A fingerprint minutia point labeling method, device and electronic equipment

Abstract

The application discloses a fingerprint minutia point marking method and device and electronic equipment, and the method comprises the following steps: acquiring a to-be-marked fingerprint image, and searching for a marked fingerprint image of the same finger from a preset fingerprint image database according to finger information of the to-be-marked fingerprint image; mapping the minutia point information that has been marked in the marked fingerprint image to the to-be-marked fingerprint image; generating a new marked fingerprint image based on the mapped to-be-marked fingerprint image, and storing the new marked fingerprint image in the fingerprint image database. The technical scheme provided by the application can improve the efficiency and accuracy of fingerprint minutia point marking.

Description

Technical Field

[0001] This invention relates to the field of data processing, and more specifically to a method, apparatus, and electronic device for fingerprint minutiae annotation. Background Technology

[0002] Fingerprint images are one of the most important biometric information sources for identity verification. Most existing automatic fingerprint recognition technologies are based on machine learning, relying on the learning and discrimination of fingerprint features. Supervised learning is an important learning method, which requires sample images and labeled information as input to train the system. The accuracy of the labeled information directly affects the system's performance. Therefore, sample information labeling is a crucial task.

[0003] Minutos are an important feature of fingerprints. Most existing high-accuracy automatic fingerprint recognition technologies are based on minutos. Minutos typically possess multiple attributes, including coordinates, orientation, type, and quality, and are easily affected by the quality of the fingerprint image acquisition. Minutos annotation is highly complex, often requiring experienced experts, and variations in annotation methods can lead to discrepancies. Furthermore, a typical planar fingerprint image usually contains dozens of minutos, while a large rolling fingerprint may have over a hundred, making minutos annotation a daunting and inefficient task. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a fingerprint minutiae annotation method, apparatus, and electronic device, thereby improving the efficiency and accuracy of fingerprint minutiae annotation.

[0005] According to a first aspect, embodiments of the present invention provide a fingerprint minutiae annotation method, the method comprising: acquiring a fingerprint image to be annotated, and searching for an annotated fingerprint image of the same finger from a preset fingerprint image database based on the finger information identifier of the fingerprint image to be annotated; mapping the annotated minutiae information in the annotated fingerprint image to the fingerprint image to be annotated; generating a new annotated fingerprint image based on the mapped fingerprint image to be annotated, and storing the new annotated fingerprint image in the fingerprint image database.

[0006] Optionally, mapping the marked minutiae information in the marked fingerprint image to the fingerprint image to be marked includes: aligning the marked fingerprint image and the fingerprint image to be marked according to fingerprint correlation, and calculating the transformation matrix between the marked fingerprint image and the fingerprint image to be marked; and mapping the marked minutiae information in the marked fingerprint image to the fingerprint image to be marked using the transformation matrix.

[0007] Optionally, the minutiae information includes the coordinates of each minutiae, the orientation of each minutiae, the attribute information of each minutiae, and the neighborhood image of each minutiae; the neighborhood image is an image cropped from the labeled fingerprint image with a preset radius centered on the coordinates of the minutiae; the step of mapping the labeled minutiae information in the labeled fingerprint image to the fingerprint image to be labeled using the transformation matrix includes: mapping the coordinates of each minutiae, the orientation of each minutiae, and the corresponding neighborhood image of each minutiae from the labeled fingerprint image to the fingerprint image to be labeled using the transformation matrix; calculating the neighborhood transformation matrix between each neighborhood image and the fingerprint image to be labeled based on the alignment relationship between the neighborhood images corresponding to each minutiae; adjusting the coordinates and orientation of the minutiae mapped to the fingerprint image to be labeled using the neighborhood transformation matrix; and adding the attribute information to each minutiae mapped to the fingerprint image to be labeled.

[0008] Optionally, if multiple labeled fingerprint images of the same finger are found from a preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled, the method further includes: mapping the labeled minutiae information in each labeled fingerprint image to the fingerprint image to be labeled, obtaining multiple minutiae mapping information; calculating the similarity between each minutiae mapping information, and fusing the minutiae mapping information with a similarity greater than a preset similarity threshold; generating a new labeled fingerprint image based on the fingerprint image to be labeled with the fused minutiae mapping information, and storing the new labeled fingerprint image in the fingerprint image database.

[0009] Optionally, calculating the similarity between the various minutiae mapping information includes:

[0010] The similarity between minutiae in the two minutiae mapping information is calculated using the following formula.

[0011]

[0012] In the formula, S(m1, m2) represents the similarity between detail points m1 and m2, x1 and y1 represent the x and y coordinates of detail point m1, respectively, x2 and y2 represent the x and y coordinates of detail point m2, respectively, θ1 and θ2 represent the directions of detail points m1 and m2, respectively, p1 and p2 represent the attribute information of detail points m1 and m2, respectively, and d x d y d θ All parameters are preset.

[0013] Optionally, if no labeled fingerprint image of the same finger is found in a preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled, the method further includes: labeling the coordinates and directions of several minutiae in the fingerprint image to be labeled according to the input labeling instruction; creating a first neighborhood image for each minutiae labeled in the fingerprint image to be labeled according to the labeling instruction; extracting a second neighborhood image with the highest similarity to the first neighborhood image from the minutiae information of the labeled fingerprint image; and using the attribute information of the minutiae corresponding to the second neighborhood image as pseudo-labeling information of the minutiae corresponding to the first neighborhood image.

[0014] Optionally, the method further includes: when creating a neighborhood image corresponding to the current detail point, rotating the created neighborhood image based on the direction of the current detail point to obtain another neighborhood image of the current detail point.

[0015] According to a second aspect, embodiments of the present invention provide a fingerprint minutiae annotation device, the device comprising: a fingerprint image matching module, configured to acquire a fingerprint image to be annotated, and search for an annotated fingerprint image of the same finger from a preset fingerprint image database based on the finger information identifier of the fingerprint image to be annotated; a minutiae mapping module, configured to map the minutiae information already annotated in the annotated fingerprint image to the fingerprint image to be annotated; and an annotation result saving module, configured to generate a new annotated fingerprint image based on the mapped fingerprint image to be annotated, and store the new annotated fingerprint image in the fingerprint image database.

[0016] According to a third aspect, embodiments of the present invention provide an electronic device, including: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the method described in the first aspect, or any optional embodiment of the first aspect.

[0017] According to a fourth aspect, embodiments of the present invention provide a computer-readable storage medium storing computer instructions for causing the computer to perform the method described in the first aspect, or any alternative embodiment of the first aspect.

[0018] The technical solution provided in this application has the following advantages:

[0019] The technical solution provided in this application first acquires a fingerprint image to be labeled, and then searches for labeled fingerprint images of the same finger from a preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled; then, it maps the already labeled minutiae information in the labeled fingerprint images to the fingerprint image to be labeled; thereby generating a new labeled fingerprint image based on the mapped fingerprint image to be labeled, and storing the new labeled fingerprint image in the fingerprint image database. This application utilizes the characteristic that fingerprint databases usually collect multiple samples from the same finger, and that the minutiae have high similarity, and proposes a minutiae annotation method based on already labeled minutiae to be labeled fingerprint images, making minutiae annotation more accurate and efficient. Attached Figure Description

[0020] The features and advantages of the invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way. In the drawings:

[0021] Figure 1 This diagram illustrates the steps of a fingerprint minutiae annotation method according to one embodiment of the present invention.

[0022] Figure 2 A schematic diagram of the structure of a neighborhood image is shown in one embodiment of the present invention;

[0023] Figure 3 A schematic diagram of the structure for neighborhood image alignment in one embodiment of the present invention is shown;

[0024] Figure 4 A schematic diagram of a fingerprint minutiae annotation device according to one embodiment of the present invention is shown;

[0025] Figure 5 A schematic diagram of an electronic device according to one embodiment of the present invention is shown. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Please see Figure 1 In one embodiment, a fingerprint minutiae annotation method specifically includes the following steps:

[0028] Step S101: Obtain the fingerprint image to be labeled, and search for labeled fingerprint images of the same finger from the preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled.

[0029] Specifically, in this embodiment of the invention, a fingerprint image database D with annotation information is established. F ={I i Each fingerprint image in the database I i All contain finger information identifiers of the source finger. i and annotation information of details on the image. Where, n i It is fingerprint image I i Total number of details on For the detail information of the J-th detail point, where For the coordinates of the detail points, For details, This includes attribute information for detail points, including but not limited to type and quality attributes. Type attributes include, but are not limited to, the start and end points of ridges, core points, triangulation points, bifurcation points, etc. For a fingerprint image to be labeled... Finger information identifier using the finger from which it originated. Search the fingerprint image database to find if there are any labeled fingerprint images originating from the same finger.

[0030] Step S102: Map the minutiae information already labeled in the labeled fingerprint image to the fingerprint image to be labeled.

[0031] Specifically, if there are already labeled fingerprint images of the same finger in the fingerprint image database, they are recorded together with the fingerprint image to be labeled. Marked fingerprints originating from the same finger are Algorithms such as image correlation calculation are used to align the fingerprint to be labeled with the labeled fingerprint, thereby mapping the minutiae information already labeled in the labeled fingerprint image to the fingerprint to be labeled.

[0032] Step S103: Generate a new labeled fingerprint image based on the mapped fingerprint image to be labeled, and store the new labeled fingerprint image in the fingerprint image database.

[0033] Specifically, after mapping the already labeled minutiae information onto the fingerprint image to be labeled, the mapped minutiae information is used as pseudo-labeling information for manual or algorithmic review and correction. Finally, a new labeled fingerprint image is generated and stored in the fingerprint image database. Compared to existing manual labeling methods, this approach allows for rapid labeling of fingerprint minutiae on the fingerprint image to be labeled, based on a partially labeled fingerprint image. The mapped pseudo-labeling information serves as a reference standard for minutiae labeling, undergoing further review and correction, significantly improving labeling efficiency. If the generated pseudo-labeling information is of high quality, it not only improves the efficiency of subsequent review and correction but also provides a degree of error prevention. Compared to purely manual labeling methods, it has less human influence and yields more accurate labeling results.

[0034] Specifically, in one embodiment, step S102 above includes the following steps:

[0035] Step 1: Align the labeled fingerprint image and the fingerprint image to be labeled according to the fingerprint correlation, and calculate the transformation matrix between the labeled fingerprint image and the fingerprint image to be labeled.

[0036] Specifically, in this embodiment of the invention, the fingerprint image is divided into blocks based on the fingerprint textures in the labeled fingerprint image and the fingerprint image to be labeled, and the texture correlation between each block is calculated. This aligns the labeled and unlabeled fingerprint images based on the texture correlation of each part, ensuring that identical fingerprint textures in the two images correspond as closely as possible. Then, a transformation matrix between the labeled and unlabeled fingerprint images is calculated through linear transformations such as translation, scaling, and rotation. The specific steps for calculating the transformation matrix between the images using linear transformations are existing techniques and will not be elaborated here.

[0037] Step 2: Map the minutiae information already labeled in the labeled fingerprint image to the fingerprint image to be labeled using a transformation matrix.

[0038] Specifically, by using the transformation matrix calculated in step one, the coordinates of minutiae in the labeled fingerprint image can be linearly transformed to obtain the coordinates of the minutiae in the fingerprint image to be labeled. Similarly, by using the transformation matrix to linearly transform the orientation angle of the minutiae, the orientation of the minutiae in the fingerprint image to be labeled can be obtained. Finally, the attribute information of each minutiae is added to each minutiae in the fingerprint image to be labeled. Thus, the above steps achieve rapid annotation of fingerprint minutiae, which significantly improves the annotation efficiency compared to empirical annotation methods.

[0039] Specifically, in one embodiment, the detail information of each detail point in the labeled fingerprint image also includes a neighborhood image, such as... Figure 2As shown, the neighborhood image is an image extracted from the labeled fingerprint image with a preset radius centered on the coordinates of the detail points. Therefore, step two above specifically includes the following steps:

[0040] Step 3: Map the coordinates, orientation, and neighborhood image of each minutiae from the labeled fingerprint image to the fingerprint image to be labeled using a transformation matrix.

[0041] Specifically, to further improve the accuracy of automatic minutiae annotation, this embodiment of the invention also establishes a fingerprint minutiae database D with annotation information. M For fingerprint image database D F Each detail point m k With details m k coordinates red k y k Using a preset radius R as the center, extract its neighborhood to obtain the following: Figure 2 The neighboring image is B k Then the fingerprint minutiae database can be represented as D M ={(m k B k Furthermore, in this embodiment of the invention, when creating a neighborhood image corresponding to the current detail point, the created neighborhood image is rotated based on the direction of the current detail point to obtain another neighborhood image of the current detail point. This rotation can reduce the total number of samples. For example, the directions of two detail points are different, but the neighborhood textures are similar after rotation.

[0042] Then, for each fingerprint minutiae m in the labeled fingerprint image... k =(x k y k θ k p k The transformation matrix T is used to map the current minutiae onto the fingerprint image to be labeled. Then, its neighborhood image is mapped onto the fingerprint image to be labeled.

[0043] Step 4: Calculate the neighborhood transformation matrix between each neighborhood image and the fingerprint image to be labeled by using the alignment relationship between the neighborhood images corresponding to each detail point.

[0044] Specifically, since fingerprints often exhibit local distortions, directly mapping a single minutiae to the fingerprint image to be labeled may not achieve perfect alignment. Therefore, this embodiment of the invention uses certain neighborhood features to find a better alignment position within a local area. That is, the neighborhood image of each minutiae includes not only the central minutiae to be mapped, but also other minutiae, such as... Figure 3The minutiae marked by the box appear simultaneously in the neighborhood images of minutiae A and minutiae B. Thus, the minutiae marked by the box can be used as a feature to align the neighborhood images of minutiae A and minutiae B in the fingerprint image to be labeled. Then, the aligned neighborhood images are used to make secondary adjustments to minutiae A and minutiae B, thereby ensuring the accurate mapping of minutiae A and minutiae B in the fingerprint image to be labeled.

[0045] Based on the above principles, algorithms such as image correlation calculation are used to find the optimal alignment position between the neighboring images and the fingerprint image to be labeled. After alignment, the neighborhood transformation matrix between each neighboring image and the fingerprint image to be labeled is calculated. For example, let the neighborhood transformation matrix of the current neighboring image be...

[0046] Step 5: Adjust the coordinates and orientation of the minutiae mapped to the fingerprint image to be labeled using a neighborhood transformation matrix.

[0047] Step 6: Add attribute information to each minutiae mapped to the fingerprint image to be labeled.

[0048] Specifically, using the neighborhood transformation matrix By further adjusting the coordinates and orientation of the minutiae in the fingerprint image to be labeled, and adding attribute information for each minutiae, pseudo-label information can be generated. This further improves the accuracy of detail annotation.

[0049] Specifically, in one embodiment, if multiple labeled fingerprint images of the same finger are found from a preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled, the fingerprint minutiae annotation method provided in this embodiment of the invention further includes the following steps:

[0050] Step 7: Map the minutiae information already labeled in each labeled fingerprint image to the fingerprint image to be labeled, and obtain multiple minutiae mapping information.

[0051] Step 8: Calculate the similarity between each set of detail mapping information, and fuse detail mapping information with a similarity greater than a preset similarity threshold.

[0052] Step 9: Generate a new labeled fingerprint image based on the fingerprint image to be labeled with the fused minutiae mapping information, and store the new labeled fingerprint image in the fingerprint image database.

[0053] Specifically, in this embodiment, if multiple labeled fingerprint images from the same finger are found in the fingerprint image database, steps one through six are repeated for each labeled fingerprint image to generate multiple minutiae mapping information for the fingerprint image to be labeled. To reduce data redundancy, the similarity of each minutiae mapping information is calculated pairwise. If the similarity of two minutiae mapping information is greater than a preset similarity threshold, the two minutiae mapping information are merged into one information, for example, by calculating the weighted result or average value of the coordinates of corresponding minutiae using a weighted method or an averaging method. In this embodiment, the similarity of the merged labeled information can be calculated again for multiple rounds of fusion, and this embodiment of the invention does not impose any special limitations. Finally, the fingerprint image to be labeled with the fused minutiae mapping information is used as a labeling reference for review and correction, and a new labeled fingerprint image is obtained and stored in the fingerprint image database, which can significantly reduce data redundancy.

[0054] Specifically, in one embodiment, the similarity between minutiae information is calculated using the following formula:

[0055]

[0056] In the formula, S(m1, m2) represents the similarity between detail points m1 and m2, x1 and y1 represent the x and y coordinates of detail point m1, respectively, x2 and y2 represent the x and y coordinates of detail point m2, respectively, θ1 and θ2 represent the directions of detail points m1 and m2, respectively, p1 and p2 represent the attribute information of detail points m1 and m2, respectively, and d x d y d θ All parameters are preset, and the specific size can be set by the user according to the actual situation. This embodiment of the invention does not impose any special limitations.

[0057] In the above formula, if the attributes of two detail points are the same, then the formula is used. Calculate the similarity between detail points. If two detail points have different attributes, the similarity between the two detail points is considered to be 0.

[0058] Specifically, in one embodiment, if no labeled fingerprint image of the same finger is found in a preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled, the fingerprint minutiae annotation method provided in this embodiment of the invention further includes the following steps:

[0059] Step 10: Mark the coordinates and directions of several detail points in the fingerprint image to be annotated according to the input annotation instructions.

[0060] Specifically, if no labeled fingerprint image from the same finger is found in the fingerprint image database, a rough labeling is first performed on the fingerprint image to be labeled using manual experience or other algorithm input instructions, and the coordinates and directions of the details are labeled.

[0061] Step 11: Create a first neighborhood image for each minutiae in the fingerprint image to be labeled according to the labeling instructions.

[0062] Specifically, after roughly labeling the coordinates and directions of the minutiae in the fingerprint image to be labeled, a corresponding first neighborhood image is created for each minutiae using the method mentioned in step three above.

[0063] Step 12: Extract the second neighboring image with the highest similarity to the first neighboring image from the minutiae information of the labeled fingerprint image.

[0064] Specifically, based on the created first neighboring images, the second neighboring image with the highest similarity is sequentially searched from the fingerprint minutiae database. For example, using an image similarity algorithm, the neighboring image C found in the fingerprint minutiae database has the highest similarity to the neighboring image D created from the current minutiae in the fingerprint image to be labeled.

[0065] Step 13: Use the attribute information of the corresponding minutiae in the second neighboring image as the pseudo-annotation information of the corresponding minutiae in the first neighboring image.

[0066] Specifically, the attribute information p corresponding to the detail points at the center of the searched second neighborhood image C is... c The pseudo-annotation information of the minutiae at the center of the first neighborhood image D completes one instance of attribute information annotation. Similarly, steps twelve and thirteen are performed for each coarsely annotated minutiae until the attribute information of all minutiae in the fingerprint image to be annotated is completed. Similarly, the attribute information mapped to each minutiae is then reviewed and corrected to obtain the accurate attribute information of each minutiae. Using the method provided in this embodiment of the invention, even for fingerprint images to be annotated where no previously annotated fingerprint image of the same finger has been found in the fingerprint image database, automated minutiae annotation is achieved. By using the mapped minutiae attribute information as pseudo-annotation information and referring to the pseudo-annotation information for secondary review and correction, the annotation efficiency of minutiae can be significantly improved.

[0067] Through the above steps, the technical solution provided in this application acquires a fingerprint image to be labeled, and searches for labeled fingerprint images of the same finger from a preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled; then, it maps the already labeled minutiae information in the labeled fingerprint images to the fingerprint image to be labeled; thereby generating a new labeled fingerprint image based on the mapped fingerprint image to be labeled, and storing the new labeled fingerprint image in the fingerprint image database. This application utilizes the characteristic that fingerprint databases usually collect multiple samples from the same finger, and that the minutiae have high similarity, and proposes a minutiae information labeling method based on already labeled minutiae to be labeled fingerprint images, making minutiae labeling more accurate and efficient.

[0068] like Figure 4 As shown, this embodiment also provides a fingerprint minutiae annotation device, which includes:

[0069] The fingerprint image matching module 101 is used to acquire a fingerprint image to be labeled, and search for labeled fingerprint images of the same finger from a preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled. For details, please refer to the relevant description of step S101 in the above method embodiment, which will not be repeated here.

[0070] The minutiae mapping module 102 is used to map the minutiae information already labeled in the labeled fingerprint image to the fingerprint image to be labeled. For details, please refer to the relevant description of step S 102 in the above method embodiment, which will not be repeated here.

[0071] The annotation result storage module 103 is used to store the mapped fingerprint image to be annotated as a new annotated fingerprint image into the fingerprint image database. For details, please refer to the relevant description of step S103 in the above method embodiment, which will not be repeated here.

[0072] The fingerprint minutiae annotation device provided in this embodiment of the invention is used to execute the fingerprint minutiae annotation method provided in the above embodiment. Its implementation method and principle are the same. For details, please refer to the relevant description of the above method embodiment, which will not be repeated here.

[0073] Through the collaborative efforts of the aforementioned components, the technical solution provided in this application acquires a fingerprint image to be labeled, and searches for labeled fingerprint images of the same finger from a pre-defined fingerprint image database based on the finger information identifier of the fingerprint image to be labeled; then, it maps the already labeled minutiae information in the labeled fingerprint images to the fingerprint image to be labeled; thereby generating a new labeled fingerprint image based on the mapped fingerprint image to be labeled, and storing the new labeled fingerprint image in the fingerprint image database. This application utilizes the characteristic that fingerprint databases typically collect multiple samples from the same finger, and that the minutiae have high similarity, proposing a minutiae annotation method that maps and annotates the minutiae information of the fingerprint image to be labeled based on the already labeled minutiae, making minutiae annotation more accurate and efficient.

[0074] Figure 5 An electronic device according to an embodiment of the present invention is shown. The device includes a processor 901 and a memory 902, which can be connected via a bus or other means. Figure 5 Taking the example of a connection between China and Israel via a bus.

[0075] Processor 901 can be a Central Processing Unit (CPU). Processor 901 can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0076] The memory 902, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the methods in the above method embodiments. The processor 901 executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions, and modules stored in the memory 902, thereby implementing the methods in the above method embodiments.

[0077] The memory 902 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor 901, etc. Furthermore, the memory 902 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 902 may optionally include memory remotely located relative to the processor 901, and these remote memories may be connected to the processor 901 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0078] One or more modules are stored in memory 902, and when executed by processor 901, they perform the methods described in the above method embodiments.

[0079] The specific details of the aforementioned electronic device can be understood by referring to the relevant descriptions and effects in the above method embodiments, and will not be repeated here.

[0080] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The implemented program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0081] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for fingerprint minutiae annotation, characterized in that, The method includes: acquiring a fingerprint image to be labeled, and searching for labeled fingerprint images of the same finger from a preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled; Map the minutiae information already labeled in the labeled fingerprint image to the fingerprint image to be labeled; A new labeled fingerprint image is generated based on the mapped fingerprint image to be labeled, and the new labeled fingerprint image is stored in the fingerprint image database; The step of mapping the already labeled minutiae information from the labeled fingerprint image to the fingerprint image to be labeled includes: The labeled fingerprint image and the fingerprint image to be labeled are aligned based on fingerprint correlation, and the transformation matrix between the labeled fingerprint image and the fingerprint image to be labeled is calculated. The transformation matrix is ​​used to map the already labeled minutiae information in the labeled fingerprint image to the fingerprint image to be labeled. The minutiae information includes the coordinates and orientation of each minutiae, the attribute information of each minutiae, and the neighborhood image of each minutiae; the neighborhood image is an image cropped from the labeled fingerprint image with a preset radius centered on the coordinates of the minutiae; the step of mapping the labeled minutiae information from the labeled fingerprint image to the fingerprint image to be labeled using the transformation matrix includes: The transformation matrix is ​​used to map the coordinates of each minutiae, the orientation of each minutiae, and the neighborhood image corresponding to each minutiae from the labeled fingerprint image to the fingerprint image to be labeled. By using the alignment relationship between the neighboring images corresponding to each detail point, the neighborhood transformation matrix between each neighboring image and the fingerprint image to be labeled is calculated respectively; The neighborhood transformation matrix is ​​used to adjust the coordinates and orientation of the minutiae mapped to the fingerprint image to be labeled. The attribute information is added to each detail point mapped to the fingerprint image to be labeled.

2. The method according to claim 1, characterized in that, If multiple labeled fingerprint images of the same finger are found in a preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled, the method further includes: The minutiae information already labeled in each labeled fingerprint image is mapped to the fingerprint image to be labeled, resulting in multiple minutiae mapping information; Calculate the similarity between each set of detail mapping information, and fuse detail mapping information with a similarity greater than a preset similarity threshold; A new labeled fingerprint image is generated based on the fingerprint image to be labeled with the fused minutiae mapping information, and the new labeled fingerprint image is stored in the fingerprint image database.

3. The method according to claim 2, characterized in that, The calculation of the similarity between each set of detail mapping information includes: The similarity between minutiae in the two minutiae mapping information is calculated using the following formula. ； In the formula, Let x1 and y1 represent the similarity between minutiae m1 and m2, respectively, and let x2 and y2 represent the x and y coordinates of minutiae m2. and Let m1 and m2 represent the directions of detail points m1 and m2, respectively, and p1 and p2 represent the attribute information of detail points m1 and m2, respectively. , , All parameters are preset.

4. The method according to claim 1, characterized in that, If no labeled fingerprint image of the same finger is found in the preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled, the method further includes: According to the input annotation instructions, the coordinates and directions of several minutiae are annotated in the fingerprint image to be annotated; For each minutiae in the fingerprint image to be labeled according to the labeling instructions, a first neighborhood image is created; Extract the second neighboring image that has the highest similarity to the first neighboring image from the details of the labeled fingerprint image; The attribute information of the corresponding minutiae in the second neighboring image is used as the pseudo-annotation information of the corresponding minutiae in the first neighboring image.

5. The method according to claim 4, characterized in that, The method further includes: When creating a neighborhood image corresponding to the current detail point, the created neighborhood image is rotated based on the direction of the current detail point to obtain another neighborhood image of the current detail point.

6. A fingerprint minutiae marking device, characterized in that, The device includes: The fingerprint image matching module is used to acquire a fingerprint image to be labeled, and search for labeled fingerprint images of the same finger from a preset fingerprint image database based on the finger information identifier of the fingerprint image to be labeled; The minutiae mapping module is used to map the minutiae information that has been annotated in the annotated fingerprint image to the fingerprint image to be annotated; The annotation result saving module is used to generate a new labeled fingerprint image based on the mapped fingerprint image to be labeled, and to store the new labeled fingerprint image into the fingerprint image database; The detail mapping module is specifically used for: The labeled fingerprint image and the fingerprint image to be labeled are aligned based on fingerprint correlation, and the transformation matrix between the labeled fingerprint image and the fingerprint image to be labeled is calculated. The transformation matrix is ​​used to map the already labeled minutiae information in the labeled fingerprint image to the fingerprint image to be labeled. The minutiae information includes the coordinates and orientation of each minutiae, the attribute information of each minutiae, and the neighborhood image of each minutiae; the neighborhood image is an image cropped from the labeled fingerprint image with a preset radius centered on the coordinates of the minutiae; the step of mapping the labeled minutiae information from the labeled fingerprint image to the fingerprint image to be labeled using the transformation matrix includes: The transformation matrix is ​​used to map the coordinates of each minutiae, the orientation of each minutiae, and the neighborhood image corresponding to each minutiae from the labeled fingerprint image to the fingerprint image to be labeled. By using the alignment relationship between the neighboring images corresponding to each detail point, the neighborhood transformation matrix between each neighboring image and the fingerprint image to be labeled is calculated respectively; The neighborhood transformation matrix is ​​used to adjust the coordinates and orientation of the minutiae mapped to the fingerprint image to be labeled. The attribute information is added to each detail point mapped to the fingerprint image to be labeled.

7. An electronic device, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the method as described in any one of claims 1-5.

Images