μs. During the synchronization process, the transparent clock mode is used to process the edge device to compensate for the error introduced by the switch residence time; The FPGA (field programmable gate array) synchronous trigger pulse generation process starts after the clock synchronization is completed. The digital phase-locked loop configured inside the FPGA locks the 1PPS signal provided by IEEE 1588 as a reference. The trigger pulse timing is controlled by a state machine written in VHDL (hardware description language), generating a TTL level synchronization signal with a period of 100ms and a pulse width of 10ns, and the rising edge time jitter is controlled within 200ps. The synchronization signal is transmitted to the three-dimensional face feature acquisition unit, palm vein feature acquisition unit and voiceprint feature acquisition unit simultaneously through a low-voltage differential signal; After receiving the synchronous trigger pulse, the 3D facial feature acquisition unit starts the structured light projector to project an 850nm infrared speckle pattern, synchronously triggers two global shutter CMOS sensors to collect binocular infrared images at a frame rate of 30fps, and the exposure time is strictly aligned with the rising edge of the trigger pulse. The palm vein feature acquisition unit turns on the 940nm near-infrared LED array under the synchronous pulse trigger, and uses a rolling shutter CMOS sensor to collect palm vein images at a frame rate of 15fps. The exposure start time maintains a fixed phase relationship with the trigger pulse. The voiceprint feature acquisition unit starts the microphone array with a sampling rate of 48kHz on the rising edge of the synchronous pulse and continuously collects 200ms voice signals; The collected 3D facial feature data includes a binocular infrared image pair with a resolution of 1024×768, the palm vein feature data includes a single near-infrared image with a resolution of 640×480, and the voiceprint feature data includes 8-channel PCM waveform data with a sampling rate of 48kHz; It should also be noted that a PTP (Precision Time Protocol) communication link is established between the master clock and the multimodal biometric terminal as follows: The network is configured to use the UDP / IPv4 transport protocol and 224.0.1.129 is set as the multicast communication address. The master clock acts as the Grandmaster (master clock source) to periodically send Sync messages and Follow_Up messages. Each terminal uses the Delay_Request / Delay_Response interaction mechanism to accurately measure the network path delay. The measured data is input into the PID (proportional-integral-differential) control algorithm to dynamically adjust the local clock phase and frequency. At the same time, the Transparent Clock function is enabled to compensate for the residence time error introduced by the network switching device in real time. The above mechanisms work together to finally establish a stable PTP communication link.

[0032] The environmental parameter matrix is ​​decomposed into compensation parameters, and the Gaussian weighted average algorithm is used to calculate the three-dimensional Euclidean distance. The three-dimensional Euclidean distance is used to calculate the spatial weight, and the effective value is output, and the user's biometric characteristics are compensated by the effective value.

[0033] The compensated user biometrics are fused, and Z-score normalization, timestamp splicing and position encoding are performed to generate a temporary biometric code using the SHA3-256 algorithm.

[0034] It should be explained that the 1024-dimensional facial feature vector, the 256-dimensional palm vein feature vector and the 128-dimensional voiceprint feature vector are concatenated into a 1408-dimensional composite feature vector, and the 1408-dimensional composite feature vector is subjected to Z-score standardization. The mean and standard deviation of each feature dimension are calculated using the standardization formula (eigenvalue - mean) / standard deviation. The standardized 1408-dimensional feature vector is concatenated with the 64-bit IEEE 1588 timestamp (accuracy 100ns) in big-endian byte order, and the timestamp occupies the first 8 bytes. The standardized 1408-dimensional feature vector data is arranged in floating-point format in the last 704 bytes (1408 dimensions × 4 bytes / floating point number) to form a concatenated data block with a total length of 712 bytes. A 2-byte position code is added to the front end to identify the collection terminal number. The concatenated data stream is encrypted using the SHA3-256 hash algorithm to generate a 256-bit temporary biometric code.

[0035] S3. Decompose the environmental parameter matrix into compensation parameters, and use the Gaussian weighted average algorithm to calculate the three-dimensional Euclidean distance, use the three-dimensional Euclidean distance to calculate the spatial weight, output the effective value, and compensate the user's biometric characteristics through the effective value.

[0036] Based on the environmental parameter matrix, the covariance matrix is ​​calculated by principal component analysis, and characteristic decomposition is performed to extract temperature compensation parameters, humidity compensation parameters, light compensation parameters and air flow rate compensation parameters.

[0037] It should be explained that the environmental parameter matrix is ​​separated into independent sub-matrices according to the four parameter dimensions of temperature, humidity, light intensity and air velocity. Each independent sub-matrix retains the original spatial coordinates and timestamp. The covariance matrix is ​​calculated for each parameter sub-matrix, and the Jacobi algorithm is used for eigendecomposition to obtain eigenvalues ​​and eigenvectors. According to the eigenvalue size, the temperature compensation parameters select the first two principal components (for example, eigenvalue>1.5), the humidity compensation parameters select the first two principal components (for example, eigenvalue>1.2), the light compensation parameters select the first three principal components (for example, eigenvalue>1.0), and the air flow rate compensation parameter selects the first principal component (for example, eigenvalue>0.8). The extracted principal component eigenvectors are multiplied with the original parameter submatrix to generate temperature compensation parameters, humidity compensation parameters, light compensation parameters, and air flow rate compensation parameters.

[0038] It should be noted that for temperature and humidity parameters, due to their large range of variation, the first two principal components in the covariance matrix can explain more than 85% of the cumulative variance, so the first two principal components are selected; the illumination changes are more complex and frequent, and the local fluctuations are strong. In order to enhance the adaptability to high-frequency changes, the first three principal components are selected to ensure that more texture changes are captured; the influence of air velocity on voiceprint is mainly concentrated in the Doppler shift in a single main direction, so only the first principal component needs to be selected to obtain representative features.

[0039] The set eigenvalue thresholds (such as 1.5, 1.2, 1.0, and 0.8) are derived from statistical analysis of historical data and are empirically set based on the principle that the cumulative explained variance exceeds 85%. Through experiments, it is found that when the selected principal component meets the above eigenvalue thresholds, the generated compensation parameters can improve the stability and accuracy of subsequent recognition without losing the main feature information.

[0040] The Gaussian weighted average algorithm is used to calculate the spatial weight according to the three-dimensional Euclidean distance between each sensor and the biometric terminal. The weighted environmental parameters of the same parameter type are normalized, the weighted average value is calculated, and the effective values ​​of temperature, humidity, light and wind speed are output.

[0041] It should be noted that the Gaussian weighted average algorithm is used to calculate the three-dimensional Euclidean distance between each environmental sensor and the multimodal biometric terminal. The three-dimensional Euclidean distance is obtained by squaring the coordinate differences, adding them, and then taking the square root. The Gaussian kernel function with a standard deviation of 1.5 meters is used to calculate the spatial weight. The spatial weight calculation formula is exp(-three-dimensional Euclidean distance² / 2σ²); The four parameter types of temperature, humidity, light intensity and air flow rate are processed separately. The sensor data of the same parameter type are multiplied by the corresponding spatial weight and then summed up, and then divided by the sum of the spatial weights to achieve normalization. A 5-second sliding window is used to perform moving average filtering on the weighted parameter values ​​in the time dimension. Finally, four compensation benchmark quantities are output: effective value of temperature (example output range 20-30℃), effective value of humidity (example output range 30-70%), effective value of light (example output range 200-1000lux) and effective value of wind speed (example output range 0.1-2m / s). All effective values ​​retain the original timestamp information.

[0042] The illumination effective value and temperature effective value are used to perform illumination compensation and temperature compensation on the three-dimensional facial features, the humidity effective value is used to perform humidity compensation on the palm vein features, and the Doppler effect compensation on the voiceprint features is performed by the wind speed effective value.

[0043] It should be noted that the illumination compensation of three-dimensional facial features adopts a grayscale correction algorithm based on the Lambertian reflectance model, and uses the effective value of illumination to adjust the image pixel value. The compensation formula is: ; in, is the compensated image pixel intensity value, is the original image pixel intensity value, is the standard illumination reference value (unit: lux), is the effective value of light (unit: lux); Temperature compensation corrects the coordinates of key points on the face through the thermal expansion coefficient. The calculation formula is: ; in, is the key point coordinate correction (unit: mm), is the standard temperature (unit: °C), is the coefficient of thermal expansion (unit: 1 / °C), is the original key point coordinate value (unit: mm), is the effective value of temperature (unit: °C); The humidity compensation of palm vein characteristics uses the refractive index correction formula to adjust the tissue penetration depth of near-infrared light according to the effective value of humidity. The expression is: ; in, is the corrected refractive index, is the original refractive index, is the effective value of humidity (unit: %RH); The Doppler compensation of the voiceprint feature calculates the frequency shift according to the effective value of the wind speed and performs translation correction on the speech spectrum. The expression is: ; in, is the frequency offset (unit: Hz), is the effective value of wind speed (unit: m / s), is the speed of sound (unit: m / s), is the original sound wave frequency (unit: Hz); All compensation operations keep the dimensions and timestamps of the original feature data unchanged, and the correspondence between the compensation parameters and the effective values ​​of the environment is realized through a pre-calibrated lookup table.

[0044] It should be noted that the pre-calibrated lookup table is an environment-compensation parameter mapping table constructed by experimental calibration, which is used to quickly obtain the corresponding compensation parameters according to the effective value of the environment during operation. The specific construction method is to collect the original biometric data and real standard data under different temperatures, humidity, light intensity and air flow rate in the experimental stage, compare and analyze the impact of each environmental variable on the quality or error of feature acquisition, and use linear regression, polynomial fitting or spline interpolation methods to construct the functional relationship between environmental parameters and compensation factors, discretize the function into a lookup table with a fixed sampling step, and each lookup table entry records the corresponding compensation parameter.

[0045] S4. Collect the microscopic surface image of the target file carrier, extract the frequency domain features through fast Fourier transform, and generate the physical unclonable feature code of the target file carrier.

[0046] The microscopic texture of the target archive carrier is captured through multi-view scanning to obtain multi-view images.

[0047] The multi-view images are denoised by using non-local mean combined with adaptive median filtering. The CLAHE algorithm is used to enhance the local texture contrast. The multi-view images are registered based on SIFT feature point matching to extract ROI images of fixed size.

[0048] It should be stated that the non-local mean denoising algorithm is used, the search window example is 21×21 pixels, the similar block window example is 7×7 pixels, the noise standard deviation is estimated to be 15, the denoised image is subjected to adaptive median filtering, the initial window size example is 3×3 pixels, the maximum window size example is 7×7 pixels, the filter strength is dynamically adjusted, the CLAHE algorithm is applied to enhance the contrast, the block area example is set to 8×8 pixels, the histogram equalization limit threshold example is 2.0, and the micro-texture visibility is enhanced; Key points are extracted based on SIFT feature point detection. The feature descriptor dimension is 128 dimensions. The nearest neighbor distance ratio matching strategy is used to screen the correct matching point pairs. The matching threshold example is 0.75. The homography transformation matrix is ​​calculated by the RANSAC algorithm. The number of iterations is 2000 times, and the error threshold example is 3.0 pixels. Accurate multi-view image registration is achieved, and a fixed-size ROI area is extracted on the registered image.

[0049] The ROI image is divided into overlapping sub-blocks and subjected to fast Fourier transform to generate a complex spectrum matrix, and the frequency domain resolution is improved by zero padding.

[0050] It should be explained that the ROI image is divided into overlapping sub-blocks of 32×32 pixels, with a step size of 16 pixels to ensure that adjacent sub-blocks have a 50% overlapping area. After each sub-block is pre-processed by the Hanning window function, it is zero-filled and expanded to 64×64 pixels to improve the frequency domain resolution; A two-dimensional fast Fourier transform is performed on the expanded sub-blocks to generate a 64×64 complex spectrum matrix, in which the real part represents the amplitude spectrum and the imaginary part represents the phase spectrum. The complex spectra of all sub-blocks are arranged in the original spatial order to form a three-dimensional complex spectrum matrix (space x×space y×frequency component), maintaining the spatial correspondence between sub-blocks. The zero padding operation is implemented before the Fourier transform by symmetrically padding zeros at the image boundary to ensure that the frequency domain sampling interval is halved and the frequency resolution is doubled.

[0051] The optimized complex spectrum matrix is ​​filtered by a two-dimensional Gaussian bandpass filter, and the energy distribution, information entropy and spectrum slope of the filtered complex spectrum matrix are calculated to construct the frequency domain feature vector, and the dimension is reduced by principal component analysis to retain the previous principal components to generate frequency domain fingerprints.

[0052] It should be explained that the passband range of the two-dimensional Gaussian bandpass filter is set, the low-frequency cutoff frequency example is 0.05 times the Nyquist frequency, the high-frequency cutoff frequency example is 0.4 times the Nyquist frequency, and the standard deviation σ example is 0.1; the filter is applied to each sub-block spectrum of the complex spectrum matrix for frequency domain filtering, retaining the frequency components within the passband range of 0.05~0.4 times the Nyquist frequency; the energy distribution of the complex spectrum of each sub-block after filtering is calculated, and the energy value is obtained by summing the squares of the amplitude spectrum.

[0053] Furthermore, the two-dimensional Gaussian bandpass filter is applied to each sub-block spectrum of the complex spectrum matrix for frequency domain filtering, and the frequency components within the set passband range are retained. The passband range is to filter out DC and very low frequency components to eliminate the overall illumination effect, while suppressing high-frequency sharp noise, and focusing on retaining the intermediate frequency components to extract representative micro-texture features of the archive surface. The intermediate frequency components mainly correspond to structural details, material textures and local edge features in the image, and are usually located between 0.05 and 0.4 times the Nyquist frequency in the Fourier frequency domain.

[0054] The information entropy calculation is based on the normalized probability distribution of the amplitude spectrum. The spectrum slope is obtained by linear fitting of the logarithmic amplitude spectrum in the radial frequency direction. The three features of energy value, information entropy and spectrum slope of each sub-block are spliced ​​in a fixed order to construct the initial frequency domain feature vector set. Perform principal component analysis on the initial frequency domain feature vector set, calculate the covariance matrix and perform eigenvalue decomposition, retain the first k principal components (k=8 in the example) with a cumulative contribution rate exceeding 85%, reduce the dimension of the original feature vector through projection transformation, and finally generate a compact frequency domain fingerprint vector.

[0055] The local binary pattern histogram of the ROI image is extracted by the local binary pattern algorithm, the frequency domain fingerprint is concatenated with the local binary pattern histogram to generate a hybrid feature descriptor, and a SHA3-256 hash operation is performed to generate a physical unclonable feature code.

[0056] It should be noted that the frequency domain fingerprint vector and the local binary pattern histogram feature vector are normalized, and the Min-Max method is used to scale each eigenvalue to the interval [0,1]. The two feature vectors are directly concatenated in the order of frequency domain fingerprint first and local binary pattern histogram later to generate a 67-dimensional hybrid feature descriptor.

[0057] Furthermore, in order to extract the texture features of the microscopic surface of the archive carrier, the local binary pattern (LBP) algorithm is used to extract the local binary pattern histogram of each pixel in the ROI image. Specifically, for each pixel, a 3×3 neighborhood is sampled with it as the center; compared with the central pixel value, if the neighborhood pixel value is greater than or equal to the central pixel, it is recorded as 1, otherwise it is recorded as 0; an 8-bit binary code is obtained and converted into a decimal integer ranging from 0 to 255; the frequency of each LBP code in the entire ROI image is counted to form a 256-dimensional local binary pattern histogram, and the local binary pattern histogram is reduced in dimension by principal component analysis. The reduced local binary pattern histogram is spliced ​​with the frequency domain fingerprint vector to form a hybrid feature descriptor.

[0058] Perform SHA3-256 hash operation on the mixed feature descriptor, arrange the input mixed feature descriptor in big-endian byte order, and perform 24 rounds of iterative processing through the Keccak-f

[1600] permutation function, each round contains After five steps of bit operations, a physical unclonable signature with a length of 256 bits is finally output.

[0059] S5. Based on real-time environmental data, dynamically adjust the temporary biometric feature code weight and the physical unclonable feature code weight, calculate the feature code matching score and perform an evaluation.

[0060] Based on real-time environmental data, the Gaussian weighted average method is used to calculate the influence coefficient of each environmental factor, and the linear weighted adjustment method is used to dynamically adjust the temporary biometric signature weight and the physical unclonable signature weight based on the influence coefficient of the environmental factor.

[0061] It should be explained that based on real-time environmental data, the influence coefficient of each environmental factor is calculated using the Gaussian weighted average method, and the expression is: ; in, is the influence coefficient of environmental factors, is the value of the environmental factor, is the mean of the environmental factors, is the standard deviation of environmental factors; it should be noted that the influence coefficient of environmental factors is used to quantify the degree of interference that real-time environmental conditions may cause to recognition accuracy, including the influence coefficients of four environmental factors: temperature, humidity, light, and air flow. The influence coefficient of each environmental factor is calculated using the Gaussian weighted average method.

[0062] The influence coefficients of the four environmental factors are summed up as the total adjustment factor, and the initial weight of the temporary biometric signature code and the initial weight of the physical unclonable signature code are multiplied by ( ) to achieve linear adjustment, the expression is: ; in, is the adjusted weight, is the initial weight, is the total number of environmental factors, It is The influence coefficient of environmental factors, is the index variable for environmental factors; The adjusted temporary biometric code weight and physical unclonable code weight are normalized to ensure that their weighted sum is 1. The integrity of the original code data is retained during the weight adjustment process, and finally the dynamically adjusted temporary biometric code weight and physical unclonable code weight values ​​are output.

[0063] The dynamically adjusted temporary biometric signature code is matched with the physical unclonable signature code, and the signature code matching score is calculated using the cosine similarity method, and the matching threshold is set. , evaluate the feature code matching degree.

[0064] It should be noted that the dimensions of the dynamically adjusted temporary biometric signature and the physical unclonable signature are aligned, and the dimensions of the two are made consistent by zero padding. The signature matching score is calculated using the cosine similarity method, and the expression is: ; in, is the signature matching score, It is a temporary biometric code after dynamically adjusting the weight. It is a physical unclonable signature code after dynamically adjusting the weight. To dynamically adjust the weighted temporary biometric code vector, To dynamically adjust the weight of the physical unclonable signature vector, for The L2 norm of for The L2 norm of is the norm product, used for normalization; It should be noted that the expression for calculating the feature code matching score is the cosine similarity calculation method, which is used to measure the angle similarity between two feature code vectors. The value range is , the closer it is to 1, the more similar the temporary biometric signature vector and the physical unclonable signature vector are, and the higher the matching degree is.

[0065] Based on historical verification data, the difference between the true positive rate and the false positive rate is analyzed by the receiver operating characteristic curve (ROC) evaluation method, and the point with the largest difference is selected as the optimal critical value of the recognition system to set the matching threshold. , evaluate the feature code matching degree; It should be noted that historical verification data refers to a sample set collected during the development and debugging phase, including real identity tags of multiple users, environmental parameters, correspondingly generated temporary biometric feature codes and physical unclonable feature codes, and actual authentication results. The sample set is used to evaluate the matching performance of the method under various environmental and behavioral conditions offline. The above information is collected with the consent of the user and is used for legitimate purposes. , indicating that the identity authentication is successful, the electromagnetic lock is activated, and the user can access the archive; when , indicating that authentication failure prohibits access to the archive.

[0066] S6. Use the UWB positioning base station to track the user's location in real time to generate behavior trajectory data, dynamically adjust user permissions based on the entropy value of the residence time, generate a dynamically adjusted permission set, and dynamically update the status of the smart archive warehouse.

[0067] Based on the evaluation results, multiple UWB positioning base stations in the warehouse are used to capture the positioning tag signals worn by users when accessing archives, and the three-dimensional coordinates are calculated in real time based on the TDOA algorithm. The Kalman filter is used to eliminate positioning jump points and the cubic spline interpolation is used to complete the occluded data to generate a continuous trajectory.

[0068] It should be noted that at least four UWB positioning base stations in the smart archive warehouse periodically send ranging signals in the 6.5GHz frequency band, and the UWB positioning tags worn by users receive the signals and return response pulses. Each base station measures the signal arrival time difference (TDOA), and uses the Chan algorithm to solve the hyperbolic equation group to calculate the three-dimensional coordinates. The update frequency example is 10Hz. The original coordinate data is processed by Kalman filtering, the state transfer matrix adopts a constant velocity model, the measurement noise covariance example is 0.1m², and the positioning jump points exceeding 3σ are eliminated. For data points lost due to occlusion, the cubic spline interpolation method is used to complete the interpolation time window, and the interpolation time window is limited to 2 seconds. The final output continuous trajectory data contains timestamps (accuracy 1ms), three-dimensional coordinates (accuracy ±10cm) and motion velocity vectors.

[0069] According to the user's continuous trajectory in front of the filing cabinet, the Shannon entropy within the time window is calculated, the abnormal state of user behavior is evaluated, and the permission downgrade is automatically triggered. The permission tag is sent to the electromagnetic lock controller through the MQTT protocol, and the real-time user location and dynamic permission set are synchronized through Redis to update the status of the smart archive warehouse.

[0070] It should be explained that the continuous trajectory data of the user in front of the filing cabinet is divided into 60-second sliding time windows, and the window step size is 10 seconds. In each window, the three-dimensional space is divided into 20cm×20cm×20cm cubic grids. The proportion of the user's stay time in each grid is counted, and the Shannon entropy in the time window is calculated. The expression is: ; in, is the Shannon entropy, It is The probability of an event (such as the proportion of time a user stays in a certain area) occurring. is the index variable of the event; When the Shannon entropy value exceeds the threshold (for example, H>2.5), it is judged as abnormal behavior, and the permission downgrade instruction is published to the "access_control / permission" topic (a message channel dedicated to permission control) through the MQTT protocol. The message payload contains the user ID, downgraded permission level (for example, downgraded to read-only permission) and timestamp. The electromagnetic lock controller subscribes to the topic and immediately executes the permission change. At the same time, the user's current location coordinates and the latest permission set are written to the Redis database in the form of a key-value pair. The key name format is "user:{ID}:status", and the value is in JSON format, including the current location coordinates, current permission level and last updated timestamp fields. The data update frequency example is 1Hz. The Redis publish / subscribe channel synchronously notifies all smart archive warehouse terminal devices to update the status display, ensuring that the entire smart archive warehouse completes permission status synchronization within 300ms.

[0071] This embodiment also provides a smart archive warehouse identity data processing system, including: a data collection module, a feature code generation module, an archive processing module, an evaluation module and a warehouse authority management module; The data acquisition module is used to collect real-time environmental data, perform preprocessing, and generate an environmental parameter matrix; A feature code generation module is used to collect user biometric features through a multimodal biometric terminal, and perform environmental compensation correction based on an environmental parameter matrix to generate a temporary biometric feature code; The file processing module is used to collect the microscopic surface image of the target file carrier, extract the frequency domain features through fast Fourier transform, and generate the physical non-clonable feature code of the target file carrier; An evaluation module, used to dynamically adjust the temporary biometric signature weight and the physical unclonable signature weight based on real-time environmental data, calculate signature matching scores and perform evaluation; The warehouse authority management module is used to track the user's location in real time through the UWB positioning base station to generate behavior trajectory data, dynamically adjust the user's authority based on the entropy value of the residence time, generate a dynamically adjusted authority set, and dynamically update the status of the smart archive warehouse.

[0072] This embodiment also provides a computer device, which is suitable for the case of a smart archive warehouse identity data processing method, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the smart archive warehouse identity data processing method proposed in the above embodiment.

[0073] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0074] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for processing identity data of a smart archive warehouse proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0075] In summary, the present invention achieves high-precision identity authentication and dynamic permission control in smart archive warehouses by dynamically compensating biometrics and frequency domain fingerprints through an environmental parameter matrix, realizing a two-factor authentication mechanism based on a physically unclonable feature code, and combining the UWB behavior trajectory entropy analysis. It solves the problems of biometric distortion, easy duplication of carrier features, and difficulty in tracing abnormal behaviors in traditional methods in complex environments, reduces the unauthorized access rate, and improves authentication efficiency, forming a full-dimensional security protection system covering "user-environment-carrier-behavior".

[0076] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for processing identity data of a smart archive warehouse, characterized by: include, Collect real-time environmental data, perform preprocessing, and generate an environmental parameter matrix; Through the multimodal biometric terminal, the user's biometric characteristics are collected, and environmental compensation correction is performed based on the environmental parameter matrix to generate a temporary biometric characteristic code; Collect the microscopic surface image of the target archive carrier, extract the frequency domain features through fast Fourier transform, and generate the physical unclonable feature code of the target archive carrier; Based on real-time environmental data, the temporary biometric signature weight and the physical unclonable signature weight are dynamically adjusted, and the signature matching score is calculated and evaluated; The UWB positioning base station is used to track the user's location in real time to generate behavior trajectory data, and the user's permissions are dynamically adjusted based on the entropy value of the residence time to generate a dynamically adjusted permission set and dynamically update the status of the smart archive warehouse.

2. A method for processing identity data of a smart archive warehouse as claimed in claim 1, characterized in that: The real-time environmental data is collected and preprocessed to generate an environmental parameter matrix. The specific steps are as follows: Real-time collection of temperature, humidity, light intensity and air velocity to obtain real-time environmental data, and perform outlier removal, missing value filling and median filtering to denoise; The preprocessed real-time environmental data is compensated by a dynamic compensation algorithm and organized into a structured matrix according to the space-time dimension to generate an environmental parameter matrix.

3. A method for processing identity data of a smart archive warehouse as claimed in claim 1, characterized in that: Generate a temporary biometric code. The specific steps are as follows: The IEEE 1588 precision time protocol is used to synchronize the clock of the multimodal biometric terminal, and the FPGA is used to generate synchronous trigger pulses to synchronously collect three-dimensional facial features, palm vein features, and voiceprint features to obtain user biometric features; The environmental parameter matrix is ​​decomposed into compensation parameters, and the three-dimensional Euclidean distance is calculated using the Gaussian weighted average algorithm. The three-dimensional Euclidean distance is used to calculate the spatial weight, and the effective value is output, and the user's biometric characteristics are compensated by the effective value; The compensated user biometrics are fused, and Z-score normalization, timestamp splicing and position encoding are performed to generate a temporary biometric code using the SHA3-256 algorithm.

4. A method for processing identity data of a smart archive warehouse as claimed in claim 3, characterized in that: The environmental parameter matrix is ​​decomposed into compensation parameters, and the Gaussian weighted average algorithm is used to calculate the three-dimensional Euclidean distance. The three-dimensional Euclidean distance is used to calculate the spatial weight, and the effective value is output. The user's biometric characteristics are compensated by the effective value. The specific steps are as follows: Based on the environmental parameter matrix, the covariance matrix is ​​calculated through principal component analysis, and characteristic decomposition is performed to extract temperature compensation parameters, humidity compensation parameters, light compensation parameters and air flow rate compensation parameters; The Gaussian weighted average algorithm is used to calculate the spatial weight according to the three-dimensional Euclidean distance between each sensor and the biometric terminal, and the weighted environmental parameters of the same parameter type are normalized to calculate the weighted average value, and the effective value of temperature, humidity, light and wind speed is output; The illumination effective value and temperature effective value are used to perform illumination compensation and temperature compensation on the three-dimensional facial features, the humidity effective value is used to perform humidity compensation on the palm vein features, and the Doppler effect compensation on the voiceprint features is performed by the wind speed effective value.

5. A method for processing identity data of a smart archive warehouse as claimed in claim 1, characterized in that: Generate the physical unclonable signature code of the target archive carrier. The specific steps are as follows: Capture the micro texture of the target archive carrier through multi-view scanning and obtain multi-view images; The multi-view images are denoised using non-local mean combined with adaptive median filtering, the local texture contrast is enhanced using the CLAHE algorithm, and the multi-view images are registered based on SIFT feature point matching to extract ROI images of fixed size. The ROI image is divided into overlapping sub-blocks and fast Fourier transformed to generate a complex spectrum matrix, and the frequency domain resolution is improved by zero padding; The optimized complex spectrum matrix is ​​filtered by a two-dimensional Gaussian bandpass filter, and the energy distribution, information entropy and spectrum slope of the filtered complex spectrum matrix are calculated to construct the frequency domain feature vector, and the dimension is reduced by principal component analysis to retain the previous principal components to generate frequency domain fingerprints; The local binary pattern histogram of the ROI image is extracted by the local binary pattern algorithm, the frequency domain fingerprint is concatenated with the local binary pattern histogram to generate a hybrid feature descriptor, and a SHA3-256 hash operation is performed to generate a physical unclonable feature code.

6. A method for processing identity data of a smart archive warehouse as claimed in claim 1, characterized in that: The specific steps of dynamically adjusting the temporary biometric feature code weight and the physical unclonable feature code weight based on real-time environmental data, calculating the feature code matching score and evaluating it are as follows: Based on real-time environmental data, the influence coefficient of each environmental factor is calculated using the Gaussian weighted average method, and the weight of the temporary biometric signature code and the physical unclonable signature code are dynamically adjusted based on the influence coefficient of the environmental factor through a linear weighted adjustment method; The dynamically adjusted temporary biometric signature code is matched with the physical unclonable signature code, and the signature code matching score is calculated using the cosine similarity method, and the matching threshold is set. , evaluate the feature code matching degree.

7. A method for processing identity data of a smart archive warehouse as claimed in claim 1, characterized in that: The specific steps of real-time tracking of user locations by UWB positioning base stations to generate behavior trajectory data, dynamically adjusting user permissions in combination with the entropy value of residence time, generating dynamically adjusted permission sets, and dynamically updating the status of the smart archive warehouse are as follows: Based on the evaluation results, multiple UWB positioning base stations in the warehouse capture the positioning tag signals worn by users when accessing archives, and calculate the three-dimensional coordinates in real time based on the TDOA algorithm. Kalman filtering is used to eliminate positioning jump points and cubic spline interpolation is used to complete the occlusion data to generate a continuous trajectory. According to the user's continuous trajectory in front of the filing cabinet, the Shannon entropy within the time window is calculated, the abnormal state of user behavior is evaluated, and the permission downgrade is automatically triggered. The permission tag is sent to the electromagnetic lock controller through the MQTT protocol, and the real-time user location and dynamic permission set are synchronized through Redis to update the status of the smart archive warehouse.

8. A smart archive warehouse identity data processing system, based on a smart archive warehouse identity data processing method according to any one of claims 1 to 7, characterized in that: Including data collection module, feature code generation module, file processing module, evaluation module and warehouse authority management module; The data acquisition module is used to collect real-time environmental data, perform preprocessing, and generate an environmental parameter matrix; A feature code generation module is used to collect user biometric features through a multimodal biometric terminal, and perform environmental compensation correction based on an environmental parameter matrix to generate a temporary biometric feature code; The file processing module is used to collect the microscopic surface image of the target file carrier, extract the frequency domain features through fast Fourier transform, and generate the physical non-clonable feature code of the target file carrier; An evaluation module, used to dynamically adjust the temporary biometric signature weight and the physical unclonable signature weight based on real-time environmental data, calculate signature matching scores and perform evaluation; The warehouse authority management module is used to track the user's location in real time through the UWB positioning base station to generate behavior trajectory data, dynamically adjust user permissions based on the entropy value of the residence time, generate a dynamically adjusted permission set, and dynamically update the status of the smart archive warehouse.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a method for processing identity data of a smart archive warehouse as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for processing identity data of a smart archive warehouse as described in any one of claims 1 to 7 are implemented.

Images