Fingerprint template updating method and device, embedded device and storage medium

Abstract

The application discloses a fingerprint template updating method and device, an embedded device and a storage medium, relates to the field of fingerprint identification, and is used for an embedded fingerprint identification device. The method comprises the following steps: collecting a to-be-identified fingerprint image, extracting to-be-identified fingerprint features of the image, comparing the to-be-identified fingerprint features with a template library, and determining a target fingerprint template of a target finger; comparing the to-be-identified fingerprint features with identified fingerprint features in the target fingerprint template, determining to-be-replaced features of the target fingerprint template according to a feature comparison score and a feature template coverage area; calculating a replacement score of the target fingerprint template based on the to-be-identified fingerprint features and the to-be-replaced features, and performing template updating when the replacement score exceeds a score threshold. In the fingerprint identification process, the corresponding replacement score is determined in a manner of fingerprint feature matching and feature area comparison, the template is updated according to the score size, the fingerprint identification precision is improved, and the false recognition rate is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of fingerprint identification, and in particular to a fingerprint template updating method and device, an embedded device and a storage medium. BACKGROUND

[0002] Fingerprint identification technology has been widely used in our life due to its uniqueness and stability. For example, fingerprint attendance, fingerprint door lock, fingerprint luggage, mobile phone fingerprint login and mobile phone fingerprint payment, etc. With the further wide application of fingerprint identification technology, the market puts forward higher requirements for the speed and performance of fingerprint identification technology.

[0003] With the smaller and smaller area of the fingerprint image collected by the sensor, the traditional recognition method based on the fixed template cannot meet the actual application requirements. For the small-area fingerprint sensor, only a limited number of images can be registered for consideration of user experience during registration, which cannot well represent the characteristics of the finger. If the template is not continuously updated subsequently, the performance will continue to decline. For example, in the related art, different fingerprint identification systems will formulate comparison strategies according to their feature extraction methods for the fingerprint feature comparison module. Generally, the recognition rate is improved in the case of fixed template, and the template updating is ignored. If the template is not updated for a long time, the recognition rate will be reduced, the finger cannot be recognized when it is pressed, and the problem of too high false rejection rate occurs. SUMMARY

[0004] The present application provides a fingerprint template updating method, device, embedded device and storage medium, which solves the problem of too high false rejection rate and too high false acceptance rate caused by incomplete fingerprint template updating in the related art.

[0005] In one aspect, the present application provides a fingerprint template updating method, which is used for an embedded fingerprint identification device, and the method comprises:

[0006] Collecting a to-be-identified fingerprint image, extracting a to-be-identified fingerprint feature of the image, and comparing the to-be-identified fingerprint feature with a template library to determine a target fingerprint template of a target finger;

[0007] Comparing the to-be-identified fingerprint feature with an identified fingerprint feature in the target fingerprint template, and determining a to-be-replaced feature of the target fingerprint template according to a feature comparison score and a feature template coverage area;

[0008] Calculating a replacement score of the target fingerprint template based on the to-be-identified fingerprint feature and the to-be-replaced feature, and updating the template when the replacement score exceeds a score threshold.

[0009] Specifically, after determining the target fingerprint template, the method further comprises:

[0010] placing all identified fingerprint features in the target fingerprint template in the same coordinate system in a feature point alignment manner;

[0011] placing the identified fingerprint features with overlapping feature points in the coordinate system into the same feature set to form a continuous feature group; and establishing feature sets for the identified fingerprint features without overlapping feature points to form several independent feature groups;

[0012] when the target fingerprint template contains at least one continuous feature group and one independent feature group, performing fingerprint feature comparison; wherein the different feature groups are discontinuous with each other.

[0013] Specifically, the comparison of the to-be-identified fingerprint feature and the identified fingerprint feature in the target fingerprint template includes:

[0014] comparing all identified fingerprint features in the target fingerprint template to determine the identified fingerprint feature with the maximum comparison score and the feature group where the identified fingerprint feature is located, and determining the group as a matching group;

[0015] calculating the original template coverage area formed by the identified fingerprint features in the matching feature group; the original template coverage area is the area of the smallest circumscribed polygon with all feature points as vertices and containing all feature points after placing all features in the same coordinate system;

[0016] replacing the identified fingerprint feature in the matching feature group with the to-be-identified fingerprint feature, and calculating the replacement coverage area after replacement; when the replacement template coverage area is greater than the original coverage area, the replaced identified fingerprint feature is determined as the to-be-replaced feature.

[0017] Specifically, when the matching group is the continuous feature group; the replacement of the identified fingerprint feature in the matching feature group with the to-be-identified fingerprint feature, and the calculation of the replacement coverage area after replacement; when the replacement template coverage area is greater than the original coverage area, the replaced identified fingerprint feature is determined as the to-be-replaced feature, including:

[0018] replacing the i identified fingerprint features in the continuous feature group with the to-be-identified fingerprint feature respectively, and calculating i replacement coverage areas of the matching group after replacement respectively;

[0019] determining the identified fingerprint feature with the maximum replacement coverage area and greater than the original coverage area as the replacement area;

[0020] when the matching group is the independent feature group, directly determining the replacement coverage area as the to-be-replaced feature when the replacement coverage area is greater than the original coverage area.

[0021] Specifically, the method comprises:

[0022] determining the image quality and feature area of the to-be-replaced feature and the to-be-identified feature, respectively;

[0023] when the image quality of the to-be-identified feature is greater than an image threshold and the feature area is greater than an area threshold, determining a first replacement score according to the image quality and the feature area; and when the image quality of the to-be-replaced feature is greater than the image threshold and the feature area is greater than the area threshold, determining a second replacement score according to the image quality and the feature area;

[0024] wherein the first feature score is directly proportional to the image quality and the image area of the to-be-identified feature, and the second feature score is inversely proportional to the image quality and the image area of the to-be-replaced feature;

[0025] determining the sum of the first replacement score and the second replacement score as the replacement score of the to-be-replaced feature;

[0026] when the replacement score is greater than a score threshold, replacing the to-be-replaced feature with the to-be-identified feature.

[0027] Specifically, when the image quality of the to-be-replaced feature is less than the image threshold or the feature area is less than the area threshold, and the image quality of the to-be-identified feature is greater than the image threshold and the feature area is greater than the area threshold, directly replacing the to-be-replaced feature;

[0028] when the image quality of the to-be-identified feature is less than the image threshold or the feature area is less than the area threshold, not replacing.

[0029] Specifically, after the template update of the target fingerprint template, the method further comprises:

[0030] arranging the replaced matching group and other feature groups in the target fingerprint template according to the positional relationship, and aggregating the feature groups that overlap with each other into a new continuous feature group;

[0031] acquiring and updating the number of times of updating the target template and the number of feature groups of the target fingerprint template; when the number of times of updating is greater than an update threshold or all features are aggregated into one feature group, stopping updating the target template.

[0032] In another aspect, the present application provides a fingerprint template updating device, which is used in an embedded fingerprint identification device, and the device comprises:

[0033] The collection module is configured to collect a to-be-identified fingerprint image, extract a to-be-identified fingerprint feature of the image, and compare the to-be-identified fingerprint feature with a template library to determine a target fingerprint template of a target finger;

[0034] The determination module is configured to compare the to-be-identified fingerprint feature with an identified fingerprint feature in the target fingerprint template, and determine a to-be-replaced feature of the target fingerprint template according to a feature comparison score and a feature template coverage area.

[0035] The update module is configured to calculate a replacement score of the target fingerprint template based on the to-be-identified fingerprint feature and the to-be-replaced feature, and perform template update when the replacement score exceeds a score threshold.

[0036] In another aspect, the present application provides an embedded device, which comprises a processor and a memory, and the memory stores at least one instruction, at least one program, a code set or an instruction set, which are loaded and executed by the processor to implement the fingerprint template update method in the above aspect.

[0037] In another aspect, the present application provides a computer readable storage medium, which stores at least one instruction, at least one program, a code set or an instruction set, which are loaded and executed by a processor to implement the fingerprint template update method in the above aspect.

[0038] The technical scheme provided by the present application has at least the following beneficial effects: in the fingerprint identification, the to-be-identified fingerprint feature is taken as a replacement feature, which is compared with an identified fingerprint feature in a target fingerprint template to determine a corresponding comparison score, and the size of the template coverage area formed by the fingerprint feature is used to determine whether the introduced to-be-identified fingerprint feature introduces a new feature area, and in order to ensure the accuracy of the update and identification mechanism, a replacement score is calculated based on the to-be-identified fingerprint feature and the to-be-replaced feature, and the fingerprint template is updated according to the size of the replacement score. Through this method, matching judgment can be continuously performed, new fingerprint features can be introduced to the fingerprint template, and the fingerprint identification accuracy can be continuously improved and the false recognition rate can be reduced. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 is a flowchart of the fingerprint template update method provided by an embodiment of the present application;

[0040] Figure 2 is a flowchart of the fingerprint template update method provided by another embodiment of the present application;

[0041] Figure 3 is a schematic diagram of a fingerprint template provided by an embodiment of the present application;

[0042] Figure 4 is the original template coverage area of the matching group provided by the embodiment of the application;

[0043] Figure 5 is a flowchart of a fingerprint template updating method provided by another embodiment of the application;

[0044] Figure 6 a schematic diagram of replacement updating of a matching group is shown;

[0045] Figure 7 is a structural block diagram of a fingerprint template updating device provided by the embodiment of the application. DETAILED DESCRIPTION

[0046] To make the purpose, technical solutions and advantages of the application clearer, the embodiments of the application will be further described in detail below with reference to the drawings.

[0047] “Multiple” mentioned in the present text refers to two or more than two. “And / or” describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent the following three cases: A exists alone, A and B exist simultaneously, and B exists alone. The character “ / ” generally represents that the associated objects before and after it are in an “or” relationship.

[0048] Figure 1 is a flowchart of a fingerprint template updating method provided by the embodiment of the application, which is used for an embedded fingerprint identification device and includes the following steps:

[0049] Step 101, collect a to-be-identified fingerprint image, extract to-be-identified fingerprint features of the image, and compare the to-be-identified fingerprint features with a template library to determine a target fingerprint template of a target finger.

[0050] The premise of template updating is that fingerprint data needs to be input in advance in the embedded device, and the fingerprint data is used as a target fingerprint template. Generally, a plurality of fingerprint templates of fingers can be stored in any embedded device, and when a person presses a finger for normal identification, the embedded device first needs to perform a normal identification confirmation process, and then update the target fingerprint template under the premise of comparing the target fingerprint template with the template library.

[0051] Step 102, compare the to-be-identified fingerprint features with identified fingerprint features in the target fingerprint template, and determine to-be-replaced features of the target fingerprint template according to a feature comparison score and a feature template coverage area.

[0052] With a mobile phone with fingerprint entry and fingerprint recognition as an example for illustration, such as a side fingerprint recognition device. Because the size of the fingerprint sensor is small, only part of the area features of the finger is detected and entered each time, and when fingerprint recognition is performed, some new fingerprint features can be brought in, at which time it is necessary to update them to the target fingerprint template. For the to-be-identified fingerprint features of the fingerprint recognition, it is necessary to compare them with the identified fingerprint features in the target fingerprint template, and according to the feature comparison score, it is determined whether there is an overlap in area. At the same time, it is also necessary to ensure that the feature template coverage area cannot be reduced after the feature is updated, and according to these judgment conditions, all identified fingerprint features are screened to determine the to-be-replaced features of the target fingerprint template.

[0053] Step 103, calculate the replacement score of the target fingerprint template based on the to-be-identified fingerprint feature and the to-be-replaced feature, and perform template update when the replacement score exceeds the score threshold.

[0054] After determining the to-be-replaced feature, the to-be-identified fingerprint feature cannot be directly replaced in the feature of the target fingerprint template, and the image quality of the feature and the number of feature points of the feature also need to be considered, which calculate the replacement score. When the replacement score calculated for the to-be-replaced feature and the to-be-identified fingerprint feature is greater than the preset score threshold, it is ensured that the feature has the replacement condition, and the fingerprint feature collected this time is replaced with the corresponding feature in the template. Since it is determined according to the feature coverage area and the comparison score, more feature points can be introduced after replacement, and the original target fingerprint template is updated.

[0055] In summary, the present application identifies the fingerprint at the same time, takes the to-be-identified fingerprint feature as the replacement feature, compares it with the identified fingerprint feature in the target fingerprint template, determines the corresponding comparison score, and determines whether the to-be-identified fingerprint feature introduces new feature area according to the size of the template coverage area formed by the fingerprint feature. In order to ensure the accuracy of the update and recognition mechanism, the replacement score of the to-be-identified fingerprint feature and the to-be-replaced feature is also calculated, and the fingerprint template is updated according to the size of the replacement score. Through this method, matching judgment can be continuously performed, new fingerprint features can be introduced to the target fingerprint template, and the fingerprint recognition accuracy can be continuously improved and the false recognition rate can be reduced.

[0056] Figure 2 is a flowchart of a fingerprint template update method provided by another embodiment of the present application, comprising the following steps:

[0057] Step 201, collect a to-be-identified fingerprint image, extract a to-be-identified fingerprint feature of the image, and compare it with a template library to determine a target fingerprint template of a target finger.

[0058] The target fingerprint template in the fingerprint template library is identified by a fast search method, such as finding the corresponding features based on the feature header and template header to determine the target fingerprint template.

[0059] Step 202: Place all the identified fingerprint features in the target fingerprint template in the same coordinate system according to the feature point alignment method.

[0060] Because fingerprints are unique, the key to obtaining continuous fingerprints is finding the overlapping portions between individual fingerprint features and piecing together a complete fingerprint image based on these overlaps. Different fingerprint features contain different feature points, and this scheme identifies and confirms them based on the number and positional relationship of these feature points. A target fingerprint template may contain multiple recorded fingerprint features. By placing the identified fingerprint features in the same coordinate system according to feature point alignment, the overlapping features are determined based on the convergence of these feature points.

[0061] Step 203: Place the identified fingerprint features with overlapping feature points in the coordinate system into the same feature set to form a continuous feature group; establish feature sets for the identified fingerprint features with non-overlapping feature points to form several independent feature groups.

[0062] like Figure 3 As shown, the fingerprint template contains AH identified fingerprint features. Features A, B, and C form a continuous area, features D, E, and F form a continuous area, while features G and H have no connection relationship with the others.

[0063] This step establishes a coordinate system by aligning the center point. Identified fingerprint features with overlapping feature points in the coordinate system are placed into the same feature set, forming continuous feature groups, namely the set consisting of features A, B, and C, and the set consisting of features D, E, and F. Identified fingerprint features with non-overlapping feature points are then established into separate feature sets, forming several independent feature groups, namely the set consisting of features H and the set consisting of features G.

[0064] Step 204: When the target fingerprint template contains at least two feature groups, perform fingerprint feature comparison.

[0065] The premise of using two feature sets is to determine whether further updates are needed. This is because an incomplete fingerprint template will inevitably contain many discontinuous fingerprint features. However, if the fingerprint template contains only one feature set, it means that it has been updated to the point where further aggregation is not possible. Therefore, fingerprint feature comparison is only performed when at least two feature sets are detected.

[0066] Step 205: Compare all identified fingerprint features in the target fingerprint template, determine the identified fingerprint feature with the highest comparison score and its feature group, and determine the group as the matching group.

[0067] For all identified fingerprint features within the target fingerprint template, each is compared with the fingerprint features to be identified in this acquisition to obtain corresponding comparison scores. The identified fingerprint feature with the highest comparison score is then determined. This step aims to find the identified fingerprint feature with the largest overlap area in the template. The feature group containing this identified fingerprint feature must contain other features that overlap and are continuous with the fingerprint feature to be identified. The feature group containing the identified fingerprint feature with the highest comparison score is the matching group that needs to be matched and replaced subsequently.

[0068] Step 206: Calculate the original template coverage area formed by the identified fingerprint features in the matching feature group.

[0069] In this scheme, the original template coverage area is defined as the area of ​​the smallest bounding polygon containing all feature points, with the outermost feature point as its vertex, after placing all features in the same coordinate system. The original template coverage area is also the area of ​​the smallest bounding polygon formed by all identified fingerprint features in the matching group. This feature coverage area is used for subsequent replacement comparisons to determine the features to be replaced in the update.

[0070] Indicative, such as Figure 4 As shown, assuming the newly acquired fingerprint feature to be identified is I, and after comparison, feature B is determined to have the highest comparison score, then features A, B, and C are the determined matching group. The corresponding original template coverage area S0 is the original template coverage area.

[0071] Step 207: Replace the identified fingerprint feature in the matching feature group with the fingerprint feature to be identified, and calculate the replacement coverage area after replacement; when the replacement template coverage area is greater than the original coverage area, the replaced identified fingerprint feature is determined as the feature to be replaced.

[0072] The replacement coverage area is calculated by replacing each of the identified fingerprint features in the matching group with the fingerprint features to be identified, and then calculating the replacement coverage area formed by replacing one of them. When the replacement coverage area is greater than the original coverage area, it means that more new feature points have been introduced after the replacement, and more fingerprint area can be covered.

[0073] This step specifically includes the following steps:

[0074] a. When the matching group is a continuous feature group, replace the i identified fingerprint features in the continuous feature group with the fingerprint features to be identified, and calculate the coverage area of ​​the i replaced fingerprint features in the matching group after replacement.

[0075] by Figure 4 For example, the identified fingerprint features include features A, B, and C. Therefore, after replacing the feature to be identified, I, three replacement coverage areas are obtained.

[0076] b, the identified fingerprint feature with the largest replacement coverage area and greater than the original coverage area is determined as the feature to be replaced.

[0077] Since different feature areas are different, and the overlapping areas are also different, there may be a case that the overall area becomes smaller after the replacement of the identified fingerprint feature. This is mainly for the case that the feature area to be replaced is too small or the introduced new feature is smaller than the replaced area. Therefore, the present scheme needs to obtain the identified fingerprint feature with the largest replacement coverage area and greater than the original coverage area, and determine it as the feature to be replaced. That is, the feature B in the formula (1) is determined as the feature to be replaced. Figure 4

[0078] c, when the matching group is an independent feature group, if the replacement coverage area is greater than the original coverage area, it is directly determined as the feature to be replaced.

[0079] Since the independent feature group is a single group, the replacement is directly determined according to the size of the front and rear coverage areas.

[0080] It needs to be explained here why replacement is used instead of addition: theoretically, direct addition can more directly obtain new fingerprint features, but since the present scheme is applied to an embedded device, in the case of small sensor collection area, it is not appropriate to store more features for the same finger template in the case of limited resources, which will also increase the chip processing load and the recognition accuracy. Although the replacement method may cause the loss of a small part of the fingerprint information, the overall area of the feature is increased, that is, the introduced new feature is much larger than the lost feature.

[0081] Step 208, respectively determine the image quality and feature area of the feature to be replaced and the identified fingerprint feature.

[0082] In order to simplify the system model and improve the detection and recognition accuracy, after determining the feature to be replaced, instead of directly replacing, further judgment is made on the identified fingerprint feature and the feature to be replaced, which is specifically considered from the image quality and the feature area. Because the principle of replacement and update is one-to-one replacement, the old feature must be removed when introducing a new feature. For some identified fingerprint features with poor image quality or small feature area, directly updating them to the target fingerprint template will not improve the accuracy, but on the contrary, it will reduce the accuracy. Similarly, the same is true for the feature to be replaced. For the feature to be replaced with poor image quality or small feature area, it can be directly replaced to improve the detection accuracy.

[0083] ​Step 209, when the image quality of the to-be-identified fingerprint feature is greater than the image threshold value, and the feature area is greater than the area threshold value, a first replacement score is determined according to the image quality and the feature area; when the image quality of the to-be-replaced fingerprint feature is greater than the image threshold value, and the feature area is greater than the area threshold value, a second replacement score is determined according to the image quality and the feature area.

[0084] The first feature score is directly proportional to the image quality and the image area of the to-be-identified fingerprint feature, and the second feature score is inversely proportional to the image quality and the image area of the to-be-replaced feature.

[0085] Step 210, the sum of the first replacement score and the second replacement score is determined as the replacement score of the to-be-replaced feature.

[0086] Step 211, when the replacement score is greater than the score threshold value, the to-be-identified fingerprint feature is used to replace the to-be-replaced feature.

[0087] For the case that the feature area is below the minimum area threshold value or the image quality is below the image threshold value as mentioned above, the replacement score can not be calculated and the replacement or non-replacement can be directly performed.

[0088] In a possible implementation, when the image quality of the to-be-replaced feature is less than the image threshold value, or the feature area is less than the area threshold value, and the image quality of the to-be-identified fingerprint feature is greater than the image threshold value, and the feature area is greater than the area threshold value, the to-be-replaced feature is directly replaced;

[0089] When the image quality of the to-be-identified fingerprint feature is less than the image threshold value, or the feature area is less than the area threshold value, no replacement is performed.

[0090] In summary, the application groups the identified fingerprint features in the target fingerprint template, aligns the center points according to the positional relationship of the feature points, and determines the continuous feature group and the independent feature group according to the feature overlapping relationship; it is convenient to determine the matching group according to the comparison score of each identified fingerprint feature. After determining the matching group, the size relationship between the original coverage area formed by the matching group and the replacement coverage area after replacing the identified fingerprint features in the matching group is calculated, the identified fingerprint feature with the largest replacement coverage area is determined as the to-be-replaced feature, and finally the replacement score is obtained according to the feature area and the image quality of the to-be-identified fingerprint feature and the to-be-replaced feature, and whether to perform template updating is determined according to the replacement score. The method can update the template area, improve the fingerprint recognition accuracy, and reduce the false recognition rate in the embedded device with limited resource space.

[0091] In some embodiments, in the case of entering more fingerprint features, higher accuracy of fingerprint recognition can have been met, if the update is still continued for each fingerprint recognition process, unnecessary resource waste will still occur, for example, when the clock-in is frequent during the peak period, the recognition and update speed of the device will be affected. Therefore, some mechanism for temporarily suspending the update needs to be set, such as Figure 5 As shown, the method comprises the following steps:

[0092] Step 501, collect the image of the fingerprint to be identified, extract the fingerprint features to be identified from the image, and compare with the template library to determine the target fingerprint template of the target finger.

[0093] Step 502, compare the identified fingerprint features to be identified with the target fingerprint template, and determine the target fingerprint template to be replaced according to the feature comparison score and the feature template coverage area.

[0094] Step 503, calculate the replacement score of the target fingerprint template based on the identified fingerprint features and the features to be replaced, and update the template when the replacement score exceeds the score threshold.

[0095] Steps 501-503 refer to the foregoing process, which will not be described here.

[0096] Step 504, arrange the replaced matching group and other features in the target fingerprint template according to the positional relationship, and aggregate the mutually overlapping feature groups into new continuous feature groups.

[0097] This process is carried out after the template is updated. For the matching group that has been replaced and updated, due to the increase in area, the relationship between the continuous feature group and the independent feature group needs to be determined again, and the order is re-ordered according to the positional relationship. When the updated matching group and at least one other feature group exist, the mutually overlapping feature groups are aggregated into new continuous feature groups.

[0098] Specifically, this process includes three cases. The first case is that the replaced matching group cannot be aggregated into a whole, but is split into at least two feature groups. The second case is that there is no change in the number between the replaced matching group and other feature groups. The third case is that there is an overlap between the replaced matching group and at least one other feature group, and the number of feature groups in the aggregated template is reduced, which is also the best update result. For example, Figure 6As shown, the update schematic diagram in three cases is shown. In ①, feature I replaces feature B and no longer overlaps with the continuous feature group composed of ABC, but forms a new continuous feature group with feature DEF. In ②, feature I replaces feature B and overlaps with feature C and feature D, so that two feature groups are aggregated and the total number is reduced. In ③, feature I replaces feature C and continues to overlap with feature B, but feature A becomes an independent feature group again. It should be noted that the above is only an example for clearly describing the method, and in the actual case where the fingerprint template has been entered, the fingerprint template update will more likely aggregate the feature groups to reduce the total number, and thus the present scheme can optimize the existing fingerprint template.

[0099] In step 505, the number of times of updating the target template and the number of feature groups of the target fingerprint template are acquired and updated; when the number of times of updating is greater than an update threshold or all features are aggregated into one feature group, the updating of the target template is stopped.

[0100] As known from the above, after the fingerprint template is updated to the later stage, a good recognition rate can be basically achieved, and no further update can bring too much benefit, so it is necessary to set an update threshold in advance, and when the threshold is reached, the updating will be stopped. Meanwhile, the number of feature groups is also a judgment standard, and when the feature groups are gradually aggregated, the total number becomes smaller and smaller, and when only one feature group is left, it is indicated that there is no need to continue updating, and at this time, the updating of the template also needs to be stopped.

[0101] Figure 7 The device is a fingerprint template updating device provided by the embodiment of the present application, and the device is used in an embedded fingerprint identification device. The device comprises:

[0102] The acquisition module 701 is configured to acquire a to-be-identified fingerprint image, extract a to-be-identified fingerprint feature of the image, and determine a target fingerprint template of a target finger by comparing with a template library.

[0103] The determination module 702 is configured to compare the to-be-identified fingerprint feature with an identified fingerprint feature in the target fingerprint template, and determine a to-be-replaced feature of the target fingerprint template according to a feature comparison score and a feature template coverage area.

[0104] The updating module 703 is configured to calculate a replacement score of the target fingerprint template based on the to-be-identified fingerprint feature and the to-be-replaced feature, and perform template updating when the replacement score exceeds a score threshold.

[0105] The application further provides an embedded device, comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the fingerprint template updating method in the above aspect.

[0106] The application further provides a computer readable storage medium, wherein the readable storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the fingerprint template updating method in the above aspect.

[0107] The preferred embodiments of the application are described above; it should be understood that the application is not limited to the specific embodiments described above, and that the devices and structures not described in detail should be understood as being implemented in the ordinary way in the art; any person skilled in the art can make many possible changes and modifications, or modify equivalent embodiments without departing from the technical solution of the application, which does not affect the essential content of the application; therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the application without departing from the technical solution of the application, all still belong to the protection scope of the technical solution of the application.

Claims

1. A fingerprint template updating method, characterized in that, The method is used in an embedded fingerprint recognition device, and the method includes: Collect an image of the fingerprint to be identified, extract the fingerprint features from the image, and determine the target fingerprint template of the target finger by comparing it with a template library; The fingerprint feature to be identified is compared with the identified fingerprint features in the target fingerprint template, and the feature to be replaced in the target fingerprint template is determined based on the feature comparison score and the feature template coverage area; including: All identified fingerprint features within the target fingerprint template are compared to determine the identified fingerprint feature with the highest comparison score and its feature group, and this group is determined as the matching group. Calculate the original template coverage area formed by the identified fingerprint features in the matching group; the original template coverage area is the area of ​​the smallest bounding polygon containing all feature points after placing all features in the same coordinate system with the outermost feature point as the vertex. The fingerprint feature to be identified replaces the identified fingerprint feature in the matching group, and the replacement coverage area is calculated; when the replacement coverage area is greater than the original template coverage area, the replaced identified fingerprint feature is determined as the feature to be replaced. The replacement score of the target fingerprint template is calculated based on the fingerprint features to be identified and the features to be replaced, and the template is updated when the replacement score exceeds the score threshold.

2. The method according to claim 1, characterized in that, After determining the target fingerprint template, the method further includes: All identified fingerprint features in the target fingerprint template are placed in the same coordinate system according to the feature point alignment method; Identified fingerprint features with overlapping feature points in the coordinate system are placed into the same feature set to form a continuous feature group; identified fingerprint features with non-overlapping feature points are established into separate feature sets to form several independent feature groups. When the target fingerprint template contains at least two feature groups, fingerprint feature comparison is performed; wherein, the different feature groups are not continuous with each other.

3. The method according to claim 2, characterized in that, When the matching group is the continuous feature group; the fingerprint feature to be identified replaces the identified fingerprint feature in the matching group, and the replacement coverage area is calculated; When the replacement coverage area is larger than the original template coverage area, the identified fingerprint feature to be replaced is determined as the feature to be replaced, including: Replace the i identified fingerprint features in the continuous feature group with the fingerprint features to be identified, and calculate the replacement coverage area of ​​the i replacement matching groups after replacement. The identified fingerprint feature with the largest replacement coverage area, which is larger than the coverage area of ​​the original template, is determined as the feature to be replaced. When the matching group is the independent feature group, if the replacement coverage area is greater than the original template coverage area, it is directly determined as the feature to be replaced.

4. The method according to claim 3, characterized in that, The step of calculating the replacement score of the target fingerprint template based on the fingerprint features to be identified and the features to be replaced, and updating the template when the replacement score exceeds a score threshold, includes: Determine the image quality and feature area of ​​the feature to be replaced and the fingerprint feature to be identified, respectively; When the image quality of the fingerprint feature to be identified is greater than the image threshold and the feature area is greater than the area threshold, a first replacement score is determined based on the image quality and the feature area; when the image quality of the feature to be replaced is greater than the image threshold and the feature area is greater than the area threshold, a second replacement score is determined based on the image quality and the feature area. Among them, the first feature score is directly proportional to the image quality and image area of ​​the fingerprint feature to be identified, and the second feature score is inversely proportional to the image quality and image area of ​​the feature to be replaced; The sum of the first replacement score and the second replacement score is determined as the replacement score of the feature to be replaced; When the replacement score is greater than the score threshold, the fingerprint feature to be identified replaces the feature to be replaced.

5. The method according to claim 4, characterized in that, When the image quality of the feature to be replaced is less than the image threshold, or the feature area is less than the area threshold, while the image quality of the fingerprint feature to be identified is greater than the image threshold and the feature area is greater than the area threshold, the feature to be replaced is directly replaced. If the image quality of the fingerprint feature to be identified is less than the image threshold, or the feature area is less than the area threshold, no replacement will be performed.

6. The method according to claim 5, characterized in that, After updating the target fingerprint template, the method further includes: The replaced matching group is arranged with other feature groups in the target fingerprint template according to their positional relationship, and the overlapping feature groups are aggregated into new continuous feature groups; Obtain and update the number of updates to the target fingerprint template and the number of feature groups of the target fingerprint template; when the number of updates exceeds the update threshold, or when all features are aggregated into one feature group, stop updating the target fingerprint template.

7. A fingerprint template updating device, characterized in that, The device is used in an embedded fingerprint recognition device, and the device includes: The acquisition module is used to acquire the fingerprint image to be identified, extract the fingerprint features to be identified from the image, and determine the target fingerprint template of the target finger by comparing it with the template library. The determining module is used to compare the fingerprint feature to be identified with the identified fingerprint features in the target fingerprint template, and determine the feature to be replaced in the target fingerprint template based on the feature comparison score and the feature template coverage area; including: All identified fingerprint features within the target fingerprint template are compared to determine the identified fingerprint feature with the highest comparison score and its feature group, and this group is determined as the matching group. Calculate the original template coverage area formed by the identified fingerprint features in the matching group; the original template coverage area is the area of ​​the smallest bounding polygon containing all feature points after placing all features in the same coordinate system with the outermost feature point as the vertex. The fingerprint feature to be identified replaces the identified fingerprint feature in the matching group, and the replacement coverage area is calculated; when the replacement coverage area is greater than the original template coverage area, the replaced identified fingerprint feature is determined as the feature to be replaced. The update module is used to calculate the replacement score of the target fingerprint template based on the fingerprint features to be identified and the features to be replaced, and to update the template when the replacement score exceeds the score threshold.

8. An embedded device, characterized in that, The embedded device includes a processor and a memory, the memory storing at least one instruction, at least one program, code set, or instruction set, the at least one instruction, the at least one program, the code set, or the instruction set being loaded and executed by the processor to implement the fingerprint template update method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The readable storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the fingerprint template update method as described in any one of claims 1 to 6.

Images