Biometric data management system and method based on multi-source stereoscopic perception

Abstract

The application belongs to the technical field of data management, and in particular relates to a biological feature data management system and method based on multi-source stereoscopic perception, wherein the system comprises a multi-source data acquisition module, a data preprocessing and fusion module, a biological feature database, a data security management module and a background supervision end; the biological feature data is collected in high precision and multi-dimension by the multi-source data acquisition module, the multi-source biological feature data is preprocessed and effectively integrated by the data preprocessing and fusion module, a unified biological feature vector is formed and sent to the biological feature database for storage, the data security management module ensures the security and privacy protection of the biological feature data in the storage, transmission and processing process, and the biological feature database is analyzed for risks by the data security management module, management auxiliary analysis is performed when a low-risk data signal is generated, the storage security of the biological feature data is ensured, and the workload and supervision difficulty of the administrator are significantly reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data management, in particular to a biometric data management system and method based on multi-source stereoscopic perception. BACKGROUND

[0002] Biometric data, as a unique identifier inherent to each individual, carries our identity, privacy and rights. With the rapid development of information technology, biometric recognition technology has become a key technology in the fields of identity verification, security monitoring, etc. Traditional biometric recognition mostly relies on a single data source, such as fingerprint recognition, facial recognition, etc., which is vulnerable to environmental interference and spoofing attacks, resulting in limited recognition accuracy and security.

[0003] Currently, combining multiple biometric data for identity recognition and verification is used to ensure recognition accuracy and security. However, the processing and management of multiple biometric data face the problems of large data volume and high processing complexity, and the data security of the biometric database and the device risk of the involved storage devices cannot be reasonably analyzed and timely warned, which is not conducive to the safe preservation of biometric data.

[0004] In view of the above technical defects, a solution is proposed. SUMMARY

[0005] The purpose of the present application is to provide a biometric data management system and method based on multi-source stereoscopic perception, which solves the problem that the prior art cannot reasonably analyze and timely warn the data security of the biometric database and the device risk of the involved storage devices, which is not conducive to the safe preservation of biometric data, and the low level of intelligence.

[0006] To achieve the above purpose, the present application provides the following technical scheme:

[0007] The biometric data management system based on multi-source stereoscopic perception comprises a multi-source data acquisition module, a data preprocessing fusion module, a biometric database, a data security management module and a background supervision end. The multi-source data acquisition module integrates multiple biometric acquisition devices and introduces stereoscopic perception technology to collect biometric data with high precision and multi-dimensionality. The collected multi-source biometric data is sent to the data preprocessing fusion module.

[0008] The data preprocessing fusion module preprocesses the collected multi-source biometric data, effectively integrates biometric data from different sources using a fusion algorithm, forms a unified biometric vector, and sends the fused biometric vector to the biometric database for storage.

[0009] The data security management module adopts encryption technology, access control mechanism and data desensitization strategy to perform security management on the biometric feature vector, and performs risk analysis on the biometric feature database to generate a database high-risk signal or a database low-risk signal, and sends the database high-risk signal or the database low-risk signal to the background supervision end. When the background supervision end receives the database high-risk signal, it issues a corresponding early warning.

[0010] Further, the biometric feature acquisition equipment integrated by the multi-source data acquisition module includes a fingerprint scanner, an iris identifier, a high-definition camera and a voiceprint acquisition device.

[0011] Further, the biometric feature database is in communication connection with the intelligent verification and identification module. When identity authentication or identity recognition is needed, the intelligent verification and identification module obtains the biometric feature data to be verified, compares it with the biometric feature vectors stored in the biometric feature database in advance, calculates the similarity between the biometric features, and judges whether the matching is successful. If the similarity reaches a preset threshold or meets a specific rule, it is determined that the matching is successful, and the identity authentication or identity recognition is successful.

[0012] Further, the specific analysis process of the data security management module includes:

[0013] The biometric feature database is monitored in real time to identify illegal intrusion and network attack, and an early warning is issued when illegal intrusion or network attack is identified. A detection period is set, the amount of data lost or leaked in the biometric feature database due to illegal intrusion and network attack in the detection period is obtained and marked as a leakage detection value, the leakage detection value is compared with a preset leakage detection threshold, and if the leakage detection value exceeds the preset leakage detection threshold, a database high-risk signal is generated.

[0014] Further, if the leakage detection value does not exceed the preset leakage detection threshold, the time is counted when illegal intrusion or network attack is identified until the illegal intrusion or network attack is ended, and the intrusion attack duration is obtained accordingly.

[0015] A detection period is set, the sum of all intrusion attack durations of the biometric feature database in the detection period is calculated to obtain an intrusion attack total time value, and the intrusion attack duration is compared with a preset intrusion attack duration threshold. If the intrusion attack duration exceeds the preset intrusion attack duration threshold, the corresponding intrusion attack duration is marked as a risk duration, and the number of risk durations in the detection period is marked as a risk frequency value.

[0016] The database risk coefficient is obtained by numerical calculation of the total time value of the intrusion attack, the risk frequency value and the leakage detection value. The database risk coefficient is compared with a preset database risk coefficient threshold value. If the database risk coefficient exceeds the preset database risk coefficient threshold value, a database high-risk signal is generated. If the database risk coefficient does not exceed the preset database risk coefficient threshold value, a database low-risk signal is generated.

[0017] Further, the data security management module is in communication connection with the management auxiliary analysis module. The management auxiliary analysis module monitors all storage devices involved in the biological feature database. The data security management module sends the database low-risk signal to the management auxiliary analysis module. The management auxiliary analysis module performs management auxiliary analysis when receiving the database low-risk signal. The management auxiliary analysis module generates an auxiliary analysis qualified signal or an auxiliary analysis abnormal signal through analysis and sends the auxiliary analysis qualified signal or the auxiliary analysis abnormal signal to the background supervision end. The background supervision end issues a corresponding early warning when receiving the auxiliary analysis abnormal signal.

[0018] Further, the specific analysis process of the management auxiliary analysis includes:

[0019] The analysis is performed to determine whether there is a service timeout device. If there is a service timeout device, an auxiliary analysis abnormal signal is generated. If there is no service timeout device, the average storage speed when data is stored in the corresponding storage device and the average calling speed when data is called from the corresponding storage device during the detection period are collected and marked as storage efficiency characteristic values and calling efficiency characteristic values, respectively.

[0020] The scanning risk value is obtained by ratio calculation of the number of times of scanning for the presence of a vulnerability virus in the corresponding storage device and the total number of scans during the detection period. The management alarm coefficient is obtained by numerical calculation of the storage efficiency characteristic value, the calling efficiency characteristic value and the scanning risk value. The management alarm coefficient is compared with a preset management alarm coefficient threshold value. If the management alarm coefficient exceeds the preset management alarm coefficient threshold value, the corresponding storage device is marked as an alarm device. If there is an alarm device, an auxiliary analysis abnormal signal is generated. If there is no alarm device, an auxiliary analysis qualified signal is generated.

[0021] Further, the specific analysis process of determining whether there is a service timeout device through analysis is as follows:

[0022] The vibration amplitude of the corresponding storage device and the dust concentration of the environment in which the corresponding storage device is located are obtained and marked as vibration condition values and dust condition values. The deviation of the humidity of the environment in which the corresponding storage device is located from the preset suitable humidity value is marked as a humidity condition value. The deviation of the temperature of the corresponding storage device from the preset suitable temperature value is marked as a temperature condition value.

[0023] The life influence value is calculated by weighted sum of the vibration condition value, the ash condition value, the wet condition value and the temperature condition value, the life influence value is compared with the preset life influence threshold value, if the life influence value exceeds the preset life influence threshold value, it is judged that the corresponding storage device is in the life vulnerable state, the total time length of the corresponding storage device in the life vulnerable state in the historical stage is obtained and marked as the life loss time value;

[0024] The production date of the corresponding storage device is obtained, the interval time length between the current date and the production date is marked as the service time table value, the life loss time value and the service time table value are compared with the preset life loss time threshold value and the preset service time table threshold value respectively, if the life loss time value or the service time table value exceeds the corresponding preset threshold value, the corresponding storage device is marked as the service overtime device.

[0025] Further, the intelligent identification verification module is in communication connection with the application stability analysis module, the application stability analysis module generates an application stability qualified signal or an application stability abnormal signal by analysis, and sends the application stability qualified signal or the application stability abnormal signal to the background supervision end, when the background supervision end receives the application stability abnormal signal, the corresponding early warning is sent out; the specific analysis process of the application stability analysis module is as follows:

[0026] The number of times of system collapse and the duration of each time within the detection period that cannot perform identity verification identification are obtained and marked as the application pause value and the application stop time value respectively, and the sum of all application stop time values within the detection period is calculated to obtain the application collapse time value;

[0027] The application pause value and the application collapse time value are compared with the preset application pause threshold value and the preset application collapse time threshold value respectively, if the application pause value or the application collapse time value exceeds the corresponding preset threshold value, the application stability abnormal signal is generated;

[0028] If the application pause value and the application collapse time value do not exceed the corresponding preset threshold value, the average processing time length of the intelligent identification verification module for identity recognition verification within the detection period is obtained and marked as the identification efficiency value, and the proportion of the number of times that the processing time length of the intelligent identification verification module for identity recognition verification within the detection period exceeds the preset processing time length threshold value is marked as the identification inefficiency value;

[0029] The application stability coefficient is obtained by numerically calculating the application pause value, the application collapse time value, the identification efficiency value and the identification inefficiency value, the application stability coefficient is compared with the preset application stability coefficient threshold value, if the application stability coefficient exceeds the preset application stability coefficient threshold value, the application stability abnormal signal is generated; if the application stability coefficient does not exceed the preset application stability coefficient threshold value, the application stability qualified signal is generated.

[0030] Further, the application also proposes a multi-source stereoscopic perception-based biometric data management method, comprising the following steps:

[0031] Step one, through the integration of multiple biometric feature acquisition devices and the introduction of stereoscopic perception technology, high-precision, multi-dimensional biometric feature data acquisition is carried out;

[0032] Step two, pre-process the collected multi-source biometric feature data;

[0033] Step three, use fusion algorithm to effectively integrate biometric feature data from different sources to form a unified biometric feature vector, and store the fused biometric feature vector through the biometric feature database;

[0034] Step four, use encryption technology, access control mechanism and data desensitization strategy to manage the security of the biometric feature vector;

[0035] Step five, risk analysis is carried out on the biometric feature database, and the background supervision end is warned when the database high-risk signal is generated.

[0036] Compared with the prior art, the application has the following advantages:

[0037] 1. In the application, multi-source biometric feature data is collected, pre-processed and effectively integrated through the multi-source data acquisition module and the data preprocessing fusion module, the biometric feature database stores the biometric feature vector, the data security management module ensures the security and privacy protection of the biometric feature data during storage, transmission and processing, and the data security management module carries out risk analysis on the biometric feature database, and carries out management auxiliary analysis when the data low-risk signal is generated, which is beneficial to ensure the storage security of the biometric feature data, significantly reduces the workload and supervision difficulty of the administrator;

[0038] 2. In the application, the intelligent identification verification module is used for identity verification and identification, which solves the problems of limited recognition accuracy, vulnerability to counterfeit attacks and security risks caused by single data in the identity recognition process, and the application stability analysis module is used to analyze the identity verification and identification performance to generate application stability qualified signal or application stability abnormal signal, which reminds the administrator to investigate the cause and strengthen the subsequent application supervision when the application stability abnormal signal is generated, ensures the subsequent use efficiency and application stability, further reduces the workload and management difficulty of the administrator, and has high intelligent level. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to facilitate the understanding of those skilled in the art, the application will be further described below with reference to the accompanying drawings;

[0040] Figure 1 The system block diagram of example one and example two in the application;

[0041] Figure 2 a system block diagram of example three in the present application;

[0042] Figure 3 a method flow chart of example four in the present application. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present application.

[0044] Example one: as shown in the figure, the biological feature data management system based on multi-source stereoscopic perception provided by the present application includes a multi-source data acquisition module, a data preprocessing fusion module, a biological feature database, a data security management module and a background supervision end. Figure 1

[0045] The multi-source data acquisition module integrates various biological feature acquisition devices (including a fingerprint scanner, an iris recognizer, a high-definition camera and a voiceprint acquisition device, etc.), and introduces stereoscopic perception technology (such as structured light, a ToF depth camera, etc.) to collect biological feature data with high precision and multi-dimension, and sends the collected multi-source biological feature data to the data preprocessing fusion module.

[0046] The data preprocessing fusion module preprocesses the collected multi-source biological feature data, including steps of denoising, enhancing, feature extraction, etc., effectively integrates biological feature data of different sources by using a fusion algorithm, forms a unified biological feature vector, and sends the fused biological feature vector to the biological feature database for storage, so as to be compared and verified with biological feature data collected on site in the future.

[0047] The data security management module uses encryption technology, an access control mechanism and a data desensitization strategy to manage the biological feature vector, ensures the security and privacy protection of the biological feature data in the storage, transmission and processing process, and analyzes the biological feature database to generate a database high-risk signal or a database low-risk signal, sends the database high-risk signal or the database low-risk signal to the background supervision end, and the background supervision end sends a corresponding early warning when receiving the database high-risk signal, to remind the administrator to strengthen the safety supervision of the biological feature database in the future, and ensure the data security of the biological feature database. The specific analysis process of the data security management module is as follows:

[0048] ​Real-time monitoring is performed on the biometric database to identify illegal intrusion and cyber attacks, and a warning is issued when illegal intrusion or cyber attacks are identified, so as to realize effective monitoring and threat identification and warning of the biometric database. A detection period is set, the amount of data lost or leaked in the biometric database due to illegal intrusion and cyber attacks in the detection period is obtained and marked as a leakage detection value, the leakage detection value is compared with a preset leakage detection threshold value, and if the leakage detection value exceeds the preset leakage detection threshold value, it indicates that the data protection performance in the detection period is poor, and a database high-risk signal is generated.

[0049] Further, if the leakage detection value does not exceed the preset leakage detection threshold value, timing is performed when illegal intrusion or cyber attacks are identified until the corresponding illegal intrusion or cyber attacks end, and accordingly an intrusion attack duration is obtained.

[0050] A detection period is set, preferably, the detection period is twenty-five days; all intrusion attack durations of the biometric database in the detection period are summed to obtain an intrusion attack total time value, and the intrusion attack duration is compared with a preset intrusion attack duration threshold value, and if the intrusion attack duration exceeds the preset intrusion attack duration threshold value, it indicates that the threat brought by the corresponding illegal intrusion or cyber attack is larger, and the corresponding intrusion attack duration is marked as a risk duration, and the number of risk durations in the detection period is marked as a risk frequency value.

[0051] The intrusion attack total time value LP, the risk frequency value NF and the leakage detection value TS are calculated by the formula The database risk coefficient GX is calculated by the intrusion attack total time value LP, the risk frequency value NF and the leakage detection value TS; wherein wq, hu and my are preset weight coefficients with values greater than zero, and the larger the value of the database risk coefficient GX, the greater the comprehensive security threat of the biometric database in the detection period.

[0052] The database risk coefficient GX is compared with a preset database risk coefficient threshold value, and if the database risk coefficient GX exceeds the preset database risk coefficient threshold value, it indicates that the comprehensive security threat of the biometric database in the detection period is larger, and a database high-risk signal is generated; if the database risk coefficient GX does not exceed the preset database risk coefficient threshold value, it indicates that the comprehensive security threat of the biometric database in the detection period is smaller, and a database low-risk signal is generated.

[0053] Embodiment two: as Figure 2As shown, the difference between this embodiment and Embodiment 1 is that the data security management module communicates with the management auxiliary analysis module, the management auxiliary analysis module monitors all storage devices involved in the biometric database, the data security management module sends the low-risk signal of the database to the hidden danger auxiliary analysis module, and the management auxiliary analysis module performs management auxiliary analysis when it receives the low-risk signal of the database, and generates an auxiliary analysis qualified signal or an auxiliary analysis abnormal signal through analysis.

[0054] Furthermore, the system sends either a pass / fail signal or an abnormal signal from the auxiliary analysis to the backend monitoring terminal. Upon receiving an abnormal signal, the backend monitoring terminal issues a corresponding warning to remind the administrator to promptly inspect and repair the relevant storage device. If necessary, it can back up the data on the storage device and then replace the device, further ensuring the security of biometric data storage and significantly reducing the workload and monitoring difficulty for administrators. The specific analysis process for managing auxiliary analysis is as follows:

[0055] The vibration amplitude of the corresponding storage device and the dust concentration of the environment are obtained and marked as vibration condition value and dust condition value. The deviation of the humidity of the environment where the corresponding storage device is located from the preset suitable humidity value is marked as humidity condition value, and the deviation of the temperature of the corresponding storage device from the preset suitable temperature value is marked as temperature condition value.

[0056] Through formula The lifetime impact value YL is calculated by weighting and summing the vibration value SW, gray value TL, humidity value WF, and temperature value ZS. Among them, c1, c2, c3, and c4 are preset weight coefficients with values ​​greater than zero. The larger the value of the lifetime impact value YL, the greater the overall real-time operation risk of the corresponding storage device and the greater the damage to the lifetime of the corresponding storage device.

[0057] The lifespan impact value YL of the corresponding storage device is compared with the preset lifespan impact threshold. If the lifespan impact value YL exceeds the preset lifespan impact threshold, it indicates that the real-time operation risk of the corresponding storage device is relatively large and the damage to the lifespan of the corresponding storage device is relatively large. Then, the corresponding storage device is judged to be in a lifespan vulnerable state. The total duration of the corresponding storage device in the lifespan vulnerable state in the historical period is obtained and marked as the lifespan damage time value.

[0058] The production date of the corresponding storage device is obtained, and the interval between the current date and the production date is marked as the service life value. The life loss value and the service life value are compared with the preset life loss threshold and the preset service life threshold respectively. If the life loss value or the service life value exceeds the corresponding preset threshold, it indicates that the corresponding storage device has reached its service life and should be retired in time. Then the corresponding storage device is marked as a service time-out device.

[0059] If there is a device that has exceeded its service life, it indicates that there is a security risk in the storage of biometric data, and an auxiliary analysis abnormal signal is generated; if there is no device that has exceeded its service life, the average storage speed when storing data to the corresponding storage device and the average retrieval speed when retrieving data from the corresponding storage device during the detection period are collected and marked as the storage effectiveness characteristic value and the tuning characteristic value, respectively.

[0060] In addition, all scanning information for vulnerability and virus scanning of storage devices during the detection period is obtained (scanning of each storage device is carried out regularly or irregularly), and the scanning risk value is calculated by the ratio of the number of times vulnerability and virus were found in the corresponding storage device (i.e. the number of times vulnerability and virus were found in the corresponding storage device during the detection period) to the total number of scans during the detection period.

[0061] Through formula The management alarm coefficient XW is obtained by numerically calculating the storage performance characteristic value ZW, the tuning characteristic value HP, and the scanning risk value FX. Among them, up, rg, and eq are preset weight coefficients with values ​​greater than zero. Furthermore, the larger the value of the management alarm coefficient XW, the worse the overall performance of the corresponding storage device during the detection period.

[0062] The management alarm coefficient XW is compared with the preset management alarm coefficient threshold. If the management alarm coefficient XW exceeds the preset management alarm coefficient threshold, it indicates that the overall performance of the corresponding storage device is poor during the detection period, and the corresponding storage device is marked as an alarm device. If an alarm device exists, it indicates that it is not conducive to the safe and efficient operation of the biometric database, and an auxiliary analysis abnormal signal is generated. If no alarm device exists, it indicates that it is conducive to the safe and efficient operation of the biometric database, and an auxiliary analysis qualified signal is generated.

[0063] Example 3: Figure 2 As shown, the difference between this embodiment and Embodiments 1 and 2 is that the biometric database is connected to the intelligent verification and identification module. When identity authentication or identification is required, the intelligent identification and verification module obtains the biometric data to be verified and compares it with the biometric vectors stored in the biometric database. By calculating the similarity between the biometrics, it determines whether the match is successful. If the similarity reaches a preset threshold or meets specific rules, it is determined that the match is successful, that is, the identity authentication or identification is successful. This solves the security risks caused by limited recognition accuracy, susceptibility to forgery attacks, and single data in the identity identification process.

[0064] It should be noted that the application scenarios include airports, financial institutions, etc. For example, the application is applied in the passenger security system of an international airport. The multi-source stereoscopic perception technology is used to collect the biological feature data of passengers, such as fingerprints, irises, and facial features. The data is compared with the data in the biological feature database to realize fast and accurate identity recognition and security process optimization, while ensuring the safety of passenger personal information.

[0065] Or, the application is integrated into the online payment platform of a certain financial institution. The multi-source biological feature data is used for user identity verification to effectively prevent fraud and improve payment security. At the same time, through data encryption and access control, the safety storage and access of user biological feature data are ensured.

[0066] Further, the intelligent identification verification module is communicatively connected to the application stability analysis module. The application stability analysis module generates an application stability qualified signal or an application stability abnormal signal by analysis, and sends the application stability qualified signal or the application stability abnormal signal to the background supervision end. When the background supervision end receives the application stability abnormal signal, it issues a corresponding early warning. The application stability analysis module can reasonably analyze the performance of the identity recognition application and timely warn to remind the administrator to investigate the cause and strengthen subsequent application supervision, ensuring the subsequent use efficiency and application stability, further reducing the workload and management difficulty of the administrator, and having high intelligent level. The specific analysis process of the application stability analysis module is as follows:

[0067] The number of times of system crashes that cannot perform identity verification identification within the detection period and the duration of each time are obtained and marked as application suspension value and application stop time value, respectively. The sum of all application stop time values within the detection period is calculated to obtain the application collapse time value.

[0068] The application suspension value and the application collapse time value are compared with the preset application suspension threshold and the preset application collapse time threshold, respectively. If the application suspension value or the application collapse time value exceeds the corresponding preset threshold, it indicates that the identity recognition verification application performance within the detection period is poor, and an application stability abnormal signal is generated.

[0069] If the application suspension value and the application collapse time value do not exceed the corresponding preset threshold, the average processing time of the intelligent identification verification module for identity recognition verification within the detection period is obtained and marked as the verification efficiency value. The proportion of the number of times that the processing time of the intelligent identification verification module for identity recognition verification within the detection period exceeds the preset processing time threshold is marked as the verification inefficiency value.

[0070] The formula is The application stability coefficient HX is obtained by numerically calculating the application pause value GY, application crash value FL, verification efficiency value QM, and verification inefficiency value PN. Among them, kp, eu, sq, and ng are preset weight coefficients with values ​​greater than zero. Furthermore, the larger the value of the application stability coefficient HX, the worse the overall performance of the identity verification application during the detection period.

[0071] The application stability coefficient HX is compared with the preset application stability coefficient threshold. If the application stability coefficient HX exceeds the preset application stability coefficient threshold, it indicates that the overall performance of the identity recognition and verification application is poor during the detection period, and an application stability abnormal signal is generated. If the application stability coefficient HX does not exceed the preset application stability coefficient threshold, it indicates that the overall performance of the identity recognition and verification application is good during the detection period, and an application stability qualified signal is generated.

[0072] Example 4: Figure 3 As shown, the difference between this embodiment and Embodiments 1, 2, and 3 is that the biometric data management method based on multi-source stereo perception proposed in this invention includes the following steps:

[0073] Step 1: By integrating multiple biometric acquisition devices and introducing stereoscopic sensing technology, biometric data is acquired with high precision and in multiple dimensions.

[0074] Step 2: Preprocess the collected multi-source biometric data;

[0075] Step 3: Use a fusion algorithm to effectively integrate biometric data from different sources to form a unified biometric vector, and store the fused biometric vector in a biometric database.

[0076] Step 4: Implement secure management of biometric vectors using encryption technology, access control mechanisms, and data anonymization strategies;

[0077] Step 5: Perform risk analysis on the biometric database and issue an early warning to the backend monitoring terminal when a high-risk signal is generated in the database.

[0078] The working principle of the present application is as follows: in use, a plurality of biological feature acquisition devices are integrated by the multi-source data acquisition module, and the stereoscopic perception technology is introduced to collect biological feature data with high precision and in multiple dimensions; the data preprocessing and fusion module pre-processes and effectively integrates the collected multi-source biological feature data, forms a unified biological feature vector and sends it to the biological feature database for storage; the data security management module uses encryption technology, access control mechanism and data desensitization strategy to safely manage the biological feature vector, ensures the security and privacy protection of the biological feature data in the storage, transmission and processing process, and analyzes the risk of the biological feature database through the data security management module, strengthens the safety supervision of the biological feature database in the subsequent process when a high-risk signal of the database is generated, guarantees the data security of the biological feature database, and through the management auxiliary analysis module for management auxiliary analysis when a low-risk signal of the data is generated, checks and maintains the corresponding storage device and selectively eliminates it when an auxiliary analysis abnormal signal is generated, further guarantees the storage security of the biological feature data, and significantly reduces the workload and supervision difficulty of the administrator.

[0079] The above formulas are dimensionless values calculated, the formulas are obtained by collecting a large amount of data to simulate a formula of the nearest real situation, and the preset parameters in the formula are set by a person skilled in the art according to the actual situation. The preferred embodiments of the present application disclosed above are only used to help explain the present application. The preferred embodiments do not describe all the details and limit the present application to the specific implementation. Obviously, according to the content of the present application, many modifications and changes can be made. The present application selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present application, so that those skilled in the art can well understand and utilize the present application. The present application is limited by the claims and the entire scope and equivalents thereof.

Claims

1. A multi-source stereoscopic perception based biometric data management system, characterized in that, The system comprises a multi-source data acquisition module, a data preprocessing and fusion module, a biological feature database, a data security management module and a background supervision terminal; the multi-source data acquisition module integrates various biological feature acquisition devices and introduces stereoscopic perception technology to collect biological feature data with high precision and in multiple dimensions, and sends the collected multi-source biological feature data to the data preprocessing and fusion module; The data preprocessing and fusion module preprocesses the collected multi-source biological feature data, effectively integrates biological feature data from different sources using a fusion algorithm, forms a unified biological feature vector, and sends the fused biological feature vector to the biological feature database for storage; The data security management module uses encryption technology, access control mechanisms and data desensitization strategies to manage the biological feature vector securely, and analyzes the biological feature database to generate a database high-risk signal or a database low-risk signal, sends the database high-risk signal or the database low-risk signal to the background supervision terminal, and the background supervision terminal sends an appropriate warning when receiving the database high-risk signal; The specific analysis process of the data security management module includes: Real-time monitoring of the biological feature database to identify illegal intrusion and network attacks, and sending a warning when illegal intrusion or network attacks are identified; setting a detection period, obtaining the amount of data lost or leaked in the biological feature database due to illegal intrusion and network attacks during the detection period and marking it as a leakage detection value, comparing the leakage detection value with a preset leakage detection threshold value, and if the leakage detection value exceeds the preset leakage detection threshold value, generating a database high-risk signal; If the leakage detection value does not exceed the preset leakage detection threshold value, timing is performed when illegal intrusion or network attacks are identified until the illegal intrusion or network attacks are ended, and the intrusion attack duration is obtained accordingly; Setting a detection period, summing all intrusion attack durations of the biological feature database during the detection period to obtain an intrusion attack total time value, and comparing the intrusion attack duration with a preset intrusion attack duration threshold value, if the intrusion attack duration exceeds the preset intrusion attack duration threshold value, marking the corresponding intrusion attack duration as a risk duration, and marking the number of risk durations during the detection period as a risk frequency value; Calculating the database risk coefficient by calculating the intrusion attack total time value, the risk frequency value and the leakage detection value, comparing the database risk coefficient with a preset database risk coefficient threshold value, if the database risk coefficient exceeds the preset database risk coefficient threshold value, generating a database high-risk signal; if the database risk coefficient does not exceed the preset database risk coefficient threshold value, generating a database low-risk signal; The data security management module is in communication connection with the management auxiliary analysis module, the management auxiliary analysis module monitors all storage devices involved in the biological feature database, the data security management module sends a database low-risk signal to the management auxiliary analysis module, the management auxiliary analysis module performs management auxiliary analysis when receiving the database low-risk signal, generates an auxiliary analysis qualified signal or an auxiliary analysis abnormal signal through analysis, and sends the auxiliary analysis qualified signal or the auxiliary analysis abnormal signal to the background supervision end, and the background supervision end sends a corresponding early warning when receiving the auxiliary analysis abnormal signal; The specific analysis process of the management auxiliary analysis includes: Through analysis, it is judged whether there is a service overtime device, if there is a service overtime device, an auxiliary analysis abnormal signal is generated, if there is no service overtime device, the average storage speed when data is stored to the corresponding storage device and the average calling speed when data is called to the corresponding storage device in the detection period are collected and marked as storage efficiency characteristic value and calling efficiency characteristic value respectively; And all scanning information for vulnerability virus scanning of the storage device in the detection period is obtained, the scanning times of the corresponding storage device with vulnerability virus and the total scanning times of the detection period are compared to obtain a scanning risk value; the management alarm coefficient is obtained by numerical calculation of the storage efficiency characteristic value, the calling efficiency characteristic value and the scanning risk value, and the management alarm coefficient is compared with the preset management alarm coefficient threshold value, if the management alarm coefficient exceeds the preset management alarm coefficient threshold value, the corresponding storage device is marked as an alarm device; if there is an alarm device, an auxiliary analysis abnormal signal is generated; if there is no alarm device, an auxiliary analysis qualified signal is generated; The specific analysis process of judging whether there is a service overtime device through analysis is as follows: The vibration amplitude of the corresponding storage device and the dust concentration of the environment are obtained and marked as vibration condition value and dust condition value, and the deviation of the humidity of the environment of the corresponding storage device from the preset suitable humidity value is marked as humidity condition value, and the deviation of the temperature of the corresponding storage device from the preset suitable temperature value is marked as temperature condition value; The life influence value is obtained by weighted summation calculation of the vibration condition value, the dust condition value, the humidity condition value and the temperature condition value, and the life influence value is compared with the preset life influence threshold value, if the life influence value exceeds the preset life influence threshold value, it is judged that the corresponding storage device is in a life vulnerable state, the total length of time when the corresponding storage device is in the life vulnerable state in the historical stage is obtained and marked as life loss time value; The production date of the corresponding storage device is obtained, the interval length of time between the current date and the production date is marked as service time table value; the life loss time value and the service time table value are compared with the preset life loss time threshold value and the preset service time table threshold value respectively, if the life loss time value or the service time table value exceeds the corresponding preset threshold value, the corresponding storage device is marked as a service overtime device; The intelligent identification and verification module is connected with the biometric database in communication, and when identity authentication or identity recognition is needed, the intelligent identification and verification module obtains the biometric data to be verified, and compares the biometric data to be verified with the biometric vectors stored in the biometric database in advance. If the similarity reaches a preset threshold or meets a specific rule, it is determined that the matching is successful. The application stability analysis module is connected with the intelligent identification and verification module in communication. The application stability analysis module generates an application stability qualified signal or an application stability abnormal signal by analysis, and sends the application stability qualified signal or the application stability abnormal signal to the background supervision end. When the background supervision end receives the application stability abnormal signal, it issues a corresponding early warning. The specific analysis process of the application stability analysis module is as follows: The number of times of system crashes that cannot perform identity authentication and recognition in the detection period and the duration of each time are obtained and marked as application pause value and application stop time value respectively, and all application stop time values in the detection period are summed to obtain application collapse time value. The application pause value and the application collapse time value are compared with the preset application pause threshold and the preset application collapse time threshold respectively. If the application pause value or the application collapse time value exceeds the corresponding preset threshold, an application stability abnormal signal is generated. If the application pause value and the application collapse time value do not exceed the corresponding preset threshold, the average processing time of the intelligent identification and verification module for identity recognition and verification in the detection period is obtained and marked as identification and verification efficiency value, and the proportion of the number of times that the processing time of the intelligent identification and verification module for identity recognition and verification in the detection period exceeds the preset processing time threshold is marked as identification and verification inefficiency value. The application pause value, the application collapse time value, the identification and verification efficiency value, and the identification and verification inefficiency value are numerically calculated to obtain an application stability coefficient. The application stability coefficient is compared with the preset application stability coefficient threshold. If the application stability coefficient exceeds the preset application stability coefficient threshold, an application stability abnormal signal is generated. If the application stability coefficient does not exceed the preset application stability coefficient threshold, an application stability qualified signal is generated.

2. The multi-source stereoscopic perception based biometric data management system of claim 1, wherein, The biometric feature acquisition equipment integrated by the multi-source data acquisition module includes a fingerprint scanner, an iris recognizer, a high-definition camera, and a voiceprint acquisition device.

3. The method for managing biometric data based on multi-source stereoscopic perception, characterized in that, The method adopts the multi-source stereoscopic perception-based biometric data management system according to any one of claims 1-2. The method adopts the multi-source stereoscopic perception-based biometric data management system according to any one of claims 1-2.

Images