Fingerprint processing method, device and system

Abstract

The invention discloses a fingerprint processing method, device and system. The method comprises the following steps: acquiring a first fingerprint image; performing image analysis on the first fingerprint image, and determining an image type corresponding to the first fingerprint image; if the image type corresponding to the first fingerprint image is a weak template type, determining a target fingerprint image based on the first fingerprint image, and caching the target fingerprint image; if the image type corresponding to the first fingerprint image is a strong template type, registering the first fingerprint image as a strong fingerprint template, and updating current state data corresponding to the strong fingerprint template; and when the current state data meets a registration ending condition, registering the target fingerprint image as a weak fingerprint template, and completing fingerprint registration based on the strong fingerprint template and the weak fingerprint template. According to the method, on the basis of not influencing the image quality of the registered fingerprint template and the subsequent identification success rate, the first fingerprint image is classified, the registration efficiency is improved, and the user registration experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of fingerprint recognition technology, and in particular to a fingerprint processing method, device and system. Background Art

[0002] Fingerprint recognition technology is used in electronic devices to authenticate users based on fingerprints. Before fingerprint recognition, fingerprint registration is generally required, and the collected fingerprint image is registered in the fingerprint template library. When in use, the real-time collected fingerprint image is compared with the fingerprint template library to achieve fingerprint recognition.

[0003] In the existing fingerprint registration process, it is necessary to collect multiple frames of fingerprint images and evaluate whether each fingerprint image meets the registration standard. The fingerprint images that meet the registration standard are registered into the fingerprint template library. The process is time-consuming and the registration efficiency is low, which affects the user registration experience. Summary of the invention

[0004] The embodiments of the present invention provide a fingerprint processing method, device and system to solve the problem that the existing fingerprint registration process is time-consuming, the registration efficiency is low, and the user registration experience is affected.

[0005] A fingerprint processing method, comprising: Acquire a first fingerprint image, where the first fingerprint image is a fingerprint image collected at a current moment; Performing image analysis on the first fingerprint image to determine an image type corresponding to the first fingerprint image; If the image type corresponding to the first fingerprint image is a weak template type, determining a target fingerprint image based on the first fingerprint image, and caching the target fingerprint image; If the image type corresponding to the first fingerprint image is a strong template type, registering the first fingerprint image as a strong fingerprint template, and updating the current state data corresponding to the strong fingerprint template; When the current state data corresponding to the strong fingerprint template meets the registration end condition, the target fingerprint image is registered as a weak fingerprint template, and the fingerprint registration is completed based on the strong fingerprint template and the weak fingerprint template.

[0006] Preferably, performing image analysis on the first fingerprint image to determine the image type corresponding to the first fingerprint image includes: Performing motion blur recognition based on the first fingerprint image and the second fingerprint image to determine the motion blur type corresponding to the first fingerprint image, where the second fingerprint image is a fingerprint image collected at a previous moment; Performing effective area and image quality detection on the first fingerprint image to determine a first detection result corresponding to the first fingerprint image; Perform repetition rate detection based on the first fingerprint image and the third fingerprint image to determine a second detection result corresponding to the first fingerprint image, wherein the third fingerprint image is a fingerprint image whose image type is a strong template type or a weak template type determined before the current moment; An image type corresponding to the first fingerprint image is determined based on the motion blur type corresponding to the first fingerprint image, the first detection result, and the second detection result.

[0007] Preferably, performing motion blur recognition based on the first fingerprint image and the second fingerprint image to determine the motion blur type corresponding to the first fingerprint image includes: Performing motion blur recognition on the first fingerprint image and the second fingerprint image to determine a target motion blur value corresponding to the first fingerprint image; Based on the target motion blur value, a motion blur type corresponding to the first fingerprint image is determined.

[0008] Preferably, the performing effective area and image quality detection on the first fingerprint image to determine a first detection result corresponding to the first fingerprint image includes: Performing effective area and image quality detection on the first fingerprint image to determine an effective area and a quality score corresponding to the first fingerprint image; If the effective area is greater than the target area threshold, and the quality score is greater than the target quality threshold, determining that the first detection result corresponding to the first fingerprint image is detection passed; If the effective area is not greater than the target area threshold, or the quality score is not greater than the target quality threshold, it is determined that the first detection result corresponding to the first fingerprint image is a detection failure.

[0009] Preferably, before performing effective area and image quality detection on the first fingerprint image and determining a first detection result corresponding to the first fingerprint image, the fingerprint processing method further includes: Performing motion blur recognition on the first fingerprint image and the second fingerprint image to determine a target motion blur value corresponding to the first fingerprint image; Based on the target motion blur value, determining the target area threshold and the target quality threshold; The target area threshold is positively correlated with the target motion blur value, and the target quality threshold is positively correlated with the target motion blur value.

[0010] Preferably, the third fingerprint image includes a fourth fingerprint image, and the fourth fingerprint image is a fingerprint image whose image type determined before the current moment is a strong template type; The performing repetition rate detection based on the first fingerprint image and the third fingerprint image to determine a second detection result corresponding to the first fingerprint image includes: Perform repetition rate detection based on the first fingerprint image and all the fourth fingerprint images to determine a first measured repetition rate; Perform repetition rate detection based on the first fingerprint image and a third fingerprint image closest to the current moment to determine a second measured repetition rate; If the first measured repetition rate is less than a first repetition rate threshold, and the second measured repetition rate is less than a second repetition rate threshold, determining that the second detection result corresponding to the first fingerprint image is detection passed; If the first measured repetition rate is not less than the first repetition rate threshold, or the second measured repetition rate is not less than the second repetition rate threshold, it is determined that the second detection result corresponding to the first fingerprint image is detection failure.

[0011] Preferably, the first measured repetition rate is the maximum value of the single repetition rates corresponding to all the fourth fingerprint images, and the single repetition rate is the repetition rate between the fourth fingerprint image and the first fingerprint image; Alternatively, the first measured repetition rate is the repetition rate between the spliced ​​fingerprint image and the first fingerprint image, and the spliced ​​fingerprint image is a fingerprint image obtained by splicing all the fourth fingerprint images.

[0012] Preferably, before performing repetition rate detection based on the first fingerprint image and the third fingerprint image to determine a second detection result corresponding to the first fingerprint image, the fingerprint processing method further includes: Acquire a first cumulative number, where the first cumulative number is the number of first fingerprint images whose image type is a weak template type acquired continuously; Determining the first repetition rate threshold and the second repetition rate threshold based on the first accumulated number; Among them, the first repetition rate threshold is negatively correlated with the first cumulative number; the second repetition rate threshold is negatively correlated with the first cumulative number.

[0013] Preferably, determining the image type corresponding to the first fingerprint image based on the motion blur type corresponding to the first fingerprint image, the first detection result, and the second detection result includes: If the motion blur type is a non-blur type, the first detection result is a detection pass, and the second detection result is a detection pass, determining that the image type corresponding to the first fingerprint image is a strong template type; If the motion blur type is a non-blur type, the first detection result is a detection pass, and the second detection result is a detection fail, determining that the image type corresponding to the first fingerprint image is a weak template type; Alternatively, if the motion blur type is a semi-blur type, the first detection result is a detection pass, and the second detection result is a detection pass, it is determined that the image type corresponding to the first fingerprint image is a weak template type.

[0014] Preferably, determining a target fingerprint image based on the first fingerprint image includes: Obtaining the number of images corresponding to a fifth fingerprint image, wherein the fifth fingerprint image is a fingerprint image whose image type is a weak template type determined before the current moment; If the number of images corresponding to the fifth fingerprint image is less than a first number threshold, the first fingerprint image and all the fifth fingerprint images are determined as target fingerprint images; If the number of images corresponding to the fifth fingerprint image is not less than a first number threshold, the first fingerprint image and all the fifth fingerprint images are analyzed to determine a target fingerprint image.

[0015] Preferably, analyzing the first fingerprint image and all the fifth fingerprint images to determine the target fingerprint image includes: Determine the priority corresponding to the first fingerprint image; If the priority corresponding to the first fingerprint image is less than the minimum priority of all the fifth fingerprint images, the first fingerprint image is deleted, and all the fifth fingerprint images are determined as target fingerprint images; If the priority corresponding to the first fingerprint image is not less than the minimum priority of all the fifth fingerprint images, the fifth fingerprint image corresponding to the minimum priority is deleted, and the first fingerprint image and the remaining fifth fingerprint images are determined as target fingerprint images.

[0016] Preferably, determining the priority corresponding to the first fingerprint image includes: Performing motion blur recognition based on the first fingerprint image and the second fingerprint image to determine a target motion blur value, and determining a first indicator score corresponding to the first fingerprint image based on the target motion blur value, wherein the second fingerprint image is a fingerprint image collected at a previous moment; Performing effective area and image quality detection on the first fingerprint image to determine an effective area and a quality score, and determining a second indicator score corresponding to the first fingerprint image based on the effective area and the quality score; Perform repetition rate detection based on the first fingerprint image and the third fingerprint image to determine the measured repetition rate, and determine the third indicator score corresponding to the first fingerprint image based on the measured repetition rate, wherein the third fingerprint image is a fingerprint image whose image type determined before the current moment is a strong template type or a weak template type; The priority corresponding to the first fingerprint image is determined based on the first indicator score, the second indicator score, and the third indicator score corresponding to the first fingerprint image.

[0017] Preferably, determining the priority corresponding to the first fingerprint image based on the first indicator score, the second indicator score and the third indicator score corresponding to the first fingerprint image includes: Determining a fingerprint collection method corresponding to the first fingerprint image; If the fingerprint collection method is a pressing method, a first indicator score, a second indicator score, and a third indicator score corresponding to the first fingerprint image are weighted based on a first weight combination to determine a priority corresponding to the first fingerprint image; in the first weight combination, the weight of the first indicator score is less than the weight of the third indicator score; If the fingerprint collection method is a sliding method, the first indicator score, the second indicator score and the third indicator score corresponding to the first fingerprint image are weighted based on the second weight combination to determine the priority corresponding to the first fingerprint image; in the second weight combination, the weight of the first indicator score is greater than the weight of the third indicator score.

[0018] Preferably, the current status data includes a second cumulative number, which is the number of templates that have been registered as strong fingerprint templates; the registration end condition includes that the second cumulative number is greater than a second number threshold; Alternatively, the current status data includes an effective area ratio, which is the ratio of the overlapping area of ​​all strong fingerprint templates after splicing to the area of ​​the standard fingerprint template; the registration end condition includes that the effective area ratio is greater than a preset ratio threshold.

[0019] A fingerprint processing device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the fingerprint processing method is implemented when the processor executes the computer program.

[0020] A fingerprint processing system comprises a fingerprint sensor and the above-mentioned fingerprint processing device, wherein the fingerprint processing device is connected to the fingerprint sensor.

[0021] The above-mentioned fingerprint processing method, device and system perform image analysis on the first fingerprint image to determine its corresponding image type so as to adopt different processing methods according to the image type. Specifically, when the image type is a weak template type, the target fingerprint image to be cached is determined based on the first fingerprint image, and a cache operation is performed first, so that when the registration end condition is met, the target fingerprint image is registered as a weak fingerprint template, which helps to improve its processing efficiency; when the image type is a strong template type, the first fingerprint image can be directly registered as a strong fingerprint template. In this solution, by serially processing the first fingerprint image of the strong template type and parallel processing the first fingerprint image of the weak template type, a processing strategy combining serial and parallel is adopted, and the first fingerprint image is classified and processed without affecting the image quality of the registered fingerprint template and the subsequent recognition success rate, thereby improving the registration efficiency and thus improving the user registration experience. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0023] Figure 1 It is a schematic diagram of an ultrasonic fingerprint system; Figure 2 is a flow chart of a fingerprint processing method in one embodiment of the present invention; Figure 3 yes Figure 2 A flow chart of step S2; Figure 4 yes Figure 2 A flow chart of step S3 in FIG. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] The fingerprint processing method provided in the embodiment of the present invention can be applied to a fingerprint processing system. The fingerprint processing system includes a fingerprint processing device and a fingerprint sensor. The fingerprint processing device is connected to the fingerprint sensor and can obtain a fingerprint image collected by the fingerprint sensor. The fingerprint registration operation can be quickly completed based on the fingerprint image. On the basis of taking into account the image quality of the registered fingerprint template, its registration efficiency is improved to enhance the user registration experience.

[0026] As an example, the fingerprint sensor may be a press-type fingerprint sensor that can capture fingerprint images based on the user's pressing operation; or, the fingerprint sensor may be a sliding fingerprint sensor that can capture fingerprint images based on the user's sliding or pushing operation; or, the fingerprint sensor may be an ultrasonic fingerprint sensor in an ultrasonic fingerprint system that can capture corresponding fingerprint images by triggering ultrasonic signals.

[0027] As an example, the fingerprint processing system may be an ultrasonic fingerprint system, such as Figure 1 As shown, the fingerprint sensor of the ultrasonic fingerprint system is an ultrasonic fingerprint sensor. The ultrasonic fingerprint system is a biometric technology based on ultrasonic technology. It obtains the corresponding fingerprint image by transmitting and receiving ultrasonic signals, and combines the difference in acoustic impedance between the screen, finger and air to distinguish the valleys and ridges on the fingerprint, so as to obtain fingerprint features for identity recognition.

[0028] The fingerprint processing device of the ultrasonic fingerprint system may include a controller, an ultrasonic fingerprint sensor, a data processor, an ADC converter, and an algorithm processor. In this example, the controller is connected to the ultrasonic fingerprint sensor, the data processor, the ADC converter, and the algorithm processor to control the operation of each module. The operation process is as follows: the controller controls the ultrasonic fingerprint sensor to generate and receive ultrasonic signals; the ADC converter performs digital-to-analog conversion on the received ultrasonic signals, and the converted signals are rearranged and packaged through the data processor, and the converted data is sent to the algorithm processor for algorithm processing to complete fingerprint registration and identification. In this example, the fingerprint processing device may include a controller, or may include a controller and a data processor, an ADC converter, and an algorithm processor connected thereto, which may be determined according to specific circumstances.

[0029] During the fingerprint registration process of the ultrasonic fingerprint system, the fingerprint pattern of the finger needs to be collected as the fingerprint template for identification. The fingerprint template has an important influence on identification. During the fingerprint template registration process, the ultrasonic frame rate is generally adjusted to collect fingerprint images without motion blur. Generally speaking, the higher the frame rate, the more likely it is to collect fingerprint images without motion blur, thereby improving the accuracy of registration and user experience. However, this method of improving the ultrasonic frame rate has the following shortcomings: First, increasing the frame rate will increase the processing burden of the device, resulting in a decrease in device performance and increasing power consumption; second, the frame rate will be limited by factors such as the processing power of the entire system, ambient light, and finger status, so that increasing the frame rate cannot guarantee that a clear fingerprint image can be collected every time, that is, the image quality of the collected fingerprint image cannot be guaranteed. Therefore, in order to improve the user experience and efficiency during application, during the working process of the system, the finger covers the fingerprint pattern collection area and slides. Through the processing of various modules of the system, the fingerprint image of the finger in motion is collected and analyzed. By using a step-by-step processing strategy for the data, the template collection efficiency is improved without affecting the recognition efficiency, effectively reducing the registration process time.

[0030] The embodiment of the present invention provides a fingerprint processing method, which is described by taking the application of the method in a fingerprint processing device as an example. Figure 2 As shown, the fingerprint processing method includes: S1: Acquire a first fingerprint image, where the first fingerprint image is a fingerprint image collected at the current moment; S2: Performing image analysis on the first fingerprint image to determine the image type corresponding to the first fingerprint image; S3: If the image type corresponding to the first fingerprint image is a weak template type, a target fingerprint image is determined based on the first fingerprint image, and the target fingerprint image is cached; S4: If the image type corresponding to the first fingerprint image is a strong template type, register the first fingerprint image as a strong fingerprint template, and update the current state data corresponding to the strong fingerprint template; S5: When the current state data corresponding to the strong fingerprint template meets the registration end condition, the target fingerprint image is registered as a weak fingerprint template, and the fingerprint registration is completed based on the strong fingerprint template and the weak fingerprint template.

[0031] As an example, in step S1, the fingerprint processing device is connected to the fingerprint sensor, and can obtain the first fingerprint image collected by the fingerprint sensor in real time. In this example, the fingerprint processing device can obtain the first fingerprint image formed by the press-type fingerprint sensor collecting the finger pressing through a specific area, or the fingerprint processing device can obtain the first fingerprint image formed by the sliding fingerprint sensor collecting the finger sliding or dragging through a specific area, or the fingerprint processing device can obtain the first fingerprint image formed by the ultrasonic fingerprint sensor collecting the finger sliding through a specific area.

[0032] As an example, in step S2, when the fingerprint processing device obtains the first fingerprint image, it can call the built-in image analysis logic to perform image analysis on the first fingerprint image to determine the image type corresponding to the first fingerprint image. In this example, the image type can be any one of a strong template type, a weak template type, and an invalid type, and the strong template type and the weak template type are two valid types with different degrees of effectiveness; the strong template type here refers to a type that can be used as a strong fingerprint template, the weak template type refers to a type that can be used as a weak fingerprint template, and the invalid type refers to a type that cannot be used as a fingerprint template. Among them, a strong fingerprint template refers to a valid fingerprint image that has completely passed the screening. A weak fingerprint template refers to a valid fingerprint image that has partially passed the screening, and is a fingerprint template used to assist a strong fingerprint template in fingerprint identification. A valid fingerprint image refers to a fingerprint image that can be stored in a fingerprint template library for fingerprint identification.

[0033] The target fingerprint image is the fingerprint image that needs to be cached.

[0034] As an example, in step S3, when the image type corresponding to the first fingerprint image is a weak template type, the fingerprint processing device can determine that it has partially passed the screening, that is, it is determined that it meets some of the screening conditions that can be registered as a fingerprint template. At this time, the first fingerprint image can be directly determined as the target fingerprint image and cached; the first fingerprint image collected at the current moment and the target fingerprint image cached before the current moment can also be analyzed and processed to update the target fingerprint image that can be cached as a weak fingerprint template. The specific details can be determined independently according to actual conditions. In this example, when the image type of the first fingerprint image is a weak template type, the target fingerprint image is first updated based on the first fingerprint image, and the target fingerprint image is cached without registration processing, which helps to improve its processing efficiency and thus improve the user registration experience.

[0035] The current state data corresponding to the strong fingerprint template is data related to the strong fingerprint template, specifically data used to evaluate whether the registration process needs to be terminated. The registration termination condition is a pre-set condition for terminating the current fingerprint registration operation.

[0036] As an example, in step S4, when the image type corresponding to the first fingerprint image is a strong template type, the fingerprint processing device can determine that it has completely passed the screening, that is, it is determined that it meets all the screening conditions that can be registered as a fingerprint template. Therefore, the first fingerprint image is registered as a strong fingerprint template, and the current state data corresponding to the strong fingerprint template is updated so that the current state data corresponding to the strong fingerprint template can be compared with the pre-set registration end condition, and the subsequent steps are executed according to the comparison result.

[0037] As an example, in step S5, when the current state data corresponding to the strong fingerprint template meets the registration end condition, the fingerprint processing device can register the cached target fingerprint image as a weak fingerprint template, and then complete the fingerprint registration operation based on all strong fingerprint templates and all weak fingerprint templates, that is, all strong fingerprint templates and all weak fingerprint templates are determined as registered fingerprint templates, and the registered fingerprint templates are stored in the fingerprint template library, so that fingerprint recognition can be performed on the fingerprint image collected in real time based on the registered fingerprint templates in the future. In this example, the weak fingerprint template can assist the strong fingerprint template in completing the fingerprint recognition operation to ensure the recognition success rate of subsequent fingerprint recognition.

[0038] In this example, when the fingerprint processing device determines that the image type corresponding to the first fingerprint image is an invalid type, it can be determined that the first fingerprint image cannot be used as a registered fingerprint template. Therefore, the first fingerprint image needs to be deleted. For example, it can be deleted after motion blur recognition is performed on the fingerprint image collected at the next moment, and the first fingerprint image is no longer processed, which helps to save processing resources and improve the efficiency of collecting registered fingerprint templates. At this time, it is necessary to repeat the acquisition of the first fingerprint image and subsequent steps.

[0039] In this example, the first fingerprint image is analyzed to determine its corresponding image type so that different processing methods can be used according to the image type. Specifically, when the image type is a weak template type, the target fingerprint image to be cached is determined based on the first fingerprint image, and the cache operation is performed first, so that when the registration end condition is met, the target fingerprint image is registered as a weak fingerprint template, which helps to improve its processing efficiency; when the image type is a strong template type, the first fingerprint image can be directly registered as a strong fingerprint template. In this example, by serially processing the first fingerprint image of the strong template type and parallel processing the first fingerprint image of the weak template type, a processing strategy combining serial and parallel is adopted, and the first fingerprint image is classified and processed without affecting the image quality of the registered fingerprint template and the subsequent recognition success rate, thereby improving the registration efficiency and thus improving the user registration experience.

[0040] In one embodiment, if Figure 3 As shown, step S2, i.e., performing image analysis on the first fingerprint image to determine the image type corresponding to the first fingerprint image, includes: S21: performing motion blur recognition based on the first fingerprint image and the second fingerprint image to determine the motion blur type corresponding to the first fingerprint image, where the second fingerprint image is a fingerprint image collected at the last moment; S22: Performing effective area and image quality detection on the first fingerprint image to determine a first detection result corresponding to the first fingerprint image; S23: performing repetition rate detection based on the first fingerprint image and the third fingerprint image to determine a second detection result corresponding to the first fingerprint image, wherein the third fingerprint image is a fingerprint image of a strong template type or a weak template type determined before the current moment; S24: Determine the image type corresponding to the first fingerprint image based on the motion blur type corresponding to the first fingerprint image, the first detection result, and the second detection result.

[0041] The second fingerprint image is the fingerprint image collected at the last moment. The second fingerprint image is cached in the memory and can be directly called when motion blur recognition is required. Motion blur refers to the phenomenon of deformation, distortion, and smearing of the fingerprint during the movement of the finger, which can reflect the mismatch between the collected fingerprint and the actual fingerprint.

[0042] As an example, in step S21, after acquiring the first fingerprint image, the fingerprint processing device may call the pre-set motion blur recognition logic to recognize the first fingerprint image and the second fingerprint image, and determine the motion blur type corresponding to the first fingerprint image, which can reflect the fingerprint pattern changes of the two frames of fingerprint images collected before and after the finger slides, drags or other movements. In this example, the motion blur type is any one of the full blur type, the semi-blur type and the non-blur type. The full blur type here refers to a completely blurred state, that is, a state where the fingerprint pattern is deformed and the deformation degree is large; the non-blur type refers to a state where no blur occurs, that is, a normal state of the fingerprint pattern; the semi-blur type is a state between the full blur type and the non-blur type, that is, a state where the fingerprint pattern is deformed but the deformation degree is small.

[0043] As an example, in step S22, after acquiring the first fingerprint image, the fingerprint processing device may perform effective area and image quality detection on the first fingerprint image respectively to determine the effective area and image quality corresponding to the first fingerprint image, and then compare the two evaluation indicators of effective area and image quality with the pre-set fingerprint validity standard; if the two evaluation indicators meet the fingerprint validity standard at the same time, the first detection result corresponding to the first fingerprint image is determined to be a passed detection; conversely, if at least one of the two evaluation indicators does not meet the fingerprint validity standard, the first detection result corresponding to the first fingerprint image is determined to be a failed detection. The fingerprint validity standard here can be understood as a standard for evaluating whether the collected fingerprint image is valid so that it meets the subsequent fingerprint recognition requirements. In this example, the first detection result corresponding to the first fingerprint image is used to reflect whether the first fingerprint image itself can meet the fingerprint recognition requirements.

[0044] Among them, the third fingerprint image is a fingerprint image whose image type is a strong template type or a weak template type determined before the current moment, that is, a valid fingerprint image determined before the current moment. A valid fingerprint image refers to a fingerprint image that can be stored in a fingerprint template library for fingerprint recognition, that is, a valid fingerprint image is a fingerprint image whose image type is a strong template type or a weak template type. In this example, the third fingerprint image can be one or more, which can be a strong fingerprint template determined before the current moment, or a target fingerprint image determined before the current moment.

[0045] As an example, in step S23, after acquiring the first fingerprint image, the fingerprint processing device may read the third fingerprint image from the memory, and then perform a repetition rate detection based on the first fingerprint image and the third fingerprint image to determine the overlapping area of ​​the two; then determine the ratio of the two based on the overlapping area and the area of ​​the first fingerprint image, and determine the ratio of the two as the measured repetition rate; compare the measured repetition rate with the preset repetition rate threshold; if the measured repetition rate is less than the repetition rate threshold, it means that the overlapping part of the first fingerprint image and the third fingerprint image accounts for a small proportion, and the first fingerprint image needs to be retained to ensure the integrity and comprehensiveness of the fingerprint features in the registered fingerprint template, and therefore, the second detection result corresponding to the first fingerprint image is determined to be a passed detection; conversely, if the measured repetition rate is not less than the repetition rate threshold, it means that the overlapping part of the first fingerprint image and the third fingerprint image accounts for a large proportion, and not retaining the first fingerprint image will not affect the integrity and comprehensiveness of the fingerprint features in the registered fingerprint template, and therefore, the second detection result corresponding to the first fingerprint image is determined to be a failed detection.

[0046] As an example, in step S24, after determining the motion blur type, the first detection result, and the second detection result corresponding to the first fingerprint image, the fingerprint processing device may determine the image type corresponding to the first fingerprint image based on the specific combination formed by the motion blur type, the first detection result, and the second detection result. In this example, the motion blur type may be any one of a full blur type, a semi-blur type, and a non-blur type; the first detection result may be any one of a detection pass and a detection fail; the second detection result may be any one of a detection pass and a detection fail; the three may form any one of 12 combinations, and the specific combination formed by them may be queried in a pre-set combination type mapping table to determine its corresponding image type. The combination type mapping table is a data table used to reflect the mapping relationship between different combinations and their corresponding image types.

[0047] In this example, motion blur recognition is performed based on the first fingerprint image and the second fingerprint image. According to the difference between the two frames of fingerprint images, the fingerprint pattern changes are analyzed to determine the motion blur type that reflects the fingerprint pattern changes; the first fingerprint image is tested for effective area and image quality to evaluate whether it can meet the quality standard of being a fingerprint template based on the quality of the fingerprint image itself, thereby determining its corresponding first test result; the first fingerprint image and the third fingerprint image are tested for repetition rate, and the fingerprint feature repetition of the two is analyzed to determine their corresponding second test result. Based on the motion blur type corresponding to the first fingerprint image, the first test result, and the second test result, the corresponding image type is determined, and the corresponding image type can be comprehensively determined from multiple dimensions to determine whether it can be used as a fingerprint template based on the image type, thereby ensuring the recognition success rate of fingerprint recognition.

[0048] In one embodiment, step S21, i.e., performing motion blur recognition based on the first fingerprint image and the second fingerprint image to determine the motion blur type corresponding to the first fingerprint image, includes: S211: performing motion blur recognition on the first fingerprint image and the second fingerprint image to determine a target motion blur value corresponding to the first fingerprint image; S212: Determine the motion blur type corresponding to the first fingerprint image based on the target motion blur value.

[0049] The target motion blur value is a specific value ultimately used to evaluate the motion blur type.

[0050] As an example, in step S211, after acquiring the first fingerprint image, the fingerprint processing device may use a preset motion blur algorithm to calculate the first fingerprint image and the second fingerprint image, and determine the output result as the initial motion blur value. The initial motion blur value may be directly determined as the target motion blur value; or, according to the specific situation, a target correction coefficient may be determined, and the initial motion blur value may be corrected using the target correction coefficient to determine the target motion blur value corresponding to the first fingerprint image.

[0051] For example, the motion blur algorithm can be S=std(diff(phase1-phase2)) / signal, where phase1 and phase2 are the first fingerprint image and the second fingerprint image, respectively, which can be understood as two frames of fingerprint images with the same configuration collected at the beginning and end of image acquisition; std is a function used to calculate the spatial standard deviation of the data; diff is a function used to calculate the difference between the two; signal is the signal quantity of the fingerprint image, which is used to reflect the fingerprint pressure and signal size; S is the initial motion blur value, which can reflect the degree of motion blur of the two frames of fingerprint images.

[0052] In this example, step S211, i.e., performing motion blur recognition on the first fingerprint image and the second fingerprint image to determine the target motion blur value corresponding to the first fingerprint image, specifically includes: S2111: Perform motion blur recognition on the first fingerprint image and the second fingerprint image to determine an initial motion blur value corresponding to the first fingerprint image.

[0053] S2112: Perform quality analysis on the first fingerprint image to determine a quality score corresponding to the first fingerprint image. The quality score is a score determined by performing quality analysis on the first fingerprint image and is used to reflect the contrast and clarity of the fingerprint.

[0054] S2113: Correcting the initial motion blur value using the quality score to determine a target motion blur value corresponding to the first fingerprint image.

[0055] Since fingerprint images with high quality scores have better anti-motion blur capabilities, the quality score corresponding to the first fingerprint image can be used to correct the initial motion blur value to ensure that the corrected target motion blur value can more easily pass the motion blur control, thereby helping to improve fingerprint registration efficiency.

[0056] Furthermore, the system pre-sets two quality thresholds, namely a first quality threshold and a second quality threshold, the first quality threshold is greater than the second quality threshold, for example, the first quality threshold can be set to 50%, and the second quality threshold can be set to 35%. In the above step S2113, the initial motion blur value is corrected by using the quality score to determine the target motion blur value corresponding to the first fingerprint image, specifically including: S21131: If the quality score is greater than the first quality threshold, the product of the initial motion blur value corresponding to the first fingerprint image and the first correction coefficient is determined as the target motion blur value corresponding to the first fingerprint image. The first correction coefficient is a value between 0 and 1. For example, the first correction coefficient can be set to 1 / 5, which is used to indicate that when the quality score of the first fingerprint image is large, the initial motion blur value is reduced by 5 times to determine the target motion blur value corresponding to the first fingerprint image, so that the first fingerprint image with better image quality can more easily pass the motion blur card control, so as to improve the fingerprint registration efficiency.

[0057] S21132: If the quality score is not greater than the first quality threshold, and the quality score is greater than the second quality threshold, the product of the initial motion blur value corresponding to the first fingerprint image and the second correction coefficient is determined as the target motion blur value corresponding to the first fingerprint image. The first correction coefficient is a value between 0 and 1, and the second correction coefficient is greater than the first correction coefficient. For example, the second correction coefficient can be 1 / 2, which is used to indicate that when the quality score of the first fingerprint image is moderate, the initial motion blur value is reduced by 2 times to determine the target motion blur value corresponding to the first fingerprint image, so that the first fingerprint image with moderate image quality can more easily pass the motion blur control to improve the fingerprint registration efficiency.

[0058] S21133: If the quality score is not greater than the second quality threshold, the initial motion blur value corresponding to the first fingerprint image is determined as the target motion blur value corresponding to the first fingerprint image, so that the first fingerprint image with poor image quality is more difficult to pass the motion blur control, and then determine its image type as an invalid type, so that it can be directly deleted later to save computing and processing resources.

[0059] In this example, the initial motion blur value is corrected based on the quality score corresponding to the first fingerprint image, so that the corrected target motion blur value combines the image quality score and the anti-motion blur capability, so that the first fingerprint image with higher image quality can pass the motion blur control more easily, and conversely, the first fingerprint image with lower image quality is more difficult to pass the motion blur control, so as to ensure the fingerprint registration efficiency.

[0060] As an example, in step S212, after determining the target motion blur value corresponding to the first fingerprint image, the fingerprint processing device may compare the target motion blur value with different pre-set blur thresholds to determine the blur threshold range to which it belongs, and determine the motion blur type corresponding to the blur threshold range as the motion blur type corresponding to the first fingerprint image, that is, determine whether it belongs to any one of the full blur type, semi-blur type and non-blur type.

[0061] In this example, the system pre-sets two blur thresholds for defining different types of motion blur, which are set as the first blur threshold and the second blur threshold. The first blur threshold is a blur threshold for defining whether the fingerprint pattern is normal, which can be specifically understood as a blur threshold for defining the non-blur type and the semi-blur type. The second blur threshold is a blur threshold for defining the degree of deformation of the fingerprint pattern, which can be specifically understood as a blur threshold for defining the semi-blur type and the full-blur type. The first blur threshold is smaller than the second blur threshold. For example, the first blur threshold can be set to 20, and the second blur threshold can be set to 50.

[0062] Step S212, i.e. determining the motion blur type corresponding to the first fingerprint image based on the target motion blur value, specifically includes: S2121: If the target motion blur value is less than the first blur threshold, the motion blur type corresponding to the first fingerprint image is determined to be a non-blur type. That is, when the target motion blur value corresponding to the first fingerprint image is small, it is determined that the fingerprint pattern is normal, and at this time, the motion blur type is determined to be a non-blur type.

[0063] S2122: If the target motion blur value is not less than the first blur threshold value, and the target motion blur value is less than the second blur threshold value, the motion blur type corresponding to the first fingerprint image is determined to be a semi-blur type. That is, when the target motion blur value corresponding to the first fingerprint image is moderate, it is determined that the fingerprint pattern is deformed, but the degree of deformation is small and does not meet the standard for determining fingerprint pattern abnormality. At this time, the motion blur type is determined to be a semi-blur type.

[0064] S2123: If the target motion blur value is not less than the second blur threshold, the motion blur type corresponding to the first fingerprint image is determined to be a full blur type, that is, the target motion blur value corresponding to the first fingerprint image is larger and meets the standard for determining fingerprint pattern abnormality. At this time, its motion blur type is determined to be a full blur type.

[0065] In one embodiment, step S22, i.e., performing effective area and image quality detection on the first fingerprint image to determine a first detection result corresponding to the first fingerprint image, includes: S221: Performing effective area and image quality detection on the first fingerprint image to determine an effective area and a quality score corresponding to the first fingerprint image; S222: If the effective area is greater than the target area threshold, and the quality score is greater than the target quality threshold, determining that the first detection result corresponding to the first fingerprint image is detection passed; S223: If the effective area is not greater than the target area threshold, or the quality score is not greater than the target quality threshold, determining that the first detection result corresponding to the first fingerprint image is detection failure.

[0066] Among them, the effective area is the ratio of the area of ​​the effective pressing area in the fingerprint image to the total area of ​​the fingerprint image. Generally speaking, when collecting fingerprint images, the fingerprint recognition area corresponding to the fingerprint sensor will collect fingerprint features. When the finger does not completely cover the fingerprint recognition area, the collected fingerprint image includes the regional image corresponding to the effective pressing area and the background image outside the effective pressing area. The ratio of the regional area corresponding to the regional image and the total area of ​​the fingerprint image is calculated to determine the effective area. The target area threshold is a threshold used to evaluate whether the effective area meets the fingerprint validity standard. The target area threshold can be a pre-set fixed value or a dynamic value determined according to actual conditions.

[0067] The quality score is a score determined by quality analysis of the first fingerprint image, and is used to reflect the contrast and clarity of the fingerprint, etc. The target quality threshold is a threshold used to evaluate whether the quality score reaches the fingerprint validity standard. The target quality threshold can be a preset fixed value or a dynamic value determined according to actual conditions.

[0068] As an example, in step S221, after acquiring the first fingerprint image, the fingerprint processing device may calculate the effective area of ​​the first fingerprint image, determine the effective area corresponding to the first fingerprint image, and compare the effective area with a preset or dynamically determined target area threshold; and perform image quality analysis on the first fingerprint image to determine the quality score corresponding to the first fingerprint image, and compare the quality score with a preset or dynamically determined target quality threshold, so as to determine the corresponding first detection result based on the two comparison results.

[0069] As an example, in step S222, when the effective area is greater than the target area threshold and the quality score is greater than the target quality threshold, the fingerprint processing device can determine that the effective area of ​​the first fingerprint image is larger, that is, the area image of the finger in the effective pressing area accounts for a larger proportion, and the probability of collecting a fingerprint image corresponding to an invalid type is smaller, and the overall image quality of the collected first fingerprint image is better. It is determined that both the effective area and image quality fingerprint validity standards are met. Therefore, the corresponding first detection result is determined to be a passed detection.

[0070] As an example, in step S223, when the effective area is not greater than the target area threshold, or the quality score is not greater than the target quality threshold, the fingerprint processing device may determine that the effective area of ​​the first fingerprint image is small, that is, the area image of the finger in the effective pressing area accounts for a small proportion; or, the overall image quality of the collected first fingerprint image is poor, and the probability of collecting a fingerprint image corresponding to an invalid type is high, and therefore, the corresponding first detection result is determined to be a detection failure.

[0071] In this example, based on the comparison result of the effective area corresponding to the first fingerprint image and the target area threshold, as well as the comparison result of the quality score corresponding to the first fingerprint image and the target quality threshold, it is determined from two evaluation dimensions whether it meets the preset fingerprint validity standard and the corresponding first detection result, thereby ensuring the accuracy of the first detection result.

[0072] In one embodiment, before step S22, that is, before performing effective area and image quality detection on the first fingerprint image and determining a first detection result corresponding to the first fingerprint image, the fingerprint processing method further includes: S221 ': Perform motion blur recognition on the first fingerprint image and the second fingerprint image to determine the target motion blur value corresponding to the first fingerprint image; S222': Determine a target area threshold and a target quality threshold based on the target motion blur value; Among them, the target area threshold is positively correlated with the target motion blur value, and the target quality threshold is positively correlated with the target motion blur value.

[0073] As an example, in step S221', after acquiring the first fingerprint image, the fingerprint processing device can use a preset motion blur algorithm to calculate the first fingerprint image and the second fingerprint image, and determine the output result as the initial motion blur value, which can be directly determined as the target motion blur value; or, according to the specific situation, determine the target correction coefficient, use the target correction coefficient to correct the initial motion blur value, and determine the target motion blur value corresponding to the first fingerprint image. The processing process is the same as step S211, and to avoid repetition, it is not described here one by one.

[0074] As an example, in step S222', after obtaining the target motion blur value corresponding to the first fingerprint image, the fingerprint processing device can dynamically determine the target area threshold corresponding to the target motion blur value based on the mapping relationship table or mapping function between the motion blur value and the area threshold, so as to subsequently evaluate whether the effective area of ​​the first fingerprint image meets the fingerprint validity standard based on the target area threshold. In this example, the target area threshold is positively correlated with the target motion blur value, that is, the smaller the target motion blur value, the clearer the first fingerprint image (that is, close to the non-blurred type). At this time, it is necessary to set the target area threshold to be smaller so that the clearer first fingerprint image is easier to pass the card control of the effective area, so as to improve the fingerprint registration efficiency and improve the user registration experience; on the contrary, the larger the target motion blur value, the blurrier the first fingerprint image (that is, close to the full blur type). At this time, it is necessary to set the target area threshold to be larger so that the blurrier first fingerprint image is more difficult to pass the card control of the effective area, so as to ensure the validity of the final registered fingerprint template.

[0075] Accordingly, after obtaining the target motion blur value corresponding to the first fingerprint image, the fingerprint processing device can dynamically determine the target quality threshold corresponding to the target motion blur value based on the mapping relationship table or mapping function between the motion blur value and the quality threshold, so as to subsequently evaluate whether the quality score corresponding to the first fingerprint image meets the fingerprint validity standard based on the target quality threshold. In this example, the target quality threshold is positively correlated with the target motion blur value, that is, the smaller the target motion blur value, the clearer the first fingerprint image (that is, close to the non-blurred type). At this time, it is necessary to set the target quality threshold to be smaller, so that the clearer first fingerprint image is easier to pass the image quality control, so as to improve the fingerprint registration efficiency and improve the user registration experience; on the contrary, the larger the target motion blur value, the blurrier the first fingerprint image (that is, close to the full blur type). At this time, it is necessary to set the target quality threshold to be larger, so that the blurrier first fingerprint image is more difficult to pass the image quality control, so as to ensure the validity of the final registered fingerprint template.

[0076] In one embodiment, the third fingerprint image includes a fourth fingerprint image, and the fourth fingerprint image is a fingerprint image whose image type determined before the current moment is a strong template type; Step S23, i.e., performing repetition rate detection based on the first fingerprint image and the third fingerprint image to determine a second detection result corresponding to the first fingerprint image, includes: S231: Perform repetition rate detection based on the first fingerprint image and all fourth fingerprint images to determine a first measured repetition rate; S232: performing repetition rate detection based on the first fingerprint image and a third fingerprint image closest to the current moment to determine a second measured repetition rate; S233: If the first measured repetition rate is less than the first repetition rate threshold, and the second measured repetition rate is less than the second repetition rate threshold, determining that the second detection result corresponding to the first fingerprint image is detection passed; S234: If the first measured repetition rate is not less than the first repetition rate threshold, or the second measured repetition rate is not less than the second repetition rate threshold, determine that the second detection result corresponding to the first fingerprint image is detection failure.

[0077] As an example, in step S231, after acquiring the first fingerprint image, the fingerprint processing device needs to read all fourth fingerprint images from the memory, where the fourth fingerprint image is a fingerprint image of a strong template type determined before the current moment, where there is at least one strong fingerprint template; and then perform repetition rate detection on the first fingerprint image and the fourth fingerprint image to determine the first measured repetition rate. In this example, the fingerprint processing device may first determine the single repetition rate corresponding to the first fingerprint image and each fourth fingerprint image, and then process the single repetition rates corresponding to all fourth fingerprint images to determine the first measured repetition rate, where the single repetition rate may be first calculating the overlapping area of ​​the first fingerprint image and a fourth fingerprint image, and then calculating the ratio of the overlapping area to the area of ​​the first fingerprint image, and determining the ratio as the single repetition rate corresponding to the fourth fingerprint image.

[0078] As an example, the first measured repetition rate is the maximum value of the single repetition rates corresponding to all fourth fingerprint images, and the single repetition rate is the repetition rate between the fourth fingerprint image and the first fingerprint image.

[0079] In this example, the fingerprint processing device may calculate the repetition rate of at least one fourth fingerprint image and the first fingerprint image, and determine the single repetition rate corresponding to each fourth fingerprint image. For example, the overlapping area of ​​each fourth fingerprint image and the first fingerprint image is first calculated, and then the ratio of the overlapping area to the area of ​​the first fingerprint image is calculated, and the ratio is determined as the single repetition rate corresponding to the fourth fingerprint image. Then, the fingerprint processing device may compare the single repetition rates corresponding to at least one fourth fingerprint image, and determine the maximum value thereof as the first measured repetition rate. The larger the first measured repetition rate, the larger the overlapping area between the first fingerprint image and one of the strong fingerprint templates, and the less it should be retained as a registered fingerprint template, so as to avoid the inclusion of multiple fingerprint images with a large degree of fingerprint feature repetition in the registered fingerprint library, which affects the recognition success rate of subsequent fingerprint recognition; conversely, the smaller the first measured repetition rate, the more fingerprint features it contains that are different from the fourth fingerprint image with the largest overlapping area, and the more it should be retained as a registered fingerprint template to ensure the recognition success rate of subsequent fingerprint recognition.

[0080] As an example, the first measured repetition rate is the repetition rate between the spliced ​​fingerprint image and the first fingerprint image, and the spliced ​​fingerprint image is the fingerprint image obtained by splicing all the fourth fingerprint images.

[0081] In this example, after reading the fourth fingerprint image, the fingerprint processing device may perform splicing processing on all the fourth fingerprint images, for example, perform union processing on the fingerprint features corresponding to all the fourth fingerprint images to obtain a spliced ​​fingerprint image. Then, the fingerprint processing device may calculate the repetition rate of the spliced ​​fingerprint image and the first fingerprint image to determine the first measured repetition rate corresponding thereto. For example, the overlapping area of ​​the spliced ​​fingerprint image and the first fingerprint image may be calculated, and then the ratio of the overlapping area to the area of ​​the first fingerprint image may be calculated, and the ratio may be determined as the first measured repetition rate. The larger the first measured repetition rate, the larger the overlapping area between the first fingerprint image and the spliced ​​fingerprint image, and the less it should be retained as a registered fingerprint template; conversely, the smaller the first measured repetition rate, the more fingerprint features it contains that are different from the spliced ​​fingerprint image, and the more it should be retained as a registered fingerprint template to ensure the recognition success rate of subsequent fingerprint recognition.

[0082] As an example, in step S232, after acquiring the first fingerprint image, the fingerprint processing device may determine the third fingerprint image closest to the current moment from all the third fingerprint images (which may be a fingerprint image of a strong template type or a fingerprint image of a weak template type); and then perform a repetition rate detection on the first fingerprint image and the third fingerprint image closest to the current moment to determine the second measured repetition rate. In this example, the fingerprint processing device may first determine the overlapping area between the first fingerprint image and the third fingerprint image closest to the current moment, and then calculate the ratio of the overlapping area to the total area of ​​the first fingerprint image, and determine the ratio as the second measured repetition rate.

[0083] The first repetition rate threshold and the second repetition rate threshold are two preset repetition rate thresholds, and the two repetition rate thresholds may be the same or different. For example, both may be specific values ​​between 70% and 90%.

[0084] As an example, in step S233, when the first measured repetition rate is less than the first repetition rate threshold and the second measured repetition rate is less than the second repetition rate threshold, the fingerprint processing device can determine that there are fewer repeated fingerprint features between the first fingerprint image and all fourth fingerprint images, and there are fewer repeated fingerprint features between the first fingerprint image and the third fingerprint image closest to the current moment, indicating that the first fingerprint image contains more fingerprint features that are not included in the third fingerprint image, and should be retained as a registered fingerprint template to ensure the accuracy and effectiveness of subsequent identification. Therefore, the second detection result corresponding to the first fingerprint image is determined to be a passed detection, so that it passes the repetition rate control.

[0085] As an example, in step S234, when the first measured repetition rate is not less than the first repetition rate threshold, or the second measured repetition rate is not less than the second repetition rate threshold, the fingerprint processing device can determine that there are more repeated fingerprint features between the first fingerprint image and the fourth fingerprint image, or there are more repeated fingerprint features between the first fingerprint image and the third fingerprint image closest to the current moment, indicating that the first fingerprint image contains fewer fingerprint features that are not included in the third fingerprint image, and should not be retained as a registered fingerprint template to avoid the limited number of fingerprint features that can be recognized by the effective fingerprint template, affecting the accuracy and effectiveness of subsequent recognition. Therefore, the second detection result corresponding to the first fingerprint image is determined to be a detection failure, so that it cannot pass the repetition rate control.

[0086] In one embodiment, before step S23, that is, performing repetition rate detection based on the first fingerprint image and the third fingerprint image to determine a second detection result corresponding to the first fingerprint image, the fingerprint processing method further includes: S231': Acquire a first cumulative number, where the first cumulative number is the number of first fingerprint images whose image type is a weak template type acquired continuously; S232': Determine a first repetition rate threshold and a second repetition rate threshold based on the first accumulated number; The first repetition rate threshold is negatively correlated with the first cumulative number; and the second repetition rate threshold is negatively correlated with the first cumulative number.

[0087] As an example, in step S231', the fingerprint processing device acquires the first fingerprint image each time, performs image analysis on the first fingerprint image, and determines the image type corresponding to the first fingerprint image. Then, based on the image type corresponding to the first fingerprint image, it is necessary to update the first cumulative number, which is the number of first fingerprint images whose image type is a weak template type that are acquired continuously. In this example, if the image type corresponding to the first fingerprint image is a weak template type, N=N+1 is used to update the first cumulative number N; if the image type corresponding to the first fingerprint image is not a weak template type, that is, its image type is a weak template type or an invalid type, N=0 is used to update the first cumulative number N to count the number of fingerprint images corresponding to the weak template type that are acquired continuously.

[0088] As an example, in step S232', after obtaining the first cumulative number, the fingerprint processing device can dynamically determine the first repetition rate threshold and the second repetition rate threshold corresponding to the first cumulative number based on the mapping relationship table or mapping function between the cumulative number and the repetition rate threshold. In this example, the first repetition rate threshold is negatively correlated with the first cumulative number, and the second repetition rate threshold is negatively correlated with the first cumulative number, that is, the larger the first cumulative number is, the more fingerprint images of the weak template type are continuously collected. At this time, it is necessary to increase the first repetition rate threshold and the second repetition rate threshold to improve the card control requirement of the repetition rate, so as to avoid the number of images of the fingerprint images of the weak template type collected reaching the first number threshold (the first number threshold is the number of weak fingerprint templates that need to be collected in advance, for example, it can be set to 20) quickly, resulting in the subsequent determination of the processing efficiency of the cached target fingerprint image; on the contrary, the smaller the first cumulative number is, the fewer fingerprint images of the weak template type are continuously collected. At this time, it is necessary to reduce the first repetition rate threshold and the second repetition rate threshold to reduce the card control requirement of the repetition rate.

[0089] In one embodiment, step S24, i.e., determining the image type corresponding to the first fingerprint image based on the motion blur type corresponding to the first fingerprint image, the first detection result, and the second detection result, includes: S241: If the motion blur type is a non-blur type, the first detection result is a pass, and the second detection result is a pass, determining that the image type corresponding to the first fingerprint image is a strong template type; S242: If the motion blur type is a non-blur type, the first detection result is a passed detection, and the second detection result is a failed detection, then the image type corresponding to the first fingerprint image is determined to be a weak template type; or, if the motion blur type is a semi-blur type, the first detection result is a passed detection, and the second detection result is a passed detection, then the image type corresponding to the first fingerprint image is determined to be a weak template type.

[0090] As an example, in step S241, when the motion blur type corresponding to the first fingerprint image is a non-blur type, the first detection result corresponding to the first fingerprint image is a detection pass, and the second detection result corresponding to the first fingerprint image is a detection pass, the fingerprint pattern of the first fingerprint image can be determined to be normal, the validity determined based on the effective area and image quality is high, and the repetition rate with the third fingerprint image is low, and the preset conditions for determining it as a strong fingerprint template are met, therefore, the image type of the first fingerprint image can be determined to be a strong template type. That is to say, when the first fingerprint image has high clarity, high validity, and low repetition rate with other registered fingerprint templates, it can be registered as a strong fingerprint template.

[0091] As an example, in step S242, when the motion blur type corresponding to the first fingerprint image is a non-blur type, the first detection result corresponding to the first fingerprint image is a detection pass, and the second detection result corresponding to the first fingerprint image is a detection fail, the fingerprint pattern of the first fingerprint image can be determined to be normal, and the validity determined based on the effective area and image quality is high, but the repetition rate with the third fingerprint image is high, and the preset condition for determining it as a weak fingerprint template is met, so the image type of the first fingerprint image can be determined to be a weak template type. That is to say, when the first fingerprint image has high clarity and high validity, and has a high repetition rate with other registered fingerprint templates, in order to avoid missing fingerprint features in part of the effective pressing area, it can be used as a target fingerprint image cache so that it can be registered as a weak fingerprint template later.

[0092] Alternatively, when the motion blur type corresponding to the first fingerprint image is a semi-blur type, the first detection result corresponding to the first fingerprint image is a detection pass, and the second detection result corresponding to the first fingerprint image is a detection pass, the fingerprint processing device can determine that the fingerprint pattern of the first fingerprint image is deformed but the deformation degree is small, the validity determined based on the effective area and image quality is high, and the repetition rate with the third fingerprint image is low, which meets the preset conditions for determining it as a weak fingerprint template. Therefore, it can be determined that the image type of the first fingerprint image is a weak template type. In other words, when the clarity of the first fingerprint image is moderate, the validity is high, and the repetition rate with other registered fingerprint templates is low, it can be registered as a strong fingerprint template. In order to avoid missing fingerprint features in some effective pressing areas, it can be used as a target fingerprint image cache so that it can be registered as a weak fingerprint template later.

[0093] Furthermore, when the motion blur type, the first detection result and the second detection result corresponding to the first fingerprint image do not meet the above-mentioned preset conditions for being identified as a strong template type or a weak template type, the fingerprint processing device determines that the corresponding image type is an invalid type, so as to subsequently delete the first fingerprint image and no longer process the first fingerprint image, which helps to save processing resources and repeat the acquisition of the first fingerprint image and subsequent operations.

[0094] In one embodiment, if Figure 4 As shown, step S3, i.e. determining a target fingerprint image based on the first fingerprint image, includes: S31: Acquire the number of images corresponding to the fifth fingerprint image, where the fifth fingerprint image is a fingerprint image whose image type is a weak template type determined before the current moment; S32: if the number of images corresponding to the fifth fingerprint image is less than the first number threshold, the first fingerprint image and all the fifth fingerprint images are determined as target fingerprint images; S33: If the number of images corresponding to the fifth fingerprint image is not less than the first number threshold, the first fingerprint image and all fifth fingerprint images are analyzed to determine a target fingerprint image.

[0095] The first quantity threshold is the preset number of weak fingerprint templates that need to be collected, and can also be understood as the maximum number of target fingerprint images that can be cached. For example, the first quantity threshold can be set to 20.

[0096] As an example, in step S31, when the fingerprint processing device determines that the image type corresponding to the first fingerprint image is a weak template type, it is necessary to first determine the number of images corresponding to the fifth fingerprint image pre-cached in the memory, compare the number of images with the pre-set first number threshold, and determine the target fingerprint image cached this time based on the comparison result.

[0097] As an example, in step S32, when the number of images corresponding to the fifth fingerprint image is less than the first number threshold, the fingerprint processing device can determine that the fifth fingerprint image cached in the memory at the current moment is small, and can directly determine the first fingerprint image as the target fingerprint image that needs to be cached. That is, in this case, the first fingerprint image and all the fifth fingerprint images can be determined as the target fingerprint images that need to be cached.

[0098] As an example, in step S33, when the number of images corresponding to the fifth fingerprint image is not less than the first number threshold, the fingerprint processing device may determine that there are many fifth fingerprint images cached in the memory at the current moment, and it is necessary to analyze and process the first fingerprint image and all the fifth fingerprint images to delete one of the fingerprint images and retain the remaining fingerprint images as the target fingerprint images that need to be cached. That is, in this case, according to the analysis results, the first fingerprint image may be deleted, or one of the fifth fingerprint images may be deleted, and the remaining fingerprint images may be determined as the target fingerprint images.

[0099] For example, when the number of images corresponding to the fifth fingerprint image is not less than the first number threshold, the fingerprint processing device can obtain the cache time of all fifth fingerprint images, delete the fifth fingerprint image with the longest cache time, and determine the remaining fifth fingerprint images and the first fingerprint image as target fingerprint images that need to be cached, so that when the fingerprint registration is completed, all target fingerprint images are registered as weak fingerprint templates.

[0100] In one embodiment, step S33, i.e., analyzing the first fingerprint image and all fifth fingerprint images to determine the target fingerprint image, includes: S331: Determine the priority corresponding to the first fingerprint image; S332: If the priority corresponding to the first fingerprint image is less than the minimum priority of all fifth fingerprint images, the first fingerprint image is deleted, and all fifth fingerprint images are determined as target fingerprint images; S333: If the priority corresponding to the first fingerprint image is not less than the minimum priority of all fifth fingerprint images, the fifth fingerprint image corresponding to the minimum priority is deleted, and the first fingerprint image and the remaining fifth fingerprint images are determined as target fingerprint images.

[0101] As an example, in step S331, the fingerprint processing device may use a preset priority determination logic to analyze and process the first fingerprint image to determine the priority of the first fingerprint image, which may be determined based on at least one of the image attributes of the fingerprint image, such as clarity, effective area, image quality, and repetition rate. Then, the priority of the first fingerprint image is compared with the minimum priority corresponding to all fifth fingerprint images in the cache to determine the fingerprint image to be deleted based on the comparison result. In this example, the fingerprint processing device analyzes and stores the corresponding priority each time it obtains the first fingerprint image, so as to perform subsequent priority comparison.

[0102] As an example, in step S332, when the priority corresponding to the first fingerprint image is less than the minimum priority of all fifth fingerprint images, the fingerprint processing device determines that the priority of the first fingerprint image is less than the priority of all fifth fingerprint images. Therefore, the first fingerprint image can be deleted and all fifth fingerprint images can be determined as target fingerprint images to achieve the purpose of protecting fingerprint images with higher priorities.

[0103] As an example, in step S333, when the priority corresponding to the first fingerprint image is greater than or equal to the minimum priority of all fifth fingerprint images, the fingerprint processing device deletes the fifth fingerprint image corresponding to the minimum priority, and determines the first fingerprint image and the remaining fifth fingerprint images as target fingerprint images, so as to achieve the purpose of protecting fingerprint images with higher priorities and protecting the target fingerprint image with the cache time closest to the current moment.

[0104] In one embodiment, step S331, i.e. determining the priority corresponding to the first fingerprint image, includes: S3311: Perform motion blur recognition based on the first fingerprint image and the second fingerprint image to determine a target motion blur value, and determine a first indicator score corresponding to the first fingerprint image based on the target motion blur value, where the second fingerprint image is a fingerprint image collected at the last moment; S3312: Perform effective area and image quality detection on the first fingerprint image to determine an effective area and a quality score, and determine a second indicator score corresponding to the first fingerprint image based on the effective area and the quality score; S3313: Perform repetition rate detection based on the first fingerprint image and the third fingerprint image to determine the measured repetition rate, and determine the third indicator score corresponding to the first fingerprint image based on the measured repetition rate, where the third fingerprint image is a fingerprint image whose image type determined before the current moment is a strong template type or a weak template type; S3314: Determine the priority corresponding to the first fingerprint image based on the first indicator score, the second indicator score, and the third indicator score corresponding to the first fingerprint image.

[0105] As an example, in step S3311, the fingerprint processing device may use a preset motion blur algorithm to analyze the first fingerprint image and the second fingerprint image to determine the target motion blur value corresponding to the first fingerprint image. The processing process is similar to step S21, and is not described here one by one to avoid repetition. After determining the target motion blur value corresponding to the first fingerprint image, the fingerprint processing device may directly determine the target motion blur value as the first indicator score, or may perform normalization or other processing on the target motion blur value to determine the first indicator score for evaluating the priority.

[0106] As an example, in step S3312, the fingerprint processing device can calculate the effective area of ​​the first fingerprint image, determine the effective area corresponding to the first fingerprint image, and perform image quality analysis on the first fingerprint image to determine the quality score corresponding to the first fingerprint image. The processing process is similar to step S22, and is not described here to avoid repetition. After determining the effective area and image quality corresponding to the first fingerprint image, the fingerprint processing device can normalize the effective area and image quality corresponding to the first fingerprint image to determine the second indicator score for evaluating the priority; or, based on the effective area and image quality corresponding to the first fingerprint image, query a pre-set mapping table for reflecting the combination of the two and the indicator score to determine the second indicator score for evaluating the priority.

[0107] As an example, in step S3313, the fingerprint processing device may perform repetition rate detection on the first fingerprint image and the third fingerprint image to determine the overlapping area between the two; then determine the ratio of the two based on the overlapping area and the area of ​​the first fingerprint image, and determine the ratio of the two as the measured repetition rate. The processing process is similar to step S23, and to avoid repetition, it is not described here one by one. After determining the measured repetition rate corresponding to the first fingerprint image, the fingerprint processing device may normalize the measured repetition rate to determine the third indicator score for evaluating the priority; or, based on the measured repetition rate, query a pre-set mapping table for reflecting the repetition rate and the indicator score to determine the third indicator score for evaluating the priority.

[0108] As an example, in step S3314, after determining the first indicator score, the second indicator score and the third indicator score corresponding to the first fingerprint image, the fingerprint processing device can directly add the first indicator score, the second indicator score and the third indicator score corresponding to the first fingerprint image to determine the priority of the first fingerprint image; it can also perform weighted processing on the first indicator score, the second indicator score and the third indicator score corresponding to the first fingerprint image to determine the priority of the first fingerprint image. During the weighted processing, the weight of each indicator score can be a fixed value or a dynamic value determined according to actual conditions.

[0109] In this example, motion blur recognition is performed based on the first fingerprint image and the second fingerprint image, and the first indicator score used to reflect the change of fingerprint lines is determined based on the determined target motion blur value; the effective area and image quality of the first fingerprint image are detected to determine the second indicator score used to reflect whether the validity standard is met based on the quality of the fingerprint image itself; the repetition rate of the first fingerprint image and the third fingerprint image is detected to determine the third indicator score reflecting the repetition of fingerprint features. Based on the first indicator score, the second indicator score and the third indicator score corresponding to the first fingerprint image, their priorities are dynamically determined, so as to achieve the comprehensive determination of their corresponding priorities from multiple dimensions, so as to ensure the validity of the target fingerprint image retained later, so as to improve the success of fingerprint recognition as a weak fingerprint template.

[0110] In one embodiment, step S3314, i.e., determining the priority corresponding to the first fingerprint image based on the first indicator score, the second indicator score, and the third indicator score corresponding to the first fingerprint image, includes: S33141: Determine a fingerprint collection method corresponding to the first fingerprint image; S33142: If the fingerprint collection method is the pressing method, based on the first weight combination, the first indicator score, the second indicator score and the third indicator score corresponding to the first fingerprint image are weighted to determine the priority corresponding to the first fingerprint image; in the first weight combination, the weight of the first indicator score is less than the weight of the third indicator score; S33143: If the fingerprint collection method is a sliding method, the first indicator score, the second indicator score and the third indicator score corresponding to the first fingerprint image are weighted based on the second weight combination to determine the priority corresponding to the first fingerprint image; in the second weight combination, the weight of the first indicator score is greater than the weight of the third indicator score.

[0111] As an example, in step S33141, the fingerprint processing device can determine the fingerprint collection method corresponding to each first fingerprint image, which can be determined based on its built-in configuration information, or based on analyzing multiple consecutive frames of first fingerprint images, that is, the fingerprint collection method can be determined independently according to actual conditions.

[0112] Among them, the first weight combination refers to the weight combination for weighting the indicator scores under the pressing mode, and the first weight combination includes the weight of the first indicator score, the weight of the second indicator score and the weight of the third indicator score, and the sum of the weights of the three indicator scores is 1.

[0113] As an example, in step S33142, when the fingerprint processing device determines that the fingerprint collection method is the pressing method, that is, the user's finger is pressed to the fingerprint sensor, and the fingerprint sensor collects the corresponding first fingerprint image, when evaluating the priority of the first fingerprint image, it is necessary to perform weighted processing on the first indicator score, the second indicator score and the third indicator score corresponding to the first fingerprint image based on the first weight combination to determine the priority corresponding to the first fingerprint image; because in the first weight combination, the weight of the first indicator score is less than the weight of the third indicator score, the indicator score of the blurred motion factor in the first fingerprint image collected by the pressing method is reduced, and the indicator score of the repeatability factor in the first fingerprint image collected by the pressing method is increased to adapt to the specific scenario.

[0114] Among them, the second weight combination refers to the weight combination for weighting the indicator scores in a sliding manner, and the second weight combination includes the weight of the first indicator score, the weight of the second indicator score and the weight of the third indicator score, and the sum of the weights of the three indicator scores is 1.

[0115] As an example, in step S33143, when the fingerprint processing device determines that the fingerprint collection method is a sliding method, that is, the user's finger slides or drags across the fingerprint sensor, the fingerprint sensor collects the corresponding first fingerprint image. When evaluating the priority of the first fingerprint image, it is necessary to perform weighted processing on the first indicator score, the second indicator score and the third indicator score corresponding to the first fingerprint image based on the second weight combination to determine the priority corresponding to the first fingerprint image; because in the first weight combination, the weight of the first indicator score is greater than the weight of the third indicator score, the indicator score of the blurred motion factor in the first fingerprint image collected by pressing is increased, and the indicator score of the repeatability factor in the first fingerprint image collected by pressing is reduced to adapt to the specific scenario.

[0116] In one embodiment, the current status data includes a second cumulative number, the second cumulative number being the number of templates that have been registered as strong fingerprints; The registration end condition includes that the second accumulated quantity is greater than a second quantity threshold.

[0117] The second accumulated number is the number of templates that have been registered as strong fingerprints. The second number threshold is a number threshold used to evaluate whether to end the fingerprint registration operation.

[0118] As an example, when the fingerprint processing device registers the first fingerprint image as a strong fingerprint template, it is necessary to use M=M+1 to update the second cumulative number M corresponding to the strong fingerprint template; then, the second cumulative number is compared with the second number threshold; if the second cumulative number is greater than the second number threshold, it is determined that the registration end condition has been met, and the target fingerprint image can be registered as a weak fingerprint template, and the fingerprint registration is completed based on all strong fingerprint templates and all weak fingerprint templates; conversely, if the second cumulative number is not greater than the second number threshold, it is determined that the registration end condition has not been met, and it is necessary to repeat the image acquisition, that is, repeat step S1.

[0119] In one embodiment, the current state data includes an effective area ratio, which is a ratio of the overlapping area of ​​all the strong fingerprint templates after splicing to the area of ​​the standard fingerprint template; The registration termination conditions include that the effective area ratio is greater than the preset ratio threshold.

[0120] Among them, the effective area ratio is the ratio of the overlapping area of ​​all strong fingerprint templates after splicing to the area of ​​the standard fingerprint template. Specifically, it means that after the first fingerprint image is registered as a strong fingerprint template, all strong fingerprint templates are spliced ​​to determine their corresponding overlapping area. The overlapping area can be understood as the overlapping area of ​​the first fingerprint image and its previously determined spliced ​​fingerprint image. The preset ratio threshold is a pre-set ratio threshold for evaluating whether to end the fingerprint registration process. The standard fingerprint template is a pre-set fingerprint template, which can be a fingerprint template with a total area determined by a pre-calibrated test.

[0121] As an example, after registering the first fingerprint image as a strong fingerprint template, the fingerprint processing device needs to determine the overlapping area based on all the strong fingerprint templates in the fingerprint template library and the splicing of all the strong fingerprint templates, and then calculate the overlapping area with the area of ​​the standard fingerprint template to determine the effective area ratio; then, compare the effective area ratio with the preset ratio threshold; if the effective area ratio is greater than the preset ratio threshold, it is determined that the registration end condition has been met, and the target fingerprint image can be registered as a weak fingerprint template, and the fingerprint registration is completed based on all the strong fingerprint templates and all the weak fingerprint templates; conversely, if the effective area ratio is not greater than the preset ratio threshold, it is determined that the registration end condition has not been met, and it is necessary to repeat the image acquisition, that is, repeat step S1.

[0122] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0123] In one embodiment, a fingerprint processing device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the fingerprint processing method in the above embodiment is implemented, for example Figure 2 S1-S5 as shown, or Figure 3 to Figure 4 To avoid repetition, it will not be described here.

[0124] In one embodiment, a fingerprint processing system is provided, including a fingerprint sensor and the above-mentioned fingerprint processing device, wherein the fingerprint processing device is connected to the fingerprint sensor, and will not be described again here to avoid repetition.

[0125] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A fingerprint processing method, characterized in that: include: Acquire a first fingerprint image, where the first fingerprint image is a fingerprint image collected at a current moment; Performing image analysis on the first fingerprint image to determine an image type corresponding to the first fingerprint image; If the image type corresponding to the first fingerprint image is a weak template type, determining a target fingerprint image based on the first fingerprint image, and caching the target fingerprint image; If the image type corresponding to the first fingerprint image is a strong template type, registering the first fingerprint image as a strong fingerprint template, and updating the current state data corresponding to the strong fingerprint template; When the current state data corresponding to the strong fingerprint template meets the registration end condition, the target fingerprint image is registered as a weak fingerprint template, and the fingerprint registration is completed based on the strong fingerprint template and the weak fingerprint template.

2. The fingerprint processing method according to claim 1, characterized in that: The performing image analysis on the first fingerprint image to determine the image type corresponding to the first fingerprint image includes: Performing motion blur recognition based on the first fingerprint image and the second fingerprint image to determine the motion blur type corresponding to the first fingerprint image, where the second fingerprint image is a fingerprint image collected at a previous moment; Performing effective area and image quality detection on the first fingerprint image to determine a first detection result corresponding to the first fingerprint image; Perform repetition rate detection based on the first fingerprint image and the third fingerprint image to determine a second detection result corresponding to the first fingerprint image, wherein the third fingerprint image is a fingerprint image whose image type is a strong template type or a weak template type determined before the current moment; An image type corresponding to the first fingerprint image is determined based on the motion blur type corresponding to the first fingerprint image, the first detection result, and the second detection result.

3. The fingerprint processing method according to claim 2, characterized in that: The performing motion blur recognition based on the first fingerprint image and the second fingerprint image to determine the motion blur type corresponding to the first fingerprint image includes: Performing motion blur recognition on the first fingerprint image and the second fingerprint image to determine a target motion blur value corresponding to the first fingerprint image; Based on the target motion blur value, a motion blur type corresponding to the first fingerprint image is determined.

4. The fingerprint processing method according to claim 2, characterized in that: The performing effective area and image quality detection on the first fingerprint image to determine a first detection result corresponding to the first fingerprint image includes: Performing effective area and image quality detection on the first fingerprint image to determine an effective area and a quality score corresponding to the first fingerprint image; If the effective area is greater than the target area threshold, and the quality score is greater than the target quality threshold, determining that the first detection result corresponding to the first fingerprint image is detection passed; If the effective area is not greater than the target area threshold, or the quality score is not greater than the target quality threshold, it is determined that the first detection result corresponding to the first fingerprint image is a detection failure.

5. The fingerprint processing method according to claim 4, characterized in that: Before performing effective area and image quality detection on the first fingerprint image and determining a first detection result corresponding to the first fingerprint image, the fingerprint processing method further includes: Performing motion blur recognition on the first fingerprint image and the second fingerprint image to determine a target motion blur value corresponding to the first fingerprint image; Based on the target motion blur value, determining the target area threshold and the target quality threshold; The target area threshold is positively correlated with the target motion blur value, and the target quality threshold is positively correlated with the target motion blur value.

6. The fingerprint processing method according to claim 2, characterized in that: The third fingerprint image includes a fourth fingerprint image, and the fourth fingerprint image is a fingerprint image whose image type determined before the current moment is a strong template type; The performing repetition rate detection based on the first fingerprint image and the third fingerprint image to determine a second detection result corresponding to the first fingerprint image includes: Perform repetition rate detection based on the first fingerprint image and all the fourth fingerprint images to determine a first measured repetition rate; Perform repetition rate detection based on the first fingerprint image and a third fingerprint image closest to the current moment to determine a second measured repetition rate; If the first measured repetition rate is less than a first repetition rate threshold, and the second measured repetition rate is less than a second repetition rate threshold, determining that the second detection result corresponding to the first fingerprint image is detection passed; If the first measured repetition rate is not less than the first repetition rate threshold, or the second measured repetition rate is not less than the second repetition rate threshold, it is determined that the second detection result corresponding to the first fingerprint image is detection failure.

7. The fingerprint processing method according to claim 6, characterized in that: The first measured repetition rate is the maximum value among the single repetition rates corresponding to all the fourth fingerprint images, and the single repetition rate is the repetition rate between the fourth fingerprint image and the first fingerprint image; Alternatively, the first measured repetition rate is the repetition rate between the spliced ​​fingerprint image and the first fingerprint image, and the spliced ​​fingerprint image is a fingerprint image obtained by splicing all the fourth fingerprint images.

8. The fingerprint processing method according to claim 6, characterized in that: Before performing repetition rate detection based on the first fingerprint image and the third fingerprint image to determine a second detection result corresponding to the first fingerprint image, the fingerprint processing method further includes: Acquire a first cumulative number, where the first cumulative number is the number of first fingerprint images whose image type is a weak template type acquired continuously; Determining the first repetition rate threshold and the second repetition rate threshold based on the first accumulated number; Among them, the first repetition rate threshold is negatively correlated with the first cumulative number; the second repetition rate threshold is negatively correlated with the first cumulative number.

9. The fingerprint processing method according to claim 2, characterized in that: The determining, based on the motion blur type corresponding to the first fingerprint image, the first detection result, and the second detection result, the image type corresponding to the first fingerprint image includes: If the motion blur type is a non-blur type, the first detection result is a detection pass, and the second detection result is a detection pass, determining that the image type corresponding to the first fingerprint image is a strong template type; If the motion blur type is a non-blur type, the first detection result is a detection pass, and the second detection result is a detection fail, determining that the image type corresponding to the first fingerprint image is a weak template type; Alternatively, if the motion blur type is a semi-blur type, the first detection result is a detection pass, and the second detection result is a detection pass, it is determined that the image type corresponding to the first fingerprint image is a weak template type.

10. The fingerprint processing method according to claim 1, characterized in that: The step of determining a target fingerprint image based on the first fingerprint image includes: Obtaining the number of images corresponding to a fifth fingerprint image, wherein the fifth fingerprint image is a fingerprint image whose image type is a weak template type determined before the current moment; If the number of images corresponding to the fifth fingerprint image is less than a first number threshold, the first fingerprint image and all the fifth fingerprint images are determined as target fingerprint images; If the number of images corresponding to the fifth fingerprint image is not less than a first number threshold, the first fingerprint image and all the fifth fingerprint images are analyzed to determine a target fingerprint image.

11. The fingerprint processing method according to claim 10, characterized in that: The step of analyzing the first fingerprint image and all the fifth fingerprint images to determine a target fingerprint image includes: Determine the priority corresponding to the first fingerprint image; If the priority corresponding to the first fingerprint image is less than the minimum priority of all the fifth fingerprint images, the first fingerprint image is deleted, and all the fifth fingerprint images are determined as target fingerprint images; If the priority corresponding to the first fingerprint image is not less than the minimum priority of all the fifth fingerprint images, the fifth fingerprint image corresponding to the minimum priority is deleted, and the first fingerprint image and the remaining fifth fingerprint images are determined as target fingerprint images.

12. The fingerprint processing method according to claim 11, characterized in that: The determining the priority corresponding to the first fingerprint image includes: Performing motion blur recognition based on the first fingerprint image and the second fingerprint image to determine a target motion blur value, and determining a first indicator score corresponding to the first fingerprint image based on the target motion blur value, wherein the second fingerprint image is a fingerprint image collected at a previous moment; Performing effective area and image quality detection on the first fingerprint image to determine an effective area and a quality score, and determining a second indicator score corresponding to the first fingerprint image based on the effective area and the quality score; Perform repetition rate detection based on the first fingerprint image and the third fingerprint image to determine the measured repetition rate, and determine the third indicator score corresponding to the first fingerprint image based on the measured repetition rate, wherein the third fingerprint image is a fingerprint image whose image type determined before the current moment is a strong template type or a weak template type; The priority corresponding to the first fingerprint image is determined based on the first indicator score, the second indicator score, and the third indicator score corresponding to the first fingerprint image.

13. The fingerprint processing method according to claim 12, characterized in that: The determining the priority corresponding to the first fingerprint image based on the first indicator score, the second indicator score, and the third indicator score corresponding to the first fingerprint image includes: Determining a fingerprint collection method corresponding to the first fingerprint image; If the fingerprint collection method is a pressing method, a first indicator score, a second indicator score, and a third indicator score corresponding to the first fingerprint image are weighted based on a first weight combination to determine a priority corresponding to the first fingerprint image; in the first weight combination, the weight of the first indicator score is less than the weight of the third indicator score; If the fingerprint collection method is a sliding method, the first indicator score, the second indicator score and the third indicator score corresponding to the first fingerprint image are weighted based on the second weight combination to determine the priority corresponding to the first fingerprint image; in the second weight combination, the weight of the first indicator score is greater than the weight of the third indicator score.

14. The fingerprint processing method according to claim 1, characterized in that: The current state data includes a second cumulative number, which is the number of fingerprint templates that have been registered as strong fingerprint templates; the registration end condition includes that the second cumulative number is greater than a second number threshold; Alternatively, the current status data includes an effective area ratio, which is the ratio of the overlapping area of ​​all strong fingerprint templates after splicing to the area of ​​the standard fingerprint template; the registration end condition includes that the effective area ratio is greater than a preset ratio threshold.

15. A fingerprint processing device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the fingerprint processing method according to any one of claims 1 to 14 is implemented.

16. A fingerprint processing system, characterized in that: It comprises a fingerprint sensor and the fingerprint processing device as claimed in claim 15, wherein the fingerprint processing device is connected to the fingerprint sensor.

Images