A door access warning system based on biometric information

Abstract

The present invention relates to a door access warning system based on biometric information, and relates to the field of system security technology. The system comprises: a processor and a memory storing a computer program. When the computer program is executed by the processor, the following steps are implemented: obtaining a candidate recognition time corresponding to a target door access device, determining a final operation type by using a first neural network model when the candidate recognition time meets a first preset condition, determining a final operation type by using a second neural network model when the candidate recognition time meets a second preset condition, obtaining a target recognition time list corresponding to a target user ID list, and judging whether the target door access device issues an early warning. The present invention confirms a method for door access identity recognition based on characteristic information of a door access installation area, confirms an image processing method in a door access device by using a model generated by training based on sample data, and simultaneously compares relevant data of the model with relevant data of the door access device to realize abnormal situation monitoring and early warning, thereby improving the security of the door access system.

Description

Technical Field

[0001] The present invention relates to the technical field of system security, and in particular to an access control early warning system based on biometric information. Background Art

[0002] With the continuous advancement of science and technology and the increasing demand for social security, the demand for access control is increasing and the management method of access control is becoming more and more important. Traditional access control systems such as keys and passwords are at risk of being easily copied and lost, which poses a great hidden danger to security management. Biometric information recognition technology has gradually been widely used in the field of access control management with its unique advantages. Biometric information recognition technology collects individual biometric information, such as fingerprints, facial features, irises, etc., and compares and verifies it to achieve the purpose of identity recognition. It can not only verify identity, but also warn of abnormal situations. Therefore, it is of great significance to develop an access control warning system based on biometric information.

[0003] The method for access control warning based on biometric information in the prior art is to capture the fingerprint image through a fingerprint identifier and compare it with the fingerprint template in the database, or to capture the iris image through an iris recognition device and match it with the iris template in the database, and issue a warning when the match is unsuccessful. The above is limited to using one identity recognition method, and does not confirm the access control identity recognition method based on the characteristic information of the access control installation area, and does not use a model to confirm the image processing method in the access control device. It is not possible to detect and warn of abnormal situations, which reduces the security of the access control system. Summary of the invention

[0004] In view of the above technical problems, the technical solution adopted by the present invention is: a biometric information-based access control warning system, the system comprising: a processor and a memory storing a computer program, when the computer program is executed by the processor, the following steps are implemented:

[0005] S100, obtaining a target access control device and a candidate identification time duration corresponding to the target access control device, wherein the candidate identification time duration is an upper limit time duration for identifying and matching a person determined based on a security level of a target geographical area where the target access control device is installed.

[0006] S200, when the candidate recognition time meets the first preset condition, determining the final operation type corresponding to the target access control device using a first neural network model, wherein the first neural network model is a neural network model obtained by model training using data associated with fingerprint recognition based on fingerprint images, wherein S200 also includes the following steps:

[0007] S201, obtaining a candidate priority corresponding to a target access control device, wherein the candidate priority is a lower limit value of the accuracy rate of the target access control device in identifying a user identity, which is determined based on a security level of a target geographical area where the target access control device is installed.

[0008] S202: Input the candidate priorities into the first neural network model to obtain a final operation type, wherein the final operation type is the first target operation type obtained by using the first neural network model.

[0009] S300, when the candidate recognition time meets the second preset condition, determining the final operation type corresponding to the target access control device using a second neural network model, wherein the second neural network model is a neural network model obtained by model training using data associated with iris recognition based on iris images, wherein S300 also includes the following steps:

[0010] S301, obtaining the candidate priority corresponding to the target access control device.

[0011] S302: Input the candidate priority into the second neural network model to obtain a final operation type, wherein the final operation type is a second target operation type obtained by using the second neural network model.

[0012] S400, based on the final operation type, obtain a target recognition duration list T={T1, ..., T e , ..., T f}, T e is the target recognition duration corresponding to the e-th target user ID, e=1…f, and f is the number of target user IDs in the target user ID list.

[0013] S500, when there is |T e -T 0 |≥λ, the target access control device issues an early warning, where T 0 is the preset time threshold, and λ is the preset time difference threshold.

[0014] Compared with the prior art, the present invention has obvious beneficial effects. By means of the above technical solution, the access control warning system based on biometric information provided by the present invention can achieve considerable technical advancement and practicality, and has wide industrial utilization value, and has at least the following beneficial effects:

[0015] The present invention is a biometric information-based access control early warning system, comprising: a processor and a memory storing a computer program, when the computer program is executed by the processor, the following steps are implemented: obtaining a target access control device and a candidate identification time corresponding to the target access control device, when the candidate identification time meets a first preset condition, using a first neural network model to determine a final operation type corresponding to the target access control device, wherein the candidate priority corresponding to the target access control device is obtained, the candidate priority is input into the first neural network model, and the final operation type is obtained, when the candidate identification time meets a second preset condition, using a second neural network model to determine the final operation type corresponding to the target access control device, wherein the target access control device is obtained. The candidate priority level corresponding to the access control device is input into the second neural network model to obtain the final operation type. Based on the final operation type, a target recognition time list corresponding to the target user ID list is obtained. The target recognition time is compared with a preset time threshold to determine whether the target access control device issues an early warning. The present invention is not limited to one identity recognition method. A method for confirming access control identity recognition based on characteristic information of an access control installation area is used. A model is trained based on sample data. A method for confirming image processing in an access control device is confirmed using the generated model. At the same time, relevant data of the model is compared with relevant data of the access control device to realize abnormal situation monitoring and early warning, thereby improving the security of the access control system.

[0016] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the following specifically cites a preferred embodiment and describes it in detail with the accompanying drawings as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of an execution computer program of a biometric information-based access control warning system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0019] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0020] Example

[0021] This embodiment provides a biometric information-based access control warning system, the system comprising: a processor and a memory storing a computer program, when the computer program is executed by the processor, the following steps are implemented, such as: Figure 1 As shown:

[0022] S100, obtaining a target access control device and a candidate recognition duration corresponding to the target access control device.

[0023] Specifically, the target access control device is a device installed in a target geographical area.

[0024] Furthermore, the target geographical area is an area where access control of personnel is required, such as an office or other target geographical area.

[0025] Specifically, the candidate identification duration is an upper limit time duration for identifying and matching a person, which is determined based on the security level of the target geographical area where the target access control device is installed.

[0026] Furthermore, those skilled in the art know that the confirmation of the security level and the selection of the upper limit time duration can be carried out according to actual needs, which all fall within the scope of protection of the present invention and will not be elaborated here. It can be understood that: when the security level of the target geographical area is higher, the control on the entry and exit of personnel is stricter. In order to ensure more accurate identification of the target access control device, the upper limit value of the time for identifying personnel will be larger.

[0027] S200: When the candidate recognition time meets the first preset condition, the final operation type corresponding to the target access control device is determined using the first neural network model.

[0028] Specifically, the first preset condition is that the duration is within the time range of fingerprint recognition, wherein those skilled in the art know that any method for determining the time range of fingerprint recognition in the prior art falls within the protection scope of the present invention and will not be repeated here. It can be understood that: the time range of fingerprint recognition is confirmed by the time experienced when fingerprint recognition is performed through the existing fingerprint recognition technology.

[0029] Specifically, the first neural network model is a neural network model obtained by model training using data associated with fingerprint recognition based on fingerprint images.

[0030] Specifically, in S200, the first neural network model is obtained through the following steps:

[0031] S11, obtain the first target operation type list D={D1, ..., D i , ..., D n} and D corresponding to the first sample image list set A = {A1, ..., A i , ..., A n}, where A i ={A i1 , ..., A ir , ..., A is（i）}, D i is the i-th first target operation type, A ir D i The corresponding first sample image list A i The rth first sample image in D, r = 1...s(i), s(i) is i The corresponding first sample image list A i The number of the first sample images in , i=1…n, where n is the number of the first target operation type.

[0032] Specifically, the first target operation type is the type of operation performed on the fingerprint image during the fingerprint recognition process. Those skilled in the art know that the fingerprint recognition method can be selected according to actual needs, and all of them fall within the protection scope of the present invention. No further description is given here. It can be understood that: there are many ways to perform fingerprint recognition based on fingerprint images in the prior art, and one of the fingerprint recognition methods is regarded as a first target operation type. For example: the first target operation type is to refine the acquired fingerprint image so that the texture of the fingerprint image is enhanced, extract key points in the image to obtain a key point image, etc.;

[0033] Specifically, the first sample image is a fingerprint image of a fingerprint identified during a fingerprint identification process.

[0034] S12, obtain the first target duration list set TA corresponding to A = {TA1, ..., TA i,……,TA n} and the first target identifier list set PA corresponding to A = {PA1, ..., PA i ,……,PA n}, where TA i ={TA i1 ,……,TA ir ,……,TA is（i）}, P.A. i ={PA i1 ,……,PA ir ,……,PA is（i）},TA ir A ir The corresponding first target duration, PA ir A ir The corresponding first target identifier, and PA ir The value is "1" or "0".

[0035] Specifically, the first target duration is the duration of fingerprint matching when fingerprint recognition is performed using the first target operation type, wherein the fingerprint matching duration is the time from the time the fingerprint image is captured and input into the fingerprint recognition system to the time the system completes the matching of the fingerprint image with the fingerprint stored in the database and outputs the matching result.

[0036] Specifically, the first target identifier is an identifier corresponding to a matching result obtained when fingerprint recognition is performed on the first sample image using the first target operation type.

[0037] Furthermore, the first target identifier includes "0" and "1", wherein when the first target identifier is "0", it indicates that the fingerprint identity identified when the fingerprint recognition is performed on the first sample image using the first target operation type is incorrect, and when the first target is represented as "1", it indicates that the fingerprint identity identified when the fingerprint recognition is performed on the first sample image using the first target operation type is correct.

[0038] Furthermore, those skilled in the art know that any method for determining whether a fingerprint identification is correct in the existing design falls within the protection scope of the present invention and will not be described in detail herein.

[0039] S13, according to PA, obtain the first target accuracy list FD corresponding to D = {FD1, ..., FD i , ..., FD n}, FD i D i The corresponding first target accuracy, where FD i Meet the following conditions:

[0040] FD i =α i / s(i),α i For PA i The number of first targets marked as "1".

[0041] S14, training the first initial neural network model based on D, A, TA, PA and FD to obtain a first neural network model.

[0042] Specifically, those skilled in the art know that the initial neural network model can be selected according to actual needs, which falls within the protection scope of the present invention and will not be described in detail here.

[0043] Specifically, S200 also includes the following steps:

[0044] S201, obtaining a candidate priority corresponding to a target access control device, wherein the candidate priority is a lower limit value of the accuracy rate of the target access control device in identifying a user identity, which is determined based on a security level of a target geographical area where the target access control device is installed.

[0045] S202: Input the candidate priorities into the first neural network model to obtain a final operation type, wherein the final operation type is the first target operation type obtained by using the first neural network model.

[0046] S300: When the candidate recognition time meets the second preset condition, a second neural network model is used to determine a final operation type corresponding to the target access control device.

[0047] Specifically, the second preset condition is that the duration is within the duration range of iris recognition, wherein those skilled in the art know that any method for determining the duration of iris recognition in the prior art falls within the protection scope of the present invention and will not be repeated here. It can be understood that the duration range of fingerprint recognition is confirmed by the time taken for fingerprint recognition using the existing fingerprint recognition technology.

[0048] Specifically, the second neural network model is a neural network model obtained by model training using data associated with iris recognition based on iris images.

[0049] Specifically, in S300, the second neural network model is obtained through the following steps:

[0050] S21, obtain a second target operation type list H={H1, ..., H j , ..., H m The second sample image list set B corresponding to H = {B1, ..., B j , ..., B m}, where B j ={B j1 , ..., B jv , ..., Bjq（j）}, H j is the jth second target operation type, B jv H j The corresponding second sample image list B j The vth second sample image in H, v = 1...q(j), q(j) is j The corresponding second sample image list B j The number of the second sample images in , j=1…m, m is the number of the second target operation types.

[0051] Specifically, the second target operation type is the type of operation performed on the iris image during the iris recognition process. Those skilled in the art know that the iris recognition method can be selected according to actual needs, all of which fall within the protection scope of the present invention and will not be described in detail herein. It can be understood that: there are many ways to perform iris recognition based on iris images in the prior art, and one of the iris recognition methods is regarded as a second target operation type, for example: a second target operation type such as a method for performing iris recognition based on an iris segmentation method based on projection constraints and randomly sampled Hough transform.

[0052] Specifically, the second sample image is an iris image used to identify an iris during an iris recognition process.

[0053] S22, obtain the second target duration list set TB corresponding to B = {TB1, ..., TB j , ..., TB m} and the second target identifier list set PB corresponding to B = {PB1, ..., PB j , ..., PB m}, where TB j ={TB j1 , ..., TB jv , ..., TB jq（j）}, PB j ={PB j1 , ..., PB jv , ..., PB jq（j）}, TB jv For B jv The corresponding second target duration, PB jv For B jv The corresponding second target identifier, and PB jv The value is "1" or "0".

[0054] Specifically, the second target duration is the duration of iris matching when iris recognition is performed using the second target operation type, wherein the duration of iris matching is the time from when the iris image is captured and input into the iris recognition system to when the system completes the matching of the iris image with the irises stored in the database and outputs the matching result.

[0055] Specifically, the second target identifier is an identifier corresponding to a matching result obtained when iris recognition is performed on the second sample image using the second target operation type.

[0056] Furthermore, the second target identifier includes "0" and "1", wherein when the second target identifier is "0", it indicates that the iris identity identified when iris recognition is performed on the second sample image using the second target operation type is incorrect, and when the second target is represented as "1", it indicates that the iris identity identified when iris recognition is performed on the second sample image using the second target operation type is correct.

[0057] Furthermore, those skilled in the art know that any method for determining whether iris recognition is correct in existing designs falls within the protection scope of the present invention and will not be described in detail herein.

[0058] S23, according to PB, obtain the second target accuracy list FH corresponding to H = {FH1, ..., FH j , ..., FH m}, FH j H j The corresponding second target accuracy, where FH j Meet the following conditions:

[0059] FH j =β j / q(j),β j PB j The number of second target identifiers "1" in .

[0060] S24, training the second initial neural network model based on H, B, TB, PB and FH to obtain a second neural network model.

[0061] Specifically, those skilled in the art know that the initial neural network model can be selected according to actual needs, which falls within the protection scope of the present invention and will not be described in detail here.

[0062] Specifically, S300 also includes the following steps:

[0063] S301, obtaining the candidate priority corresponding to the target access control device.

[0064] S302: Input the candidate priority into the second neural network model to obtain a final operation type, wherein the final operation type is a second target operation type obtained by using the second neural network model.

[0065] S400, based on the final operation type, obtain a target recognition duration list T={T1, ..., T e , ..., T f}, T e is the target recognition duration corresponding to the e-th target user ID, e=1…f, and f is the number of target user IDs in the target user ID list.

[0066] Specifically, the target user ID is a unique identifier that characterizes the identity of the target user. Those skilled in the art know that any method for obtaining a user identity in the prior art falls within the protection scope of the present invention and will not be described in detail herein.

[0067] Furthermore, the target user is a person who is about to enter the target geographical area, for example, an office employee or other target user.

[0068] Specifically, the target identification duration is the duration of identity matching when the target access control device identifies the target user based on the final operation type.

[0069] Furthermore, the duration of the identity matching is the time required from the time when the target access control device captures the image corresponding to the target user to the time when the target access control device outputs the result.

[0070] S500, when there is |T e -T 0 |≥λ, the target access control device issues an early warning, where T 0 is the preset time threshold, and λ is the preset time difference threshold.

[0071] Specifically, in S500, T is obtained by the following steps: 0 :

[0072] S501, when the final operation type is the same as D i When the corresponding first target operation type is consistent, obtain T 0 , where T 0 Meet the following conditions:

[0073] .

[0074] S502, when the final operation type is the same as H j When the corresponding second target operation type is consistent, obtain T 0 , where T 0 Meet the following conditions:

[0075] .

[0076] Specifically, the value range of λ is 3 seconds to 5 seconds. Those skilled in the art know that λ can be selected according to actual needs, which falls within the protection scope of the present invention and will not be described in detail here.

[0077] The present embodiment provides a biometric information-based access control warning system, including: a processor and a memory storing a computer program. When the computer program is executed by the processor, the following steps are implemented: obtaining a target access control device and a candidate recognition time corresponding to the target access control device, when the candidate recognition time meets a first preset condition, using a first neural network model to determine a final operation type corresponding to the target access control device, wherein the candidate priority corresponding to the target access control device is obtained, the candidate priority is input into the first neural network model, and the final operation type is obtained, when the candidate recognition time meets a second preset condition, using a second neural network model to determine the final operation type corresponding to the target access control device, wherein the target access control device is obtained. The candidate priority corresponding to the access control device is input into the second neural network model to obtain the final operation type. Based on the final operation type, the target recognition time list corresponding to the target user ID list is obtained, and the target recognition time is compared with the preset time threshold to determine whether the target access control device is warned. The present invention is not limited to one identity recognition method, a method for confirming access control identity recognition based on characteristic information of the access control installation area, training a model based on sample data, and confirming an image processing method in the access control device using the generated model. At the same time, the relevant data of the model is compared with the relevant data of the access control device to realize abnormal situation monitoring and early warning, thereby improving the security of the access control system.

[0078] Although some specific embodiments of the present invention have been described in detail by way of example, it will be appreciated by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It will also be appreciated by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. A biometric access control warning system, characterized in that: The system comprises: a processor and a memory storing a computer program, and when the computer program is executed by the processor, the following steps are implemented: S100, obtaining a target access control device and a candidate identification duration corresponding to the target access control device, wherein the candidate identification duration is an upper limit time duration for identifying and matching a person determined based on a security level of a target geographical area where the target access control device is installed; S200, when the candidate recognition time meets the first preset condition, determine the final operation type corresponding to the target access control device using the first neural network model, the first neural network model is a neural network model obtained by model training using data associated with fingerprint recognition based on fingerprint images, the first preset condition is that the time is within the time range of fingerprint recognition, wherein S200 also includes the following steps: S201, obtaining a candidate priority corresponding to a target access control device, wherein the candidate priority is a lower limit value of the accuracy rate of the target access control device in identifying a user's identity, which is determined based on the security level of the target geographical area where the target access control device is installed; S202, inputting the candidate priority into the first neural network model to obtain a final operation type, wherein the final operation type is the first target operation type obtained by using the first neural network model; In S200, the first neural network model is obtained through the following steps: S11, obtain the first target operation type list D={D1, ..., D i , ..., D n } and D corresponding to the first sample image list set A = {A1, ..., A i , ..., A n }, where A i ={A i1 , ..., A ir , ..., A is（i） }, D i is the i-th first target operation type, A ir D i The corresponding first sample image list A i The rth first sample image in D, r = 1...s(i), s(i) is i The corresponding first sample image list A i The number of the first sample images in , i=1...n, n is the number of the first target operation type; S12, obtain the first target duration list set TA corresponding to A = {TA1, ..., TA i ,……,TA n } and the first target identifier list set PA corresponding to A = {PA1, ..., PA i ,……,PA n }, where TA i ={TA i1 ,……,TA ir ,……,TA is（i） }, P.A. i ={PA i1 ,……,PA ir ,……,PA is（i） },TA ir A ir The corresponding first target duration, PA ir A ir The corresponding first target identifier, and PA ir The value is "1" or "0"; S13, according to PA, obtain the first target accuracy list FD corresponding to D = {FD1, ..., FD i , ..., FD n }, FD i D i The corresponding first target accuracy, where FD i Meet the following conditions: FD i =α i / s(i),α i For PA i The number of first targets marked as "1" in ; S14, training the first initial neural network model based on D, A, TA, PA and FD to obtain a first neural network model; S300, when the candidate recognition time meets the second preset condition, the final operation type corresponding to the target access control device is determined using the second neural network model, the second neural network model is a neural network model obtained by model training using data associated with iris recognition based on iris images, and the second preset condition is that the time is within the time range of iris recognition, wherein S300 also includes the following steps: S301, obtaining the candidate priority corresponding to the target access control device; S302, inputting the candidate priority into the second neural network model to obtain a final operation type, wherein the final operation type is a second target operation type obtained by using the second neural network model; In S300, the second neural network model is obtained through the following steps: S21, obtain a second target operation type list H={H1, ..., H j , ..., H m The second sample image list set B corresponding to H = {B1, ..., B j , ..., B m }, where B j ={B j1 , ..., B jv , ..., B jq（j） }, H j is the jth second target operation type, B jv H j The corresponding second sample image list B j The vth second sample image in H, v = 1...q(j), q(j) is j The corresponding second sample image list B j The number of the second sample images in , j=1…m, m is the number of the second target operation types; S22, obtain the second target duration list set TB corresponding to B = {TB1, ..., TB j , ..., TB m } and the second target identifier list set PB corresponding to B = {PB1, ..., PB j , ..., PB m }, where TB j ={TB j1 , ..., TB jv , ..., TB jq（j） }, PB j ={PB j1 , ..., PB jv , ..., PB jq（j） }, TB jv For B jv The corresponding second target duration, PB jv For B jv The corresponding second target identifier, and PB jv The value is "1" or "0"; S23, according to PB, obtain the second target accuracy list FH corresponding to H = {FH1, ..., FH j , ..., FH m }, FH j H j The corresponding second target accuracy, where FH j Meet the following conditions: FH j =β j / q(j),β j PB j The number of second targets marked as "1" in ; S24, training the second initial neural network model based on H, B, TB, PB and FH to obtain a second neural network model; S400, based on the final operation type, obtain a target recognition duration list T={T1, ..., T e , ..., T f }, T e is the target recognition duration corresponding to the e-th target user ID, e=1…f, f is the number of target user IDs in the target user ID list; S500, when there is |T e -T 0 |≥λ, the target access control device issues an early warning, where T 0 is a preset time threshold, λ is a preset time difference threshold, wherein T is obtained by the following steps in S500 0 : S501, when the final operation type is the same as D i When the corresponding first target operation type is consistent, obtain T 0 , where T 0 Meet the following conditions: ； S502, when the final operation type is the same as H j When the corresponding second target operation type is consistent, obtain T 0 , where T 0 Meet the following conditions: 。 2. The access control warning system based on biometric information according to claim 1 is characterized in that: The first target operation type is the type of operation performed on the fingerprint image during the fingerprint recognition process.

3. The access control warning system based on biometric information according to claim 1 is characterized in that: The second target operation type is the type of operation performed on the iris image during iris recognition.

4. The access control warning system based on biometric information according to claim 1 is characterized in that: The value range of λ is 3 seconds to 5 seconds.

Images