Fingerprint feature recognition method, system and device based on Seaweed FS and SSD and medium

Abstract

The invention provides a fingerprint feature recognition method, system and device based on a SeaweedFS and an SSD and a medium. The method according to the invention comprises the following steps: acquiring fingerprint data and storing the fingerprint data in a pre-configured SeaweedFS cluster; adopting a preset fingerprint preliminary screening algorithm to generate a candidate fingerprint list by performing primary comparison on the fingerprint data; fingerprint detail features in fingerprint data are loaded in a SeaweedFS cluster in real time, and secondary comparison and sorting are carried out on the fingerprint data based on a preset fingerprint fine sorting algorithm in combination with the candidate fingerprint list and the fingerprint detail features to obtain a final candidate result. According to the technical scheme provided by the invention, the problems of high memory occupation, performance bottleneck and poor expandability in a traditional fingerprint identification method are solved, a solid technical basis is provided for further development of a fingerprint identification system, and the method has a wide application prospect.

Description

Technical Field

[0001] This document relates to the field of distributed fingerprint recognition technology, and in particular to a fingerprint feature recognition method, system, device and medium based on SeaweedFS and SSD. Background Art

[0002] Fingerprint recognition technology is widely used in various fields such as identity authentication and security monitoring due to its uniqueness and stability. The current fingerprint recognition system mainly relies on two types of features: traditional detail features and deep learning features. Traditional detail features include small bifurcation points and endpoints in fingerprints. These feature information is fine and the data volume is large; while deep learning features extract high-level features through algorithms such as convolutional neural networks. Usually, the data volume is small but contains rich information.

[0003] In the prior art, the storage and loading methods of fingerprint features mainly have the following problems:

[0004] 1. High memory usage: The traditional implementation method is to preload all fingerprint details into memory for comparison. Since the amount of detail feature data is huge, generally 2 to 10 times that of deep learning features, this method will result in extremely high memory usage. When the fingerprint database is large, memory resources will quickly become a bottleneck, limiting the system's scalability.

[0005] 2. Performance bottleneck: In practical applications, fingerprint matching needs to respond quickly to meet real-time requirements. The traditional implementation uses a process of first screening through deep learning features and then performing secondary fine sorting through detail features. Although the number of candidates after the initial screening is reduced, the detail features of these candidates still need to be accurately compared. If these features are pre-loaded into the memory, the overall performance of the system will be affected. In addition, due to the limited hard disk I / O speed, frequent reading and writing of large amounts of fingerprint data may also cause performance bottlenecks.

[0006] 3. Poor scalability: As the fingerprint database continues to grow, how to effectively manage and expand storage space has become a major challenge. Traditional file systems are inefficient when processing a large number of small files and have difficulty meeting the needs of efficient storage and retrieval of massive data. In this case, the maintenance cost and complexity of the system continue to increase, affecting the availability and reliability of the system.

[0007] In view of the above problems, the present invention proposes a fingerprint feature recognition method, system, device and medium based on SeaweedFS and SSD, which provides a new solution for the fingerprint recognition system. Summary of the invention

[0008] The present invention provides a fingerprint feature recognition method, system, device and medium based on SeaweedFS and SSD, aiming to solve the above problems.

[0009] The embodiment of the present invention provides a fingerprint feature recognition method based on SeaweedFS and SSD, comprising:

[0010] S1, obtain fingerprint data and store it in the pre-configured SeaweedFS cluster;

[0011] S2. Using a preset fingerprint preliminary screening algorithm to perform a preliminary comparison on the fingerprint data to generate a candidate fingerprint list;

[0012] S3, loading fingerprint details in the fingerprint data in the SeaweedFS cluster in real time, performing secondary comparison and sorting on the fingerprint data based on a preset fingerprint sorting algorithm in combination with the candidate fingerprint list and the fingerprint details to obtain a final candidate result.

[0013] The embodiment of the present invention provides a fingerprint feature recognition system based on SeaweedFS and SSD, including:

[0014] Fingerprint feature storage module, used to obtain fingerprint data and store it in a pre-configured SeaweedFS cluster;

[0015] The initial comparison module uses a preset fingerprint initial screening algorithm to generate a candidate fingerprint list by performing an initial comparison on the fingerprint data;

[0016] The final result module is used to load the fingerprint details in the fingerprint data in the SeaweedFS cluster in real time, and perform secondary comparison and sorting on the fingerprint data based on a preset fingerprint sorting algorithm in combination with the candidate fingerprint list and the fingerprint details to obtain the final candidate result.

[0017] According to an embodiment of the present invention, an electronic device is provided, including:

[0018] processor; and,

[0019] A memory arranged to store computer executable instructions, which, when executed, cause the processor to perform the steps of the fingerprint feature recognition method based on SeaweedFS and SSD as described above.

[0020] An embodiment of the present invention provides a storage medium for storing computer executable instructions, which, when executed, implement the steps of the fingerprint feature recognition method based on SeaweedFS and SSD as described above.

[0021] By adopting the embodiment of the present invention, SeaweedFS is used for distributed storage of fingerprint feature data, avoiding the performance and scalability problems existing in traditional databases or file systems. In the fingerprint matching process, the initial screening is achieved through deep learning features, and only the detail features corresponding to the candidate results are downloaded from SeaweedFS in real time instead of loading all of them into the memory, thereby significantly reducing memory usage. The detail features are only compared in the candidate set after the initial screening to achieve accurate sorting and improve the matching efficiency. The initial screening uses deep learning features, with high matching accuracy and relatively low calculation amount; the fine sorting uses traditional detail features to improve the accuracy of the final matching. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate one or more embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0023] Figure 1 It is a flow chart of a fingerprint feature recognition method based on SeaweedFS and SSD according to an embodiment of the present invention;

[0024] Figure 2 A schematic diagram of a fingerprint feature recognition system based on SeaweedFS and SSD according to an embodiment of the present invention;

[0025] Figure 3 It is a specific implementation flow chart of the fingerprint feature recognition method based on SeaweedFS and SSD according to an embodiment of the present invention;

[0026] Figure 4 A flowchart of storing fingerprint features according to an embodiment of the present invention;

[0027] Figure 5 This is a flow chart of initial comparison of fingerprint data according to an embodiment of the present invention;

[0028] Figure 6 A flowchart of performing secondary comparison and sorting of fingerprint data according to an embodiment of the present invention;

[0029] Figure 7 This is a flow chart of a request for adding, deleting or updating fingerprint data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the following will be combined with the drawings in one or more embodiments of this specification to clearly and completely describe the technical solutions in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this document.

[0031] Glossary

[0032] 1. SeaweedFS

[0033] SeaweedFS is an open source distributed file system suitable for application scenarios that require large-scale file storage and high concurrent access. Its open source nature also allows users to perform secondary development and customization according to their own needs.

[0034] It has the following features:

[0035] (1) High scalability: SeaweedFS can be easily expanded to hundreds or even thousands of servers, providing high-capacity and high-performance file storage.

[0036] (3) Fault tolerance: The system uses redundant storage and data replication technology to ensure the reliability and availability of files, and files can be accessed normally even when some servers fail.

[0037] (3) Easy to use: SeaweedFS provides a simple API and command line tools to facilitate users to upload, download, delete and other operations.

[0038] (4) High performance: By optimizing data storage and access methods, SeaweedFS can provide fast file read and write performance, suitable for scenarios with large-scale data storage and high concurrent access.

[0039] (5) Cross-platform: It can run on multiple operating systems, including Linux, Windows, and macOS.

[0040] (6) Data consistency: The consistency of files is guaranteed, and data inconsistency will not occur even in the case of concurrent access.

[0041] (7) Customizability: Users can customize and extend the system according to their needs, such as adding new storage backends or implementing specific functions.

[0042] 2. SeaweedFS cluster

[0043] The steps to build a SeaweedFS cluster are as follows:

[0044] (1) Prepare servers: Select multiple servers as cluster nodes. Ensure that the servers have sufficient computing resources (CPU, memory) and storage capacity;

[0045] (2) Install SeaweedFS: Install SeaweedFS software on each server;

[0046] (3) Configure the master node: Select a server as the master node to manage the entire cluster. Perform the following configuration on the master node: Set the IP address and port of the master node; configure the data storage directory and other related parameters;

[0047] (4) Start the master node: Start the SeaweedFS service on the master node;

[0048] (5) Configure slave nodes: Configure slave nodes on other servers, including setting the IP address and port of the slave nodes and specifying the address of the master node;

[0049] (6) Start the slave node: Start the SeaweedFS service on the slave node;

[0050] (7) Data replication and redundancy: SeaweedFS automatically replicates data between the master node and slave nodes to achieve data redundancy and high availability;

[0051] (8) Client connection: Use the client application to connect to the SeaweedFS cluster. You can connect by specifying the address of the master node;

[0052] (9) Testing and optimization: Test to ensure that the cluster is functioning properly, and perform performance optimization and adjustments based on actual needs.

[0053] Method Embodiment

[0054] According to an embodiment of the present invention, a fingerprint feature recognition method based on SeaweedFS and SSD is provided. Figure 1 is a flow chart of a fingerprint feature recognition method based on SeaweedFS and SSD according to an embodiment of the present invention. Figure 1 As shown, the fingerprint feature recognition method based on SeaweedFS and SSD in the embodiment of the present invention specifically includes:

[0055] S1, obtain fingerprint data and store it in the pre-configured SeaweedFS cluster;

[0056] Specifically, the SSD hard disk and SeaweedFS are used to optimize the access performance of small files to ensure efficient feature loading in real-time comparison. The embodiment of the present invention combines SSD and SeaweedFS to improve the overall storage performance and response time. The operation is specifically performed in the following steps:

[0057] 1. Install SeaweedFS: First, you need to install SeaweedFS on the server or storage node and follow the installation guide to install and configure it.

[0058] 2. Configure SeaweedFS to use SSD: In the SeaweedFS configuration file, specify that the data will be stored on the SSD. Usually, this can be achieved by setting the storage path or storage media. Make sure the SSD device is mounted correctly and its path is configured into SeaweedFS.

[0059] By combining SSD with SeaweedFS, you can use the high-speed read and write performance of SSD to accelerate file access and improve the overall performance and response time of the storage system. At the same time, the distributed nature of SeaweedFS can provide scalability and data redundancy to ensure data reliability and availability.

[0060] like Figure 4 The figure shows a storage flow chart of fingerprint features of an embodiment of the present invention. First, the fingerprint data is loaded, the detail features and deep learning features are extracted, and it is determined whether they need to be stored in the SeaweedFS cluster. If necessary, the detail features and deep learning features are stored in the SeaweedFS cluster.

[0061] S2. Using a preset fingerprint preliminary screening algorithm to perform a preliminary comparison on the fingerprint data to generate a candidate fingerprint list;

[0062] Figure 5 This is a flowchart of the initial comparison of fingerprint data in an embodiment of the present invention. Figure 5 It can be seen that the fingerprint initial screening algorithm of the embodiment of the present invention specifically includes:

[0063] First, obtain the fingerprint to be compared, extract the deep learning features obtained in S1, use the deep learning features for preliminary screening, and generate candidate fingerprints. Determine whether further detailed comparison is needed, that is, whether a secondary comparison is needed. If necessary, enter S3 for secondary comparison and sorting to obtain the final candidate results.

[0064] Among them, the step of extracting the deep learning features obtained in S1 to generate candidate fingerprints includes: inputting the fingerprint image to be identified into a trained deep learning model.

[0065] The trained deep learning model automatically extracts primary features of the fingerprint to be identified and compares them with the fingerprint features in the SeaweedFS cluster;

[0066] According to the comparison results, a candidate fingerprint list is generated through the trained deep learning model, and the candidate fingerprint list is sorted from high to low according to the similarity.

[0067] S3, loading fingerprint details in the fingerprint data in the SeaweedFS cluster in real time, performing secondary comparison and sorting on the fingerprint data based on a preset fingerprint sorting algorithm in combination with the candidate fingerprint list and the fingerprint details to obtain a final candidate result.

[0068] like Figure 6 The flowchart of performing secondary comparison and sorting of fingerprint data in an embodiment of the present invention first determines whether it is necessary to obtain fingerprint detail features of a candidate fingerprint list, and if necessary, loads fingerprint detail features in the candidate fingerprint list in real time in the SeaweedFS cluster;

[0069] If not needed, the fingerprint details of the candidate fingerprint list in the memory are directly used;

[0070] The fingerprint image to be identified is accurately compared with the fingerprint detail features to obtain the final comparison result.

[0071] The fingerprint feature recognition method based on SeaweedFS and SSD in the embodiment of the present invention further includes:

[0072] In response to the system's fingerprint feature addition, deletion and update instructions, the fingerprint feature is added, deleted and updated, such as Figure 7 The flowchart of adding, deleting or updating fingerprint data request according to an embodiment of the present invention is shown, which specifically includes:

[0073] If the instruction is to add a new fingerprint feature, the new fingerprint feature is stored in the SeaweedFS cluster, and the index value of the new fingerprint feature is updated;

[0074] If the instruction is to delete a fingerprint feature, the corresponding fingerprint feature is deleted in the SeaweedFS cluster, and the index values ​​of the remaining fingerprint features are updated;

[0075] If it is determined that the instruction is a fingerprint feature update, the fingerprint feature in the SeaweedFS cluster is updated, and the index value of the corresponding fingerprint feature is updated.

[0076] Figure 3 is a specific implementation flow chart of the fingerprint feature recognition method based on SeaweedFS and SSD according to an embodiment of the present invention, Figure 3The main controller in the SeaweedFS cluster is used to coordinate and manage the entire process; Fingerprint Feature Extraction processes fingerprint data and stores it in the SeaweedFS cluster; Fingerprint Preliminary Matching obtains data from the SeaweedFS cluster for preliminary matching; Generate Candidate Fingerprint List generates a candidate fingerprint list; Fingerprint Detailed Matching loads detailed features from the SeaweedFS cluster in real time for accurate matching and ranking; Generate Candidate Results generates the final candidate results based on the results of the detailed matching program; SeaweedFS Cluster is responsible for storing all fingerprint data and features; Fingerprint Data and Feature Storage manages the data content stored in the SeaweedFS cluster.

[0077] The embodiments of the present invention have the following beneficial effects:

[0078] The embodiment of the present invention only downloads detail features from SeaweedFS when necessary, rather than pre-loading all detail features into the memory. Specifically, deep learning features are first used for preliminary screening to narrow the number of potential candidate fingerprints to a smaller range. Then, for the candidate fingerprints after the preliminary screening, only when a secondary fine sorting is required, the corresponding detail features are instantly downloaded from SeaweedFS for comparison. Since the number of candidates after the preliminary screening is usually small, and SeaweedFS and SSD hard drives are optimized for small file access performance, this method can not only meet the requirements of real-time comparison, but also significantly reduce memory usage.

[0079] The use of the distributed file system SeaweedFS also brings additional advantages. SeaweedFS can easily expand storage capacity and support distributed deployment, which can improve the maintainability and scalability of the system. Whether facing the growth of the fingerprint database or distributed deployment in different geographical locations, it can cope well, thereby improving the reliability and flexibility of the entire fingerprint recognition system.

[0080] The present invention realizes an efficient fingerprint feature storage and loading method by cleverly combining SeaweedFS and SSD hard disk. It not only solves the problems of high memory usage, performance bottleneck and poor scalability in traditional methods, but also provides a solid technical foundation for the further development of fingerprint recognition system and has broad application prospects.

[0081] System Example

[0082] According to an embodiment of the present invention, a fingerprint feature recognition system based on SeaweedFS and SSD is provided. Figure 2 is a schematic diagram of a fingerprint feature recognition system based on SeaweedFS and SSD according to an embodiment of the present invention. Figure 2 As shown, the fingerprint feature recognition system based on SeaweedFS and SSD in the embodiment of the present invention specifically includes:

[0083] Fingerprint feature storage module 20, used to obtain fingerprint data and store it in a pre-configured SeaweedFS cluster;

[0084] The initial comparison module 22 uses a preset fingerprint initial screening algorithm to perform an initial comparison on the fingerprint data to generate a candidate fingerprint list;

[0085] The final result module 24 is used to load the fingerprint details in the fingerprint data in the SeaweedFS cluster in real time, and perform secondary comparison and sorting on the fingerprint data based on the preset fingerprint sorting algorithm in combination with the candidate fingerprint list and the fingerprint details to obtain the final candidate result.

[0086] The fingerprint feature recognition system based on SeaweedFS and SSD further includes: a fingerprint change module, which is used to respond to the system's fingerprint feature addition, deletion and update instructions to add, delete and update the fingerprint features, specifically including:

[0087] If the instruction is to add a new fingerprint feature, the new fingerprint feature is stored in the SeaweedFS cluster, and the index value of the new fingerprint feature is updated;

[0088] If the instruction is to delete a fingerprint feature, the corresponding fingerprint feature is deleted in the SeaweedFS cluster, and the index values ​​of the remaining fingerprint features are updated;

[0089] If it is determined that the instruction is a fingerprint feature update, the fingerprint feature in the SeaweedFS cluster is updated, and the index value of the corresponding fingerprint feature is updated.

[0090] Device Example 1

[0091] According to an embodiment of the present invention, there is provided an electronic device, including:

[0092] processor; and,

[0093] A memory is arranged to store computer executable instructions, which when executed cause the processor to perform the steps of the above method embodiments.

[0094] Device Example 2

[0095] According to an embodiment of the present invention, a storage medium is provided for storing computer executable instructions, wherein the computer executable instructions implement the steps of the above method embodiment when executed.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fingerprint feature recognition method based on SeaweedFS and SSD, characterized in that include: S1, obtain fingerprint data and store it in the pre-configured SeaweedFS cluster; S2. Using a preset fingerprint preliminary screening algorithm to perform a preliminary comparison on the fingerprint data to generate a candidate fingerprint list; S3, loading fingerprint details in the fingerprint data in the SeaweedFS cluster in real time, performing secondary comparison and sorting on the fingerprint data based on a preset fingerprint sorting algorithm in combination with the candidate fingerprint list and the fingerprint details to obtain a final candidate result.

2. The method according to claim 1, characterized in that: The preset fingerprint initial screening algorithm specifically includes: Input the fingerprint image to be identified into the trained deep learning model. The trained deep learning model automatically extracts primary features of the fingerprint to be identified and compares them with the fingerprint features in the SeaweedFS cluster; According to the comparison results, a candidate fingerprint list is generated through the trained deep learning model, and the candidate fingerprint list is sorted from high to low according to the similarity.

3. The method according to claim 1, characterized in that The preset fingerprint sorting algorithm specifically includes: Determine whether it is necessary to obtain fingerprint detail features of the candidate fingerprint list, and if necessary, load the fingerprint detail features in the candidate fingerprint list in real time in the SeaweedFS cluster; If not needed, the fingerprint details of the candidate fingerprint list in the memory are directly used; The fingerprint image to be identified is accurately compared with the fingerprint detail features to obtain the final comparison result.

4. The method according to claim 1, characterized in that: The fingerprint feature recognition method based on SeaweedFS and SSD further includes: responding to the system's fingerprint feature addition, deletion and update instructions, adding, deleting and updating the fingerprint feature, specifically including: If the instruction is to add a new fingerprint feature, the new fingerprint feature is stored in the SeaweedFS cluster, and the index value of the new fingerprint feature is updated; If the instruction is to delete a fingerprint feature, the corresponding fingerprint feature is deleted in the SeaweedFS cluster, and the index values ​​of the remaining fingerprint features are updated; If it is determined that the instruction is a fingerprint feature update, the fingerprint feature in the SeaweedFS cluster is updated, and the index value of the corresponding fingerprint feature is updated.

5. A fingerprint feature recognition system based on SeaweedFS and SSD, characterized in that: include: Fingerprint feature storage module, used to obtain fingerprint data and store it in a pre-configured SeaweedFS cluster; The initial comparison module uses a preset fingerprint initial screening algorithm to generate a candidate fingerprint list by performing an initial comparison on the fingerprint data; The final result module is used to load the fingerprint details in the fingerprint data in the SeaweedFS cluster in real time, and perform secondary comparison and sorting on the fingerprint data based on a preset fingerprint sorting algorithm in combination with the candidate fingerprint list and the fingerprint details to obtain the final candidate result.

6. The system according to claim 5, characterized in that The initial comparison module is specifically used for: Input the fingerprint image to be identified into the trained deep learning model. The trained deep learning model automatically extracts primary features of the fingerprint to be identified and compares them with the fingerprint features in the SeaweedFS cluster; According to the comparison results, a candidate fingerprint list is generated through the trained deep learning model, and the candidate fingerprint list is sorted from high to low according to the similarity.

7. The system according to claim 5, characterized in that The final result module is specifically used for: Determine whether it is necessary to obtain fingerprint detail features of the candidate fingerprint list, and if necessary, load the fingerprint detail features in the candidate fingerprint list in real time in the SeaweedFS cluster; If not needed, the fingerprint details of the candidate fingerprint list in the memory are directly used; The fingerprint image to be identified is accurately compared with the fingerprint detail features to obtain the final comparison result.

8. The system according to claim 5, characterized in that The fingerprint feature recognition system based on SeaweedFS and SSD further includes: a fingerprint change module, which is used to respond to the system's fingerprint feature addition, deletion and update instructions to add, delete and update the fingerprint features, specifically including: If the instruction is to add a new fingerprint feature, the new fingerprint feature is stored in the SeaweedFS cluster, and the index value of the new fingerprint feature is updated; If the instruction is to delete a fingerprint feature, the corresponding fingerprint feature is deleted in the SeaweedFS cluster, and the index values ​​of the remaining fingerprint features are updated; If it is determined that the instruction is a fingerprint feature update, the fingerprint feature in the SeaweedFS cluster is updated, and the index value of the corresponding fingerprint feature is updated.

9. An electronic device, comprising: processor; as well as, A memory arranged to store computer executable instructions, which, when executed, cause the processor to perform the steps of the fingerprint feature recognition method based on SeaweedFS and SSD as claimed in any one of claims 1 to 4.

10. A storage medium for storing computer executable instructions, wherein the computer executable instructions, when executed, implement the steps of the fingerprint feature recognition method based on SeaweedFS and SSD as claimed in any one of claims 1 to 4.

Images