CN112084893B

CN112084893B - Method, device, equipment and storage medium for detecting abnormality of biological identification terminal

Info

Publication number: CN112084893B
Application number: CN202010856967.5A
Authority: CN
Inventors: 窦逸辛; 柴培林; 赖嘉伟; 卞凯; 傅宜生
Original assignee: China Unionpay Co Ltd
Current assignee: China Unionpay Co Ltd
Priority date: 2020-08-24
Filing date: 2020-08-24
Publication date: 2024-05-17
Anticipated expiration: 2040-08-24
Also published as: CN112084893A

Abstract

The application discloses a method, a device, equipment and a storage medium for detecting abnormality of a biological identification terminal, and belongs to the field of data processing. The method comprises the following steps: acquiring historical identification data of a target biological identification terminal, wherein the historical identification data comprises first biological characteristic data and similarity parameter values of the first biological characteristic data and sample biological characteristic data in an identification database; based on the historical identification data, obtaining first distribution characteristics associated with first biological characteristic data under each similarity parameter value; and under the condition that the first distribution characteristic does not accord with any preset standard distribution condition, determining that the target biological identification terminal is abnormal, wherein the standard distribution condition is related to the distribution characteristic of the second biological characteristic data of the biological identification terminal which is not abnormal under each similarity parameter value. According to the embodiment of the application, the detection of the abnormal biological identification terminal can be realized.

Description

Method, device, equipment and storage medium for detecting abnormality of biological identification terminal

Technical Field

The application belongs to the field of data processing, and particularly relates to a method, a device, equipment and a storage medium for detecting abnormality of a biological identification terminal.

Background

The biometric technology refers to a technology for identity authentication through biometric features. The applicable scenarios of biometric identification are increasing, such as transaction scenarios, attendance scenarios, traffic scenarios, etc.

In order to avoid abnormality of the biological recognition, the biological recognition model is trained by introducing abundant and various biological characteristic samples at the present stage, so that the stability of the biological recognition model is improved, and the possibility of abnormality in the biological recognition process is reduced. But the biometric terminal may also be abnormal due to hardware or external influence factors, etc. Because of large variability among software, hardware, external influencing factors, etc. of different biometric terminals performing biometric identification, it is difficult to detect the biometric terminal having abnormality.

Disclosure of Invention

The embodiment of the application provides a method, a device, equipment and a storage medium for detecting abnormality of a biological identification terminal, which can realize the detection of the biological identification terminal with abnormality.

In one aspect, an embodiment of the present application provides a method for detecting abnormality of a biometric terminal, including: acquiring historical identification data of a target biological identification terminal, wherein the historical identification data comprises first biological characteristic data and similarity parameter values of the first biological characteristic data and sample biological characteristic data in an identification database; based on the historical identification data, obtaining first distribution characteristics associated with first biological characteristic data under each similarity parameter value; and under the condition that the first distribution characteristic does not accord with any preset standard distribution condition, determining that the target biological identification terminal is abnormal, wherein the standard distribution condition is related to the distribution characteristic of the second biological characteristic data of the biological identification terminal which is not abnormal under each similarity parameter value.

In a second aspect, an embodiment of the present application provides a device for detecting abnormality of a biometric terminal, including: the data acquisition module is used for acquiring historical identification data of the target biological identification terminal, wherein the historical identification data comprises first biological characteristic data and similarity parameter values of the first biological characteristic data and sample biological characteristic data in an identification database; the first processing module is used for obtaining first distribution characteristics associated with first biological characteristic data under each similarity parameter value based on the historical identification data; the abnormality judgment module is used for determining that the target biological identification terminal is abnormal under the condition that the first distribution characteristic does not accord with any preset standard distribution condition, and the standard distribution condition is related to the distribution characteristic of the biological identification terminal which is not abnormal and is related to the second biological characteristic data under each similarity parameter value in the standard condition.

In a third aspect, an embodiment of the present application provides a biometric terminal abnormality detection apparatus, including: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements the method for detecting abnormality of the biometric terminal in the first aspect.

In a fourth aspect, an embodiment of the present application provides a computer storage medium, on which computer program instructions are stored, which when executed by a processor implement the method for detecting an abnormality of a biometric terminal in the first aspect.

According to the method, the device, the equipment and the storage medium for detecting the abnormality of the biological identification terminal, the first distribution characteristics associated with the first biological characteristic data under the similarity parameter value are obtained by utilizing the historical identification data of the target biological identification terminal. The standard distribution condition can be obtained by utilizing the distribution characteristics of the second biological characteristic data of the biological identification terminal which is not abnormal under the similarity parameter value in advance. If the target biometric terminal is not abnormal, the distribution situation of the target biometric terminal related to the first biometric data under each similarity parameter value should have a similar point or be close to the distribution situation of the second biometric data under each similarity parameter value of the biometric terminal which is not abnormal. Therefore, the standard distribution condition can be used as a basis for judging whether the target biometric terminal is abnormal or not. And under the condition that the first distribution characteristic does not accord with any preset standard distribution condition, representing the distribution condition of the target biological identification terminal related to the first biological characteristic data under each similarity parameter value, wherein the distribution condition of the second biological characteristic of the biological identification terminal which is not abnormal under each similarity parameter value is greatly different from the distribution condition of the second biological characteristic of the biological identification terminal which is not abnormal, and the difference is caused by the abnormality of the target biological identification terminal. Therefore, whether the target biological recognition terminal is abnormal or not can be determined according to whether the first distribution characteristics meet the standard distribution conditions, so that the abnormal biological recognition terminal can be detected.

Drawings

In order to more clearly illustrate the technical solution of the embodiments of the present application, the drawings that are needed to be used in the embodiments of the present application will be briefly described, and it is possible for a person skilled in the art to obtain other drawings according to these drawings without inventive effort.

FIG. 1 is a schematic diagram of a biometric identification system according to an embodiment of the present application;

FIG. 2 is a flowchart of a method for detecting abnormality of a biometric terminal according to an embodiment of the present application;

FIG. 3 is a schematic diagram illustrating an example of a distribution curve of first biometric data under similarity parameter values according to an embodiment of the present application;

FIG. 4 is a schematic diagram illustrating an example of a distribution curve of second biometric data under similarity parameter values according to an embodiment of the present application;

FIG. 5 is a schematic diagram of another example of a distribution curve of first biometric data under a similarity parameter value according to an embodiment of the present application;

FIG. 6 is a schematic diagram of another example of a distribution curve of first biometric data under similarity parameter values according to an embodiment of the present application;

FIG. 7 is a schematic diagram of another example of a distribution curve of first biometric data under similarity parameter values according to an embodiment of the present application;

FIG. 8 is a schematic diagram of an example of a similarity parameter value distribution between user identification success and user identification failure according to an embodiment of the present application;

FIG. 9 is a flowchart of a method for detecting abnormality of a biometric terminal according to another embodiment of the present application;

Fig. 10 is a flowchart of a method for detecting abnormality of a biometric terminal according to still another embodiment of the present application;

FIG. 11 is a schematic diagram of a device for detecting abnormality of a biometric terminal according to an embodiment of the present application;

fig. 12 is a schematic structural diagram of an abnormality detection device for a biometric terminal according to another embodiment of the present application;

fig. 13 is a schematic structural diagram of a device for detecting abnormality of a biometric terminal according to an embodiment of the present application.

Detailed Description

Features and exemplary embodiments of various aspects of the present application will be described in detail below, and in order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described in further detail below with reference to the accompanying drawings and the detailed embodiments. It should be understood that the particular embodiments described herein are meant to be illustrative of the application only and not limiting. It will be apparent to one skilled in the art that the present application may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the application by showing examples of the application.

The biometric technology refers to a technology for identity authentication through biometric features, and can be applied to, for example, transaction, attendance checking, traffic, etc., but is not limited thereto. Fig. 1 is a schematic diagram of a biological recognition system according to an embodiment of the present application. As shown in fig. 1, the biometric system may include a biometric terminal 11 and a biometric server 12. Each biometric server 12 may communicate with one biometric terminal 11 or may communicate with a plurality of biometric terminals 11, which is not limited herein. The type of the biometric terminal 11 may be determined according to the application scenario, and is not limited herein. For example, in a transaction scenario, the biometric terminal may pay for a face vending machine, automated teller machine, or the like. For another example, in the attendance scenario, the biometric terminal 11 may be a face recognition attendance machine or the like. For another example, in a traffic scenario, the biometric terminal 11 may be a face recognition traffic gate or the like.

The biometric terminal 11 receives the biometric trigger request, and collects biometric characteristics of the user under test in response to the biometric trigger request. Based on the biometric characteristics, biometric data is generated and uploaded to the biometric server 12. The biometric server 12 is provided with an identification database. The identification database includes sample biometric data, each sample biometric data corresponding to a user having rights. The biometric server 12 calculates the similarity of the uploaded biometric data with the sample biometric data. If the similarity is greater than or equal to a threshold value defining success and failure of recognition, the biometric server 12 determines that the biometric of the user under test is successful, and considers the user with authority to include the user under test. If the similarity is below a threshold value defining success and failure of recognition, biometric server 12 determines that the biometric of the user under test failed, and considers that the user with authority does not include the user under test.

In some cases, the biometric terminal 11 and the biometric server 12 may be integrated into the same device, which is not limited herein.

However, the biological recognition terminals are possibly attacked maliciously due to large differences among the camera types, the image acquisition algorithms and the surrounding environments of the different biological recognition terminals. These differences and malicious attacks may affect the biometric algorithm, possibly confuse the biometric algorithm recognition result, or misidentify the two different persons as the same person and miss-identify the user with authority as the user without authority, thereby reducing the accuracy of the biometric recognition and causing the biometric terminal to be abnormal. But such anomalies are difficult to find at this stage.

The embodiment of the application provides a method, a device, equipment and a storage medium for detecting abnormality of a biological identification terminal, which can detect whether the biological identification terminal is abnormal or not so as to facilitate the follow-up taking of corresponding measures. The method for detecting abnormality of the biometric terminal in the embodiment of the present application can be applied to a biometric server or a biometric terminal and a biometric server integrated together, but is not limited thereto.

Fig. 2 is a flowchart of a method for detecting abnormality of a biometric terminal according to an embodiment of the present application. As shown in fig. 2, the biometric terminal abnormality detection method may include steps S201 to S203.

In step 201, history identification data of a target biometric terminal is acquired.

The history identification data may be data acquired by the target biometric terminal over a period of time. Specifically, the historical identification data includes data that may include the first biometric data, and a similarity parameter value of the first biometric data to sample biometric data in the identification database. In an embodiment of the present application, the biometric data includes, but is not limited to, facial feature data, fingerprint feature data, palm print feature data, iris feature data, and the like. The first biological characteristic data is data corresponding to biological characteristics collected by the target biological identification terminal. And the biological characteristic data corresponds to the biological characteristic acquired by the target biological identification terminal at one time. For example, if the target biometric terminal performs three biometric acquisitions on the same user a, three first biometric data are correspondingly generated.

The similarity parameter value is used for representing the similarity between the two. The higher the similarity parameter value of the two is, the higher the similarity between the two is. In some examples, the similarity parameter value may be expressed in terms of a percentage.

The identification database includes at least one sample biometric data. Each sample biometric data corresponds to a user having rights associated with the biometric terminal. For example, in a payment scenario, each sample biometric data in the identification database corresponds to a registered user who is able to make a biometric payment through the target biometric terminal. And under the condition that the similarity parameter value of certain first biological characteristic data and one sample biological characteristic data is higher than or equal to the identification success threshold value, the user corresponding to the first biological characteristic data and the user corresponding to the sample biological characteristic data are considered to be the same user. And under the condition that the similarity parameter value of certain first biological characteristic data and one sample biological characteristic data is lower than the recognition success threshold value, the user corresponding to the first biological characteristic data and the user corresponding to the sample biological characteristic data are considered to be different users. In the case where the identification database includes two or more sample biometric data, the similarity parameter values of the first biometric data and each sample biometric data may be calculated separately.

In step 202, a first distribution feature associated with first biometric data at each similarity parameter value is derived based on the historical identification data.

The first distribution feature is used to characterize a distribution of similarity between the first biometric data and the sample biometric data in the identification database. In particular, the first distribution feature may be used to characterize the distribution of the first biometric data under the similarity parameter value, and may also be used to characterize the distribution of the identification performed using the first biometric data under the similarity parameter value, which is not limited herein.

In some examples, a distribution curve may be employed to embody a distribution of similarity between the first biometric data and the sample biometric data. For example, fig. 3 is a schematic diagram of an example of a distribution curve of first biometric data under a similarity parameter value according to an embodiment of the present application. As shown in fig. 3, the abscissa of the coordinate system in which the distribution curve is located is the similarity parameter value, and the ordinate is the distribution number of the first biometric data. Any one point on the distribution curve may represent the number of the first biometric data corresponding to the similarity parameter value in the history identification data of the target biometric terminal. As can be seen from fig. 3, the distribution curve can be divided into two parts according to the recognition success threshold, and the distribution part of the first biometric data under each similarity parameter value lower than the recognition success threshold is located on the left side of the broken line L1, and the distribution part of the first biometric data under each similarity parameter value higher than or equal to the recognition success threshold is located on the right side of the broken line L1.

In step 203, if the first distribution feature does not meet any preset standard distribution condition, it is determined that the target biometric terminal is abnormal.

The standard distribution condition is related to the distribution characteristics of the second biological characteristic data of the biological identification terminal without abnormality under each similarity parameter value. The biological recognition terminal which is not abnormal is a biological recognition terminal which is normal in software, hardware, surrounding environment and the like and is not attacked by malicious. The second biological characteristic data is generated by the biological recognition terminal without abnormality according to the collected biological characteristics. The distribution of the second biological characteristic data of the biological identification terminal without abnormality has a certain rule under each similarity parameter value. The similarity parameter value related to the biological recognition terminal without abnormality is the similarity parameter value of the second biological feature data and the sample biological feature data in the recognition database.

In some examples, a distribution curve may be employed to embody a distribution of similarity between the second biometric data and the sample biometric data. Fig. 4 is a schematic diagram of an example of a distribution curve of second biometric data under a similarity parameter value according to an embodiment of the present application. As shown in fig. 4, the abscissa of the coordinate system in which the distribution curve is located is the similarity parameter value, and the ordinate is the distribution number of the second biometric data. Any one point on the distribution curve may represent the number of second biometric data corresponding to the similarity parameter value among the history identification data of the biometric terminal in which no abnormality has occurred. As can be seen from fig. 4, the distribution curve can be divided into two parts according to the recognition success threshold, and the distribution part of the second biometric data under each similarity parameter value lower than the recognition success threshold is located on the left side of the broken line L1, and the distribution part of the second biometric data under each similarity parameter value higher than or equal to the recognition success threshold is located on the right side of the broken line L1.

As can be seen from the distribution curve associated with the biometric terminals in which no abnormality has occurred shown in fig. 4, the distribution curve has alternately occurring peak points and valley points. For example, the distribution curve in fig. 4 includes a peak point P1, a peak point P2, a valley point D1, a valley point D2, and a valley point D3. The similarity parameter corresponding to the peak point and the similarity parameter corresponding to the valley point have a certain limit range. The change trend of the distribution curve is regular. In the embodiment of the present application, in addition to the normal valley point, when the value of the ordinate of the point in the distribution curve is smaller as the minimum value of the abscissa is closer to the distribution curve, one point of the distribution curve in the region close to the minimum value of the abscissa may be regarded as the valley point. That is, in addition to the normal valley value, if the number of biometric data decreases as the similarity parameter value decreases in the process of decreasing the similarity parameter value to the minimum value, the number of biometric data corresponding to the similarity parameter value close to the minimum value may be taken as the valley value. Similarly, in addition to the normal valley point, in the case where the closer the maximum value of the abscissa is in the distribution curve, the smaller the value of the ordinate is of the points in the distribution curve, one point of the distribution curve in the region near the maximum value of the abscissa may be taken as the valley point. That is, in addition to the normal valley, if the number of biometric data decreases as the similarity parameter value increases in the process of increasing the similarity parameter value to the maximum value, the number of biometric data corresponding to the similarity parameter value close to the maximum value may be taken as the valley.

And obtaining distribution characteristics according to the distribution of the second biological characteristic data of the biological identification terminal without abnormality under each similarity parameter value. Based on the distribution characteristics, standard distribution conditions can be set. The standard distribution condition may be used to distinguish between a biometric terminal in which an abnormality has occurred and a biometric terminal in which no abnormality has occurred. The first distribution characteristics meet all standard distribution conditions, represent the distribution condition of the target biological identification terminal related to the first biological characteristic data under each similarity parameter value, have similar points or are close to the distribution condition of the second biological characteristics of the biological identification terminal without abnormality under each similarity parameter value, and can determine that the target biological identification terminal is not abnormal. The first distribution characteristics do not accord with any standard distribution condition, represent the distribution condition of the target biological identification terminal related to the first biological characteristic data under each similarity parameter value, and have larger difference with the distribution condition of the second biological characteristics of the biological identification terminal without abnormality under each similarity parameter value, so that the abnormality of the target biological identification terminal can be determined.

The standard distribution condition may be one or two or more, and is not limited herein. If the first distribution characteristic meets the standard distribution condition, determining that the target biological identification terminal is not abnormal; if the first distribution characteristics do not meet the standard distribution conditions, the target biological identification terminal can be determined to be abnormal. If the first distribution characteristics meet all the standard distribution conditions under the condition that the standard distribution conditions are more than two, determining that the target biological identification terminal is not abnormal; if the first distribution characteristic does not meet at least one of the standard distribution conditions, it can be determined that the target biometric terminal is abnormal.

In the embodiment of the application, the first distribution characteristics associated with the first biological characteristic data under the similarity parameter value are obtained by utilizing the historical identification data of the target biological identification terminal. The standard distribution condition can be obtained by utilizing the distribution characteristics of the second biological characteristic data of the biological identification terminal which is not abnormal under the similarity parameter value in advance. If the target biometric terminal is not abnormal, the distribution situation of the target biometric terminal related to the first biometric data under each similarity parameter value should have a similar point or be close to the distribution situation of the second biometric data under each similarity parameter value of the biometric terminal which is not abnormal. Therefore, the standard distribution condition can be used as a basis for judging whether the target biometric terminal is abnormal or not. And under the condition that the first distribution characteristic does not accord with any preset standard distribution condition, representing the distribution condition of the target biological identification terminal related to the first biological characteristic data under each similarity parameter value, wherein the distribution condition of the second biological characteristic of the biological identification terminal which is not abnormal under each similarity parameter value is greatly different from the distribution condition of the second biological characteristic of the biological identification terminal which is not abnormal, and the difference is caused by the abnormality of the target biological identification terminal. Therefore, whether the target biological recognition terminal is abnormal or not can be determined according to whether the first distribution characteristics meet the standard distribution conditions, so that the abnormal biological recognition terminal is detected, and the safety of biological recognition is improved.

The first distribution feature, the standard distribution condition corresponding to the first distribution feature, and specifically how to determine that the target biometric terminal is abnormal will be described below with several examples.

In a first example, the first distribution feature includes a similarity parameter value corresponding to the first peak. The first peak value is the peak value of the number of the first biological characteristic data corresponding to each similarity parameter value. The number of first biometric data corresponds to the number of identifications. For example, fig. 5 is a schematic diagram of another example of a distribution curve of the first biometric data under the similarity parameter value provided in the embodiment of the present application. As shown in fig. 5, the distribution curve includes a peak point PA1 and a peak point PA2. The similarity parameter value corresponding to the peak point PA1, that is, the abscissa value of the peak point PA1 is xa1, and the first peak value corresponding to the peak point PA1, that is, the ordinate value of the peak point PA1 is ya1. The first peak value is ya1, which means that the number of first biometric data is ya1, i.e., the number of recognition times is ya1, at the similarity parameter value xa 1. The similarity parameter value corresponding to the peak point PA2, that is, the abscissa value of the peak point PA2 is xa2, and the first peak value corresponding to the peak point PA2, that is, the ordinate value of the peak point PA2 is ya2. The first peak value is ya2, which means that the number of first biometric data is ya2, i.e., the number of recognition times is ya2, at the similarity parameter value xa 2.

Correspondingly, the standard distribution condition comprises that the absolute value of the difference value between the similarity parameter value corresponding to the first peak value and the similarity parameter value corresponding to the standard peak value is smaller than a first preset threshold value. Under the abnormal conditions that the hardware defect of the target biological identification terminal has a large influence on the collected biological characteristics, the defect of the biological characteristic collection algorithm of the target biological identification terminal has a large influence on the collected biological characteristics, the surrounding environment of the target biological identification terminal has a large influence on the collected biological characteristics, the target biological identification terminal is attacked maliciously and is broken, and the like, the similarity parameter value corresponding to the first peak value of the target biological identification terminal is greatly changed compared with the similarity parameter value corresponding to the standard peak value. Therefore, the similarity parameter value corresponding to the first peak value can be utilized to detect whether the target biological recognition terminal is abnormal.

In some examples, where the distribution curve of the first biometric data includes a plurality of peak points, e.g., two peak points, the first peak of the first occurring peak point and the standard peak of the first occurring peak point may be compared, and the first peak of the second occurring peak point and the standard peak of the second occurring peak point may be compared, in order of the similarity parameter value from small to large.

The standard peak includes a peak value of the number of second biometric data at each similarity parameter value. For example, as shown in fig. 4, peak points corresponding to standard peaks include P1 and P2. The similarity parameter value corresponding to the peak point P1, that is, the abscissa value of the peak point P1 is x1, and the first peak value corresponding to the peak point P1, that is, the ordinate value of the peak point P1 is y1. The similarity parameter value corresponding to the peak point P2, that is, the abscissa value of the peak point P2 is x2, and the first peak value corresponding to the peak point P2, that is, the ordinate value of the peak point P2 is y2. The first preset threshold may be set according to the working scenario and the working requirement, which is not limited herein. For ease of description, the first preset threshold is denoted by z 1.

According to fig. 4 and 5, if |xa1-x1|gtoreq.z1 and/or |xa2-x2|gtoreq.z1, the similarity parameter value corresponding to the first peak value and the similarity parameter value corresponding to the standard peak value are excessively different, it can be determined that the target biological identification terminal is abnormal; if the value of the similarity parameter corresponding to the first peak value is smaller than z1 and the value of the similarity parameter corresponding to the standard peak value is smaller than z1, the target biological identification terminal can be determined to be abnormal.

In a second example, the first distribution feature includes a similarity parameter value corresponding to the first valley. The first valley value is the valley value of the number of the first biological characteristic data corresponding to each similarity parameter value. For example, as shown in fig. 5, the distribution curve includes a valley point DB1, a valley point DB2, and a valley point DB3. The abscissa value of the valley point DB1, which is the similarity parameter value corresponding to the valley point DB1, is xb1, and the ordinate value of the first valley value corresponding to the valley point DB1, which is the valley point DB1, is yb1. The first valley value is yb1, which means that the number of first biometric data is yb1, i.e., the number of recognition times is yb1, at the similarity parameter value xb 1. The abscissa value of the valley point DB2, which is the similarity parameter value corresponding to the valley point DB2, is xb2, and the ordinate value of the first valley value corresponding to the valley point DB2, which is the valley point DB2, is yb2. The first valley value is yb2, which means that under the similarity parameter value xb2, the number of first biometric data is yb2, that is, the number of recognition times is yb2. The abscissa value of the valley point DB3, which is the similarity parameter value corresponding to the valley point DB3, is xb3, and the ordinate value of the first valley value corresponding to the valley point DB3, which is the valley point DB3, is yb3. The first valley value is yb3, which means that under the similarity parameter value xb3, the number of first biometric data is yb3, that is, the number of recognition times is yb3.

Correspondingly, the standard distribution condition comprises that the absolute value of the difference value between the similarity parameter value corresponding to the first valley value and the similarity parameter value corresponding to the standard valley value is smaller than a second preset threshold value. Under the abnormal conditions that the hardware defect of the target biological recognition terminal has a large influence on the collected biological characteristics, the defect of the biological characteristic collection algorithm of the target biological recognition terminal has a large influence on the collected biological characteristics, the surrounding environment of the target biological recognition terminal has a large influence on the collected biological characteristics, the target biological recognition terminal is attacked maliciously and is broken, and the like, the similarity parameter value corresponding to the first valley value of the target biological recognition terminal is changed greatly compared with the similarity parameter value corresponding to the standard valley value. Therefore, the similarity parameter value corresponding to the first valley value can be utilized to detect whether the target biological recognition terminal is abnormal.

In some examples, where the distribution curve of the first biometric data includes a plurality of dip, e.g., three dip, the first dip of the first occurrence and the standard dip of the first occurrence may be compared, the first dip of the second occurrence and the standard dip of the second occurrence may be compared, and the first dip of the third occurrence and the standard dip of the third occurrence may be compared, in order of the similarity parameter value from small to large.

The standard valley includes a valley of the number of second biometric data at each similarity parameter value. For example, as shown in fig. 4, the valley points corresponding to the standard valleys include D1, D2, and D3. The similarity parameter value corresponding to the valley point D1, that is, the abscissa value of the valley point D1 is x3, and the first valley value corresponding to the valley point D1, that is, the ordinate value of the valley point D1 is y3. The similarity parameter value corresponding to the valley point D2, that is, the abscissa value of the valley point D2 is x4, and the first valley value corresponding to the valley point D2, that is, the ordinate value of the valley point D2 is y4. The similarity parameter value corresponding to the valley point D3, that is, the abscissa value of the valley point D3 is x5, and the first valley value corresponding to the valley point D3, that is, the ordinate value of the valley point D3 is y5. The second preset threshold may be set according to the working scenario and the working requirement, which is not limited herein. For ease of description, the second preset threshold is denoted by z 2.

According to fig. 4 and 5, if |xb1-x3|gtoreq.z2, |xb2-x4|gtoreq.z2 and/or |xb2-x5|gtoreq.z2, the similarity parameter value corresponding to the first valley value is excessively different from the similarity parameter value corresponding to the standard valley value, and it can be determined that the target biological identification terminal is abnormal; if the absolute value of the similarity parameter value corresponding to the first valley value is smaller than z2, the absolute value of the similarity parameter value corresponding to the standard valley value is smaller than z2, and the fact that the target biological recognition terminal is abnormal can be determined.

In a third example, the first distribution feature includes a number of first biometric data corresponding to the first similarity parameter value. The first similarity parameter value is a similarity parameter value corresponding to the standard valley value. The standard valley is the valley of the number of the second biometric data corresponding to each similarity parameter value. For example, as shown in fig. 4, the valley points corresponding to the standard valleys include a valley point D1, a valley point D2, and a valley point D3. The first similarity parameter value corresponding to the valley point D1 is x3, the first similarity parameter value corresponding to the valley point D2 is x4, and the first similarity parameter value corresponding to the valley point D3 is x5. For ease of illustration, one valley point in the profile shown in fig. 5 is illustrated, and other valley points in the profile shown in fig. 5 may be detected in this manner. As shown in fig. 5, x4 corresponds to the number of first biometric data as y4'.

Correspondingly, the standard distribution condition comprises that the difference value between the number of the first biological characteristic data corresponding to the first similarity parameter value and the standard valley value is smaller than a third preset threshold value. The target biological recognition terminal can not collect accurate biological characteristics due to the hardware defects of the target biological recognition terminal, the influence of the surrounding environment of the target biological recognition terminal and the like, if clear face images can not be collected, the number of the first biological characteristic data corresponding to the original first similarity parameter value is greatly increased, namely the recognition times corresponding to the original first similarity parameter value is greatly increased. Therefore, the number of the first biological characteristic data corresponding to the first similarity parameter value can be utilized to detect whether the target biological identification terminal is abnormal.

The third preset threshold may be set according to the working scenario and the working requirement, which is not limited herein. For ease of description, the third preset threshold is denoted by z 3.

According to fig. 4 and 5, if |y4' -y4|z3 is not less than or equal to z3, the number of the first biometric data corresponding to the first similarity parameter value is excessively different from the standard valley value, so that it can be determined that the target biometric terminal is abnormal; if the value of the |y4' -y4| < z3 indicates that the difference between the number of the first biological characteristic data corresponding to the first similarity parameter value and the standard valley value is in a normal range, it can be determined that the target biological identification terminal is not abnormal. It should be noted that other valley points in the distribution curve shown in fig. 5 may also be detected in this manner, and will not be described herein.

In a fourth example, the first distribution characteristic includes a sum of occurrences of the first peak. The first peak value is the peak value of the number of the first biological characteristic data corresponding to each similarity parameter value. The sum of the occurrence times of the first peak value is the number of peak points in the distribution curve of the first biological characteristic data under the similarity parameter value. Note that, when a plurality of peak points exist in the distribution curve of the first biometric data in the similarity parameter value, the number of first biometric data corresponding to the plurality of peak points, that is, the first peak values may be equal or unequal, and the present invention is not limited thereto. The equal first peak appears twice, and the statistics of the occurrence times of the first peak is twice. For example, as shown in fig. 5, the distribution curve includes two peak points, i.e., the sum of the occurrence times of the first peak is 2.

Correspondingly, the standard distribution condition includes that the sum of the occurrence times of the first peak is equal to 2. In the case that no abnormality occurs in the target biometric terminal, the distribution curve of the first biometric data under the similarity parameter value should include two peak points. Under the conditions that the biological characteristics cannot be acquired due to the failure of the target biological identification terminal hardware, the biological characteristics cannot be acquired due to the influence of the surrounding environment of the target biological identification terminal, and the target biological identification terminal is broken through by malicious attack, due peak points disappear. Therefore, the detection of whether or not the abnormality occurs in the target biometric terminal can be performed using the sum of the occurrence times of the first peak, that is, the number of peak points.

For example, the number of peak points in the distribution curve shown in fig. 5 is equal to 2, that is, the number of occurrences of the first peak is equal to 2, and the standard distribution condition is met, it can be determined that no abnormality occurs in the target biometric terminal.

For another example, fig. 6 is a schematic diagram of another example of a distribution curve of the first biometric data under the similarity parameter value provided in the embodiment of the present application. As shown in fig. 6, in the case of an abnormality such as a failure of the target biometric terminal hardware, which fails to collect the biometric feature, the number of the first biometric feature data corresponding to the similarity parameter value close to the minimum value is greatly increased, so that a part of the peak points in the distribution curve disappear. The peak points in the distribution curve shown in fig. 6 include only the peak point PA3, that is, the number of peak points in the distribution curve shown in fig. 6 is less than 2, that is, the sum of the occurrence times of the first peak is less than 2, and it can be determined that abnormality occurs in the target biometric terminal.

For another example, fig. 7 is a schematic diagram of still another example of a distribution curve of the first biometric data under the similarity parameter value according to the embodiment of the present application. As shown in fig. 7, in the case where the target biometric terminal is subjected to a malicious attack or the like such that the first biometric data of the user who is not identical to the user corresponding to the sample biometric data is successfully identified, the number of first biometric data corresponding to the similarity parameter value near the maximum value is caused to be greatly increased, so that a partial peak point of the distribution curve disappears. The peak points in the distribution curve shown in fig. 7 include only the peak point PA4, that is, the number of peak points in the distribution curve shown in fig. 7 is less than 2, that is, the sum of the occurrence times of the first peak is less than 2, and it can be determined that abnormality occurs in the target biometric terminal.

In a fifth example, the first distribution feature includes a sum of occurrences of the first valley. The first valley value is the valley value of the number of the first biological characteristic data corresponding to each similarity parameter value. The sum of the occurrence times of the first valley is the number of valley points in the distribution curve of the first biological characteristic data under the similarity parameter value. Note that, when there are a plurality of dips in the distribution curve of the first biometric data in the similarity parameter value, the number of the first biometric data corresponding to the plurality of dips, that is, the first dips may be equal or unequal, and the present invention is not limited thereto. The equal first valley appears twice, and the statistics of the occurrence times of the first valley is twice. For example, as shown in fig. 5, the distribution curve includes three dips, i.e., the sum of the occurrence times of the first dips is 3.

Correspondingly, the standard distribution condition includes that the sum of the occurrence times of the first valley is equal to 3. In the case that no abnormality occurs in the target biometric terminal, the distribution curve of the first biometric data under the similarity parameter value should include three dip points. Under the conditions that the biological characteristics cannot be acquired due to the failure of the target biological identification terminal hardware, the accuracy of the biological characteristics acquired by the target biological identification terminal under the influence of the surrounding environment is low, and the target biological identification terminal is broken through by malicious attack, due valley points disappear. Therefore, the detection of whether or not the abnormality of the target biometric terminal has occurred can be performed using the sum of the occurrence times of the first valley, that is, the number of valley points.

For example, the number of valley points in the distribution curve shown in fig. 5 is equal to 3, that is, the number of occurrences of the first valley value is equal to 3, and the standard distribution condition is met, it may be determined that no abnormality occurs in the target biometric terminal.

For another example, as shown in fig. 6, in the case of an abnormality such as a failure of the target biometric terminal hardware, which fails to collect the biometric feature, the number of the first biometric data corresponding to the similarity parameter value close to the minimum value is greatly increased, so that a part of the valley point disappears in the distribution curve. The valley points in the distribution curve shown in fig. 6 only include the valley points DB4 and the valley points DB5, that is, the number of valley points in the distribution curve shown in fig. 6 is less than 3, that is, the sum of the occurrence times of the first valley is less than 3, and it can be determined that the abnormality occurs in the target biometric terminal.

As another example, as shown in fig. 7, in the case where the target biometric terminal is subjected to a malicious attack or the like such that the identification of the first biometric data of the user who is not the same as the user corresponding to the sample biometric data is successful, the number of the first biometric data corresponding to the similarity parameter value near the maximum value is greatly increased, so that a partial valley point in the distribution curve disappears. The valley points in the distribution curve shown in fig. 7 only include the valley points DB6 and the valley points DB7, that is, the number of valley points in the distribution curve shown in fig. 7 is less than 3, that is, the sum of the occurrence times of the first valley is less than 3, and it can be determined that the abnormality occurs in the target biometric terminal.

In a sixth example, the first distribution feature includes a normalized number of samples. The number of samples at the second similarity parameter value includes a sum of the number of sample biometric data with each of the first biometric data for which the similarity parameter value is the second similarity parameter value. The second similarity parameter value may be any similarity parameter value below a recognition success threshold, and is not limited herein.

The target biological recognition terminal acquires first biological feature data and calculates similarity parameter values of the first biological feature data and the biological feature data of each sample in the recognition database respectively. In the case that no abnormality occurs in the target biometric terminal, at most one first biometric data can have a similarity parameter value with one sample biometric data greater than or equal to the recognition success threshold.

For example, fig. 8 is a schematic diagram of an example of a similarity parameter value distribution between user identification success and user identification failure according to an embodiment of the present application. As shown in fig. 8, a thin solid line indicates a user whose identification is successful, and a thick solid line indicates a user whose identification is failed. The abscissa is the number of the sample biological feature data, and the ordinate is the similarity parameter value of the first biological feature data and the sample biological feature data. Since there is only one sample biometric data in each identification that has a similarity parameter value with the first biometric data that is greater than the identification success threshold. Therefore, the distribution of the sum of the numbers of the sample biometric data having the similarity parameter value higher than or equal to the recognition success threshold value with respect to each first biometric data should be consistent in overall trend with the distribution of the second biometric data of the biometric terminal in which no abnormality occurs at each similarity parameter value lower than the recognition success threshold value, with the distribution of the sum of the numbers of the sample biometric data having the similarity parameter value lower than the recognition success threshold value with respect to each first biometric data being smaller. That is, the distribution of the number of samples should be consistent in overall trend with the distribution of the second biometric data at each similarity parameter value below the recognition success threshold value for the biometric terminal in which no abnormality has occurred.

For comparison, the number of samples may be normalized so that the distribution range of the number of samples after normalization coincides with the distribution range of the number of second biometric data at each similarity parameter value below the recognition success threshold.

Correspondingly, the standard distribution condition comprises that the absolute value of the difference value between the number of samples after normalization processing and the number of second biological characteristic data under the second similarity parameter value is smaller than a fourth preset threshold value. The fourth preset threshold may be specifically set according to the working scenario and the working requirement, which is not limited herein. In the case that the target biometric terminal is abnormal, the distribution of the number of samples and the distribution of the second biometric data of the biometric terminal without the abnormality at each similarity parameter value lower than the recognition success threshold will be greatly different in overall trend. Therefore, the detection of whether or not abnormality occurs in the target biometric terminal can be performed using the number of samples after normalization processing.

The absolute value of the difference between the number of samples after normalization processing and the number of the second biological feature data under the second similarity parameter value is smaller than a fourth preset threshold value, and the distribution of the number of samples is smaller in overall trend difference and is in an acceptable range when the distribution of the number of samples is lower than the similarity parameter value of the recognition success threshold value of the non-abnormal biological recognition terminal. The absolute value of the difference between the number of samples after normalization processing and the number of the second biological feature data under the second similarity parameter value is larger than or equal to a fourth preset threshold value, and the distribution of the number of samples is larger in overall trend difference with the distribution of the second biological feature data of the biological recognition terminal without abnormality under each similarity parameter value lower than the recognition success threshold value.

In a seventh example, the first distribution characteristic includes a first duty cycle. The first duty cycle is a ratio of the number of first negative-sample biometric data to the number of first biometric data. The first negative-sample biometric data is first biometric data having a similarity parameter value less than a recognition success threshold. The similarity parameter value refers to the similarity parameter value of the first biological feature data and the sample biological feature data.

Correspondingly, the standard distribution condition comprises that the difference value between the first duty ratio and the standard duty ratio threshold value is smaller than or equal to a fifth preset threshold value. The standard duty threshold is an average value of first duty ratios of the plurality of target biometric terminals, or a ratio of the number of second negative-sample biometric data to the number of second biometric data in the standard condition of the biometric terminal in which no abnormality has occurred. The second negative-sample biometric data is second biometric data having a similarity parameter value less than a recognition success threshold. The fifth preset threshold may be specifically set according to the working scenario and the working requirement, which is not limited herein.

Since the number of the first sample feature data having failed recognition in a single target biometric terminal is significantly too high, it means that the target biometric terminal is likely to be abnormal, such as may be being attacked or identified maliciously. The first duty cycle and the standard duty cycle threshold may be used to detect whether an abnormality has occurred in the target biometric terminal.

The first distribution feature may include, but is not limited to, one or more of a similarity parameter value corresponding to the first peak value, a similarity parameter value corresponding to the first valley value, a number of first biometric feature data corresponding to the first similarity parameter value, a sum of occurrence times of the first peak value, a sum of occurrence times of the first valley value, a number of samples after normalization processing, and a first duty ratio in the above embodiment.

If the first distribution characteristics used for judging whether the target biological identification terminal is abnormal comprise a plurality of items, the corresponding standard distribution conditions comprise a plurality of items, and when at least one of the first distribution characteristics does not accord with the corresponding standard distribution condition, the abnormal condition of the target biological identification terminal can be determined.

Fig. 9 is a flowchart of a method for detecting abnormality of a biometric terminal according to another embodiment of the present application. Fig. 9 is different from fig. 2 in that the biometric terminal abnormality detection method shown in fig. 9 may further include steps S204 to S208.

In step S204, when the similarity parameter values of the N first biometric data continuously acquired by the target biometric terminal and the sample biometric data in the recognition database are both lower than the recognition success threshold, and the similarity parameter values between the N first biometric data are higher than the similarity determination threshold, it is determined that the target biometric terminal is abnormal, and the users corresponding to the N first biometric data are marked as abnormal users.

The similarity parameter values of the N continuous first biological characteristic data and the sample biological characteristic data in the recognition database are lower than the recognition success threshold value, and the similarity parameter values among the N first biological characteristic data are higher than the similarity judgment threshold value, so that the situation that the same user continuously tries to continue recognition under the condition of recognition failure is very likely, and the possibility that the user carries out malicious attack on the target biological recognition terminal exists. N is an integer greater than 1. On the one hand, the abnormality of the target biological identification terminal can be determined; on the other hand, the user can be marked as an abnormal user, and the abnormal detection of the user operation behavior of the target biometric terminal can also be realized.

The similarity determination threshold is a threshold for defining whether the two biometric feature data indicate the same user, and may be specifically set according to the working scenario and the working requirement, but is not limited thereto.

In step S205, an abnormal user database is built using the first biometric data of the abnormal user.

I.e. the abnormal user database comprises first biometric data of the abnormal user. In the case of each tagged obtaining of an abnormal user, the first biometric data of the tagged abnormal user may be stored into the abnormal user database, thereby enabling the abnormal user database to be used to identify the abnormal user.

In step S206, a biometric request of the target biometric terminal is received.

Wherein the biometric request includes third biometric data.

In step S207, similarity parameter values of the third biometric data and the first biometric data of the abnormal user in the abnormal user database are calculated, respectively.

Before calculating the similarity parameter values of the third biological characteristic data and the sample biological characteristic data in the identification database, calculating the similarity parameter values of the third biological characteristic data and the first biological characteristic data of the abnormal user in the abnormal user database respectively, so that when the abnormal user requests for biological identification, certain measures can be taken for the abnormal user to limit the biological identification operation of the abnormal user.

In step S208, in the case where the similarity parameter value of the third biometric data and the first biometric data of the abnormal user in the abnormal user database is higher than the similarity determination threshold, the recognition success threshold for determining that the biometric recognition is successful is raised.

The similarity parameter value of the third biological characteristic data and the first biological characteristic data of the abnormal user in the abnormal user database is higher than a similarity judgment threshold value, which indicates that the user corresponding to the third biological characteristic data is the abnormal user, and in order to limit the biological recognition of the abnormal user, the recognition success threshold value for judging the success of the biological recognition can be improved.

In some examples, the recognition success threshold for the target biometric terminal may be increased. For example, the recognition success threshold of the target biometric terminal is 80%; and under the condition that the similarity parameter value of the third biological characteristic data and the first biological characteristic data of the abnormal user in the abnormal user database is higher than the similarity judgment threshold value, the recognition success threshold value of the target biological recognition terminal is increased to 90%.

In other examples, the recognition success threshold for the anomalous user may be increased. For example, the recognition success threshold of the target biometric terminal is 80%; in the case where the similarity parameter value of the third biometric data with the first biometric data of the abnormal user in the abnormal user database is higher than the similarity determination threshold, the recognition success threshold for recognition with the third biometric data with the sample biometric data in the recognition database is increased to 90%, but the recognition success threshold for recognition with the sample biometric data in the recognition database remains 80% for other third biometric data whose similarity parameter value with the first biometric data of the abnormal user in the abnormal user database is lower than or equal to the similarity determination threshold.

Because the abnormal users may have malicious attack users and normal users, the probability of successful identification of the malicious attack users can be reduced by improving the identification success threshold, and the normal users can be ensured to be successful in identification to a certain extent.

Fig. 10 is a flowchart of a method for detecting abnormality of a biometric terminal according to still another embodiment of the present application. Fig. 10 is different from fig. 2 in that the biometric terminal abnormality detection method shown in fig. 10 may further include step S209 and/or step S210.

In step S209, in the case where a biometric request of the target biometric terminal for which abnormality is determined to occur is received, the recognition success threshold for determining that biometric is successful is raised.

By improving the recognition success threshold of the target biological recognition terminal with the abnormality, the risk of the abnormal biological recognition can be reduced to a certain extent, and the safety of the biological recognition can be improved.

In step S210, when a biometric request of an abnormally associated biometric terminal is received, a recognition success threshold for determining that biometric is successful is raised.

The abnormality related biometric terminal is a biometric terminal with the same attribute as that of the target biometric terminal with the abnormality. The attributes here are attributes of the biometric terminal. For example, the attributes may include a terminal model, a terminal operator, and the like. In the case that the biometric terminal is applied to a transaction scenario, the attribute may further include a merchant to which the terminal belongs, a acquirer corresponding to the terminal transaction, and the like. The attributes are not limited herein.

The probability of abnormality of the abnormality association biometric terminal is higher than that of other biometric terminals, and the security of biometric identification of the abnormality association biometric terminal is improved by improving the identification success threshold of the abnormality association biometric terminal.

The embodiment of the application also provides a device for detecting the abnormality of the biological identification terminal. Fig. 11 is a schematic structural diagram of an abnormality detection device for a biometric terminal according to an embodiment of the present application. As shown in fig. 11, the biometric terminal abnormality detection device 300 may include a data acquisition module 301, a first processing module 302, and an abnormality determination module 303.

The data acquisition module 301 may be configured to acquire historical identification data of the target biometric terminal, where the historical identification data includes first biometric data and a similarity parameter value of the first biometric data and sample biometric data in the identification database.

The first processing module 302 may be configured to obtain, based on the historical identification data, a first distribution feature associated with the first biometric data at each similarity parameter value.

The abnormality determination module 303 is configured to determine that the target biometric terminal is abnormal if the first distribution feature does not meet any preset standard distribution condition.

The standard distribution condition is related to the distribution characteristics of the biological recognition terminal which is not abnormal and is associated with the second biological characteristic data under the similarity parameter values in the standard condition.

In some examples, the first distribution feature includes a similarity parameter value corresponding to the first peak. The first peak value is the peak value of the number of the first biological characteristic data corresponding to each similarity parameter value.

Correspondingly, the standard distribution condition comprises that the absolute value of the difference value between the similarity parameter value corresponding to the first peak value and the similarity parameter value corresponding to the standard peak value is smaller than a first preset threshold value. The standard peak includes a peak value of the number of second biometric data at each similarity parameter value.

In some examples, the first distribution feature includes a similarity parameter value corresponding to the first valley. The first valley value is the valley value of the number of the first biological characteristic data corresponding to each similarity parameter value.

Correspondingly, the standard distribution condition comprises that the absolute value of the difference value between the similarity parameter value corresponding to the first valley value and the similarity parameter value corresponding to the standard valley value is smaller than a second preset threshold value. The standard valley includes a valley of the number of second biometric data at each similarity parameter value.

In some examples, the first distribution feature includes a number of first biometric data corresponding to the first similarity parameter value. The first similarity parameter value is a similarity parameter value corresponding to the standard valley value. The standard valley is the valley of the number of the second biometric data corresponding to each similarity parameter value.

Correspondingly, the standard distribution condition comprises that the difference value between the number of the first biological characteristic data corresponding to the first similarity parameter value and the standard valley value is smaller than a third preset threshold value.

In some examples, the first distribution characteristic includes a sum of occurrences of the first peak. The first peak value is the peak value of the number of the first biological characteristic data corresponding to each similarity parameter value.

Correspondingly, the standard distribution condition includes that the sum of the occurrence times of the first peak is equal to 2.

In some examples, the first distribution feature includes a sum of occurrences of the first valley. The first valley value is the valley value of the number of the first biological characteristic data corresponding to each similarity parameter value.

Correspondingly, the standard distribution condition includes that the sum of the occurrence times of the first valley is equal to 3.

In some examples, the first distribution characteristic includes a normalized number of samples. The number of samples at the second similarity parameter value includes a sum of the number of sample biometric data with each of the first biometric data for which the similarity parameter value is the second similarity parameter value. The second similarity parameter value is below the recognition success threshold. The distribution range of the number of samples after normalization processing coincides with the distribution range of the number of second biometric data at each similarity parameter value lower than the recognition success threshold.

Correspondingly, the standard distribution condition comprises that the absolute value of the difference value between the number of samples after normalization processing and the number of second biological characteristic data under the second similarity parameter value is smaller than a fourth preset threshold value.

In some examples, the first distribution characteristic includes a first duty cycle. The first duty cycle is a ratio of the number of first negative-sample biometric data to the number of first biometric data. The first negative-sample biometric data is first biometric data having a similarity parameter value less than a recognition success threshold.

The standard distribution condition includes that the difference between the first duty cycle and the standard duty cycle threshold is less than or equal to a fifth preset threshold. The standard duty threshold is a ratio of the number of second negative-sample biometric data to the number of second biometric data in the standard condition of the average value of the first duty of the plurality of target biometric terminals or the biometric terminal in which no abnormality has occurred. The second negative-sample biometric data is second biometric data having a similarity parameter value less than a recognition success threshold.

Fig. 12 is a schematic structural diagram of an abnormality detection device for a biometric terminal according to another embodiment of the present application. Fig. 12 is different from fig. 11 in that the biometric terminal abnormality detection apparatus shown in fig. 12 may further include an abnormality user marking module 304, a database creation module 305, a reception module 306, a second processing module 307, and a threshold adjustment module 308.

The abnormal user marking module 304 may be configured to determine that the target biometric terminal is abnormal when the similarity parameter values of the N first biometric data continuously acquired by the target biometric terminal and the sample biometric data in the recognition database are both lower than the recognition success threshold, and the similarity parameter values between the N first biometric data are higher than the similarity determination threshold, and mark the users corresponding to the N first biometric data as abnormal users.

The database creation module 305 may create an abnormal user database using the first biometric data of the abnormal user.

The receiving module 306 may be configured to receive a biometric request of the target biometric terminal, the biometric request including third biometric data.

The second processing module 307 may be configured to calculate similarity parameter values between the third biometric data and the first biometric data of the abnormal user in the abnormal user database, respectively.

The threshold adjustment module 308 may be configured to increase the recognition success threshold for determining that the biometric recognition is successful in the event that the similarity parameter value of the third biometric data and the first biometric data of the abnormal user in the abnormal user database is greater than the similarity determination threshold.

In some examples, threshold adjustment module 308 may also be configured to increase an identification success threshold for determining that the biometric identification was successful upon receipt of a biometric identification request from a target biometric identification terminal that determined that an anomaly occurred.

In some examples, threshold adjustment module 308 may also be configured to increase an identification success threshold for determining that the biometric identification was successful upon receipt of a biometric request from an abnormally associated biometric terminal.

The abnormality related biometric terminal is a biometric terminal with the same attribute as that of the target biometric terminal with the abnormality.

The embodiment of the application also provides a device for detecting the abnormality of the biological identification terminal. Fig. 13 is a schematic structural diagram of a device for detecting abnormality of a biometric terminal according to an embodiment of the present application. As shown in fig. 13, the biometric terminal abnormality detection apparatus 400 includes a memory 401, a processor 402, and a computer program stored on the memory 401 and executable on the processor 402.

In one example, the processor 402 may include a Central Processing Unit (CPU), or an Application SPECIFIC INTEGRATED Circuit (ASIC), or may be configured as one or more integrated circuits that implement embodiments of the present application.

Memory 401 may include mass storage for data or instructions. By way of example, and not limitation, memory 401 may include a hard disk drive (HARD DISK DRIVE, HDD), a floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or a universal serial bus (Universal Serial Bus, USB) drive, or a combination of two or more of these.

The Memory may include Read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic disk storage media devices, optical storage media devices, flash Memory devices, electrical, optical, or other physical/tangible Memory storage devices. Thus, in general, the memory comprises one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software comprising computer-executable instructions and when the software is executed (e.g., by one or more processors) it is operable to perform the operations described with reference to the biometric terminal anomaly detection method in accordance with the present application.

The processor 402 runs a computer program corresponding to the executable program code by reading the executable program code stored in the memory 401 for realizing the biometric terminal abnormality detection method in the above-described embodiment.

In one example, biometric terminal abnormality detection device 400 may also include a communication interface 403 and a bus 404. As shown in fig. 13, the memory 401, the processor 402, and the communication interface 403 are connected to each other by a bus 404 and perform communication with each other.

The communication interface 403 is mainly used to implement communication between each module, device, unit and/or apparatus in the embodiment of the present application. Input devices and/or output devices may also be accessed through communication interface 403.

The bus 404 includes hardware, software, or both, and couples the components of the biometric terminal abnormality detection apparatus 400 to each other. By way of example, and not limitation, bus 404 may include an accelerated graphics Port (ACCELERATED GRAPHICS Port, AGP) or other graphics Bus, an enhanced industry Standard architecture (Enhanced Industry Standard Architecture, EISA) Bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an industry Standard architecture (Industrial Standard Architecture, ISA) Bus, an Infiniband interconnect, a Low Pin Count (LPC) Bus, a memory Bus, a micro channel architecture (Micro Channel Architecture, MCA) Bus, a peripheral component interconnect (PERIPHERAL COMPONENT INTERCONNECT, PCI) Bus, a PCI-Express (PCI-X) Bus, a serial advanced technology attachment (SERIAL ADVANCED Technology Attachment, SATA) Bus, a video electronics standards Association local (Video Electronics Standards Association Local Bus, VLB) Bus, or other suitable Bus, or a combination of two or more of the above. Bus 404 may include one or more buses, where appropriate. Although embodiments of the application have been described and illustrated with respect to a particular bus, the application contemplates any suitable bus or interconnect.

The embodiment of the application also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, can implement the method for detecting abnormality of a biometric terminal in the above embodiment, and achieve the same technical effects, and in order to avoid repetition, no detailed description is given here. The computer readable storage medium may include, but is not limited to, read-Only Memory (ROM), random access Memory (Random Access Memory RAM), magnetic disk, optical disk, and the like.

It should be understood that, in the present specification, each embodiment is described in an incremental manner, and the same or similar parts between the embodiments are all referred to each other, and each embodiment is mainly described in a different point from other embodiments. For apparatus embodiments, device embodiments, computer readable storage medium embodiments, the relevant points may be found in the description of method embodiments. The application is not limited to the specific steps and structures described above and shown in the drawings. Those skilled in the art will appreciate that various alterations, modifications, and additions may be made, or the order of steps may be altered, after appreciating the spirit of the present application. Also, a detailed description of known method techniques is omitted here for the sake of brevity.

Aspects of the present application are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions/acts specified in the flowchart and/or block diagram block or blocks. Such a processor may be, but is not limited to being, a general purpose processor, a special purpose processor, an application specific processor, or a field programmable logic circuit. It will also be understood that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware which performs the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Those skilled in the art will appreciate that the above-described embodiments are exemplary and not limiting. The different technical features presented in the different embodiments may be combined to advantage. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in view of the drawings, the description, and the claims. In the claims, the term "comprising" does not exclude other means or steps; the word "a" does not exclude a plurality; the terms "first," "second," and the like, are used for designating a name and not for indicating any particular order. Any reference signs in the claims shall not be construed as limiting the scope. The functions of the various elements presented in the claims may be implemented by means of a single hardware or software module. The presence of certain features in different dependent claims does not imply that these features cannot be combined to advantage.

Claims

1. A method for detecting abnormalities in a biometric terminal, comprising:

Acquiring historical identification data of a target biological identification terminal, wherein the historical identification data comprises first biological characteristic data and similarity parameter values of the first biological characteristic data and sample biological characteristic data in an identification database;

based on the history identification data, obtaining first distribution characteristics associated with the first biological characteristic data under each similarity parameter value;

Determining that the target biological identification terminal is abnormal under the condition that the first distribution characteristic does not accord with any preset standard distribution condition, wherein the standard distribution condition is related to the distribution characteristic of second biological characteristic data of the biological identification terminal without the abnormality under each similarity parameter value;

The first distribution feature comprises similarity parameter values corresponding to first peaks, the first peaks are peaks of the number of the first biological feature data corresponding to each similarity parameter value,

The standard distribution condition comprises that the absolute value of the difference value between the similarity parameter value corresponding to the first peak value and the similarity parameter value corresponding to the standard peak value is smaller than a first preset threshold value, and the standard peak value comprises the peak value of the number of second biological feature data under each similarity parameter value;

the first distribution feature comprises similarity parameter values corresponding to first valley values, the first valley values are the valley values of the number of the first biological feature data corresponding to each similarity parameter value,

The standard distribution condition includes that the absolute value of the difference value between the similarity parameter value corresponding to the first valley value and the similarity parameter value corresponding to the standard valley value is smaller than a second preset threshold value, and the standard valley value includes the valleys of the number of second biological feature data under each similarity parameter value.

2. The method of claim 1, wherein the first distribution feature comprises a number of the first biometric data corresponding to a first similarity parameter value, the first similarity parameter value being a similarity parameter value corresponding to a standard valley, the standard valley being a valley of a number of the second biometric data corresponding to each similarity parameter value,

The standard distribution condition comprises that the difference value between the number of the first biological characteristic data corresponding to the first similarity parameter value and the standard valley value is smaller than a third preset threshold value.

3. The method of claim 1, wherein the first distribution characteristic comprises a sum of a number of occurrences of a first peak, the first peak being a peak of a number of the first biometric data corresponding to each similarity parameter value,

The standard distribution condition includes that the sum of the occurrence times of the first peak is equal to 2.

4. The method of claim 1, wherein the first distribution feature comprises a sum of a number of occurrences of a first valley, the first valley being a valley of a number of the first biometric data corresponding to each similarity parameter value,

The standard distribution condition includes that the sum of the number of occurrences of the first valley is equal to 3.

5. The method of claim 1, wherein the first distribution feature comprises a normalized number of samples, the number of samples at a second similarity parameter value comprising a sum of the number of sample biometric data for which the similarity parameter value for each of the first biometric data is the second similarity parameter value, the second similarity parameter value being below a recognition success threshold, the distribution range of the normalized number of samples being consistent with the distribution range of the number of second biometric data for each of the similarity parameter values below the recognition success threshold;

The standard distribution condition comprises that the absolute value of the difference value between the number of samples after normalization processing and the number of second biological characteristic data under the second similarity parameter value is smaller than a fourth preset threshold value.

6. The method of claim 1, wherein the first distribution feature comprises a first duty cycle that is a ratio of a number of first negative sample biometric data to a number of the first biometric data, the first negative sample biometric data being the first biometric data having a similarity parameter value less than a recognition success threshold,

The standard distribution condition comprises that the difference value between the first duty ratio and a standard duty ratio threshold is smaller than or equal to a fifth preset threshold, the standard duty ratio threshold is an average value of the first duty ratios of a plurality of target biological identification terminals or a ratio of the number of second negative sample biological characteristic data to the number of second biological characteristic data in the standard condition, and the second negative sample biological characteristic data is the second biological characteristic data with a similarity parameter value smaller than the identification success threshold.

7. The method as recited in claim 1, further comprising:

And determining that the target biological recognition terminal is abnormal under the condition that the similarity parameter values of N pieces of first biological feature data continuously acquired by the target biological recognition terminal and the sample biological feature data in the recognition database are lower than a recognition success threshold value and the similarity parameter values of the N pieces of first biological feature data are higher than a similarity judgment threshold value, and marking the users corresponding to the N pieces of first biological feature data as abnormal users, wherein N is an integer larger than 1.

8. The method as recited in claim 7, further comprising:

And establishing an abnormal user database by utilizing the first biological characteristic data of the abnormal user.

9. The method as recited in claim 8, further comprising:

Receiving a biometric request of the target biometric terminal, the biometric request including third biometric data;

respectively calculating similarity parameter values of the third biological characteristic data and the first biological characteristic data of the abnormal user in the abnormal user database;

And in the case that the similarity parameter value of the third biometric data and the first biometric data of the abnormal user in the abnormal user database is higher than the similarity judgment threshold, increasing the recognition success threshold for judging that the biometric recognition is successful.

10. The method as recited in claim 1, further comprising:

And in the case of receiving a biometric request of the target biometric terminal for which the occurrence of the abnormality is determined, increasing a recognition success threshold for judging that the biometric is successful.

11. The method as recited in claim 1, further comprising:

in the case of receiving a biometric request of an abnormally associated biometric terminal, increasing a recognition success threshold for judging that biometric recognition is successful,

The abnormality related biometric terminal is a biometric terminal having the same attribute as the target biometric terminal having the abnormality.

12. A biometric terminal abnormality detection device, characterized by comprising:

the data acquisition module is used for acquiring historical identification data of the target biological identification terminal, wherein the historical identification data comprises first biological characteristic data and similarity parameter values of the first biological characteristic data and sample biological characteristic data in an identification database;

The first processing module is used for obtaining first distribution characteristics associated with the first biological characteristic data under each similarity parameter value based on the historical identification data;

the abnormality judgment module is used for determining that the target biological identification terminal is abnormal under the condition that the first distribution characteristic does not accord with any preset standard distribution condition, and the standard distribution condition is related to the distribution characteristic of the biological identification terminal which is not abnormal and is related to the second biological characteristic data under each similarity parameter value in the standard condition;

13. A biometric terminal abnormality detection apparatus, characterized by comprising: a processor and a memory storing computer program instructions;

the processor, when executing the computer program instructions, implements the biometric terminal abnormality detection method according to any one of claims 1 to 11.

14. A computer storage medium having stored thereon computer program instructions which, when executed by a processor, implement the biometric terminal anomaly detection method of any one of claims 1 to 11.