A fingerprint confidential file cabinet based on the Internet of Things

Abstract

The present invention belongs to the field of file cabinet management technology, and specifically discloses a fingerprint confidentiality file cabinet based on the Internet of Things, including a confidentiality file cabinet body and a confidentiality file cabinet management system. The confidentiality file cabinet management system is placed on the confidentiality file cabinet body, and is composed of a fingerprint acquisition module, a fingerprint recognition module, a fingerprint record storage library, a fingerprint recognition processing terminal, and a fingerprint recognition execution terminal. When performing fingerprint recognition, the present invention improves the setting accuracy of the fingerprint recognition algorithm by setting the fingerprint image matching evaluation compensation weight, and at the same time, by combining the record fingerprint recognition data table of each historical fingerprint recognition, the fingerprint recognition change judgment and the confirmation of the fingerprint recognition change information are carried out, so as to realize the long-term tracking of the fingerprint recognition process, effectively solve the current fingerprint recognition only focuses on the security level. The deficiency is convenient to improve the accuracy, accuracy and recognition efficiency of subsequent recognition, thereby effectively improving the user experience.

Description

Technical Field

[0001] The invention belongs to the technical field of file cabinet management and relates to a fingerprint confidentiality file cabinet based on the Internet of Things. Background Art

[0002] Fingerprint confidential file cabinets use advanced fingerprint recognition technology to ensure that only authorized personnel can access files. It is a device for protecting file security and is widely used in places that require high confidentiality and access control, such as financial institutions and some commercial enterprises.

[0003] The prior art, such as the Chinese invention patent application with application publication number CN114575704A, discloses a highly secure confidentiality file cabinet, including a confidentiality cabinet body, a supporting chassis, a sealed cabinet door and a connecting rod, wherein the bottom of the confidentiality cabinet body is fixedly connected to the supporting chassis, the front left side of the confidentiality cabinet body is rotatably provided with a connecting rod, and the middle of the connecting rod is fixedly connected to the sealed cabinet door. The finger of the present invention is placed on the electromagnetic lock, and the electromagnetic lock stops limiting the sealed cabinet door after receiving the set fingerprint, and then pulls the force grip plate to reverse, and the locking column moves inward and separates from the confidentiality cabinet body, and people can open the sealed cabinet door through the force grip rod to take out the documents, so that the documents and archives can be stored safely and the confidentiality is high.

[0004] Another example of the prior art is a multifunctional file confidentiality management cabinet disclosed in a Chinese invention patent application with application publication number CN116671733A, which includes a mounting body, a cabinet body, a drawer, a vertical board, a movable vertical partition, a support board and a telescopic device. Only when the fingerprint is adapted can the drawer be pulled outward to expose the confidentiality space, thereby improving the confidentiality effect.

[0005] For the above two technical solutions, the current confidentiality settings for confidential files belong to the settings in the file extraction process, such as the settings of confidential space. There are still the following aspects in the management of fingerprint recognition: 1. Only focusing on the security level, without further setting the fingerprint recognition efficiency on the basis of security, and identifying according to fixed fingerprint recognition modes and algorithms, there is a problem of low recognition efficiency.

[0006] 2. Confidential file cabinets usually involve multiple levels of authority, that is, there are multiple types of weighted file access personnel. Currently, only whether the authority is met is identified, and targeted fingerprint recognition control is not performed based on the personnel's authority level and file access situation, and the strictness of recognition cannot be adjusted. It is not convenient to improve user experience and operational efficiency under the premise of ensuring security.

[0007] 3. Currently, the fingerprint recognition sensor is not updated based on the recorded data of the recognition process. The speed and accuracy of fingerprint image acquisition and processing cannot be guaranteed, and the fingerprint response efficiency cannot be guaranteed, and the fingerprint characteristics of different users cannot be adapted. Summary of the invention

[0008] In view of this, in order to solve the problems raised in the above background technology, a fingerprint confidentiality file cabinet based on the Internet of Things is now proposed.

[0009] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides a fingerprint confidential file cabinet based on the Internet of Things, including a confidential file cabinet body and a confidential file cabinet management system. The confidential file cabinet management system is placed on the confidential file cabinet body and includes: a fingerprint acquisition module, which is used to collect the fingerprint image of the current user when the user performs fingerprint recognition.

[0010] The fingerprint recognition module is used to import the currently set fingerprint pass permission level, and then perform fingerprint recognition on the current user, and output the recognition result, which is one of recognition pass and recognition fail.

[0011] The fingerprint record repository is used to store the fingerprint image, access permission level, historical fingerprint recognition times and fingerprint recognition data table of each registered user, store the setting update frequency of the fingerprint recognition sensor, and store the reference fingerprint recognition times for each access permission level corresponding to a single file access.

[0012] The fingerprint recognition processing terminal is used to make fingerprint recognition change judgments and confirm the fingerprint recognition change information when the judgment result is a change requirement.

[0013] The fingerprint recognition execution terminal is used to open the locking mechanism of the confidential file cabinet and authorize the user to access the file when the recognition result is passed. When the recognition result is not passed, it returns to the fingerprint collection module to perform secondary fingerprint information collection, and at the same time receives fingerprint recognition change information and makes corresponding changes according to the fingerprint recognition change information.

[0014] Compared with the prior art, the present invention has the following beneficial effects: (1) When performing fingerprint recognition, the present invention sets the fingerprint image match evaluation compensation weight to compensate for the recognition deviation when the fingerprint is incomplete, thereby improving the setting accuracy of the fingerprint recognition algorithm and ensuring the reliability and effectiveness of the fingerprint recognition result.

[0015] (2) The present invention combines the fingerprint recognition data table recorded during each historical fingerprint recognition to determine fingerprint recognition changes and confirm fingerprint recognition change information, thereby achieving long-term tracking of the fingerprint recognition process, effectively solving the problem that current fingerprint recognition only focuses on the security level, and facilitating the improvement of the accuracy, precision and efficiency of subsequent recognition, thereby better managing and updating fingerprint information, and greatly reducing the need for users to frequently re-register or re-verify, thereby effectively improving the user experience.

[0016] (3) The present invention determines the fingerprint recognition change by counting the identification setting change needs of each registered user and the overall fingerprint recognition setting change needs, thereby achieving a dual analysis of individual users and the fingerprint recognition sensor as a whole, thereby facilitating the improvement of the authenticity of the fingerprint recognition change judgment results, and also improving the timeliness of the fingerprint recognition setting changes.

[0017] (4) The present invention realizes the update setting of the fingerprint recognition sensor by confirming the change object and, when the change object is the fingerprint recognition sensor, confirming the change setting update frequency of the fingerprint recognition sensor, thereby ensuring not only the accuracy of fingerprint recognition, but also the efficiency of fingerprint recognition, while avoiding the shortcomings of the current fixed fingerprint recognition mode and recognition algorithm, and ensuring the subsequent fingerprint image acquisition and processing speed and processing accuracy, thereby further improving the response efficiency of fingerprint recognition.

[0018] (5) When the object of change is a user, the present invention can adapt to the fingerprint characteristics of different users by confirming the changed user and the change identification threshold of each changed user, fully considering the identification characteristics of users with different authority levels, and realizing targeted control of corresponding fingerprint identification of different users and dynamic regulation of the strictness of corresponding identification of different users, thereby effectively improving user experience and operation efficiency under the premise of ensuring security. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for describing the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 This is a schematic diagram of the structure of the fingerprint security file cabinet of the present invention.

[0021] Figure 2 This is a schematic diagram of the connection of the modules of the confidential file cabinet management system of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] See also Figure 1 As shown, the present invention provides a fingerprint confidentiality file cabinet based on the Internet of Things, including a confidentiality file cabinet body and a confidentiality file cabinet management system.

[0024] The confidentiality file cabinet management system is arranged on the confidentiality file cabinet body, and is composed of a fingerprint collection module, a fingerprint identification module, a fingerprint record storage library, a fingerprint identification processing terminal and a fingerprint identification execution terminal.

[0025] See also Figure 2 As shown, in the above, the fingerprint recognition module is respectively connected to the fingerprint collection module, the fingerprint record storage and the fingerprint recognition execution terminal, and the fingerprint recognition processing terminal is respectively connected to the fingerprint recognition execution terminal and the fingerprint record storage.

[0026] The fingerprint collection module is used to collect the fingerprint image of the current user when the user performs fingerprint recognition.

[0027] The fingerprint recognition module is used to import the currently set fingerprint pass permission level, and then perform fingerprint recognition on the current user, and output a recognition result, wherein the recognition result is one of recognition pass and recognition fail.

[0028] Specifically, fingerprint identification of the current user includes: S1, extracting fingerprint images and access authority levels of each registered user from a fingerprint record repository.

[0029] S2. Record the currently set fingerprint access permission level as the target level, and compare it with the access permission level of each registered user, and select each registered user whose access permission level is equal to or higher than the target level as each analysis user.

[0030] S3. Compare the fingerprint image of the current user with the fingerprint images of each analyzed user. If the fingerprint image of the current user is consistent with the fingerprint image of a certain analyzed user, the recognition is passed as the recognition result.

[0031] S4. If the fingerprint image of the current user is inconsistent with the fingerprint images of each analyzed user, the degree of consistency between the corresponding fingerprint images of the current user and each analyzed user is calculated.

[0032] S5. If the matching degree between the current user and the corresponding fingerprint image of a certain analyzed user is greater than or equal to the set fingerprint image matching degree, The recognition result is recognition passed.

[0033] S6. If the matching degree between the current user and the fingerprint images of each analyzed user is less than the set matching degree, the recognition failure is regarded as the recognition result.

[0034] Furthermore, in step S4, the degree of consistency between the fingerprint image of the current user and the fingerprint image of each analyzed user is counted, including: S41, the fingerprint image of the current user and the fingerprint image of each analyzed user are respectively identified by image recognition technology, and the identification feature data is output, and the fingerprint feature data set A of the current user and the fingerprint feature data set B of each analyzed user are constructed accordingly. j , j represents the analysis user number, j=1,2,......m.

[0035] It should be added that image recognition technology is a relatively mature technology, and its specific recognition process will not be described here. The recognition feature data includes but is not limited to the number of feature points, the location of each feature point, the number of stripe lines, the direction of each stripe line and the horizontal spacing between each stripe line.

[0036] S42, extracting the number of feature points from the identification feature data, and setting the fingerprint image matching evaluation compensation weight accordingly, denoted as

[0037] S43: Count the matching degree between the current user and the fingerprint images of each analyzed user ∪ represents the intersection operator, and ∩ represents the union operator.

[0038] Furthermore, step S42 sets the fingerprint image matching evaluation compensation weight, including: step 1, recording the number of feature points of the current user as M a , and set the feature point number threshold M y for comparison.

[0039] Step 2: If M a ≥M y , 0 is used as the compensation weight for fingerprint image matching evaluation.

[0040] Step 3: If M a <M y ,Will As the compensation weight for fingerprint image matching evaluation, it is recorded as β, where e is a natural constant and k y To set the reference feature point deviation ratio, so as to obtain the compensation weight of fingerprint image matching evaluation The value is 0 or β, β>0.

[0041] In a specific embodiment, the fingerprint is composed of a large number of feature points, including termination points and bifurcation points, where the termination point is the end of the fingerprint line, and the bifurcation point is the place where the fingerprint line separates. In fingerprint identification, it is generally necessary to identify a certain number of feature points for effective comparison. If the number of identifiable feature points in a fingerprint sample is lower than a certain threshold, it can be considered incomplete. At this time, there is obviously a large error in the calculation based on the extracted feature data through the similarity algorithm. Therefore, the fingerprint image match evaluation compensation weight is set according to the number of feature points to improve the authenticity and accuracy of the fingerprint image match analysis results.

[0042] In another specific embodiment, generally, the number of identifiable feature points in a fingerprint is generally between 50 and 100. These feature points are usually called Caltron points. In actual applications, the specific threshold may be adjusted according to different standards and practices. For example, when the actual application is fingerprint unlocking, at least 8 to 12 clear and repeatable feature points are found for basic comparison and confirmation. Specifically, 12 can be selected as the threshold of the number of feature points, that is, M y =12.

[0043] It should be added that the fewer the number of feature points, the more incomplete the fingerprint is. In this case, the greater the error in the fingerprint image matching calculation result, that is, the greater the corresponding fingerprint image matching evaluation compensation weight. y The specific value can be 2, that is, when M a Tend to M y When the weight is half of the value, the weight is the middle value. a Below M y half of , the weight tends to be maximum, that is, tends to be 1.

[0044] It should also be added that the setting formula of the compensation weight β of the fingerprint image match evaluation is specifically changed with reference to the Sigmoid function. Its purpose is to map the value of the weight to between 0 and 1 to quantify the specific compensation degree and ensure the effectiveness and consistency of the weight, thereby realizing the correction of the statistical results of the fingerprint image match.

[0045] When performing fingerprint recognition, the embodiment of the present invention sets the fingerprint image match evaluation compensation weight to compensate for the recognition deviation when the fingerprint is incomplete, thereby improving the setting accuracy of the fingerprint recognition algorithm and ensuring the reliability and effectiveness of the fingerprint recognition result.

[0046] The fingerprint record repository is used to store the fingerprint image, access authority level, historical fingerprint recognition times and fingerprint recognition data table of each registered user, store the setting update frequency of the fingerprint recognition sensor, and store the reference fingerprint recognition times for each access authority level corresponding to a single file access.

[0047] The fingerprint recognition processing terminal is used to make a fingerprint recognition change judgment, and when the judgment result is a change requirement, confirm the fingerprint recognition change information.

[0048] Specifically, the fingerprint recognition change judgment includes: H1, extracting the cumulative fingerprint recognition time and recognition date from the record fingerprint recognition data table, and based on this, counting the recognition setting change demand of each registered user, recorded as BG i , i represents the registered user number, i=1,2,......n.

[0049] H2. Extract the number of fingerprint identification times from the record fingerprint identification data table, and count the overall setting change demand of fingerprint identification, which is recorded as ZG.

[0050] H3. The existence of a registered user's identification setting change requirement greater than 0 is used as judgment condition 1, and the overall setting change requirement of the fingerprint indicator is greater than 0 is used as judgment condition 2.

[0051] H4. If judgment condition 1 or judgment condition 2 is met, the required change is used as the judgment result of the fingerprint recognition change judgment. If judgment condition 1 and judgment condition 2 are both not met, no change is needed as the judgment result of the fingerprint recognition change judgment.

[0052] The embodiment of the present invention realizes long-term tracking of the fingerprint recognition process by combining the fingerprint recognition data table recorded during each historical fingerprint recognition to judge the fingerprint recognition change and confirm the fingerprint recognition change information, effectively solving the deficiency that the current fingerprint recognition only focuses on the security level, and is convenient for improving the accuracy, precision and recognition efficiency of subsequent recognition, thereby better managing and updating fingerprint information, and greatly reducing the need for users to frequently re-register or verify, thereby effectively improving the user experience.

[0053] Furthermore, in step H1, the need for changing the identification settings of each registered user is counted, including: H11, selecting the identification date of the first fingerprint identification in history and the identification date of the last fingerprint identification in history from the identification dates of each historical fingerprint identification, and recording them as the starting identification date and the ending identification date respectively.

[0054] H12. Compare the start identification date and the end identification date to obtain the number of days between the start identification date and the end identification date of each registered user as the identification cycle days, and record it as Ti , and record the number of historical fingerprint recognitions of each registered user as N i .

[0055] H13. Compare the identification dates of each fingerprint indicator to obtain the interval days between each fingerprint identification. Calculate the average of the interval days between each historical fingerprint identification and the cumulative fingerprint identification time of each registered user to obtain the average identification interval days D of each registered user. i And the average cumulative fingerprint recognition time T i .

[0056] H14. Count the fingerprint recognition frequency of each registered userξ i , D i ′ and k i ′ respectively represent the reference identification interval days and reference identification frequency of the i-th registered user.

[0057] It should be added that represents the recognition frequency of the i-th registered user, D i ′ and k i The specific value of ′ is extracted from the fingerprint recognition evaluation technical table. For example, according to the access permission level of each registered user, the frequently defined identification interval days and the defined identification frequency of fingerprint identification of each registered user corresponding to the corresponding access permission level are extracted from the fingerprint recognition evaluation technical table as the reference identification interval days and reference identification frequency of each registered user, wherein the identification interval days is an analysis element for identification frequency analysis based on the local time dimension, and the identification frequency is an analysis element for identification frequency analysis based on the overall time dimension. The combination of the local and the overall increases the detailed performance of the identification frequency analysis, thereby avoiding the one-sidedness of only evaluating the identification frequency from the overall perspective.

[0058] H15. The fingerprint recognition frequency of each registered user is matched and compared with the corresponding reference fingerprint recognition frequency of each suitable fingerprint recognition time, and the matching cumulative fingerprint recognition time of each registered user is obtained, which is recorded as T. i ', and then count the identification setting change needs of each registered user BG i ,

[0059] It should be added that the reference fingerprint recognition frequencies corresponding to the appropriate fingerprint recognition durations are obtained by comprehensive settings based on experience and a fingerprint recognition evaluation technology table.

[0060] Furthermore, in step H2, the overall setting change requirement of fingerprint identification is counted, including: H21, counting the number of registered users whose identification setting change requirement is greater than 0 as the number of identification change users, and recording the ratio of the number of identification change users to the number of registered users as the identification change ratio, recorded as k g .

[0061] H22. Extract the reference fingerprint recognition times for a single file access corresponding to the access permission level of each registered user from the fingerprint record storage repository as the appropriate fingerprint recognition times for each registered user.

[0062] H23, subtract the number of fingerprint recognition times of each registered user from the number of suitable fingerprint recognition times, and record the difference as the recognition times difference, and determine the recognition deviation ratio based on this, recorded as k p .

[0063] Understandably, confirming the recognition deviation ratio includes: X1, counting the historical fingerprint recognition times whose recognition times difference is greater than 0, as the deviation fingerprint recognition times, and recording the ratio of the deviation fingerprint recognition times to the historical fingerprint recognition times as the deviation times ratio.

[0064] X2. Using the historical recognition order as the horizontal coordinate and the deviation fingerprint recognition times as the vertical coordinate, construct a deviation index recognition times curve for each registered user, extract the slope of the curve, and set the deviation correction factor η for each registered user based on the slope of the curve. i .

[0065] It should be added that the deviation correction factor for each registered user is set, including: if the slope of the deviation indicator identification frequency curve of a registered user is less than or equal to 0, 0 is used as the deviation correction factor of the registered user.

[0066] If the slope of the deviation index identification times curve of a registered user is greater than 0, the slope of the deviation identification curve corresponding to the registered user is recorded as k v ,Will As the deviation correction factor of the registered user, and recorded as θ, k′ v To set the reference deviation change rate, set the deviation correction factor η for each registered user i , η i The value is 0 or θ.

[0067] In a specific embodiment, k′ v The specific value can be 0.45.

[0068] X3, record the deviation ratio of each registered user as (k c ) i , calculate the correction deviation ratio of each registered user (kc ′) i , (k c ′) i =(k c ) i *(1+η i ).

[0069] X4. Compare the correction deviation frequency ratio with the set reference correction deviation frequency ratio, count the number of registered users whose correction deviation frequency ratio is greater than the set reference correction deviation frequency ratio, and compare it with the number of registered users, and use the ratio as the identification deviation ratio.

[0070] It should be added that the specific value of setting the reference correction deviation ratio can be 0.3. H24, the overall setting change requirement ZG of statistical fingerprint recognition, r 0 and r 1 are the weights corresponding to the recognition change ratio and recognition deviation ratio, respectively, k′ g and k′ p Set the reference recognition change ratio and recognition deviation ratio respectively.

[0071] Understandably, Indicates the overall need for setting changes at the identified change level, Indicates the overall need for setting changes at the level of identification deviations, Indicates that when k g ≤k′ g When k g >k′ g At this time, the overall setting change requirement value of the identification change level is Indicates that when k p ≤k′ p When k p >k′ p At this time, the overall setting change requirement value of the identification change level is

[0072] It should be added that the statistical formula for the overall setting change requirement is based on the linear regression function, and the statistical result is a non-negative number. The overall setting change requirement at the setting identification change level and the overall setting change requirement at the identification deviation level are non-negative values ​​to prevent the interference of the inconsistent value signs of the two on the overall setting change requirement value result. For example: when the overall setting change requirement at the identification change level is -10 and the overall setting change requirement at the identification deviation level is 5, the overall value is less than 0, which ignores the impact of the identification deviation level on the overall setting change demand.

[0073] It should be added that, in a specific embodiment, the recognition change ratio reflects the proportion of registered users whose fingerprint recognition threshold needs to be changed, and the recognition deviation ratio reflects the proportion of registered users whose fingerprint recognition accuracy deviation is set. The recognition change ratio parameter indicator is set to ensure the overall optimization of the effectiveness of the subsequent fingerprint recognition threshold setting. The recognition deviation ratio is set to be an overall optimization based on the current actual recognition accuracy deviation level, that is, the recognition change ratio indirectly affects the size of the overall setting change requirement, and the recognition deviation ratio directly affects the size of the overall setting change requirement. Therefore, the weight corresponding to the recognition deviation ratio is set to be greater than the weight corresponding to the recognition change ratio. For example, r 0 and r 1 The values ​​are 0.45 and 0.55 respectively.

[0074] In another specific embodiment, k′ g and k′ p The specific value of is mainly set according to the weights corresponding to the recognition change ratio and the recognition deviation ratio. For example, k′ g The value can be 0.5, k′ p The value can be 0.3.

[0075] The embodiment of the present invention judges fingerprint recognition changes by counting the recognition setting change needs of each registered user and the overall setting change needs of fingerprint recognition, thereby achieving dual analysis of individual users and the fingerprint recognition sensor as a whole, thereby facilitating the improvement of the authenticity of the fingerprint recognition change judgment results, and also improving the timeliness of the fingerprint recognition setting changes.

[0076] In another specific embodiment, confirming the fingerprint recognition change information includes: W1. If judgment condition 1 is met, the user is taken as the change object; if judgment condition 2 is met, the fingerprint recognition sensor is taken as the change object.

[0077] W2. If the change object is user identification, confirm each changed user and the change identification threshold of each changed user, and use the fingerprint identification change information under the change object being user identification.

[0078] W3. If the changed object is the fingerprint recognition sensor, extract the setting update frequency of the fingerprint recognition sensor from the fingerprint record repository, denoted as p 0 .

[0079] W4. Match and compare the overall setting change requirement of fingerprint recognition with the setting change requirement corresponding to each increase and update ratio, and obtain the matching increase and update ratio, which is recorded as k. ψ , and then p 0 *(1+k ψ ) as the change setting update frequency pω , p ω The object of change is the fingerprint recognition change information of the fingerprint recognition sensor.

[0080] It should be added that the setting change requirement of each set increase update ratio corresponding to the reference is obtained by extracting from the update technology table corresponding to the setting of the fingerprint recognition sensor.

[0081] Furthermore, in step W2, confirming each changed user and the change identification threshold of each changed user includes: W21, identifying each registered user whose setting change requirement is greater than 0 as each changed user.

[0082] W22, record the identification setting change requirement of each changed user as BG d , d represents the changed user number, d=1,2,......u.

[0083] W23, will As a benchmark identification indicator, calculate the change identification threshold of each changed user To set the matching degree of the reference security identification fingerprint image corresponding to the d-th changed user, R is a set constant.

[0084] The embodiment of the present invention confirms the change object and, when the change object is a fingerprint recognition sensor, confirms the change setting update frequency of the fingerprint recognition sensor, thereby realizing the update setting of the fingerprint recognition sensor, which not only ensures the accuracy of fingerprint recognition, but also ensures the efficiency of fingerprint recognition, while avoiding the shortcomings of the current fixed fingerprint recognition mode and recognition algorithm, and can ensure the subsequent fingerprint image acquisition and processing speed and processing accuracy, thereby further improving the response efficiency of fingerprint recognition.

[0085] In a specific embodiment, R is a constant greater than 1, and for example, it can be 2. The consistency of the reference security identification fingerprint image corresponding to each changed user is The specific value of is set according to the access permission level of each changed user. For example, according to the access permission level of each changed user, the fingerprint security trigger recognition similarity of each changed user corresponding to the access permission level is extracted from the recognition technology table of the fingerprint recognition sensor, and then used as the reference security recognition fingerprint image matching degree.

[0086] The embodiment of the present invention can adapt to the fingerprint characteristics of different users by confirming the changed user and the change identification threshold of each changed user when the changed object is a user, fully considers the identification characteristics of users with different authority levels, and realizes targeted control of corresponding fingerprint identification of different users and dynamic regulation of the strictness of corresponding identification of different users, thereby effectively improving user experience and operational efficiency under the premise of ensuring security.

[0087] The fingerprint recognition execution terminal is used to open the locking mechanism of the confidential file cabinet and authorize the user to access the file when the recognition result is passed. When the recognition result is not passed, it returns to the fingerprint collection module to perform secondary fingerprint information collection, and at the same time receives fingerprint recognition change information and makes corresponding changes according to the fingerprint recognition change information.

[0088] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A fingerprint confidentiality file cabinet based on the Internet of Things, comprising a confidentiality file cabinet body, characterized in that: It also includes a confidentiality file cabinet management system, which is placed on the confidentiality file cabinet body and includes: Fingerprint collection module, used to collect the fingerprint image of the current user; Fingerprint recognition module, used to import the currently set fingerprint to pass the permission level, and then perform fingerprint recognition on the current user and output the recognition result; Perform fingerprint recognition on the current user, including: extracting fingerprint images and access rights levels of each registered user from a fingerprint record repository; Recording each registered user whose access authority level is equal to or higher than the target level as each analysis user; The fingerprint image of the current user is compared with the fingerprint images of each analyzed user. If the fingerprint image of the user is consistent with the fingerprint image of a certain analyzed user, the recognition is passed as the recognition result. Otherwise, the consistency between the current user and the corresponding fingerprint images of each analyzed user is counted. ; If the matching degree between the current user and the corresponding fingerprint image of a certain analyzed user is greater than or equal to the set fingerprint image matching degree , the recognition result is regarded as passed, otherwise the recognition result is regarded as failed; Said The analysis method is: The fingerprint image of the current user and the fingerprint images of each analyzed user are identified through image recognition technology, and the identification feature data is output, based on which the fingerprint feature data set of the current user is constructed. And the fingerprint feature dataset of each analyzed user , Indicates the analysis user number, ; Extract the number of feature points from the identification feature data and set the fingerprint image matching evaluation compensation weight accordingly ; Count the matching degree between the current user and the corresponding fingerprint images of each analyzed user , , Represents the intersection operator symbol, Represents the union operator symbol; Said The analysis method is: The number of feature points of the current user is recorded as , and set the feature point number threshold Make a comparison; like , 0 is used as the compensation weight for fingerprint image matching evaluation; like ,Will As the compensation weight for fingerprint image matching evaluation, it is recorded as ,in, is a natural constant, To set the reference feature point deviation ratio, so as to obtain the compensation weight of fingerprint image matching evaluation , The value is 0 or , ; Fingerprint record repository, used to store fingerprint images, access rights level, historical fingerprint recognition times and fingerprint recognition data tables of each registered user, store the setting update frequency of the fingerprint recognition sensor, and store the reference fingerprint recognition times for each access rights level corresponding to a single file access; The fingerprint recognition processing terminal is used to judge the fingerprint recognition change and confirm the fingerprint recognition change information when the judgment result is a change requirement; The fingerprint recognition execution terminal is used to open the locking mechanism of the confidential file cabinet and authorize the user to access the file when the recognition result is passed. When the recognition result is not passed, it returns to the fingerprint collection module to perform secondary fingerprint information collection, and at the same time receives fingerprint recognition change information and makes corresponding changes according to the fingerprint recognition change information.

2. The fingerprint confidentiality file cabinet based on the Internet of Things as claimed in claim 1, characterized in that: The fingerprint identification change judgment comprises: The accumulated fingerprint recognition time and recognition date are extracted from the record fingerprint recognition data table, and the recognition setting change demand of each registered user is counted based on the accumulated fingerprint recognition time and recognition date, which is recorded as , Indicates the registered user ID. ; Extract the number of fingerprint identification times from the record fingerprint identification data table, and count the overall setting change demand of fingerprint identification, which is recorded as ; The existence of a registered user's identification setting change requirement greater than 0 is used as judgment condition 1, and the overall setting change requirement of the fingerprint indicator is greater than 0 is used as judgment condition 2; If judgment condition 1 or judgment condition 2 is met, the required change is used as the judgment result of the fingerprint recognition change judgment. If judgment condition 1 and judgment condition 2 are both not met, no change is needed as the judgment result of the fingerprint recognition change judgment.

3. The fingerprint security file cabinet based on the Internet of Things as claimed in claim 2, characterized in that: The counting of the identification setting change needs of each registered user includes: The identification date of the first fingerprint identification in history and the identification date of the last fingerprint identification in history are selected from the identification dates of each fingerprint identification in history, and are recorded as the starting identification date and the ending identification date respectively; Compare the start identification date and the end identification date to obtain the number of days between the start identification date and the end identification date of each registered user as the identification cycle days, and record it as , and record the historical fingerprint recognition times of each registered user as ; Compare the identification dates of each fingerprint indicator to get the number of days between each fingerprint identification. Calculate the average of the number of days between each fingerprint identification and the cumulative fingerprint identification time for each registered user to get the average number of days between each registered user. And the average cumulative fingerprint recognition time ; Statistics on the frequency of fingerprint recognition for each registered user , , and Respectively expressed as The number of days between reference identification and the frequency of reference identification for each registered user; The fingerprint recognition frequency of each registered user is matched and compared with the corresponding reference fingerprint recognition frequency of each suitable fingerprint recognition time, and the matching cumulative fingerprint recognition time of each registered user is obtained, which is recorded as , and then count the recognition setting change needs of each registered user , .

4. The fingerprint confidentiality file cabinet based on the Internet of Things as claimed in claim 2, characterized in that: The overall setting change requirement of the statistical fingerprint recognition includes: The number of registered users whose identification setting change demand is greater than 0 is counted as the number of identified changed users, and the ratio of the number of identified changed users to the number of registered users is recorded as the identification change ratio, recorded as ; Extracting from the fingerprint record storage library the reference fingerprint identification number of single file access corresponding to the access permission level of each registered user as the appropriate fingerprint identification number of each registered user; The fingerprint recognition times of each registered user in each fingerprint recognition history are subtracted from the appropriate fingerprint recognition times, and the difference is recorded as the recognition times difference. Based on this, the recognition deviation ratio is determined and recorded as ; Statistical fingerprint recognition overall configuration change needs , , and Set the weights corresponding to the recognition change ratio and recognition deviation ratio respectively. and Set the reference recognition change ratio and recognition deviation ratio respectively.

5. The fingerprint security file cabinet based on the Internet of Things as claimed in claim 4, characterized in that: The confirmation recognition deviation ratio includes: Count the historical fingerprint recognition times whose recognition times difference is greater than 0 as the deviation fingerprint recognition times, and record the ratio of the deviation fingerprint recognition times to the historical fingerprint recognition times as the deviation times ratio; With the historical recognition order as the horizontal axis and the deviation fingerprint recognition times as the vertical axis, a deviation index recognition times curve of each registered user is constructed, the slope of the curve is extracted, and based on the slope of the curve, the deviation correction factor of each registered user is set. ; The deviation ratio of each registered user is recorded as , calculate the correction deviation ratio of each registered user , ; The correction deviation frequency ratio is compared with the set reference correction deviation frequency ratio, the number of registered users whose correction deviation frequency ratio is greater than the set reference correction deviation frequency ratio is counted, and compared with the number of registered users, and the ratio is used as the identification deviation ratio.

6. The fingerprint security file cabinet based on the Internet of Things as claimed in claim 2, characterized in that: The confirmation of fingerprint identification change information includes: If judgment condition 1 is met, the user is taken as the change object, and if judgment condition 2 is met, the fingerprint recognition sensor is taken as the change object; If the object of change is user identification, confirm each changed user and the change identification threshold of each changed user, and use the fingerprint identification change information under the change object of user identification as the change; If the changed object is the fingerprint sensor, extract the setting update frequency of the fingerprint sensor from the fingerprint record repository and record it as ; The overall setting change requirement of fingerprint recognition is matched and compared with the setting change requirement of each set increase and update ratio corresponding to the reference, and the matched increase and update ratio is obtained, which is recorded as , and then As a change to set the update frequency ,Will The object of change is the fingerprint recognition change information of the fingerprint recognition sensor.

7. The fingerprint security file cabinet based on the Internet of Things as claimed in claim 6, characterized in that: The confirming of each changed user and the change identification threshold of each changed user includes: Each registered user whose setting change requirement is greater than 0 is identified as each change user; The identification setting change requirement of each changing user is recorded as , Indicates changing the user ID. ; Will As a benchmark identification indicator, calculate the change identification threshold of each changed user , , To set the Change the user's corresponding reference security identification fingerprint image matching degree, To set a constant.

Images