Fingerprint verification method and device, storage medium and electronic device

Abstract

Embodiments of the present application provide a fingerprint verification method and device, a storage medium and an electronic device, wherein the method comprises: extracting a first feature of a fingerprint to be verified; determining a first similarity between a first global feature of each first minutia included in the first feature and a second global feature of each second minutia included in a second feature of each fingerprint included in a fingerprint library, wherein the second feature of each fingerprint is obtained by performing feature fusion on features in multiple images of the fingerprint; determining a second similarity between the fingerprint to be verified and each fingerprint included in the fingerprint library based on the first similarity; and in a case where there is a similarity in the second similarity that satisfies a predetermined condition, determining that the fingerprint to be verified is verified successfully. Through the present application, the problem of inaccurate fingerprint verification in the related art is solved, and the effect of improving the accuracy of fingerprint verification is achieved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the field of image recognition, and in particular, to a fingerprint verification method and device, a storage medium and an electronic device. BACKGROUND

[0002] With the development and progress of society, the actual demand for fast, effective and automatic personal identification is increasingly urgent. As an important topic of biometric technology, fingerprint recognition technology is increasingly valued by people because of its high uniqueness and strong stability. Fingerprint recognition technology is initially applied to the field of criminal investigation, and in recent years, it has gradually been popularized in people's daily life, such as attendance, access control, safes, etc. Traditional fingerprint recognition technology usually includes steps of fingerprint preprocessing, fingerprint minutia extraction, fingerprint minutia matching, etc.

[0003] However, in the related art, there is an inaccurate problem of fingerprint verification.

[0004] In view of the above problems existing in the related art, no effective solution has been proposed so far. SUMMARY

[0005] Embodiments of the present application provide a fingerprint verification method and device, a storage medium and an electronic device to at least solve the problem of inaccurate fingerprint verification in the related art.

[0006] According to an embodiment of the present application, a fingerprint verification method is provided, comprising: extracting a first feature of a fingerprint to be verified; determining a first similarity between a first global feature of each first minutia included in the first feature and a second global feature of each second minutia in a second feature of each fingerprint included in a fingerprint library, wherein the second feature of each fingerprint is a feature obtained by feature fusion on features in multiple images of the fingerprint; determining a second similarity between the fingerprint to be verified and each fingerprint included in the fingerprint library based on the first similarity; and in the case that there is a similarity satisfying a predetermined condition in the second similarity, determining that the fingerprint to be verified is verified successfully.

[0007] According to another embodiment of the present application, there is provided a fingerprint verification device, comprising: an extraction module configured to extract first features of a fingerprint to be verified; a first determination module configured to determine a first similarity between a first global feature of each first minutia included in the first features and a second global feature of each second minutia included in second features of each fingerprint included in a fingerprint library, wherein the second features of each fingerprint are features obtained by performing feature fusion on features in multiple images of the fingerprint; a second determination module configured to determine a second similarity between the fingerprint to be verified and each fingerprint included in the fingerprint library based on the first similarity; and a verification module configured to determine that the fingerprint to be verified is verified successfully in a case where there is a similarity satisfying a predetermined condition in the second similarity.

[0008] According to still another embodiment of the present application, there is also provided a computer readable storage medium having a computer program stored therein, wherein the computer program is configured to perform the steps in any of the method embodiments above when executed.

[0009] According to still another embodiment of the present application, there is also provided an electronic device comprising a memory and a processor, wherein the memory has a computer program stored therein, and the processor is configured to execute the computer program to perform the steps in any of the method embodiments above.

[0010] According to the present application, the first features of the fingerprint to be verified are extracted, the first similarity between the first global feature of each first minutia included in the first features and the second global feature of each second minutia included in the second features of each fingerprint included in the fingerprint library is determined, the second similarity between the fingerprint to be verified and each fingerprint included in the fingerprint library is determined based on the first similarity, the second features of each fingerprint are features obtained by performing feature fusion on features in multiple images of the fingerprint, and the fingerprint to be verified is determined to be verified successfully in a case where there is a similarity satisfying a predetermined condition in the second similarity. Since the second features of each fingerprint included in the fingerprint library are features obtained by performing feature fusion on features in multiple images of the fingerprint, the randomness and contingency of a single fingerprint image are reduced, and the accuracy of the determined first similarity is improved. When verifying the fingerprint, the second similarity between the fingerprints is determined based on the first similarity between the first global feature of each first minutia and the second global feature of each second minutia, i.e., the features of each first minutia of the fingerprint to be verified are sufficiently fused when determining the second similarity, and the accuracy of the determined second similarity is improved. Therefore, the problem of inaccurate fingerprint verification in the related art can be solved, and the effect of improving the accuracy of fingerprint verification is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0011] Figure 1This is a hardware structure block diagram of a mobile terminal for a fingerprint verification method according to an embodiment of the present invention.

[0012] Figure 2 This is a flowchart of a fingerprint verification method according to an embodiment of the present invention;

[0013] Figure 3 This is a flowchart of fingerprint preprocessing according to an exemplary embodiment of the present invention;

[0014] Figure 4 According to an exemplary embodiment of the present invention, the second similarity between the fingerprint to be verified and each fingerprint included in the fingerprint database is determined.

[0015] Figure 5 This is a schematic diagram of feature fusion according to an exemplary embodiment of the present invention;

[0016] Figure 6 This is a flowchart of a fingerprint verification method according to a specific embodiment of the present invention;

[0017] Figure 7 This is a flowchart of feature verification according to a specific embodiment of the present invention;

[0018] Figure 8 This is a structural block diagram of a fingerprint verification device according to an embodiment of the present invention. Detailed Implementation

[0019] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and examples.

[0020] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0021] The methods and embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a fingerprint verification method according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1more or less components than those shown, or configured differently from those shown, as is Figure 1 described.

[0022] The memory 104 is operable to store a computer program, such as a software program of an application and modules, such as a computer program corresponding to the method for verifying a fingerprint according to an embodiment of the present application. The processor 102 is configured to perform various functions and data processing by running the computer program stored in the memory 104, i.e., to implement the method described above. The memory 104 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 104 can further include a memory remotely disposed relative to the processor 102, which can be connected to the mobile terminal through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0023] The transmission device 106 is configured to receive or send data via a network. Examples of the network include, but are not limited to, a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (NIC), which can be connected to other network devices through a base station so as to be able to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is configured to communicate with the Internet in a wireless manner.

[0024] In the present embodiment, a method for verifying a fingerprint is provided, Figure 2 which is a flowchart of the method for verifying a fingerprint according to an embodiment of the present application, as shown in Figure 2 which includes the following steps:

[0025] In step S202, a first feature of a fingerprint to be verified is extracted.

[0026] In step S204, a first similarity between a first global feature of each first minutia included in the first feature and a second global feature of each second minutia included in a second feature of each fingerprint included in a fingerprint library is determined, wherein the second feature of each fingerprint is a feature obtained by performing feature fusion on features in multiple images of the fingerprint.

[0027] In step S206, a second similarity between the fingerprint to be verified and each fingerprint included in the fingerprint library is determined based on the first similarity.

[0028] Step S208: If there is a similarity in the second similarity that meets the predetermined conditions, it is determined that the fingerprint to be verified has been successfully verified.

[0029] In the above embodiments, the fingerprint to be verified can be a fingerprint collected by a target device. The target device can include smart terminals, fingerprint scanners, and other devices with integrated fingerprint collection functions, such as smartphones, smartwatches, tablets, smart locks, and safes. The fingerprint database can be a fingerprint database stored in the storage unit of the target device. The fingerprint database can include one or more fingerprints and fingerprint features corresponding to each fingerprint. The fingerprints in the fingerprint database can be fingerprints pre-registered with the target device. During fingerprint registration, the first fingerprint collected by the target device can be determined as the reference fingerprint. Fingerprint collection is performed again, and it is verified whether the currently collected fingerprint is the same as the reference fingerprint. For example, the similarity between the collected fingerprint and the reference fingerprint can be used to determine whether the currently collected fingerprint is the same as the reference fingerprint. If the similarity is greater than a preset threshold, they can be considered to be the same fingerprint. When it is determined that they are the same fingerprint, the features of the currently collected fingerprint and the features of the reference fingerprint are fused, and the fused features are determined as the features of the reference fingerprint. After feature fusion, fingerprints can be collected again, and it can be verified whether the collected fingerprint is the same as the reference fingerprint. If they are the same fingerprint, fingerprint feature fusion is performed. When the number of fused fingerprints reaches a predetermined number, the final feature can be identified as the feature of the reference fingerprint and stored in the fingerprint database.

[0030] In the above embodiments, fingerprint images can be acquired by pressing on a target device, such as a fingerprint scanner, according to the acquisition requirements. After fingerprint acquisition is completed, the acquired fingerprint image can be preprocessed. Fingerprint preprocessing may include steps such as segmentation, orientation field and frequency field calculation, enhancement, binarization, and thinning. A flowchart of the fingerprint preprocessing process can be found in the appendix. Figure 3 After fingerprint preprocessing, feature extraction can be performed on the preprocessed fingerprint image. Fingerprint feature extraction typically refers to extracting and recording minutiae information. That is, the first feature includes the minutiae information of each minutiae and the global feature of each minutiae. Minutiae information can be represented as M = {m1, m2, ... m}. k}, m i ={x i y i θ i Let M represent the set of extracted minutiae, m represent a single minutiae, x represent the x-coordinate of a minutiae, y represent the y-coordinate of a minutiae, and θ represent the direction of the minutiae. Global features can be features determined based on minutiae information.

[0031] In the above embodiment, after the first feature of the to-be-verified fingerprint is extracted, feature comparison can be performed. That is, a first similarity between a first global feature of each first minutia included in the first feature and a second global feature of each second minutia included in a second feature of each fingerprint in the fingerprint library is determined, and a second similarity between the to-be-verified fingerprint and each fingerprint in the fingerprint library is determined according to the first similarity. When a maximum similarity in the second similarity is greater than a preset similarity, it can be considered that there is a similarity in the second similarity that meets a predetermined condition, and then it is determined that the to-be-verified fingerprint is verified successfully. That is, after the feature comparison, a similarity set S and a maximum similarity Stop1 of the to-be-verified fingerprint and all fingerprints in the fingerprint library can be obtained, and then the maximum similarity is compared with a set similarity threshold. If the maximum similarity is greater than the threshold, the verification is passed; otherwise, the verification is failed. The verification manner can be

[0032] S = {S1, S2,... S k}, S top1 = max (S)

[0033] is represented as

[0034] The execution subject of the above steps can be a target device, a processor, or a device integrated with a fingerprint collection device and a data processing device, but is not limited thereto.

[0035] According to the present application, the first feature of the to-be-verified fingerprint is extracted, the first similarity between the first global feature of each first minutia included in the first feature and the second global feature of each second minutia included in the second feature of each fingerprint in the fingerprint library is determined, the second similarity between the to-be-verified fingerprint and each fingerprint included in the fingerprint library is determined according to the first similarity, the second feature of each fingerprint is a feature obtained by feature fusion on features in multiple images of the fingerprint, and in the case that there is a similarity in the second similarity that meets a predetermined condition, it is determined that the to-be-verified fingerprint is verified successfully. Since the second feature of each fingerprint in the fingerprint library fuses the features of multiple images of the fingerprint, the randomness and contingency of a single fingerprint image are reduced, and the accuracy of the determined first similarity is improved. When verifying the fingerprint, the second similarity between the fingerprints is determined according to the first similarity between the first global feature of each first minutia and the second global feature of each second minutia, that is, when determining the second similarity, the features of each first minutia of the to-be-verified fingerprint are fully fused, and the accuracy of the determined second similarity is improved. Therefore, the problem of inaccurate fingerprint verification in the related art can be solved, and the effect of improving the accuracy of fingerprint verification is achieved.

[0036] In an exemplary embodiment, determining a first similarity between a first global feature of each first minutiae included in the first feature and a second global feature of each second minutiae in a second feature of each fingerprint included in a fingerprint database includes: performing the following operations for each first minutiae included in the first feature to obtain a first global feature similar to each first minutiae: determining a second minutiae included in the first feature that is closest to the first minutiae; determining a first distance between the first minutiae and the second minutiae; determining a first angle between a first line connecting the first minutiae and the second minutiae and the direction of the first minutiae; and determining a second similarity between the first line connecting the first minutiae and the direction of the second minutiae. The first distance, the first included angle, and the second included angle are determined as the first global feature. For each first detail point, the following operations are performed on the first global feature to obtain the first similarity between the first global feature and each second global feature: The first arithmetic square root of the square of the difference between the first distance and the second distance included in the second global feature is determined; the second arithmetic square root of the square of the difference between the first included angle and the third included angle included in the second global feature is determined; the third arithmetic square root of the square of the difference between the second included angle and the fourth included angle included in the second global feature is determined; and the first similarity is determined based on the first arithmetic square root, the second arithmetic square root, and the third arithmetic square root. In this embodiment, for each first detail point in the first feature, a first global feature for each first detail point can be determined. When determining the first global feature, the second detail point closest to the first detail point can be found in the first feature, and the first distance, the first included angle, and the second included angle between the first detail point and the second detail point are determined as the first global feature. For example, for detail point m... i Find the nearest detail point m. j Then, the distance d, the included angle α, and the included angle β are used as detail points m. i global features f i ={d i α i ,β i After determining the global features, the first similarity of all minutiae pairs of two fingerprints can be calculated based on the global features of the minutiae. The first similarity can be determined based on the first square root of the square of the difference between the first distance and the second distance included in the second global features, the second square root of the square of the difference between the first angle and the third angle included in the second global features, and the third square root of the square of the difference between the second angle and the fourth angle included in the second global features. The first square root can be expressed as... The second arithmetic square root can be expressed as... The third arithmetic square root can be expressed as...

[0037] In an example embodiment, determining the first similarity based on the first arithmetical square root, the second arithmetical square root and the third arithmetical square root comprises: determining a first sum value of the first arithmetical square root, the second arithmetical square root and the third arithmetical square root; determining the first similarity as the first sum value. In this embodiment, the first similarity can be expressed as s(i, j) = d(i, j) + a(i, j) + b(i, j).

[0038] In an example embodiment, determining the second similarity between the to-be-verified fingerprint and each fingerprint included in the fingerprint library based on the first similarity comprises: performing the following operations for each first fingerprint included in the fingerprint library to determine the second similarity between the to-be-verified fingerprint and the first fingerprint: matching the second minutia included in the first fingerprint with the first minutia based on the first similarity to obtain a plurality of first target minutia pairs; determining a target number of the first target minutia pairs; determining a third similarity corresponding to each of the first target minutia pairs included in the first similarity; determining a second sum value of the third similarities; and determining the second similarity as a ratio of the second sum value to the target number. In this embodiment, the matching result of the minutia of the two fingerprints can be obtained by using the Hungarian matching according to the first similarity of the calculated minutia pairs, and the similarity of the two fingerprints can be calculated according to the matching result. The second minutia matched with the first minutia among the minutia in each first fingerprint can be determined according to the Hungarian matching, and the first target minutia pair can be determined by the first minutia and the second minutia. The similarity between the first minutia and the second minutia can be greater than the similarity between the other first minutia and other minutia in the first fingerprint except the second minutia. Alternatively, the similarity between the first minutia and the second minutia can be greater than the similarity between the other first minutia and other minutia in the first fingerprint except the second minutia, and the similarity is greater than a preset threshold.

[0039] In the above embodiment, after the plurality of first target minutia pairs are determined, the target number of the first target minutia pairs can be determined, the third similarity corresponding to each of the first target minutia pairs can be determined, and the second similarity can be determined as a ratio of a second sum value of the plurality of third similarities to the target number. The second similarity can be expressed as wherein n and m respectively represent the number of the minutia of the two fingerprints, and p represents the minimum value of n and m, i.e., the target number. The flowchart of determining the second similarity between the to-be-verified fingerprint and each fingerprint included in the fingerprint library can be seen in FIG. 2. Figure 4 .

[0040] In one example embodiment, after determining that the to-be-verified fingerprint verification is successful, the method further comprises: determining a second fingerprint corresponding to the maximum similarity included in the second similarity in the fingerprint library; matching minutiae included in the second fingerprint with the first minutiae included in the to-be-verified fingerprint to obtain a second target minutiae pair; determining a fourth similarity corresponding to each of the second target minutiae pair included in the first similarity; determining a fifth similarity included in the fourth similarity and less than a first predetermined threshold, and a sixth similarity included in the fourth similarity and greater than or equal to the first predetermined threshold; determining a third target minutiae pair corresponding to the fifth similarity and a fourth target minutiae pair corresponding to the sixth similarity in the second target minutiae pair; adding the features of the minutiae of the to-be-verified fingerprint included in the third target minutiae pair to the second features of the second fingerprint; fusing the features corresponding to the minutiae included in the fourth target minutiae pair to obtain first fused features; and updating the features corresponding to the fourth target minutiae pair included in the second features to the first fused features. In this embodiment, after the to-be-verified fingerprint is verified successfully, the features of the to-be-verified fingerprint can be fused with the features of the greatest similarity in the feature library. The fused features replace the corresponding original features in the feature library. The feature fusion schematic diagram can be referred to in the description of the feature fusion schematic diagram. Figure 5 As shown in Figure 5 , the dashed line represents the minutiae matching relationship, and the numerical value represents the minutiae similarity. For the minutiae in the to-be-fused features and less than the first predetermined threshold, the minutiae are directly added to the reference features. For the minutiae greater than or equal to the first predetermined threshold, the minutiae are fused with the corresponding minutiae in the reference features. Until all the minutiae pairs in the matching result are traversed, the feature fusion is completed, and the final fusion result is obtained.

[0041] In the above embodiment, the feature update strategy is used after verification, the feature template is updated in real time, the stability of the feature template is enhanced, the recognition rate is improved, and the false recognition rate is reduced.

[0042] In an example embodiment, fusing the features corresponding to the minutia pairs included in the fourth target minutia pair to obtain the first fused feature includes: determining a third feature of a first sub-minutia and a fourth feature of a second sub-minutia included in the fourth target minutia pair; determining a first average value of a first horizontal coordinate included in the third feature and a second horizontal coordinate included in the fourth feature; determining a second average value of a first vertical coordinate included in the third feature and a second vertical coordinate included in the fourth feature; determining a third average value of a first direction angle included in the third feature and a second direction angle included in the fourth feature; and determining a feature including the first average value, the second average value, and the third average value as the first fused feature. In this embodiment, the feature fusion can be the fusion of the minutia information of the minutia, i.e., the third feature and the fourth feature can be the minutia information. The third feature can be represented as m i =(x i ,y i ,θ i ), and the fourth feature can be represented as m j =(x j ,y j ,θ j ). Then the first fused feature can be represented as

[0043]

[0044] In one example embodiment, before determining the first similarity between the first global feature of each first minutia included in the first feature and the second global feature of each second minutia in the second feature of each fingerprint included in the fingerprint library, the method further comprises: obtaining a to-be-registered fingerprint feature of a to-be-registered fingerprint; determining a seventh similarity between the to-be-registered fingerprint and a reference fingerprint based on the to-be-registered fingerprint feature, wherein the reference fingerprint is a fingerprint collected for the first time in a registration process; in a case where the seventh similarity is greater than a third predetermined threshold, fusing the to-be-registered fingerprint feature and a reference fingerprint feature of the reference fingerprint to obtain a second fused feature; updating the second fused feature as the reference fingerprint feature; and registering the reference fingerprint and the reference fingerprint feature into the fingerprint library. In this embodiment, fingerprint registration can be performed before fingerprint verification. First, fingerprint collection is performed, and it is determined whether the collected fingerprint is the first fingerprint. When the collected fingerprint is the first fingerprint, the fingerprint is determined as the reference fingerprint, and the feature of the fingerprint is determined as the reference fingerprint feature. When the collected fingerprint is not the first fingerprint, the fingerprint is determined as the to-be-registered fingerprint, and a seventh similarity between a to-be-registered fingerprint feature of the to-be-registered fingerprint and the reference fingerprint feature is determined. That is, if the extracted feature is the feature of the first fingerprint, the feature is taken as the reference feature and directly verified. Otherwise, the feature is taken as the to-be-verified feature and compared with the reference feature, and the similarity obtained by the feature comparison is compared with a set threshold. If the similarity is greater than the threshold, the verification is passed. Otherwise, the verification is failed, until the number of features that pass the verification reaches the number of features required for feature fusion. Then, the to-be-registered fingerprint feature is fused with the reference fingerprint feature one by one, the fused feature is determined as the reference fingerprint feature, and the reference fingerprint feature is registered into the fingerprint library.

[0045] The fingerprint verification method will be described below in combination with a specific embodiment:

[0046] Figure 6 is a flowchart of the fingerprint verification method according to the specific embodiment of the present application, as shown in the figure, the flowchart includes a registration stage and a verification stage, and includes: Figure 6

[0047] 1. Registration stage

[0048] 1.1 Fingerprint collection

[0049] The fingerprint image is obtained by pressing on the fingerprint collection instrument according to the collection requirements.

[0050] 1.2 Preprocessing

[0051] ​After fingerprint acquisition, the acquired fingerprint image needs to be preprocessed. Fingerprint preprocessing typically includes steps such as segmentation, orientation field and frequency field calculation, enhancement, binarization, and thinning.

[0052] 1.3 Feature Extraction

[0053] After fingerprint preprocessing, feature extraction is required from the preprocessed fingerprint image. Fingerprint feature extraction typically refers to extracting and recording minutiae information:

[0054] M = {m1, m2, ... m} k}, m i ={x i y i θ i}, i = 1, 2, ..., k,

[0055] Where M represents the set of extracted minutiae, m represents a single minutiae, x represents the x-coordinate of a minutiae, y represents the y-coordinate of a minutiae, and θ represents the direction of a minutiae.

[0056] 1.4 Feature Verification

[0057] The feature verification process can be found in the attached document. Figure 7 ,like Figure 7 As shown, if the extracted feature is the same as the first fingerprint feature, then this feature is used as the baseline feature and passes the verification directly. Otherwise, this feature is used as the feature to be verified and compared with the baseline feature. Then, the similarity obtained from the feature comparison is compared with a set threshold. If the similarity is greater than the threshold, the verification passes; otherwise, the verification fails. This process continues until the number of verified features reaches the number of features required for feature fusion.

[0058] 1.4.1 Feature Comparison

[0059] The feature comparison flowchart can be found in the appendix. Figure 4 First, global features for minutiae are constructed for each of the two features (for minutiae m). i Find the nearest detail point m. j Then, the distance d, the included angle α, and the included angle β are used as global features f of the detail point mi. i ):

[0060] f i ={d i α i ,β i}

[0061] Then, based on the global features of the minutiae, the similarity of all pairs of minutiae between the two fingerprints is calculated:

[0062]

[0063] s(i, j) = d(i, j) + a(i, j) + β(i, j)

[0064] Then according to the similarity of the minutia pairs calculated, the matching results of the two fingerprint minutia are obtained by using the Hungarian matching and recorded, and finally the similarity of the two fingerprints is calculated according to the matching results:

[0065] Wherein, n and m represent the number of minutia of two fingerprints respectively, and p represents the minimum value in n and m.

[0066] 1.5 Feature fusion

[0067] According to the matching results of the two fingerprint minutia recorded by the feature comparison (wherein, the dashed line represents the minutia matching relationship, and the numerical value represents the minutia similarity), for the minutia in the to-be-fused feature whose minutia similarity is less than the threshold value (0.6), it is directly added to the reference feature, and for the minutia greater than the threshold value, it is fused with the corresponding minutia in the reference feature), the fusion mode is as follows:

[0068] m i = {x i , y i , θ i}, m j = {x j , y j , θ j}

[0069]

[0070] Until all the minutia pairs in the matching results are traversed, the feature fusion is completed, and the final fusion result is obtained.

[0071] 2. Verification stage

[0072] 2.1 Fingerprint collection

[0073] This step is the same as step 1.1.

[0074] 2.2 Preprocessing

[0075] This step is the same as step 1.2.

[0076] 2.3 Feature extraction

[0077] This step is the same as step 1.3.

[0078] 2.4 Feature comparison

[0079] Compare the features of the verification fingerprint with all the features of the fingerprint in the feature library. The specific process of feature comparison is the same as step 1.4.1.

[0080] 2.5 Decision

[0081] After the feature comparison, the similarity set S and the maximum similarity Stop1 between the verification fingerprint and all the fingerprints in the feature library can be obtained. Then the maximum similarity is compared with the set similarity threshold. If the maximum similarity is greater than the threshold, the verification is passed; otherwise, the verification is failed.

[0082] 2.6 Feature fusion

[0083] After the fingerprint verification is passed, the verification fingerprint feature is fused with the feature with the maximum similarity in the feature library. The specific process of the feature fusion is the same as step 1.5.

[0084] 2.7 Feature update

[0085] The fused feature is replaced with the corresponding original feature in the feature library.

[0086] In the foregoing embodiments, the multi-fingerprint fusion registration scheme is adopted, the randomness and contingency in the registration stage are reduced, the recognition rate is improved, and the false recognition rate is reduced. The check mechanism is introduced in the multi-fingerprint fusion registration scheme, and the fusion method independent of the minutia alignment is adopted, the accuracy of the fused template is enhanced, the recognition rate is improved, and the false recognition rate is reduced. After the verification is passed, the feature update strategy is used to update the feature template in real time, the stability of the feature template is enhanced, the recognition rate is improved, and the false recognition rate is reduced.

[0087] Those skilled in the art can clearly understand the method according to the foregoing embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as a ROM / RAM, a magnetic disk, or an optical disk) and includes a plurality of instructions for causing a terminal device (which can be a mobile phone, a computer, a server, or a network device) to execute the methods described in the various embodiments of the present application.

[0088] In the present embodiment, a fingerprint verification device is also provided, which is used to implement the foregoing embodiments and preferred embodiments, and will not be described again. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably realized in software, hardware, or a combination of software and hardware is also possible and contemplated.

[0089] Figure 8 is a structure block diagram of the fingerprint verification device according to the embodiments of the present application, asFigure 8 The apparatus includes, as shown in the figure,

[0090] The extraction module 82 is configured to extract a first feature of a to-be-verified fingerprint;

[0091] The first determination module 84 is configured to determine a first similarity between a first global feature of each first minutia included in the first feature and a second global feature of each second minutia included in a second feature of each fingerprint included in a fingerprint library, wherein the second feature of each fingerprint is a feature obtained by performing feature fusion on features in multiple images of the fingerprint;

[0092] The second determination module 86 is configured to determine a second similarity between the to-be-verified fingerprint and each fingerprint included in the fingerprint library based on the first similarity;

[0093] The verification module 88 is configured to determine that the to-be-verified fingerprint is verified successfully in a case where there is a similarity satisfying a predetermined condition in the second similarity.

[0094] In an example embodiment, the first determination module 84 can determine the first similarity between the first global feature of each first minutia included in the first feature and the second global feature of each second minutia included in the second feature of each fingerprint included in the fingerprint library by performing the following operations for each first minutia included in the first feature to obtain the first global feature of each first minutia: determining a second minutia included in the first feature closest to the first minutia, determining a first distance between the first minutia and the second minutia, determining a first included angle between a first line connecting the first minutia and the second minutia and a direction of the first minutia, and determining a second included angle between the first line and a direction of the second minutia, determining the first distance, the first included angle, and the second included angle as the first global feature; performing the following operations for the first global feature of each first minutia to obtain the first similarity between the first global feature and each second global feature: determining a first arithmetic square root of a square of a difference between the first distance and a second distance included in the second global feature, determining a second arithmetic square root of a square of a difference between the first included angle and a third included angle included in the second global feature, determining a third arithmetic square root of a square of a difference between the second included angle and a fourth included angle included in the second global feature, and determining the first similarity based on the first arithmetic square root, the second arithmetic square root, and the third arithmetic square root.

[0095] In an example embodiment, the first determining module 84 can determine the first similarity based on the first arithmetical square root, the second arithmetical square root and the third arithmetical square root by determining a first sum value of the first arithmetical square root, the second arithmetical square root and the third arithmetical square root, and determining the first sum value as the first similarity.

[0096] In an example embodiment, the second determining module 86 can determine the second similarity between the to-be-verified fingerprint and each fingerprint included in the fingerprint library based on the first similarity by performing the following operations for each first fingerprint included in the fingerprint library to determine the second similarity between the to-be-verified fingerprint and the first fingerprint: matching the second minutia included in the first fingerprint with the first minutia based on the first similarity to obtain a plurality of first target minutia pairs; determining a target number of the first target minutia pairs; determining a third similarity corresponding to each of the first target minutia pairs included in the first similarity; determining a second sum value of the third similarity; and determining a ratio of the second sum value to the target number as the second similarity.

[0097] In an example embodiment, the apparatus can be configured to, after determining that the to-be-verified fingerprint is verified successfully, determine a second fingerprint included in the fingerprint library and corresponding to a maximum similarity included in the second similarity; match minutia included in the second fingerprint with the first minutia included in the to-be-verified fingerprint to obtain a second target minutia pair; determine a fourth similarity corresponding to each of the second target minutia pair included in the first similarity; determine a fifth similarity smaller than a first predetermined threshold and a sixth similarity greater than or equal to the first predetermined threshold included in the fourth similarity; determine a third target minutia pair corresponding to the fifth similarity and a fourth target minutia pair corresponding to the sixth similarity included in the second target minutia pair; add a feature of minutia of the to-be-verified fingerprint included in the third target minutia pair to the second feature of the second fingerprint; fuse features corresponding to minutia included in the fourth target minutia pair to obtain a first fused feature; and update the feature corresponding to the fourth target minutia pair included in the second feature as the first fused feature.

[0098] In an example embodiment, the apparatus can realize the fusion of the features corresponding to the minutia pair included in the fourth target minutia pair to obtain a first fused feature by determining a third feature of a first sub-minutia and a fourth feature of a second sub-minutia included in the fourth target minutia pair, determining a first average of a first abscissa included in the third feature and a second abscissa included in the fourth feature, determining a second average of a first ordinate included in the third feature and a second ordinate included in the fourth feature, determining a third average of a first direction angle included in the third feature and a second direction angle included in the fourth feature, and determining a feature including the first average, the second average, and the third average as the first fused feature.

[0099] In an example embodiment, the apparatus can further be configured to, before determining the first similarity between the first global feature of each first minutia included in the first feature and the second global feature of each second minutia included in the second feature of each fingerprint included in the fingerprint library, acquire a to-be-registered fingerprint feature of a to-be-registered fingerprint, determine a seventh similarity between the to-be-registered fingerprint and a reference fingerprint based on the to-be-registered fingerprint feature, wherein the reference fingerprint is a fingerprint collected for the first time in a registration process, and in a case where the seventh similarity is greater than a third predetermined threshold, fuse the to-be-registered fingerprint feature and a reference fingerprint feature of the reference fingerprint to obtain a second fused feature, update the second fused feature as the reference fingerprint feature, and register the reference fingerprint and the reference fingerprint feature into the fingerprint library.

[0100] It should be noted that each of the above modules can be realized by software or hardware, and for the latter, the following implementation manners can be used, but are not limited thereto: all of the above modules are located in the same processor; or each of the above modules is located in a different processor in an arbitrary combination.

[0101] Embodiments of the present application also provide a computer readable storage medium having a computer program stored therein, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.

[0102] In an example embodiment, the above computer readable storage medium can include, but is not limited to, a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0103] The embodiment of the present application also provides an electronic device, comprising a memory and a processor, the memory stores a computer program, and the processor is arranged to run the computer program to execute the steps in any of the above method embodiments.

[0104] In an exemplary embodiment, the electronic device described above can further comprise a transmission device connected with the processor and an input and output device connected with the processor.

[0105] The specific examples in the embodiment can refer to the examples described in the above embodiments and exemplary implementation manners, and the embodiment will not be described here again.

[0106] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be realized by general computing devices, which can be concentrated on a single computing device or distributed on a network composed of multiple computing devices, which can be realized by program codes executable by the computing devices, so that they can be stored in storage devices and executed by the computing devices, and in some cases, the steps shown or described can be executed in different order, or they can be manufactured into individual integrated circuit modules, or multiple modules or steps among them can be manufactured into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.

[0107] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A fingerprint verification method, characterized in that, include: Extract the first feature of the fingerprint to be verified; Determine a first similarity between a first global feature of each first minutiae included in the first feature and a second global feature of each second minutiae included in the fingerprint database, wherein the second feature of each fingerprint is a feature obtained by feature fusion of features from multiple images of the fingerprint; A second similarity is determined between the fingerprint to be verified and each fingerprint included in the fingerprint database based on the first similarity. If there is a similarity in the second similarity that meets the predetermined conditions, it is determined that the fingerprint to be verified has been successfully verified; After confirming that the fingerprint to be verified has been successfully verified, the method further includes: determining a second fingerprint in the fingerprint database that corresponds to the maximum similarity included in the second similarity; matching the minutiae included in the second fingerprint with the first minutiae included in the fingerprint to be verified to obtain a second target minutiae pair; determining a fourth similarity included in the first similarity that corresponds to each second target minutiae pair; determining a fifth similarity included in the fourth similarity that is less than a first predetermined threshold, and a sixth similarity included in the fourth similarity that is greater than or equal to the first predetermined threshold; determining a third target minutiae pair included in the second target minutiae pair that corresponds to the fifth similarity, and a fourth target minutiae pair that corresponds to the sixth similarity; adding the features of the minutiae of the fingerprint to be verified included in the third target minutiae pair to the second features of the second fingerprint; fusing the features corresponding to the minutiae included in the fourth target minutiae pair to obtain a first fused feature; and updating the features included in the second feature that correspond to the fourth target minutiae pair to the first fused feature.

2. The method according to claim 1, characterized in that, Determining the first similarity between the first global feature of each first minutiae included in the first feature and the second global feature of each second minutiae included in the second feature of each fingerprint in the fingerprint database includes: For each first detail point included in the first feature, the following operations are performed to obtain the first global feature with each first detail point: determine the second detail point included in the first feature that is closest to the first detail point, determine the first distance between the first detail point and the second detail point, determine the first angle between the first line connecting the first detail point and the second detail point and the direction of the first detail point, and determine the second angle between the first line connecting the first detail point and the direction of the second detail point, and determine the first distance, the first angle and the second angle as the first global feature; For each of the first global features of the first detail point, the following operations are performed to obtain the first similarity between the first global feature and each of the second global features; determine the first arithmetic square root of the square of the difference between the first distance and the second distance included in the second global feature; determine the second arithmetic square root of the square of the difference between the first included angle and the third included angle included in the second global feature; determine the third arithmetic square root of the square of the difference between the second included angle and the fourth included angle included in the second global feature; and determine the first similarity based on the first arithmetic square root, the second arithmetic square root, and the third arithmetic square root.

3. The method according to claim 2, characterized in that, Determining the first similarity based on the first arithmetic square root, the second arithmetic square root, and the third arithmetic square root includes: Determine the first sum of the first arithmetic square root, the second arithmetic square root, and the third arithmetic square root; The first sum is determined as the first similarity.

4. The method according to claim 1, characterized in that, Determining a second similarity between the fingerprint to be verified and each fingerprint included in the fingerprint database based on the first similarity includes: For each first fingerprint included in the fingerprint database, the following operations are performed to determine the second similarity between the fingerprint to be verified and the first fingerprint: Based on the first similarity, the second minutiae included in the first fingerprint are matched with the first minutiae to obtain multiple first target minutiae pairs; Determine the number of targets in the first target detail point pair; Determine the third similarity corresponding to each pair of first target detail points included in the first similarity; Determine the second sum value of the third similarity; The ratio of the second sum to the target quantity is determined as the second similarity.

5. The method according to claim 1, characterized in that, The first fused feature, obtained by fusing the features corresponding to the minutiae points included in the fourth target minutiae pair, includes: Determine the third feature of the first sub-minutive point and the fourth feature of the second sub-minutive point included in the fourth target minutiae pair; Determine the first average value of the first horizontal coordinate included in the third feature and the second horizontal coordinate included in the fourth feature; Determine the second average value of the first ordinate included in the third feature and the second ordinate included in the fourth feature; Determine the third average value of the first direction angle included in the third feature and the second direction angle included in the fourth feature; The feature including the first average value, the second average value, and the third average value is determined as the first fusion feature.

6. The method according to claim 1, characterized in that, Before determining the first similarity between the first global feature of each first minutiae included in the first feature and the second global feature of each second minutiae included in the second feature of each fingerprint in the fingerprint database, the method further includes: Obtain the fingerprint features to be registered; The seventh similarity between the fingerprint to be registered and the reference fingerprint is determined based on the features of the fingerprint to be registered, wherein the reference fingerprint is the fingerprint collected for the first time during a registration process; If the seventh similarity is greater than the third predetermined threshold, the fingerprint feature to be registered is fused with the reference fingerprint feature of the reference fingerprint to obtain the second fused feature; Update the second fused feature with the reference fingerprint feature; The reference fingerprint and its features are registered in the fingerprint database.

7. A fingerprint verification device, characterized in that, include: The extraction module is used to extract the first feature of the fingerprint to be verified; The first determining module is used to determine a first similarity between a first global feature of each first minutiae included in the first feature and a second global feature of each second minutiae included in the second feature of each fingerprint in the fingerprint database, wherein the second feature of each fingerprint is a feature obtained by feature fusion of features in multiple images of the fingerprint; The second determining module is used to determine a second similarity between the fingerprint to be verified and each fingerprint included in the fingerprint database based on the first similarity. The verification module is used to determine that the fingerprint to be verified is successfully verified if there is a similarity in the second similarity that meets a predetermined condition. The device is further configured to, after determining that the fingerprint to be verified has been successfully verified: determine a second fingerprint in the fingerprint database that corresponds to the maximum similarity included in the second similarity; match the minutiae included in the second fingerprint with the first minutiae included in the fingerprint to be verified to obtain a second target minutiae pair; determine a fourth similarity included in the first similarity that corresponds to each second target minutiae pair; determine a fifth similarity included in the fourth similarity that is less than a first predetermined threshold, and a sixth similarity included in the fourth similarity that is greater than or equal to the first predetermined threshold; determine a third target minutiae pair included in the second target minutiae pair that corresponds to the fifth similarity, and a fourth target minutiae pair that corresponds to the sixth similarity; add the features of the minutiae of the fingerprint to be verified included in the third target minutiae pair to the second features of the second fingerprint; fuse the features corresponding to the minutiae included in the fourth target minutiae pair to obtain a first fused feature; and update the features included in the second feature that correspond to the fourth target minutiae pair to the first fused feature.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program is configured to perform the method described in any one of claims 1 to 6 when executed.

9. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the method as described in any one of claims 1 to 6.

Images