Fingerprint identification access control method and system

Abstract

The application provides a fingerprint identification access control method and system. The fingerprint identification access control method comprises the following steps: collecting a fingerprint image for fingerprint identification of a same user in real time; obtaining a first gray threshold and a second gray threshold corresponding to a current fingerprint image according to a gray value of a pixel block of a fingerprint part of the current fingerprint image; obtaining a plurality of to-be-identified image blocks by using the first gray threshold and the second gray threshold, and obtaining a comprehensive similarity value of the current fingerprint image by combining the gray value of the to-be-identified image blocks; and determining that the fingerprint image of the current user passes the fingerprint verification when the comprehensive similarity value is not lower than a preset similarity threshold. The system comprises modules corresponding to the method steps.

Description

Technical Field

[0001] This invention relates to the field of fingerprint recognition technology, and in particular to a fingerprint recognition access control method and system. Background Technology

[0002] Fingerprint recognition technology is a biometric authentication technology characterized by uniqueness and stability, and it has wide applications in the field of security. Currently, most fingerprint recognition products on the market use capacitive or optical sensors to capture fingerprint images, then use specific algorithms to extract features for comparison to determine whether the user is legitimate. These products typically require users to provide specific pressing postures and pressure, and are greatly affected by the moisture level of the fingers, resulting in a less than ideal user experience.

[0003] Furthermore, in the mobile internet era, people have increasingly higher requirements for identity authentication, hoping to complete the process quickly and accurately, while also expecting the authentication process to be as simple and convenient as possible, without placing too much burden on users. This necessitates that fingerprint recognition systems provide a good user experience while ensuring recognition accuracy. Summary of the Invention

[0004] This invention provides a fingerprint recognition access control method and system to solve the problem that traditional fingerprint recognition requires full feature recognition of the entire fingerprint image. Due to the large area of ​​the corresponding fingerprint, the processing efficiency and recognition response speed are low. While extracting local fingerprint image blocks for recognition can reduce the recognition area, the recognition accuracy is low. This invention extracts fingerprint image blocks from the fingerprint image. However, according to traditional image block extraction methods, it is impossible to select the image blocks with the strongest fingerprint feature representation. Therefore, the fingerprint image block selection has weak representation, which easily leads to a high probability of fingerprint recognition failure and low fingerprint recognition accuracy.

[0005] This invention proposes a fingerprint recognition access control method, which includes:

[0006] S1: Real-time acquisition of fingerprint images for fingerprint recognition of the same user;

[0007] S2: Based on the grayscale values ​​of the pixel blocks of the fingerprint portion of the currently acquired fingerprint image, obtain the first grayscale threshold and the second grayscale threshold corresponding to the current fingerprint image; wherein, the first grayscale threshold is greater than the second grayscale threshold.

[0008] S3: Use the first grayscale threshold and the second grayscale threshold to obtain multiple image blocks to be identified, and combine the grayscale values ​​of the image blocks to be identified to obtain the comprehensive similarity value of the current fingerprint image.

[0009] S4: When the overall similarity value is not lower than the preset similarity threshold, it is determined that the current user's fingerprint image has passed fingerprint verification.

[0010] Furthermore, based on the grayscale values ​​of the pixel blocks of the fingerprint portion in the currently acquired fingerprint image, a first grayscale threshold and a second grayscale threshold corresponding to the current fingerprint image are obtained, including:

[0011] S21: Extract the grayscale values ​​of the pixel blocks of the fingerprint portion of the current fingerprint image;

[0012] S22: Extract the gray values ​​of the pixel blocks in the fingerprint portion that are lower than the optimal threshold corresponding to the fingerprint image, and use them as the first gray value data;

[0013] S23: Extract the grayscale values ​​of the pixel blocks in the fingerprint portion that are not lower than the optimal threshold corresponding to the fingerprint image, and use them as the second grayscale value data;

[0014] S24: Use the first grayscale value data to set the first candidate grayscale threshold for the fingerprint portion;

[0015] S25: Use the second grayscale value data to set the second candidate grayscale threshold for the fingerprint portion;

[0016] S26: Compare the first candidate grayscale threshold with the second candidate grayscale threshold, and take the larger of the two candidate grayscale thresholds as the first grayscale threshold, and take the smaller of the two candidate grayscale thresholds as the second grayscale threshold.

[0017] Furthermore, multiple image blocks to be identified are obtained using the first grayscale threshold and the second grayscale threshold, and the comprehensive similarity value of the current fingerprint image is obtained by combining the grayscale values ​​of the image blocks to be identified, including:

[0018] S31: The fingerprint image is segmented using the first grayscale threshold and the second grayscale threshold to obtain multiple first image blocks to be identified and second image blocks to be identified;

[0019] S32: The plurality of first image blocks to be identified and second image blocks to be identified are sequentially compared with the fingerprint positions corresponding to the fingerprint reference images that the user has entered into the database to obtain the similarity values ​​corresponding to the plurality of first image blocks to be identified and second image blocks to be identified respectively.

[0020] S33: The comprehensive similarity value of the current fingerprint image is obtained by combining the similarity values ​​of the multiple first and second image blocks to be identified with the grayscale value of the image to be identified.

[0021] Furthermore, the fingerprint image is segmented using the first grayscale threshold and the second grayscale threshold to obtain multiple image blocks to be identified, including:

[0022] S311: Extract the grayscale values ​​of all pixel blocks in the fingerprint portion of the fingerprint image;

[0023] S312: Extract pixel blocks whose grayscale values ​​exceed the first grayscale threshold and use them as the first type of pixel blocks;

[0024] S313: Retrieve a target region within the area of ​​the first target region from the first type of pixel blocks, where the number of blocks containing the first type of pixel blocks exceeds a preset first block number threshold, and use it as the first target region;

[0025] S314: Segment the fingerprint image according to the position of the first target region to obtain a first image block to be identified;

[0026] S315: Extract pixel blocks whose grayscale values ​​exceed the second grayscale threshold but do not exceed the first grayscale threshold, and classify them as second type pixel blocks;

[0027] S316: In the second type of pixel blocks, retrieve a target region within the area of ​​the second target region where the number of blocks containing the second type of pixel blocks exceeds a preset second block number threshold, and use it as the second target region;

[0028] S317: Segment the fingerprint image according to the position of the second target region to obtain a second image block to be identified.

[0029] Furthermore, the area of ​​the first target region is smaller than the area of ​​the second target region, and the ratio between the areas of the first target region and the second target region is in the range of 1:1.7 to 1:2.3; at the same time, the first block number threshold is smaller than the second block number threshold, and the ratio between the first block number threshold and the second block number threshold is in the range of 1:2.1 to 1:2.8.

[0030] This invention proposes a fingerprint recognition access control system, which includes:

[0031] Image acquisition module: Real-time acquisition of fingerprint images for fingerprint recognition of the same user;

[0032] Threshold acquisition module: Based on the grayscale values ​​of the pixel blocks of the fingerprint portion of the currently acquired fingerprint image, acquire the first grayscale threshold and the second grayscale threshold corresponding to the current fingerprint image; wherein, the first grayscale threshold is greater than the second grayscale threshold;

[0033] Image block acquisition module: acquires multiple image blocks to be identified using the first grayscale threshold and the second grayscale threshold, and obtains the comprehensive similarity value of the current fingerprint image by combining the grayscale values ​​of the image blocks to be identified;

[0034] Similarity judgment module: When the comprehensive similarity value is not lower than the preset similarity threshold, the fingerprint image of the current user is determined to pass fingerprint verification.

[0035] Furthermore, the threshold acquisition module includes:

[0036] Grayscale extraction module: Extracts the grayscale values ​​of pixel blocks in the fingerprint portion of the current fingerprint image;

[0037] First grayscale data module: Extracts the grayscale values ​​of pixel blocks in the fingerprint portion that are lower than the optimal threshold corresponding to the fingerprint image, and uses them as the first grayscale data;

[0038] Second grayscale data module: Extracts the grayscale values ​​of pixel blocks in the fingerprint portion that are not lower than the optimal threshold corresponding to the fingerprint image, and uses them as second grayscale data;

[0039] First candidate grayscale threshold module: Sets the first candidate grayscale threshold for the fingerprint portion using the first grayscale value data;

[0040] Second candidate grayscale threshold module: Sets the second candidate grayscale threshold for the fingerprint portion using the second grayscale value data;

[0041] Threshold comparison module: compares the first candidate grayscale threshold with the second candidate grayscale threshold, takes the larger of the two candidate grayscale thresholds as the first grayscale threshold, and takes the smaller of the two candidate grayscale thresholds as the second grayscale threshold.

[0042] Furthermore, the image patch acquisition module includes:

[0043] Image block acquisition module: The fingerprint image is segmented using the first grayscale threshold and the second grayscale threshold to acquire multiple first image blocks to be identified and second image blocks to be identified;

[0044] Similarity value comparison module: The plurality of first image blocks to be identified and second image blocks to be identified are sequentially compared with the fingerprint positions corresponding to the fingerprint reference images that have been entered into the database by the user to obtain the similarity values ​​corresponding to the plurality of first image blocks to be identified and second image blocks to be identified respectively.

[0045] Similarity value acquisition module: The module uses the similarity values ​​corresponding to the multiple first and second image blocks to be identified, combined with the grayscale value of the image to be identified, to obtain the comprehensive similarity value of the current fingerprint image.

[0046] Furthermore, the image patch acquisition module includes:

[0047] Grayscale extraction module: Extracts the grayscale values ​​of all pixel blocks in the fingerprint portion of the fingerprint image;

[0048] First type of pixel block module: Extract pixel blocks whose grayscale values ​​exceed the first grayscale threshold and use them as first type of pixel blocks;

[0049] Target region setting module: retrieves a target region within the area of ​​the first target region whose number of blocks containing the first type of pixel blocks exceeds a preset first block number threshold, and uses it as the first target region;

[0050] First image block to be identified module: Segment the fingerprint image according to the position of the first target region to obtain a first image block to be identified;

[0051] Second-class pixel block determination module: Extracts pixel blocks whose grayscale values ​​exceed the second grayscale threshold but do not exceed the first grayscale threshold as second-class pixel blocks;

[0052] Area region retrieval module: Retrieves a target region within the area of ​​the second target region whose number of blocks containing the second type of pixel blocks exceeds a preset second block number threshold, and uses it as the second target region;

[0053] Second target region module: Segment the fingerprint image according to the position of the second target region to obtain a second image block to be identified.

[0054] Furthermore, the area of ​​the first target region is smaller than the area of ​​the second target region, and the ratio between the areas of the first target region and the second target region is in the range of 1:1.7 to 1:2.3; at the same time, the first block number threshold is smaller than the second block number threshold, and the ratio between the first block number threshold and the second block number threshold is in the range of 1:2.1 to 1:2.8.

[0055] The beneficial effects of this invention are as follows: The technical solution proposed in this invention reduces the possibility of false and missed identification by real-time acquisition and comparison of multiple fingerprint images provided by the user, thereby improving the accuracy and security of fingerprint recognition. By adopting a grayscale value comparison method, the complex manual feature extraction and matching process is avoided, greatly saving time and computing resources. This invention is fast and convenient, providing a good user experience and is suitable for application in various access control systems. By extracting local fingerprint image blocks for recognition, the recognition area can be reduced, thereby reducing processing and response time and improving recognition efficiency. By extracting fingerprint image blocks from fingerprint images and selecting the image blocks with the strongest fingerprint feature representation, the accuracy of fingerprint recognition can be enhanced and the probability of recognition failure can be reduced. By improving the image block extraction method and selecting the image blocks with the strongest fingerprint feature representation, the robustness of fingerprint recognition can be enhanced, enabling the fingerprint recognition system to have better performance when faced with different fingerprint images. By improving the efficiency and accuracy of fingerprint recognition, the demand for hardware resources can be reduced, thereby reducing the cost of the fingerprint recognition system. By improving recognition efficiency and accuracy, user waiting time can be reduced and the system response speed can be improved, thereby improving the user experience. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating the steps of a fingerprint recognition access control method according to the present invention.

[0057] Figure 2 This is a block diagram of a fingerprint recognition access control system according to the present invention. Detailed Implementation

[0058] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0059] Example 1

[0060] This embodiment provides a fingerprint recognition access control method, which includes:

[0061] S1: Real-time acquisition of fingerprint images for fingerprint recognition of the same user;

[0062] S2: Based on the grayscale values ​​of the pixel blocks of the fingerprint portion of the currently acquired fingerprint image, obtain the first grayscale threshold and the second grayscale threshold corresponding to the current fingerprint image; wherein, the first grayscale threshold is greater than the second grayscale threshold.

[0063] S3: Use the first grayscale threshold and the second grayscale threshold to obtain multiple image blocks to be identified, and combine the grayscale values ​​of the image blocks to be identified to obtain the comprehensive similarity value of the current fingerprint image.

[0064] S4: When the overall similarity value is not lower than the preset similarity threshold, it is determined that the current user's fingerprint image has passed fingerprint verification.

[0065] The working principle of the above technical solution is as follows: When a user presses their fingerprint sensor on the access control system, the system acquires the fingerprint image used for fingerprint recognition in real time. By analyzing the grayscale values ​​of the pixel blocks of the fingerprint portion in the currently acquired fingerprint image, the system obtains a first grayscale threshold and a second grayscale threshold corresponding to the current fingerprint image. These thresholds may be used to identify specific features in the fingerprint image for subsequent comparison and verification. Using the first and second grayscale thresholds, the system segments the fingerprint image into multiple image blocks to be identified and calculates the comprehensive similarity value of the current fingerprint image by combining the grayscale values ​​of these image blocks. This step may include extracting features from the fingerprint image and then comparing them with reference features stored in the database to calculate the similarity. When the comprehensive similarity value reaches or exceeds a preset similarity threshold, the system determines that the current user's fingerprint image passes fingerprint verification, thereby allowing the user to obtain access control privileges.

[0066] The effects of the above technical solution are as follows: It improves user experience by providing a novel grayscale value comparison method, avoiding complex manual feature extraction and matching processes, significantly saving time and computing resources, and conserving considerable manpower and material resources, thus improving operational efficiency and management level. Through grayscale value analysis and comprehensive similarity calculation of fingerprint images, high-precision fingerprint recognition can be achieved. This helps ensure that only authorized users can pass through the access control system, improving security. The use of grayscale thresholds can effectively combat some simple fingerprint forgery attacks, such as the use of fake fingerprints. This increases the security of the access control system, preventing deception by illegally copied fingerprints. The process of acquiring and processing fingerprint images is real-time, so users can quickly obtain the corresponding verification result after pressing their fingerprint on the access control system, improving user experience. This method dynamically obtains grayscale thresholds based on the features of the currently acquired fingerprint image and performs similarity calculations by combining the grayscale values ​​of the image block to be identified. Therefore, it can adapt to changes in fingerprint features under different environments, improving the robustness and applicability of the system. Judgment based on preset similarity thresholds can, to a certain extent, avoid false positives and false negatives, improving the reliability of fingerprint verification.

[0067] Example 2

[0068] In this embodiment, based on the grayscale values ​​of the pixel blocks of the fingerprint portion of the currently acquired fingerprint image, a first grayscale threshold and a second grayscale threshold corresponding to the current fingerprint image are obtained, including:

[0069] S21: Extract the grayscale values ​​of the pixel blocks of the fingerprint portion of the current fingerprint image;

[0070] S22: Extract the grayscale values ​​of the pixel blocks in the fingerprint portion that are lower than the optimal threshold corresponding to the fingerprint image (the optimal threshold can be obtained by the Otsu algorithm) and use them as the first grayscale value data;

[0071] S23: Extract the grayscale values ​​of the pixel blocks in the fingerprint portion that are not lower than the optimal threshold corresponding to the fingerprint image, and use them as the second grayscale value data;

[0072] S24: Use the first grayscale value data to set the first candidate grayscale threshold for the fingerprint portion;

[0073] S25: Use the second grayscale value data to set the second candidate grayscale threshold for the fingerprint portion;

[0074] S26: Compare the first candidate grayscale threshold with the second candidate grayscale threshold, and take the larger of the two candidate grayscale thresholds as the first grayscale threshold, and take the smaller of the two candidate grayscale thresholds as the second grayscale threshold.

[0075] The first candidate grayscale threshold and the second candidate grayscale threshold are obtained by the following formula:

[0076]

[0077]

[0078] Among them, H 01 H represents the first candidate grayscale threshold; z Indicates the optimal threshold; X max The value represents the inter-class variance corresponding to the optimal threshold; n1 represents the number of pixel blocks corresponding to the first grayscale value; H i α1 represents the grayscale value corresponding to the i-th pixel block; α1 represents the first threshold adjustment coefficient; L represents the total number of grayscale levels contained in the fingerprint image; P i This represents the probability of the gray level corresponding to the i-th pixel block in the first gray value data appearing;

[0079]

[0080]

[0081] Among them, H 02 H represents the second candidate grayscale threshold; z Indicates the optimal threshold; X max The value of the inter-class variance corresponding to the optimal threshold; n2 represents the number of pixel blocks corresponding to the second grayscale value; Hj α represents the grayscale value corresponding to the j-th pixel block; α2 represents the second threshold adjustment coefficient; L represents the total number of grayscale levels contained in the fingerprint image; P j This represents the probability of the gray level corresponding to the j-th pixel block in the second gray value data appearing.

[0082] The working principle of the above technical solution is as follows: extract the grayscale values ​​of the pixel blocks of the fingerprint portion of the current fingerprint image; extract the grayscale values ​​of the pixel blocks of the fingerprint portion that are lower than the optimal threshold corresponding to the fingerprint image (the optimal threshold can be obtained by the Otsu algorithm), and use them as the first grayscale value data; extract the grayscale values ​​of the pixel blocks of the fingerprint portion that are not lower than the optimal threshold corresponding to the fingerprint image, and use them as the second grayscale value data; set the first candidate grayscale threshold of the fingerprint portion using the first grayscale value data; set the second candidate grayscale threshold of the fingerprint portion using the second grayscale value data; compare the first candidate grayscale threshold with the second candidate grayscale threshold, and use the larger of the two candidate grayscale thresholds as the first grayscale threshold, and use the smaller of the two candidate grayscale thresholds as the second grayscale threshold.

[0083] The above technical solution achieves the following effects: By extracting the grayscale values ​​of pixel blocks in the fingerprint image and determining the first and second grayscale value data based on the optimal threshold, this solution can automatically adjust the grayscale threshold according to the characteristics of different fingerprint images, thus adapting to different fingerprint images; by comparing the first and second candidate grayscale thresholds and selecting the relatively suitable thresholds as the first and second grayscale thresholds, this solution can ensure high-precision processing of the current fingerprint image and improve the accuracy of fingerprint recognition; by processing based on the grayscale values ​​of pixel blocks, this solution considers more information in the process of extracting grayscale values ​​and determining grayscale thresholds, thereby improving the reliability and robustness of the fingerprint image; since the calculation process of this solution is relatively simple and does not require a large amount of computing resources, the first and second grayscale thresholds can be quickly obtained and determined, making it suitable for real-time fingerprint image processing scenarios. The above-mentioned calculation formulas for the first and second candidate grayscale thresholds improve the rationality of the threshold setting, thereby improving the accuracy of subsequent target region calibration and the fingerprint characteristic characterization of the fingerprint portion within the target region. By improving the accuracy of subsequent target region calibration and the fingerprint characteristic characterization of the fingerprint portion within the target region, this effectively improves the recognition accuracy when only local fingerprints are identified. Meanwhile, the formula uses the optimal threshold and inter-class variance values ​​as references. By adjusting the threshold adjustment coefficients α1 and α2, the first and second candidate grayscale thresholds can be adaptively determined based on the grayscale distribution of pixel blocks in the first and second grayscale data. This allows for better adaptation to the characteristics of different fingerprint images, improving the accuracy and robustness of fingerprint recognition. The formula also considers the probability P of the grayscale level of the pixel block appearing in the first and second grayscale data. i and P j This allows for a better reflection of the distribution of different gray levels in the fingerprint image, making the selected candidate gray-level thresholds more reasonable and accurate. Adjustment coefficients α1 and α2 are introduced into the formula to control the adjustment range of the candidate gray-level thresholds. By setting the adjustment coefficients appropriately, the accuracy and sensitivity of the gray-level thresholds can be further adjusted while maintaining adaptability, thereby improving the performance of the fingerprint recognition system.

[0084] Example 3

[0085] In this embodiment, multiple image blocks to be identified are obtained using the first grayscale threshold and the second grayscale threshold, and the comprehensive similarity value of the current fingerprint image is obtained by combining the grayscale values ​​of the image blocks to be identified, including:

[0086] S31: The fingerprint image is segmented using the first grayscale threshold and the second grayscale threshold to obtain multiple first image blocks to be identified and second image blocks to be identified;

[0087] S32: The plurality of first image blocks to be identified and second image blocks to be identified are sequentially compared with the fingerprint positions corresponding to the fingerprint reference images that the user has entered into the database to obtain the similarity values ​​corresponding to the plurality of first image blocks to be identified and second image blocks to be identified respectively.

[0088] S33: The comprehensive similarity value of the current fingerprint image is obtained by combining the similarity values ​​of the multiple first and second image blocks to be identified with the grayscale value of the image to be identified.

[0089] The overall similarity value of the current fingerprint image is obtained using the following formula:

[0090] SIM=0.5·(λ1·S p1 +λ2·S p2 )

[0091]

[0092] Where SIM represents the overall similarity score; S p1 S represents the average similarity value of all first image patches to be identified; p2 λ1 and λ2 represent the average similarity values ​​of all the first and second image patches to be identified, respectively; C1 and C2 represent the weight coefficients corresponding to the first and second image patches to be identified, respectively; B represents the number of the first and second image patches to be identified, respectively. s N represents the ratio between the area of ​​the first target region and the area of ​​the second target region. i M represents the number of first-class pixel blocks contained in the i-th first image block to be identified; i N represents the number of pixel blocks of the first class contained in the i-th first image block to be identified whose gray value is greater than the second gray threshold; 01 N represents the threshold number of the first block; 02 N represents the threshold number of the second block; j M represents the number of second-class pixel blocks contained in the j-th second image block to be identified; j This represents the number of pixel blocks in the second type of pixel blocks contained in the j-th second image block to be identified whose gray value is not lower than the first gray threshold.

[0093] The working principle of the above technical solution is as follows: The fingerprint image is segmented using a first grayscale threshold and a second grayscale threshold to obtain multiple first and second image blocks to be identified. These image blocks contain different feature regions in the fingerprint image. Next, each first and second image block to be identified is compared with a fingerprint reference image already entered into the user's database. Various fingerprint matching algorithms can be used for similarity comparison, such as algorithms based on feature points or ridges. Through comparison, the similarity value between each image block and the fingerprint reference image in the database can be calculated. The overall similarity value of the current fingerprint image is calculated by combining the similarity values ​​of each of the multiple first and second image blocks to be identified with their corresponding grayscale values.

[0094] The above technical solution achieves the following effects: By segmenting the fingerprint image into multiple image blocks and comparing them with fingerprint reference images in the database, the accuracy of the fingerprint recognition system can be improved. This is because each image block can match the fingerprint images in the database more precisely, reducing the overall matching error. After segmenting the fingerprint image, each image block can be independently compared for similarity, thus avoiding the impact of damage, contamination, or deformation areas in the entire fingerprint image on the overall matching result. This enhances the system's robustness to various fingerprint variations. Compared to directly matching the entire fingerprint image, segmenting it into multiple image blocks allows for parallel processing of similarity comparisons for each block, thereby improving the matching speed. Because each image block is smaller, the required computation is reduced accordingly, allowing for more efficient use of computational resources, optimizing algorithm performance, and improving the recognition rate. Combining the grayscale values ​​of the image blocks to be identified allows for a more comprehensive evaluation of the fingerprint image similarity. Grayscale values ​​reflect the brightness information within the image block; by comprehensively considering both similarity values ​​and grayscale values, the degree of fingerprint matching can be determined more accurately. The above formula improves the accuracy of similarity evaluation by integrating similarity. The weighting effectively enhances the accuracy of determining the representational importance of pixel blocks within their corresponding grayscale ranges, further improving the accuracy of obtaining the comprehensive similarity threshold. Furthermore, by introducing weight coefficients λ1 and λ2, different weights can be assigned to different image blocks to be identified. This allows for adjustments to the influence of different image blocks on the overall similarity based on actual conditions, improving the flexibility of similarity calculation; B in the formula... s This represents the ratio between the areas of the first target region and the second target region. This ratio can be used to measure the difference in the size of the fingerprint regions occupied by the two image patches to be identified. By considering the target region area ratio, a reasonable similarity assessment can be performed for image patches of different sizes; N in the formula... i and M jand represent the number of pixel blocks in the i-th first image block to be identified and the j-th second image block to be identified, respectively, and the number of pixel blocks that reach a certain grayscale threshold. This allows for the introduction of statistical information on pixel block grayscale values ​​into the similarity calculation, providing a more comprehensive consideration of the brightness characteristics within image blocks; N in the formula... 01 N 02 This is related to the first and second grayscale thresholds. These thresholds can be adjusted according to the actual situation to meet the characteristics and needs of different fingerprint images. By setting adaptive thresholds, the adaptability and accuracy of similarity calculation can be improved.

[0095] Example 4

[0096] In this embodiment, the fingerprint image is segmented using a first grayscale threshold and a second grayscale threshold to obtain multiple image blocks to be identified, including:

[0097] S311: Extract the grayscale values ​​of all pixel blocks in the fingerprint portion of the fingerprint image;

[0098] S312: Extract pixel blocks whose grayscale values ​​exceed the first grayscale threshold and use them as the first type of pixel blocks;

[0099] S313: Retrieve a target region within the area of ​​the first target region from the first type of pixel blocks, where the number of blocks containing the first type of pixel blocks exceeds a preset first block number threshold, and use it as the first target region;

[0100] S314: Segment the fingerprint image according to the position of the first target region to obtain a first image block to be identified;

[0101] S315: Extract pixel blocks whose grayscale values ​​exceed the second grayscale threshold but do not exceed the first grayscale threshold, and classify them as second type pixel blocks;

[0102] S316: In the second type of pixel blocks, retrieve a target region within the area of ​​the second target region where the number of blocks containing the second type of pixel blocks exceeds a preset second block number threshold, and use it as the second target region;

[0103] S317: Segment the fingerprint image according to the position of the second target region to obtain a second image block to be identified.

[0104] The working principle of the above technical solution is as follows: All pixel blocks are extracted from the entire fingerprint image, and their grayscale values ​​are recorded; pixel blocks with grayscale values ​​exceeding a first grayscale threshold are identified, as these pixel blocks may represent the main feature parts of the fingerprint image; within the first type of pixel blocks, a target region satisfying a preset first block number threshold is found, i.e., a region containing a sufficient number of first-type pixel blocks within a certain range. This region is considered the first target region; based on the location information of the first target region, the fingerprint image is segmented to extract a first image block to be identified, which may contain important fingerprint features; next, pixel blocks with grayscale values ​​between the first and second grayscale thresholds but not reaching the second grayscale threshold are identified, as these pixel blocks may contain secondary fingerprint features; within the second type of pixel blocks, a target region satisfying a preset second block number threshold is found, which is also a region containing a sufficient number of second-type pixel blocks, as the second target region; based on the location information of the second target region, the fingerprint image is segmented to extract a second image block to be identified, which may contain secondary fingerprint features.

[0105] The effects of the above technical solution are as follows: By segmenting and extracting regions from the fingerprint image, the image block to be identified, containing both primary and secondary fingerprint features, can be accurately extracted, which helps improve the accuracy and comprehensiveness of the fingerprint recognition system in terms of fingerprint features; by setting a first grayscale threshold and a second grayscale threshold, noise and non-fingerprint regions in the image can be effectively filtered out, thereby increasing the focus on the fingerprint part and reducing the impact of noise on the recognition results; segmenting the fingerprint image into multiple image blocks to be identified and processing the feature information of different regions separately helps increase the robustness of the system in recognizing different fingerprint features and improves the fingerprint recognition effect of the system in complex environments; through multiple segmentations and extractions, the technical solution can more comprehensively capture the feature information in the fingerprint image, thereby improving the accuracy and reliability of the fingerprint recognition system and making it more suitable for various practical application scenarios. By extracting pixel blocks in the fingerprint image that exceed a first grayscale threshold and those that do not exceed the first grayscale threshold but exceed a second grayscale threshold, fingerprint features can be extracted more accurately, improving recognition accuracy. Retrieving the target region from the first and second types of pixel blocks allows for segmentation of the fingerprint image based on the target region's location, obtaining the image block to be recognized. This avoids processing the entire image block, reducing computational complexity and improving efficiency. When processing large numbers of fingerprint images, this target region-based segmentation and recognition method significantly reduces the amount of data that needs to be stored and processed, thus optimizing storage space utilization. Furthermore, the use of a grayscale threshold-based pixel block classification method and a target region-based target selection method effectively prevents unauthorized intrusion and security vulnerabilities, improving the security of the access control system.

[0106] Example 5

[0107] In this embodiment, the area of ​​the first target region is smaller than the area of ​​the second target region, and the ratio between the areas of the first target region and the second target region is in the range of 1:1.7 to 1:2.3; at the same time, the first block number threshold is smaller than the second block number threshold, and the ratio between the first block number threshold and the second block number threshold is in the range of 1:2.1 to 1:2.8.

[0108] The working principle of the above technical solution is as follows: First, for a specific region in the fingerprint image, the areas of the first target region and the second target region are calculated, with the ratio between the areas of the first target region and the second target region ranging from 1:1.7 to 1:2.3. This means that after calculating the target region areas, the system will compare and filter, and only target regions that meet the specific ratio range will be retained; the ratio between the first block number threshold and the second block number threshold ranges from 1:2.1 to 1:2.8.

[0109] The effects of the above technical solution are as follows: By limiting and filtering the target area and numerical thresholds, the accuracy and precision of fingerprint image processing can be improved. Only target areas and values ​​that conform to a specific ratio range will be retained, thereby reducing the possibility of misjudgment and misidentification; setting specific ratio ranges and numerical thresholds can enhance the system's resistance to noise and interference. This means that even under certain interference conditions, the system can still identify and extract fingerprint features well; by flexibly setting the ratio range and numerical thresholds, this technical solution may have a certain degree of adaptability, applicable to fingerprint images of different types and qualities, and to some extent overcome the challenges brought about by changes in image quality and environment; once a suitable ratio range and numerical thresholds are determined, this technical solution may be able to process large amounts of fingerprint image data quickly and effectively, thereby improving processing efficiency and response speed.

[0110] Example 6

[0111] This embodiment discloses a fingerprint recognition access control system, which includes:

[0112] Image acquisition module: Real-time acquisition of fingerprint images for fingerprint recognition of the same user;

[0113] Threshold acquisition module: Based on the grayscale values ​​of the pixel blocks of the fingerprint portion of the currently acquired fingerprint image, acquire the first grayscale threshold and the second grayscale threshold corresponding to the current fingerprint image; wherein, the first grayscale threshold is greater than the second grayscale threshold;

[0114] Image block acquisition module: acquires multiple image blocks to be identified using the first grayscale threshold and the second grayscale threshold, and obtains the comprehensive similarity value of the current fingerprint image by combining the grayscale values ​​of the image blocks to be identified;

[0115] Similarity judgment module: When the comprehensive similarity value is not lower than the preset similarity threshold, the fingerprint image of the current user is determined to pass fingerprint verification.

[0116] The working principle of the above technical solution is as follows: When a user presses their fingerprint sensor on the access control system, the system acquires the fingerprint image used for fingerprint recognition in real time. By analyzing the grayscale values ​​of the pixel blocks of the fingerprint portion in the currently acquired fingerprint image, the system obtains a first grayscale threshold and a second grayscale threshold corresponding to the current fingerprint image. These thresholds may be used to identify specific features in the fingerprint image for subsequent comparison and verification. Using the first and second grayscale thresholds, the system segments the fingerprint image into multiple image blocks to be identified and calculates the comprehensive similarity value of the current fingerprint image by combining the grayscale values ​​of these image blocks. This step may include extracting features from the fingerprint image and then comparing them with reference features stored in the database to calculate the similarity. When the comprehensive similarity value reaches or exceeds a preset similarity threshold, the system determines that the current user's fingerprint image passes fingerprint verification, thereby allowing the user to obtain access control privileges.

[0117] The effects of the above technical solution are as follows: It improves user experience by providing a novel grayscale value comparison method, avoiding complex manual feature extraction and matching processes, significantly saving time and computing resources, and conserving considerable manpower and material resources, thus improving operational efficiency and management level. Through grayscale value analysis and comprehensive similarity calculation of fingerprint images, high-precision fingerprint recognition can be achieved. This helps ensure that only authorized users can pass through the access control system, improving security. The use of grayscale thresholds can effectively combat some simple fingerprint forgery attacks, such as the use of fake fingerprints. This increases the security of the access control system, preventing deception by illegally copied fingerprints. The process of acquiring and processing fingerprint images is real-time, so users can quickly obtain the corresponding verification result after pressing their fingerprint on the access control system, improving user experience. This method dynamically obtains grayscale thresholds based on the features of the currently acquired fingerprint image and performs similarity calculations by combining the grayscale values ​​of the image block to be identified. Therefore, it can adapt to changes in fingerprint features under different environments, improving the robustness and applicability of the system. Judgment based on preset similarity thresholds can, to a certain extent, avoid false positives and false negatives, improving the reliability of fingerprint verification.

[0118] Example 7

[0119] In this embodiment, the threshold acquisition module includes:

[0120] Grayscale extraction module: Extracts the grayscale values ​​of pixel blocks in the fingerprint portion of the current fingerprint image;

[0121] First grayscale data module: Extracts the grayscale values ​​of pixel blocks in the fingerprint portion that are lower than the optimal threshold corresponding to the fingerprint image (the optimal threshold can be obtained by the Otsu algorithm), and uses them as the first grayscale data;

[0122] Second grayscale data module: Extracts the grayscale values ​​of pixel blocks in the fingerprint portion that are not lower than the optimal threshold corresponding to the fingerprint image, and uses them as second grayscale data;

[0123] First candidate grayscale threshold module: Sets the first candidate grayscale threshold for the fingerprint portion using the first grayscale value data;

[0124] Second candidate grayscale threshold module: Sets the second candidate grayscale threshold for the fingerprint portion using the second grayscale value data;

[0125] Threshold comparison module: compares the first candidate grayscale threshold with the second candidate grayscale threshold, takes the larger of the two candidate grayscale thresholds as the first grayscale threshold, and takes the smaller of the two candidate grayscale thresholds as the second grayscale threshold.

[0126] The first candidate grayscale threshold and the second candidate grayscale threshold are obtained by the following formula:

[0127]

[0128]

[0129] Among them, H 01 H represents the first candidate grayscale threshold; z Indicates the optimal threshold; X max The value represents the inter-class variance corresponding to the optimal threshold; n1 represents the number of pixel blocks corresponding to the first grayscale value; H i α1 represents the grayscale value corresponding to the i-th pixel block; α1 represents the first threshold adjustment coefficient; L represents the total number of grayscale levels contained in the fingerprint image; P i This represents the probability of the gray level corresponding to the i-th pixel block in the first gray value data appearing;

[0130]

[0131]

[0132] Among them, H 02 H represents the second candidate grayscale threshold;z Indicates the optimal threshold; X max The value of the inter-class variance corresponding to the optimal threshold; n2 represents the number of pixel blocks corresponding to the second grayscale value; H j α represents the grayscale value corresponding to the j-th pixel block; α2 represents the second threshold adjustment coefficient; L represents the total number of grayscale levels contained in the fingerprint image; P j This represents the probability of the gray level corresponding to the j-th pixel block in the second gray value data appearing.

[0133] The working principle of the above technical solution is as follows: extract the grayscale values ​​of the pixel blocks of the fingerprint portion of the current fingerprint image; extract the grayscale values ​​of the pixel blocks of the fingerprint portion that are lower than the optimal threshold corresponding to the fingerprint image (the optimal threshold can be obtained by the Otsu algorithm), and use them as the first grayscale value data; extract the grayscale values ​​of the pixel blocks of the fingerprint portion that are not lower than the optimal threshold corresponding to the fingerprint image, and use them as the second grayscale value data; set the first candidate grayscale threshold of the fingerprint portion using the first grayscale value data; set the second candidate grayscale threshold of the fingerprint portion using the second grayscale value data; compare the first candidate grayscale threshold with the second candidate grayscale threshold, and use the larger of the two candidate grayscale thresholds as the first grayscale threshold, and use the smaller of the two candidate grayscale thresholds as the second grayscale threshold.

[0134] The above technical solution achieves the following effects: By extracting the grayscale values ​​of pixel blocks in the fingerprint image and determining the first and second grayscale value data based on the optimal threshold, this solution can automatically adjust the grayscale threshold according to the characteristics of different fingerprint images, thus adapting to different fingerprint images; by comparing the first and second candidate grayscale thresholds and selecting the relatively suitable thresholds as the first and second grayscale thresholds, this solution can ensure high-precision processing of the current fingerprint image and improve the accuracy of fingerprint recognition; by processing based on the grayscale values ​​of pixel blocks, this solution considers more information in the process of extracting grayscale values ​​and determining grayscale thresholds, thereby improving the reliability and robustness of the fingerprint image; since the calculation process of this solution is relatively simple and does not require a large amount of computing resources, the first and second grayscale thresholds can be quickly obtained and determined, making it suitable for real-time fingerprint image processing scenarios. The above-mentioned calculation formulas for the first and second candidate grayscale thresholds improve the rationality of the threshold setting, thereby improving the accuracy of subsequent target region calibration and the fingerprint characteristic characterization of the fingerprint portion within the target region. By improving the accuracy of subsequent target region calibration and the fingerprint characteristic characterization of the fingerprint portion within the target region, this effectively improves the recognition accuracy when only local fingerprints are identified. Meanwhile, the formula uses the optimal threshold and inter-class variance values ​​as references. By adjusting the threshold adjustment coefficients α1 and α2, the first and second candidate grayscale thresholds can be adaptively determined based on the grayscale distribution of pixel blocks in the first and second grayscale data. This allows for better adaptation to the characteristics of different fingerprint images, improving the accuracy and robustness of fingerprint recognition. The formula also considers the probability P of the grayscale level of the pixel block appearing in the first and second grayscale data. i and P j This allows for a better reflection of the distribution of different gray levels in the fingerprint image, making the selected candidate gray-level thresholds more reasonable and accurate. Adjustment coefficients α1 and α2 are introduced into the formula to control the adjustment range of the candidate gray-level thresholds. By setting the adjustment coefficients appropriately, the accuracy and sensitivity of the gray-level thresholds can be further adjusted while maintaining adaptability, thereby improving the performance of the fingerprint recognition system.

[0135] Example 8

[0136] In this embodiment, the image patch acquisition module includes:

[0137] Image block acquisition module: The fingerprint image is segmented using the first grayscale threshold and the second grayscale threshold to acquire multiple first image blocks to be identified and second image blocks to be identified;

[0138] Similarity value comparison module: The plurality of first image blocks to be identified and second image blocks to be identified are sequentially compared with the fingerprint positions corresponding to the fingerprint reference images that have been entered into the database by the user to obtain the similarity values ​​corresponding to the plurality of first image blocks to be identified and second image blocks to be identified respectively.

[0139] Similarity value acquisition module: The module uses the similarity values ​​corresponding to the multiple first and second image blocks to be identified, combined with the grayscale value of the image to be identified, to obtain the comprehensive similarity value of the current fingerprint image.

[0140] The overall similarity value of the current fingerprint image is obtained using the following formula:

[0141] SIM=0.5·(λ1·S p1 +λ2·S p2 )

[0142]

[0143]

[0144] Where SIM represents the overall similarity score; S p1 S represents the average similarity value of all first image patches to be identified; p2 λ1 and λ2 represent the average similarity values ​​of all the first and second image patches to be identified, respectively; C1 and C2 represent the weight coefficients corresponding to the first and second image patches to be identified, respectively; B represents the number of the first and second image patches to be identified, respectively. s N represents the ratio between the area of ​​the first target region and the area of ​​the second target region. i M represents the number of first-class pixel blocks contained in the i-th first image block to be identified; i N represents the number of pixel blocks of the first class contained in the i-th first image block to be identified whose gray value is greater than the second gray threshold; 01 N represents the threshold number of the first block; 02 N represents the threshold number of the second block; j M represents the number of second-class pixel blocks contained in the j-th second image block to be identified; j This represents the number of pixel blocks in the second type of pixel blocks contained in the j-th second image block to be identified whose gray value is not lower than the first gray threshold.

[0145] The working principle of the above technical solution is as follows: The fingerprint image is segmented using a first grayscale threshold and a second grayscale threshold to obtain multiple first and second image blocks to be identified. These image blocks contain different feature regions in the fingerprint image. Next, each first and second image block to be identified is compared with a fingerprint reference image already entered into the user's database. Various fingerprint matching algorithms can be used for similarity comparison, such as algorithms based on feature points or ridges. Through comparison, the similarity value between each image block and the fingerprint reference image in the database can be calculated. The overall similarity value of the current fingerprint image is calculated by combining the similarity values ​​of each of the multiple first and second image blocks to be identified with their corresponding grayscale values.

[0146] The above technical solution achieves the following effects: By segmenting the fingerprint image into multiple image blocks and comparing them with fingerprint reference images in the database, the accuracy of the fingerprint recognition system can be improved. This is because each image block can match the fingerprint images in the database more precisely, reducing the overall matching error. After segmenting the fingerprint image, each image block can be independently compared for similarity, thus avoiding the impact of damage, contamination, or deformation areas in the entire fingerprint image on the overall matching result. This enhances the system's robustness to various fingerprint variations. Compared to directly matching the entire fingerprint image, segmenting it into multiple image blocks allows for parallel processing of similarity comparisons for each block, thereby improving the matching speed. Because each image block is smaller, the required computation is reduced accordingly, allowing for more efficient use of computational resources, optimizing algorithm performance, and improving the recognition rate. Combining the grayscale values ​​of the image blocks to be identified allows for a more comprehensive evaluation of the fingerprint image similarity. Grayscale values ​​reflect the brightness information within the image block; by comprehensively considering both similarity values ​​and grayscale values, the degree of fingerprint matching can be determined more accurately. The above formula improves the accuracy of similarity evaluation by integrating similarity. The weighting effectively enhances the accuracy of determining the representational importance of pixel blocks within their corresponding grayscale ranges, further improving the accuracy of obtaining the comprehensive similarity threshold. Furthermore, by introducing weight coefficients λ1 and λ2, different weights can be assigned to different image blocks to be identified. This allows for adjustments to the influence of different image blocks on the overall similarity based on actual conditions, improving the flexibility of similarity calculation; B in the formula... s This represents the ratio between the areas of the first target region and the second target region. This ratio can be used to measure the difference in the size of the fingerprint regions occupied by the two image patches to be identified. By considering the target region area ratio, a reasonable similarity assessment can be performed for image patches of different sizes; N in the formula... i and M jand represent the number of pixel blocks in the i-th first image block to be identified and the j-th second image block to be identified, respectively, and the number of pixel blocks that reach a certain grayscale threshold. This allows for the introduction of statistical information on pixel block grayscale values ​​into the similarity calculation, providing a more comprehensive consideration of the brightness characteristics within image blocks; N in the formula... 01 N 02 This is related to the first and second grayscale thresholds. These thresholds can be adjusted according to the actual situation to meet the characteristics and needs of different fingerprint images. By setting adaptive thresholds, the adaptability and accuracy of similarity calculation can be improved.

[0147] Example 9

[0148] In this embodiment, the image block acquisition module to be identified includes:

[0149] Grayscale extraction module: Extracts the grayscale values ​​of all pixel blocks in the fingerprint portion of the fingerprint image;

[0150] First type of pixel block module: Extract pixel blocks whose grayscale values ​​exceed the first grayscale threshold and use them as first type of pixel blocks;

[0151] Target region setting module: retrieves a target region within the area of ​​the first target region whose number of blocks containing the first type of pixel blocks exceeds a preset first block number threshold, and uses it as the first target region;

[0152] First image block to be identified module: Segment the fingerprint image according to the position of the first target region to obtain a first image block to be identified;

[0153] Second-class pixel block determination module: Extracts pixel blocks whose grayscale values ​​exceed the second grayscale threshold but do not exceed the first grayscale threshold as second-class pixel blocks;

[0154] Area region retrieval module: Retrieves a target region within the area of ​​the second target region whose number of blocks containing the second type of pixel blocks exceeds a preset second block number threshold, and uses it as the second target region;

[0155] Second target region module: Segment the fingerprint image according to the position of the second target region to obtain a second image block to be identified.

[0156] The working principle of the above technical solution is as follows: All pixel blocks are extracted from the entire fingerprint image, and their grayscale values ​​are recorded; pixel blocks with grayscale values ​​exceeding a first grayscale threshold are identified, as these pixel blocks may represent the main feature parts of the fingerprint image; within the first type of pixel blocks, a target region satisfying a preset first block number threshold is found, i.e., a region containing a sufficient number of first-type pixel blocks within a certain range. This region is considered the first target region; based on the location information of the first target region, the fingerprint image is segmented to extract a first image block to be identified, which may contain important fingerprint features; next, pixel blocks with grayscale values ​​between the first and second grayscale thresholds but not reaching the second grayscale threshold are identified, as these pixel blocks may contain secondary fingerprint features; within the second type of pixel blocks, a target region satisfying a preset second block number threshold is found, which is also a region containing a sufficient number of second-type pixel blocks, as the second target region; based on the location information of the second target region, the fingerprint image is segmented to extract a second image block to be identified, which may contain secondary fingerprint features.

[0157] The effects of the above technical solution are as follows: By segmenting and extracting regions from the fingerprint image, the image block to be identified, containing both primary and secondary fingerprint features, can be accurately extracted, which helps improve the accuracy and comprehensiveness of the fingerprint recognition system in terms of fingerprint features; by setting a first grayscale threshold and a second grayscale threshold, noise and non-fingerprint regions in the image can be effectively filtered out, thereby increasing the focus on the fingerprint part and reducing the impact of noise on the recognition results; segmenting the fingerprint image into multiple image blocks to be identified and processing the feature information of different regions separately helps increase the robustness of the system in recognizing different fingerprint features and improves the fingerprint recognition effect of the system in complex environments; through multiple segmentations and extractions, the technical solution can more comprehensively capture the feature information in the fingerprint image, thereby improving the accuracy and reliability of the fingerprint recognition system and making it more suitable for various practical application scenarios. By extracting pixel blocks in the fingerprint image that exceed a first grayscale threshold and those that do not exceed the first grayscale threshold but exceed a second grayscale threshold, fingerprint features can be extracted more accurately, improving recognition accuracy. Retrieving the target region from the first and second types of pixel blocks allows for segmentation of the fingerprint image based on the target region's location, obtaining the image block to be recognized. This avoids processing the entire image block, reducing computational complexity and improving efficiency. When processing large numbers of fingerprint images, this target region-based segmentation and recognition method significantly reduces the amount of data that needs to be stored and processed, thus optimizing storage space utilization. Furthermore, the use of a grayscale threshold-based pixel block classification method and a target region-based target selection method effectively prevents unauthorized intrusion and security vulnerabilities, improving the security of the access control system.

[0158] Example 10

[0159] In this embodiment, the area of ​​the first target region is smaller than the area of ​​the second target region, and the ratio between the areas of the first target region and the second target region is in the range of 1:1.7 to 1:2.3; at the same time, the first block number threshold is smaller than the second block number threshold, and the ratio between the first block number threshold and the second block number threshold is in the range of 1:2.1 to 1:2.8.

[0160] The working principle of the above technical solution is as follows: First, for a specific region in the fingerprint image, the areas of the first target region and the second target region are calculated, with the ratio between the areas of the first target region and the second target region ranging from 1:1.7 to 1:2.3. This means that after calculating the target region areas, the system will compare and filter, and only target regions that meet the specific ratio range will be retained; the ratio between the first block number threshold and the second block number threshold ranges from 1:2.1 to 1:2.8.

[0161] The effects of the above technical solution are as follows: By limiting and filtering the target area and numerical thresholds, the accuracy and precision of fingerprint image processing can be improved. Only target areas and values ​​that conform to a specific ratio range will be retained, thereby reducing the possibility of misjudgment and misidentification; setting specific ratio ranges and numerical thresholds can enhance the system's resistance to noise and interference. This means that even under certain interference conditions, the system can still identify and extract fingerprint features well; by flexibly setting the ratio range and numerical thresholds, this technical solution may have a certain degree of adaptability, applicable to fingerprint images of different types and qualities, and to some extent overcome the challenges brought about by changes in image quality and environment; once a suitable ratio range and numerical thresholds are determined, this technical solution may be able to process large amounts of fingerprint image data quickly and effectively, thereby improving processing efficiency and response speed.

[0162] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A fingerprint recognition access control method, characterized in that, The fingerprint recognition access control method includes: Real-time acquisition of fingerprint images for the same user used for fingerprint recognition; Based on the grayscale values ​​of the pixel blocks of the fingerprint portion in the currently acquired fingerprint image, a first grayscale threshold and a second grayscale threshold corresponding to the current fingerprint image are obtained; wherein, the first grayscale threshold is greater than the second grayscale threshold; including: Extract the grayscale values ​​of pixel blocks in the fingerprint portion of the current fingerprint image; Extract the grayscale values ​​of the pixel blocks in the fingerprint portion that are lower than the optimal threshold corresponding to the fingerprint image, and use them as the first grayscale value data; Extract the grayscale values ​​of the pixel blocks in the fingerprint portion that are not lower than the optimal threshold corresponding to the fingerprint image, and use them as the second grayscale value data; The first candidate grayscale threshold for the fingerprint portion is set using the first grayscale value data; The second grayscale value data is used to set the second candidate grayscale threshold for the fingerprint portion; The first candidate grayscale threshold is compared with the second candidate grayscale threshold. The larger of the two candidate grayscale thresholds is taken as the first grayscale threshold, and the smaller of the two candidate grayscale thresholds is taken as the second grayscale threshold. Multiple image blocks to be identified are obtained using the first grayscale threshold and the second grayscale threshold, and the comprehensive similarity value of the current fingerprint image is obtained by combining the grayscale values ​​of the image blocks to be identified. When the overall similarity value is not lower than the preset similarity threshold, it is determined that the current user's fingerprint image has passed fingerprint verification.

2. The fingerprint recognition access control method according to claim 1, characterized in that, Multiple image blocks to be identified are obtained using the first grayscale threshold and the second grayscale threshold, and the comprehensive similarity value of the current fingerprint image is obtained by combining the grayscale values ​​of the image blocks to be identified, including: The fingerprint image is segmented using the first grayscale threshold and the second grayscale threshold to obtain multiple first image blocks to be identified and second image blocks to be identified; The plurality of first image blocks to be identified and second image blocks to be identified are sequentially compared with the fingerprint positions corresponding to the fingerprint reference images that the user has entered into the database to obtain the similarity values ​​corresponding to each of the plurality of first image blocks to be identified and second image blocks to be identified. The comprehensive similarity value of the current fingerprint image is obtained by combining the similarity values ​​corresponding to the multiple first and second image blocks to be identified with the grayscale value of the image to be identified.

3. The fingerprint recognition access control method according to claim 2, characterized in that, The fingerprint image is segmented using the first grayscale threshold and the second grayscale threshold to obtain multiple image blocks to be identified, including: Extract the grayscale values ​​of all pixel blocks in the fingerprint portion of the fingerprint image; Pixel blocks whose grayscale values ​​exceed the first grayscale threshold are extracted and classified as first-class pixel blocks; In the first type of pixel blocks, a target region whose number of blocks containing the first type of pixel blocks within the area of ​​the first target region exceeds a preset first block number threshold is selected as the first target region; The fingerprint image is segmented according to the location of the first target region to obtain a first image block to be identified; Pixel blocks whose grayscale values ​​exceed the second grayscale threshold but do not exceed the first grayscale threshold are extracted and classified as second-class pixel blocks; In the second type of pixel blocks, a target region whose number of blocks containing the second type of pixel blocks within the area of ​​the second target region exceeds a preset second block number threshold is selected as the second target region; The fingerprint image is segmented based on the location of the second target region to obtain a second image block to be identified.

4. The fingerprint recognition access control method according to claim 3, characterized in that, The area of ​​the first target region is smaller than the area of ​​the second target region, and the ratio between the areas of the first target region and the second target region is in the range of 1:1.7 to 1:2.3; at the same time, the first block number threshold is smaller than the second block number threshold, and the ratio between the first block number threshold and the second block number threshold is in the range of 1:2.1 to 1:2.

8.

5. A fingerprint recognition access control system, characterized in that, The fingerprint recognition access control system includes: Image acquisition module: Real-time acquisition of fingerprint images for fingerprint recognition of the same user; Threshold acquisition module: Based on the grayscale values ​​of pixel blocks in the fingerprint portion of the currently acquired fingerprint image, acquire a first grayscale threshold and a second grayscale threshold corresponding to the current fingerprint image; wherein, the first grayscale threshold is greater than the second grayscale threshold; including: Grayscale extraction module: Extracts the grayscale values ​​of pixel blocks in the fingerprint portion of the current fingerprint image; First grayscale data module: Extracts the grayscale values ​​of pixel blocks in the fingerprint portion that are lower than the optimal threshold corresponding to the fingerprint image, and uses them as the first grayscale data; Second grayscale data module: Extracts the grayscale values ​​of pixel blocks in the fingerprint portion that are not lower than the optimal threshold corresponding to the fingerprint image, and uses them as second grayscale data; First candidate grayscale threshold module: Sets the first candidate grayscale threshold for the fingerprint portion using the first grayscale value data; Second candidate grayscale threshold module: Sets the second candidate grayscale threshold for the fingerprint portion using the second grayscale value data; Threshold comparison module: compares the first candidate grayscale threshold with the second candidate grayscale threshold, takes the larger of the two candidate grayscale thresholds as the first grayscale threshold, and takes the smaller of the two candidate grayscale thresholds as the second grayscale threshold. Image block acquisition module: acquires multiple image blocks to be identified using the first grayscale threshold and the second grayscale threshold, and obtains the comprehensive similarity value of the current fingerprint image by combining the grayscale values ​​of the image blocks to be identified; Similarity judgment module: When the comprehensive similarity value is not lower than the preset similarity threshold, the fingerprint image of the current user is determined to pass fingerprint verification.

6. The fingerprint recognition access control system according to claim 5, characterized in that, The image patch acquisition module includes: Image block acquisition module: The fingerprint image is segmented using the first grayscale threshold and the second grayscale threshold to acquire multiple first image blocks to be identified and second image blocks to be identified; Similarity value comparison module: The plurality of first image blocks to be identified and second image blocks to be identified are sequentially compared with the fingerprint positions corresponding to the fingerprint reference images that have been entered into the database by the user to obtain the similarity values ​​corresponding to the plurality of first image blocks to be identified and second image blocks to be identified respectively. Similarity value acquisition module: The module uses the similarity values ​​corresponding to the multiple first and second image blocks to be identified, combined with the grayscale value of the image to be identified, to obtain the comprehensive similarity value of the current fingerprint image.

7. The fingerprint recognition access control system according to claim 6, characterized in that, The image patch acquisition module includes: Grayscale extraction module: Extracts the grayscale values ​​of all pixel blocks in the fingerprint portion of the fingerprint image; First type of pixel block module: Extract pixel blocks whose grayscale values ​​exceed the first grayscale threshold and use them as first type of pixel blocks; Target region setting module: retrieves a target region within the area of ​​the first target region whose number of blocks containing the first type of pixel blocks exceeds a preset first block number threshold, and uses it as the first target region; First image block to be identified module: Segment the fingerprint image according to the position of the first target region to obtain the first image block to be identified; Second-class pixel block determination module: Extracts pixel blocks whose grayscale values ​​exceed the second grayscale threshold but do not exceed the first grayscale threshold as second-class pixel blocks; Area region retrieval module: Retrieves a target region within the area of ​​the second target region whose number of blocks containing the second type of pixel blocks exceeds a preset second block number threshold, and uses it as the second target region; Second target region module: Segment the fingerprint image according to the position of the second target region to obtain a second image block to be identified.

8. The fingerprint recognition access control system according to claim 7, characterized in that, The area of ​​the first target region is smaller than the area of ​​the second target region, and the ratio between the areas of the first target region and the second target region is in the range of 1:1.7 to 1:2.3; at the same time, the first block number threshold is smaller than the second block number threshold, and the ratio between the first block number threshold and the second block number threshold is in the range of 1:2.1 to 1:2.8.

Images