Fingerprint carrier state identification method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a fingerprint carrier state recognition method and device, electronic equipment and a storage medium. The fingerprint carrier state recognition method comprises the steps that a current frame fingerprint image is acquired; inputting the feature information of the current frame of fingerprint image, the time information corresponding to the current frame of fingerprint image, and the hidden variable corresponding to the current frame of fingerprint image into a recurrent neural network model to obtain carrier state information output by the recurrent neural network model and a hidden variable corresponding to the next frame of fingerprint image, the fingerprint carrier is a medium bearing a fingerprint corresponding to the fingerprint image; and determining the state of the fingerprint carrier according to the carrier state information. According to the fingerprint carrier state identification method provided by the invention, when carrier state identification is carried out, the time information of the current frame of fingerprint image and the hidden variables obtained from the historical fingerprint images are referred, and the identification accuracy can be relatively high.

Description

Technical Field

[0001] Embodiments of the present application relate to the technical field of fingerprint recognition, and in particular, to a method, device, electronic device, and storage medium for identifying the state of a fingerprint carrier. Background Art

[0002] Fingerprint recognition is a biometric technology. Since fingerprint recognition is widely used in multiple fields such as unlocking smartphones, financial transactions, and access control, it is crucial to avoid illegal unlocking caused by fake fingerprints and foreign object attacks, as well as derivative problems such as information leakage in daily applications.

[0003] Currently, after a finger presses on the recognition area of a fingerprint module, a single-frame fingerprint image or multiple-frame fingerprint images are collected to determine whether the fingerprint carrier corresponding to the fingerprint image is a forged finger.

[0004] However, for the single-frame fingerprint image anti-counterfeiting scheme, limited by the information collected once, it may be interfered by possible local pressing, dirty fingers, etc., resulting in misjudgment. For the multi-frame fingerprint image anti-counterfeiting scheme, due to the high collection repetition rate, the anti-counterfeiting effect is poor, and the time required for fingerprint authentication is long. Therefore, the existing fingerprint anti-counterfeiting schemes have a poor user experience. Summary of the Invention

[0005] In view of this, embodiments of the present application provide a method, device, electronic device, and computer storage medium for identifying the state of a fingerprint carrier to at least partially solve the above problems.

[0006] According to a first aspect of the embodiments of the present application, a method for identifying the state of a fingerprint carrier is provided, including: obtaining a current-frame fingerprint image; inputting the feature information of the current-frame fingerprint image, the time information corresponding to the current-frame fingerprint image, and the latent variable corresponding to the current-frame fingerprint image into a recurrent neural network model to obtain the carrier state information output by the recurrent neural network model and the latent variable corresponding to the next-frame fingerprint image, where the fingerprint carrier is a medium carrying the fingerprint corresponding to the fingerprint image; determining the state of the fingerprint carrier according to the carrier state information.

[0007] In a possible implementation, the feature information of the current-frame fingerprint image includes at least one of the fingerprint signal output by the fingerprint sensor when collecting the current-frame fingerprint image, the current-frame fingerprint image, and the fingerprint feature extracted from the current-frame fingerprint image.

[0008] In a possible implementation, the time information corresponding to the current-frame fingerprint image includes the time when the current-frame fingerprint image is collected, and / or the time interval between collecting the current-frame fingerprint image and collecting the previous-frame fingerprint image.

[0009] In a possible implementation, the state of the fingerprint carrier includes at least one of whether the fingerprint carrier is a dichotomous result of a finger, the medium classification of the fingerprint carrier, and the physical state of the fingerprint carrier.

[0010] In a possible implementation, the recurrent neural network model is obtained by the following method: Obtain a plurality of fingerprint image queues, where each fingerprint image queue includes a plurality of sample fingerprint images, and the sample fingerprint images included in each fingerprint image queue are obtained by collecting fingerprints of at least one fingerprint carrier; Determine the timing information of each fingerprint image queue, where the timing information includes the time information corresponding to each sample fingerprint image in the fingerprint image queue; Determine the carrier state information of the last sample fingerprint image in each fingerprint image queue; Construct a training sample corresponding to each fingerprint image queue, where the training sample includes the fingerprint image queue, the timing information corresponding to the fingerprint image queue, and the carrier state information of the last sample fingerprint image in the fingerprint image queue; Train the model to be trained with each training sample to obtain the recurrent neural network model.

[0011] In a possible implementation, the obtaining of the plurality of fingerprint image queues includes: respectively collecting sample fingerprint images corresponding to fingerprints carried by at least two fingerprint carriers to obtain at least two sample fingerprint image libraries; extracting a plurality of sample fingerprint images from the at least two sample fingerprint image libraries at a target time interval to obtain multiple groups of sample fingerprint image queues, where the sample fingerprint image queue includes a plurality of sample fingerprint images extracted from at least one sample fingerprint library; sorting each group of sample fingerprint image queues in the extraction order to obtain a plurality of fingerprint image queues.

[0012] In a possible implementation, the training the model to be trained with each training sample to obtain the recurrent neural network model includes: respectively inputting the fingerprint image queues included in each training sample and the timing information corresponding to the fingerprint image queues into the model to be trained to obtain the recognition results corresponding to each training sample, where the recognition result is used to indicate the carrier state of the last sample fingerprint image in the fingerprint image queue of the training sample; determining the matching result corresponding to the training sample according to the recognition result corresponding to each training sample and the carrier state information of the last sample fingerprint image in the training sample; correcting the parameters of the model to be trained according to the matching results corresponding to each training sample until the matching results corresponding to each training sample meet the training completion condition to obtain the recurrent neural network model.

[0013] In a possible implementation, the step of inputting the fingerprint image queues included in each of the training samples and the timing information corresponding to the fingerprint image queues into the model to be trained to obtain the recognition results corresponding to each of the training samples includes: performing the following operations on the fingerprint image queue in each of the training samples: obtaining an initial variable; inputting the initial variable, the first-sorted sample fingerprint image in the fingerprint image queue, and the time information corresponding to the first-sorted sample fingerprint image in the fingerprint image queue into the model to be trained to obtain a first variable; inputting the (n - 1)-th variable, the n-th sorted sample fingerprint image in the fingerprint image queue, and the time information corresponding to the n-th sorted sample fingerprint image in the fingerprint image queue into the model to be trained to obtain an n-th variable; inputting the (m - 1)-th variable, the last sample fingerprint image in the fingerprint image queue, and the time information corresponding to the last sample fingerprint image in the fingerprint image queue into the model to be trained to obtain the recognition result corresponding to the fingerprint image queue, where m is the number of sample fingerprint images included in the fingerprint image queue, and n is an integer less than or equal to m - 1 and greater than or equal to 2.

[0014] In a possible implementation, the recurrent neural network model is obtained by the following method: obtaining model parameters; and modifying the parameters in the model to be trained according to the model parameters to obtain the recurrent neural network model.

[0015] According to a second aspect of the embodiments of the present application, there is provided a fingerprint carrier state recognition device, including: an acquisition unit configured to acquire a current frame fingerprint image; an input unit configured to input the feature information of the current frame fingerprint image, the time information corresponding to the current frame fingerprint image, and the latent variable corresponding to the current frame fingerprint image into a recurrent neural network model to obtain the carrier state information output by the recurrent neural network model and the latent variable corresponding to the next frame fingerprint image, where the fingerprint carrier is a medium carrying the fingerprint corresponding to the fingerprint image; and a determination unit configured to determine the state of the fingerprint carrier according to the carrier state information.

[0016] According to a third aspect of the embodiments of the present application, there is provided an electronic device, including: a processor, a memory, a communication interface, and a communication bus, where the processor, the memory, and the communication interface complete communication with each other through the communication bus; the memory is configured to store at least one executable instruction, and the executable instruction causes the processor to perform the operations corresponding to the method described in the first aspect.

[0017] According to a fourth aspect of the embodiments of the present application, there is provided a computer storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the method described in the first aspect.

[0018] According to the fingerprint carrier status recognition solution provided by the embodiments of the present application, the current frame fingerprint image is obtained, and the feature information of the current frame fingerprint image, the time information corresponding to the current frame fingerprint image, and the latent variable corresponding to the current frame fingerprint image are input into a recurrent neural network model to determine the carrier status information corresponding to the current frame fingerprint image. Thus, the fingerprint carrier status corresponding to the current frame fingerprint image can be determined, and fingerprint anti-counterfeiting and derivative operations can be performed based on the determined fingerprint carrier status. Since the content input into the recurrent neural network model includes the latent variable of the current frame fingerprint image, and the latent variable of the current frame fingerprint image is determined according to historical fingerprint images, historical fingerprint images are referred to during carrier status recognition, and the acquisition time of the current frame fingerprint image is also referred to. Compared with the prior art solution for anti-counterfeiting recognition using a single-frame fingerprint image, since the fingerprint anti-counterfeiting result is not determined solely based on a single-frame fingerprint image, the recognition accuracy can be improved. Compared with the prior art solution for anti-counterfeiting recognition using multiple-frame fingerprint images, since the current frame fingerprint image is recognized by a recurrent neural network model, less time is consumed, and the user experience can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the embodiments of the present application, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0020] Figure 1 is a flowchart of a fingerprint carrier status recognition method provided by an embodiment of the present application;

[0021] Figure 2 is a flowchart of a recurrent neural network model training method provided by an embodiment of the present application;

[0022] Figure 3 is a schematic diagram of a fingerprint image queue construction process provided by an embodiment of the present application;

[0023] Figure 4 is a flowchart of a recurrent neural network model training method provided by an embodiment of the present application;

[0024] Figure 5 is a schematic diagram of a fingerprint carrier status recognition device provided by an embodiment of the present application;

[0025] Figure 6 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art shall fall within the scope of protection of the embodiments of the present application.

[0027] The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. The singular forms "a", "the", and "said" used in the present application and the appended claims are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0028] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "when" or "while" or "in response to determining".

[0029] As mentioned above, fingerprint recognition is a biometric technology. Since fingerprint recognition is widely used in multiple fields such as smartphone unlocking, financial transactions, and access control, it is crucial to avoid problems such as illegal unlocking caused by fake fingerprints and foreign object attacks, as well as the resulting information leakage in daily applications. Currently, after a finger presses on the recognition area of the fingerprint module, it is determined whether the fingerprint carrier corresponding to the fingerprint image is a forged finger by collecting a single-frame fingerprint image or multiple-frame fingerprint images. However, for the single-frame fingerprint image anti-counterfeiting scheme, due to the limitation of the information collected once, it is interfered by possible local pressing, dirty fingers, etc., resulting in misjudgment. For the multi-frame fingerprint image anti-counterfeiting scheme, due to the high collection repetition rate, the anti-counterfeiting effect is poor, and the time required for fingerprint authentication is long. Therefore, the existing fingerprint anti-counterfeiting schemes have a poor user experience.

[0030] An embodiment of the present application provides a method for identifying the state of a fingerprint carrier. The method includes obtaining a current-frame fingerprint image, inputting the feature information of the current-frame fingerprint image, the time information corresponding to the current-frame fingerprint image, and the latent variable corresponding to the current-frame fingerprint image into a recurrent neural network model, and determining the carrier state information corresponding to the current-frame fingerprint image. Thus, the fingerprint carrier state corresponding to the current-frame fingerprint image can be determined, and fingerprint anti-counterfeiting and derivative operations can be performed based on the determined fingerprint carrier state. Since the content input into the recurrent neural network model includes the latent variable of the current-frame fingerprint image, and the latent variable of the current-frame fingerprint image is determined based on historical fingerprint images, historical fingerprint images are referred to during carrier state identification, and the acquisition time of the current-frame fingerprint image is also referred to. Compared with the prior art solution for anti-counterfeiting identification using a single-frame fingerprint image, since the fingerprint anti-counterfeiting result is not determined only based on a single-frame fingerprint image, the identification accuracy can be improved. Compared with the prior art solution for anti-counterfeiting identification using multiple-frame fingerprint images, since the current-frame fingerprint image is identified through a recurrent neural network model, less time is consumed, and the user experience can be improved.

[0031] The following describes the method for identifying the state of a fingerprint carrier provided by the present application through embodiments.

[0032] Figure 1 is a flowchart of a method for identifying the state of a fingerprint carrier provided by an embodiment of the present application. As Figure 1 shown, the method for identifying the state of a fingerprint carrier includes the following steps 101 to 103:

[0033] Step 101: Obtain a current-frame fingerprint image.

[0034] When a fingerprint is pressed on a fingerprint sensor, the current-frame fingerprint image is collected through the fingerprint sensor. The current-frame fingerprint image is a single-frame fingerprint image.

[0035] Step 102: Input the feature information of the current-frame fingerprint image, the time information corresponding to the current-frame fingerprint image, and the latent variable corresponding to the current-frame fingerprint image into a recurrent neural network model, and obtain the carrier state information output by the recurrent neural network model and the latent variable corresponding to the next-frame fingerprint image.

[0036] The fingerprint carrier is a medium that bears the fingerprint corresponding to the fingerprint image. The fingerprint carrier can be a finger, 2D printing material, 3D silicone material, etc. The latent variable corresponding to the current-frame fingerprint image can be obtained based on historically collected fingerprint images. Specifically, when a fingerprint image is collected for the first time, the latent variable corresponding to the current-frame fingerprint image is a preset value. For example, it can be set to 0. When a fingerprint image is not collected for the first time, the latent variable corresponding to the current-frame fingerprint image is the latent variable output after the previous-frame fingerprint image is input into the recurrent neural network model.

[0037] The recurrent neural network model outputs the carrier state information corresponding to the current frame fingerprint image and the hidden variable corresponding to the next frame fingerprint image according to the feature information of the current frame fingerprint image input into the model, the time information corresponding to the current frame fingerprint image, and the hidden variable corresponding to the current frame fingerprint image. The hidden variable corresponding to the next frame fingerprint image can be used for carrier state recognition of the next frame fingerprint image. In one example, the recurrent neural network can be the following formula, [Y t ,h t+1 = f(X t ,T t ,h t ), where t = 1, 2, …, M. t is used to represent the t-th frame fingerprint image. Y t is used to represent the carrier state information corresponding to the t-th frame fingerprint image. h t is used to represent the hidden variable corresponding to the t-th frame fingerprint image. X t is used to represent the feature information of the t-th frame fingerprint image. T t is used to represent the time information corresponding to the t-th frame fingerprint image.

[0038] In one example, the recurrent neural network model can adopt various time series signal processing recurrent architectures including the Recurrent Neural Network (RNN) architecture, the Gate Recurrent Unit (GRU) architecture, the Long Short-Term Memory (LSTM) architecture, and Mamba.

[0039] Optionally, the hidden variable corresponding to the next frame fingerprint image is obtained by the recurrent neural network model updating the hidden variable corresponding to the current frame fingerprint image according to the feature information of the current frame fingerprint image input into the model and the time information corresponding to the current frame fingerprint image. In one example, for the recurrent neural network with the RNN architecture, the hidden variable can be updated by the following formula, is used to represent the activation function. b, U, and W are used to represent the fully connected parameters. X t is used to represent the feature information of the t-th frame fingerprint image. T t is used to represent the time information corresponding to the t-th frame fingerprint image.

[0040] Step 103: Determine the state of the fingerprint carrier according to the carrier state information.

[0041] Determine the state of the fingerprint carrier according to the carrier state information. For example, the state of the fingerprint carrier can be determined according to the above Y t For example, it can be determined whether the fingerprint carrier is a finger.

[0042] In the embodiment of the present application, the current frame fingerprint image is acquired, and the feature information of the current frame fingerprint image, the time information corresponding to the current frame fingerprint image, and the latent variable corresponding to the current frame fingerprint image are input into a recurrent neural network model to determine the carrier state information corresponding to the current frame fingerprint image. Thus, the fingerprint carrier state corresponding to the current frame fingerprint image can be determined, and fingerprint anti-counterfeiting and derivative operations can be performed based on the determined fingerprint carrier state. Since the content input into the recurrent neural network model includes the latent variable of the current frame fingerprint image, and the latent variable of the current frame fingerprint image is determined according to historical fingerprint images, historical fingerprint images are referred to during carrier state recognition, and the acquisition time of the current frame fingerprint image is also referred to. Compared with the prior art solution for anti-counterfeiting recognition using a single-frame fingerprint image, since the fingerprint anti-counterfeiting result is not determined solely based on a single-frame fingerprint image, the recognition accuracy can be improved. Compared with the prior art solution for anti-counterfeiting recognition using multiple-frame fingerprint images, since the current frame fingerprint image is recognized by a recurrent neural network model, less time is consumed, and the user experience can be improved.

[0043] In a possible implementation manner, the feature information of the current frame fingerprint image includes at least one of the fingerprint signal output by the fingerprint sensor when the current frame fingerprint image is acquired, the current frame fingerprint image, and the fingerprint features extracted from the current frame fingerprint image.

[0044] The feature information of the current frame fingerprint image may be the fingerprint signal output by the fingerprint sensor when the current frame fingerprint image is acquired. For example: the optical imaging result of an optical fingerprint module, the capacitance impedance value of a capacitive fingerprint module, the ultrasonic echo of an ultrasonic fingerprint module, etc., in the original domain. The feature information of the current frame fingerprint image may also be the current frame fingerprint image. For example: the fingerprint image recognized by the processing unit according to the fingerprint signal output by the fingerprint sensor. The feature information of the current frame fingerprint image may also be the fingerprint features extracted from the current frame fingerprint image. For example: the fingerprint features obtained by performing operations such as convolution on the current frame fingerprint image.

[0045] In the embodiment of the present application, the feature information of the current frame fingerprint image includes at least one of the fingerprint signal output by the fingerprint sensor when the current frame fingerprint image is acquired, the current frame fingerprint image, and the fingerprint features extracted from the current frame fingerprint image. Thus, the features of the current frame fingerprint image can be accurately represented by the feature information of the current frame fingerprint image, enabling the recurrent neural network model to obtain the features of the current frame fingerprint image.

[0046] In a possible implementation manner, the time information corresponding to the current frame fingerprint image includes the time when the current frame fingerprint image is acquired, and / or the time interval between the acquisition of the current frame fingerprint image and the acquisition of the previous frame fingerprint image.

[0047] It should be noted that after the carrier state of the previous frame of fingerprint image is recognized, if the current frame of fingerprint image is collected within a very short time, the probability of change in the fingerprint carrier is relatively low. Therefore, the proportion of the hidden variable for the current frame of fingerprint image in the recurrent neural network model is relatively large. That is, when the time interval between collecting the current frame of fingerprint image and the previous frame of fingerprint image is small, the reference significance of the historical fingerprint image is relatively high. After the carrier state of the previous frame of fingerprint image is recognized and the current frame of fingerprint image is collected after a long time, the probability of change in the fingerprint carrier corresponding to the fingerprint image is relatively high. Therefore, the proportion of the hidden variable for the current frame of fingerprint image in the recurrent neural network model is relatively small. That is, when the time interval between collecting the current frame of fingerprint image and the previous frame of fingerprint image is large, the reference significance of the historical fingerprint image is relatively low.

[0048] In the embodiments of the present application, the time information corresponding to the current frame of fingerprint image includes the time when the current frame of fingerprint image is collected, and / or, the time interval between collecting the current frame of fingerprint image and the previous frame of fingerprint image. Thus, the time information of the current frame of fingerprint image can be input into the recurrent neural network model, so that the recurrent neural network model refers to the time information when performing fingerprint carrier state recognition, thereby determining the reference proportion of the hidden variable, which can improve the recognition accuracy of the fingerprint carrier state.

[0049] In a possible implementation manner, the state of the fingerprint carrier includes at least one of a binary result of whether the fingerprint carrier is a finger, a medium classification of the fingerprint carrier, and a physical state of the fingerprint carrier.

[0050] The state of the fingerprint carrier determined according to the carrier state information can be a binary result of whether the fingerprint carrier is a finger, that is, output whether the fingerprint carrier is a finger, output 1 if it is, otherwise output 0. The state of the fingerprint carrier can also be a medium classification of the fingerprint carrier. For example: the carrier is a value, the carrier is 3D silicone, etc. The state of the fingerprint carrier can also be a physical state of the fingerprint carrier. For example: taking the fingerprint carrier as a finger as an example, the dryness degree of the finger skin surface, the skin state of the finger, etc. can be output.

[0051] In an example, continuous skin state detection (such as water content, cutin layer state, scald or burn degree detection, etc.) can be performed according to the state of the fingerprint carrier, and accurate skin state classification can be achieved through the carrier state of the fingerprint carrier determined multiple times at different times.

[0052] In the embodiments of the present application, the state of the fingerprint carrier includes at least one of a binary result of whether the fingerprint carrier is a finger, a medium classification of the fingerprint carrier, and a physical state of the fingerprint carrier. Thus, fingerprint anti-counterfeiting recognition or derivative operations can be performed according to the current frame of fingerprint image, which can meet different needs of users.

[0053] Figure 2It is a flowchart of a method for training a recurrent neural network model provided by an embodiment of the present application. As Figure 2 shown, the recurrent neural network model can be obtained through the following steps 201 to 205.

[0054] Step 201: Obtain multiple fingerprint image queues.

[0055] Obtain multiple fingerprint image queues. Each fingerprint image queue includes multiple sample fingerprint images. The sample fingerprint images included in each fingerprint image queue are obtained by collecting fingerprints from at least one fingerprint carrier, that is, each fingerprint image queue includes fingerprint images corresponding to fingerprints carried by at least one fingerprint carrier. Different fingerprint image queues may include different numbers of sample fingerprint images and / or sample fingerprint images corresponding to different numbers of fingerprint carriers.

[0056] Step 202: Determine the timing information of each fingerprint image queue.

[0057] The timing information includes the time information corresponding to each sample fingerprint image in the fingerprint image queue. In one example, the timing information may be preset timing information. The time information corresponding to each sample fingerprint image may indicate the time corresponding to the sample fingerprint image, for example: the time interval from the previous frame of sample fingerprint image, the acquisition time corresponding to the sample fingerprint image, and so on.

[0058] Step 203: Determine the carrier state information of the last sample fingerprint image in each fingerprint image queue.

[0059] Each fingerprint image queue includes multiple sample fingerprint images. The multiple sample fingerprint images are arranged to form a fingerprint image queue. To determine the carrier state information of the sample fingerprint image at the end of the sorting of each fingerprint image queue, it should be noted that since the recurrent neural network model outputs the carrier state information of the last sample fingerprint image in the fingerprint queue, determining the carrier state information of the last sample fingerprint image can inform the model to be trained of the correct result.

[0060] Step 204: Construct training samples corresponding to each fingerprint image queue.

[0061] Each training sample includes a fingerprint image queue, the timing information corresponding to the fingerprint image queue, and the carrier state information of the last sample fingerprint image in the fingerprint image queue.

[0062] Step 205: Train the model to be trained with each training sample to obtain a recurrent neural network model.

[0063] In an embodiment of the present application, multiple fingerprint image queues are obtained, and multiple training samples are determined according to each fingerprint image queue, the corresponding timing information of the fingerprint image queue, and the carrier state information of the last sample fingerprint image in the fingerprint image queue. The to-be-trained model is trained with the multiple training samples to obtain a recurrent neural network model, realizing the model training process. Since the timing information corresponding to the fingerprint image queue is input as a variable into the to-be-trained model during model training, the trained recurrent neural network model refers to the time condition when identifying the carrier state information, and refers to historical fingerprint images when identifying the carrier state information of the current frame fingerprint image, which can improve the recognition accuracy of the recurrent neural network model.

[0064] In a possible implementation manner, when obtaining multiple fingerprint image queues, sample fingerprint images corresponding to fingerprints carried by at least two fingerprint carriers may be collected respectively to obtain at least two sample fingerprint image libraries. From the at least two sample fingerprint image libraries, multiple sample fingerprint images are extracted at a target time interval to obtain multiple groups of sample fingerprint image queues. Among them, a sample fingerprint image queue includes multiple sample fingerprint images extracted from at least one sample fingerprint library, and each group of sample fingerprint image queues is sorted according to the extraction order to obtain multiple fingerprint image queues.

[0065] Sample fingerprint images corresponding to fingerprints carried by at least two fingerprint carriers are collected respectively. The fingerprint carriers include but are not limited to fingers, 2D printing materials, 3D silicone materials, etc. When collecting sample fingerprint images, sample fingerprint images can be obtained by means such as continuous collection, interval collection, and virtual stitching. In an example, when constructing a sample fingerprint image library according to the sample fingerprint images, the sample fingerprint images can be preprocessed. For example, the sample fingerprint images can be preprocessed by methods such as rotation, affine transformation, displacement, flipping, denoising, enhancement, and position swapping to increase the diversity of the sample fingerprint images included in the sample fingerprint image library. It should be noted that a sample fingerprint image library only includes sample fingerprint images corresponding to one fingerprint carrier.

[0066] Taking two fingerprint image libraries as an example for illustration, that is, taking two fingerprint carriers as an example for illustration. Figure 3 is a schematic diagram of a fingerprint image queue construction process provided by an embodiment of the present application. As Figure 3 shown, from at least two sample fingerprint image libraries (for example, sample fingerprint image library 1 and sample fingerprint image library 2), according to the target time interval (for example Figure 3 the t in 1 、t 2 to t nExtract multiple sample fingerprint images (such as etc.), obtain multiple queues of sample fingerprint images. Each queue of sample fingerprint images includes multiple sample fingerprint images extracted from at least one sample fingerprint image library. For example, the queue 1 includes sample fingerprint images corresponding to one type of fingerprint carrier, and the queue 2 includes sample fingerprint images corresponding to two types of fingerprint carriers. Sort the sample fingerprint images in each queue of sample fingerprint images according to the extraction time to obtain multiple queues of fingerprint images. For example: Figure 3 Queues 1 to n in

[0067] It should be noted that the target time interval can be set manually or randomly generated. The target time interval is the timing information of the corresponding queue of fingerprint images.

[0068] In the embodiments of the present application, sample fingerprint images corresponding to fingerprints carried by at least two types of fingerprint carriers are respectively collected, at least two sample fingerprint image libraries are constructed, sample fingerprint images are extracted from at least two sample fingerprint image libraries according to a preset time sequence, and the extracted sample fingerprint images are sorted according to the extraction time to obtain multiple queues of fingerprint images. Thus, the construction of multiple queues of fingerprint images is realized. Since the sample fingerprint images in the queue of fingerprint images are sorted according to the extraction time, the process of collecting fingerprint images during daily use can be simulated. And since the queue of fingerprint images can include sample fingerprint images corresponding to at least two types of fingerprint carriers, continuous real finger presses, continuous fake fingerprint presses, alternating real and fake fingerprint presses, etc. can be simulated. Therefore, the recurrent neural network model trained according to the queue of fingerprint images can have a high recognition accuracy during daily use.

[0069] Figure 4 It is a flowchart of a method for training a recurrent neural network model provided by an embodiment of the present application. As Figure 4 shown, when training a model to be trained through each training sample to obtain a recurrent neural network model, the following steps 401 to 405 can be executed:

[0070] Step 401: Input the queue of fingerprint images included in each training sample and the timing information corresponding to the queue of fingerprint images into the model to be trained respectively, and obtain the recognition result corresponding to each training sample.

[0071] Construct a model to be trained, input the queue of fingerprint images included in each training sample and the timing information corresponding to the queue of fingerprint images into the constructed model to be trained. The recognition result can indicate the carrier state of the last sample fingerprint image in the queue of fingerprint images in this training sample.

[0072] Step 402: Determine the matching result corresponding to each training sample according to the recognition result corresponding to each training sample and the carrier state information of the last sample fingerprint image in this training sample.

[0073] Determine whether the recognition result output by the model to be trained matches the carrier status information of the last sample fingerprint image in the training sample. If they match, it proves that the recognition result output by the model to be trained is correct; if they do not match, it proves that the recognition result output by the model to be trained is incorrect.

[0074] Step 403: Determine whether the matching results corresponding to each training sample meet the training completion condition. If so, execute Step 404; otherwise, execute Step 405.

[0075] Determine whether the matching results corresponding to each training sample meet the training completion condition. The training completion condition can be set as needed. For example, the training completion condition can be set that the proportion of training samples with matching results is greater than the proportion threshold, the number of training samples with matching results exceeds the quantity threshold, and so on.

[0076] Step 404: Determine the model to be trained as a recurrent neural network model.

[0077] When the matching results corresponding to each training sample meet the training completion condition, the training is completed, and the model to be trained after the completion of training is determined as a recurrent neural network model.

[0078] Step 405: Modify the parameters of the model to be trained according to the matching results corresponding to each training sample, and execute Step 401.

[0079] When the matching results corresponding to each training sample do not meet the training completion condition, for example, the number of training samples with matching results is too low, the proportion is too low, etc. At this time, modify the parameters of the model to be trained according to the matching results corresponding to each training sample, and re-recognize after modifying the parameters of the model until the matching results corresponding to each training sample meet the training completion condition. It should be noted that when re-executing Step 401, it can be recognized through a fingerprint image queue different from the current recognition to obtain a new matching result, or it can also be recognized through the same fingerprint image queue, which is not limited here.

[0080] In the embodiments of the present application, the fingerprint image queue included in each training sample and the timing information corresponding to the fingerprint image queue are respectively input into the model to be trained, the recognition result corresponding to each training sample is obtained, the matching situation between the recognition result and the carrier state information of the fingerprint image of the last sample is determined, and according to the matching situations of each training sample, the parameters of the model to be trained are corrected until the matching results corresponding to each training sample meet the training completion condition, and a recurrent neural network model is obtained. Thus, the training of the recurrent neural network model can be realized. Since the timing information of the fingerprint queue is referred to when training the model, the recurrent neural network trained will refer to the timing information when recognizing the fingerprint image of the current frame. And since the training ends when the matching results corresponding to each training sample meet the training completion condition, the recognition accuracy of the recurrent neural network model can be guaranteed. Thus, the recurrent neural network model trained according to the fingerprint image queue can have a high recognition accuracy in daily use.

[0081] In a possible implementation manner, when the fingerprint image queue included in each training sample and the timing information corresponding to the fingerprint image queue are respectively input into the model to be trained to obtain the recognition result corresponding to each training sample, the following operations can be performed on the fingerprint image queue in each training sample:

[0082] An initial variable is obtained, the initial variable, the sample fingerprint image ranked 1st in the fingerprint image queue, and the time information corresponding to the sample fingerprint image ranked 1st in the fingerprint image queue are input into the model to be trained to obtain the 1st variable. The (n - 1)th variable, the sample fingerprint image ranked nth in the fingerprint image queue, and the time information corresponding to the sample fingerprint image ranked nth in the fingerprint image queue are input into the model to be trained to obtain the nth variable. The (m - 1)th variable, the fingerprint image of the last sample in the fingerprint image queue, and the time information corresponding to the fingerprint image of the last sample in the fingerprint image queue are input into the model to be trained to obtain the recognition result corresponding to this fingerprint image queue, where m is the number of sample fingerprint images included in the fingerprint image queue, and n is an integer less than or equal to m - 1 and greater than or equal to 2.

[0083] Since the sample fingerprint image ranked 1st in the fingerprint image queue is the first fingerprint image input into the model to be trained, the hidden variable corresponding to the sample fingerprint image ranked 1st is the initial variable, and the initial variable can be set manually. After the initial variable, the sample fingerprint image ranked 1st in the fingerprint image queue, and the time information corresponding to the sample fingerprint image ranked 1st in the fingerprint image queue are input into the model to be trained, the model to be trained will output the recognition result corresponding to the sample fingerprint image ranked 1st and the 1st variable after updating the initial variable.

[0084] Input the (n - 1)th variable, the nth sample fingerprint image sorted in the fingerprint image queue, and the time information corresponding to the nth sample fingerprint image sorted in the fingerprint image queue into the model to be trained, and obtain the recognition result corresponding to the nth sample fingerprint image output by the model to be trained and the nth variable obtained by updating the (n - 1)th variable. Here, n is an integer less than or equal to m - 1 and greater than or equal to 2, and m is the number of sample fingerprint images included in the fingerprint image queue. For example, when n = 2, input the 1st variable, the 2nd sample fingerprint image, and the time information corresponding to the 2nd sample fingerprint image sorted in the fingerprint image queue into the model to be trained to obtain the 2nd variable. When n = 5, input the 4th variable, the 5th sample fingerprint image sorted in the fingerprint image queue, and the time information corresponding to the 5th sample fingerprint image sorted in the fingerprint image queue into the model to be trained to obtain the 5th variable.

[0085] Input the (m - 1)th variable, the last sample fingerprint image in the fingerprint image queue, and the time information corresponding to the last sample fingerprint image in the fingerprint image queue into the model to be trained, and obtain the recognition result corresponding to this fingerprint image queue and the mth variable obtained by updating the (m - 1)th variable. Here, m is the number of sample fingerprint images in the fingerprint image queue, and thus obtain the recognition result corresponding to the last sample fingerprint image in the fingerprint image queue.

[0086] In the embodiments of the present application, for each training sample, successively input each sample fingerprint image in the fingerprint queue, the time information corresponding to each sample fingerprint image, and the hidden variable updated by the model to be trained according to the information corresponding to the previous sample fingerprint image into the model to be trained, and thus obtain the recognition result of the last sample fingerprint image in each fingerprint queue. Since the sample fingerprint images in the fingerprint queue are sorted according to the extraction time, it is possible to simulate the carrier state information of the current frame fingerprint image based on the hidden variable corresponding to the previous frame fingerprint image during daily use, thereby realizing the simulation of daily use.

[0087] In a possible implementation manner, the model parameters can be obtained, and the parameters in the model to be trained can be corrected according to the model parameters to obtain a recurrent neural network model.

[0088] When obtaining the recurrent neural network model, the model parameters can also be set according to experience. For example, the model parameters can be manually input, and the model parameters of the constructed model to be trained can be corrected by the manually input model parameters to obtain the recurrent neural network model.

[0089] In the embodiments of the present application, the model parameters are obtained, and the parameters in the model to be trained are corrected according to the model parameters to obtain a recurrent neural network model, thereby realizing the acquisition of the recurrent neural network model. Since there is no need to train the model to be trained, the time required to obtain the recurrent neural network by manually inputting the model parameters according to experience is short, and the model acquisition efficiency can be improved.

[0090] Figure 5 This is a schematic diagram of a fingerprint carrier state recognition device provided by an embodiment of the present application. As Figure 5 shown, the device 500 includes:

[0091] An acquisition unit 501, configured to acquire a current frame fingerprint image;

[0092] An input unit 502, configured to input the feature information of the current frame fingerprint image, the time information corresponding to the current frame fingerprint image, and the latent variable corresponding to the current frame fingerprint image into a recurrent neural network model, and obtain the carrier state information output by the recurrent neural network model and the latent variable corresponding to the next frame fingerprint image, where the fingerprint carrier is a medium carrying the fingerprint corresponding to the fingerprint image;

[0093] A determination unit 503, configured to determine the state of the fingerprint carrier according to the carrier state information.

[0094] In an embodiment of the present application, the acquisition unit 501 can be used to execute step 101 in the above method embodiment, the input unit 502 can be used to execute step 102 in the above method embodiment, and the determination unit 503 can be used to execute step 103 in the above method embodiment.

[0095] In a possible implementation manner, the feature information of the current frame fingerprint image includes at least one of the fingerprint signal output by the fingerprint sensor when acquiring the current frame fingerprint image, the current frame fingerprint image, and the fingerprint feature extracted from the current frame fingerprint image.

[0096] In a possible implementation manner, the time information corresponding to the current frame fingerprint image includes the time when the current frame fingerprint image is acquired, and / or the time interval between acquiring the current frame fingerprint image and acquiring the previous frame fingerprint image.

[0097] In a possible implementation manner, the state of the fingerprint carrier includes at least one of: a dichotomy result of whether the fingerprint carrier is a finger, a medium classification of the fingerprint carrier, and a physical state of the fingerprint carrier.

[0098] In a possible implementation, the recurrent neural network model is obtained through the following method: Obtain a plurality of fingerprint image queues, where each fingerprint image queue includes a plurality of sample fingerprint images, and the sample fingerprint images included in each fingerprint image queue are obtained by collecting fingerprints from at least one type of fingerprint carrier; Determine the timing information of each fingerprint image queue, where the timing information includes the time information corresponding to each sample fingerprint image in the fingerprint image queue; Determine the carrier state information of the last sample fingerprint image in each fingerprint image queue; Construct a training sample corresponding to each fingerprint image queue, where the training sample includes the fingerprint image queue, the timing information corresponding to the fingerprint image queue, and the carrier state information of the last sample fingerprint image in the fingerprint image queue; Train the model to be trained with each training sample to obtain a recurrent neural network model.

[0099] In a possible implementation, obtaining a plurality of fingerprint image queues includes: respectively collecting sample fingerprint images corresponding to fingerprints carried by at least two types of fingerprint carriers to obtain at least two sample fingerprint image libraries; extracting a plurality of sample fingerprint images from at least two sample fingerprint image libraries at a target time interval to obtain multiple groups of sample fingerprint image queues, where a sample fingerprint image queue includes a plurality of sample fingerprint images extracted from at least one sample fingerprint library; sorting each group of sample fingerprint image queues in the extraction order to obtain a plurality of fingerprint image queues.

[0100] In a possible implementation, training the model to be trained with each training sample to obtain a recurrent neural network model includes: respectively inputting the fingerprint image queues included in each training sample and the timing information corresponding to the fingerprint image queues into the model to be trained to obtain the recognition results corresponding to each training sample, where the recognition result is used to indicate the carrier state of the last sample fingerprint image in the fingerprint image queue of the training sample; determining the matching result corresponding to the training sample according to the recognition result corresponding to each training sample and the carrier state information of the last sample fingerprint image in the training sample; correcting the parameters of the model to be trained according to the matching results corresponding to each training sample until the matching results corresponding to each training sample meet the training completion condition to obtain a recurrent neural network model.

[0101] In a possible implementation, the fingerprint image queue included in each training sample and the corresponding timing information of the fingerprint image queue are respectively input into the model to be trained, and the recognition results corresponding to each training sample are obtained, including: performing the following operations on the fingerprint image queue in each training sample: obtaining an initial variable; inputting the initial variable, the sample fingerprint image ranked 1st in the fingerprint image queue, and the time information corresponding to the sample fingerprint image ranked 1st in the fingerprint image queue into the model to be trained to obtain a 1st variable; inputting the (n - 1)th variable, the sample fingerprint image ranked nth in the fingerprint image queue, and the time information corresponding to the sample fingerprint image ranked nth in the fingerprint image queue into the model to be trained to obtain an nth variable; inputting the (m - 1)th variable, the last sample fingerprint image in the fingerprint image queue, and the time information corresponding to the last sample fingerprint image in the fingerprint image queue into the model to be trained to obtain the recognition result corresponding to this fingerprint image queue, where m is the number of sample fingerprint images included in the fingerprint image queue, and n is an integer less than or equal to m - 1 and greater than or equal to 2.

[0102] In a possible implementation, a recurrent neural network model is obtained by the following method: obtaining model parameters; correcting the parameters in the model to be trained according to the model parameters to obtain a recurrent neural network model.

[0103] Referring to Figure 6 , a schematic structural diagram of an electronic device according to an embodiment of the present application is shown. The specific implementation of the electronic device in the specific embodiments of the present application is not limited.

[0104] As Figure 6 shown, the electronic device may include: a processor 602, a communication interface 604, a memory 606, and a communication bus 608.

[0105] Among them:

[0106] The processor 602, the communication interface 604, and the memory 606 communicate with each other through the communication bus 608.

[0107] The communication interface 604 is used to communicate with other electronic devices or servers.

[0108] The processor 602 is used to execute the program 610, and specifically can execute the relevant steps in the above-mentioned embodiment of the fingerprint carrier state recognition method.

[0109] Specifically, the program 610 may include program code, and the program code includes computer operation instructions.

[0110] The processor 602 may be a central processing unit (CPU), or a graphics processing unit (GPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application. One or more processors included in the intelligent device may be of the same type of processor, such as one or more CPUs; one or more GPUs; or may be of different types of processors, such as one or more CPUs, one or more GPUs, and one or more ASICs.

[0111] A memory 606 for storing a program 610. The memory 606 may include high-speed RAM memory and may also include non-volatile memory, such as at least one disk memory.

[0112] The program 610 may specifically be used to cause the processor 602 to execute the fingerprint carrier state recognition method in any of the foregoing embodiments.

[0113] For the specific implementation of each step in the program 610, reference may be made to the corresponding steps and descriptions in the corresponding units in any of the foregoing fingerprint carrier state recognition method embodiments, which will not be elaborated herein. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the devices and modules described above may refer to the corresponding process descriptions in the foregoing method embodiments, which will not be repeated herein.

[0114] In the embodiments of the present application, a current frame fingerprint image is obtained, and the feature information of the current frame fingerprint image, the time information corresponding to the current frame fingerprint image, and the latent variable corresponding to the current frame fingerprint image are input into a recurrent neural network model to determine the carrier state information corresponding to the current frame fingerprint image. Thus, the fingerprint carrier state corresponding to the current frame fingerprint image can be determined, and fingerprint anti-counterfeiting and derivative operations can be performed based on the determined fingerprint carrier state. Since the content input into the recurrent neural network model includes the latent variable of the current frame fingerprint image, and the latent variable of the current frame fingerprint image is determined according to historical fingerprint images, historical fingerprint images are referred to during carrier state recognition, and the acquisition time of the current frame fingerprint image is also referred to. Compared with the prior art solution for anti-counterfeiting recognition using a single-frame fingerprint image, since the fingerprint anti-counterfeiting result is not determined only based on a single-frame fingerprint image, the recognition accuracy can be improved. Compared with the prior art for anti-counterfeiting recognition using multiple-frame fingerprint images, since the current frame fingerprint image is recognized through a recurrent neural network model, less time is consumed, and the user experience can be improved.

[0115] An embodiment of the present application also provides a computer program product, including computer instructions, which direct a computing device to perform operations corresponding to any one of the above-mentioned multiple method embodiments.

[0116] It should be noted that, according to the needs of implementation, each component / step described in the embodiments of the present application can be split into more components / steps, or two or more components / steps or partial operations of a component / step can be combined into a new component / step to achieve the purpose of the embodiments of the present application.

[0117] The method according to the embodiments of the present application can be implemented in hardware, firmware, or be implemented as software or computer code that can be stored in a recording medium (such as a CD ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or be implemented as computer code originally stored in a remote recording medium or a non-transitory machine-readable medium and to be downloaded through a network and stored in a local recording medium, so that the method described herein can be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component (such as a RAM, a ROM, a flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the fingerprint carrier state recognition method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the fingerprint carrier state recognition method shown herein, the execution of the code converts the general-purpose computer into a dedicated computer for executing the fingerprint carrier state recognition method shown herein.

[0118] Those of ordinary skill in the art can realize that the units and method steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the embodiments of the present application.

[0119] The above embodiments are only used to illustrate the embodiments of the present application, rather than to limit the embodiments of the present application. Those of ordinary skill in the relevant technical field can also make various changes and modifications without departing from the spirit and scope of the embodiments of the present application. Therefore, all equivalent technical solutions also belong to the scope of the embodiments of the present application, and the patent protection scope of the embodiments of the present application shall be defined by the claims.

Claims

1. A fingerprint carrier state recognition method, characterized in that: include: Get the fingerprint image of the current frame; Input the feature information of the current frame fingerprint image, the time information corresponding to the current frame fingerprint image, and the hidden variable corresponding to the current frame fingerprint image into a recurrent neural network model, and obtain the carrier state information output by the recurrent neural network model and the hidden variable corresponding to the next frame fingerprint image, wherein the fingerprint carrier is a medium that carries the fingerprint corresponding to the fingerprint image; The state of the fingerprint carrier is determined according to the carrier state information.

2. The method according to claim 1, characterized in that The feature information of the current frame fingerprint image includes at least one of a fingerprint signal output by a fingerprint sensor when collecting the current frame fingerprint image, the current frame fingerprint image, and a fingerprint feature extracted from the current frame fingerprint image.

3. The method according to claim 1, characterized in that The time information corresponding to the current frame fingerprint image includes the time when the current frame fingerprint image is collected, and / or the time interval between collecting the current frame fingerprint image and collecting the previous frame fingerprint image.

4. The method according to claim 1, characterized in that The state of the fingerprint carrier includes at least one of: whether the fingerprint carrier is a binary differentiation result of a finger, a medium classification of the fingerprint carrier, and a physical state of the fingerprint carrier.

5. The method according to claim 1, characterized in that: The recurrent neural network model is obtained by the following method: Acquire a plurality of fingerprint image queues, wherein each of the fingerprint image queues includes a plurality of sample fingerprint images, and the sample fingerprint images included in each of the fingerprint image queues are obtained by performing fingerprint collection on at least one fingerprint carrier; Determine the timing information of each of the fingerprint image queues, wherein the timing information includes time information corresponding to each of the sample fingerprint images in the fingerprint image queue; Determining the carrier status information of the last sample fingerprint image in each of the fingerprint image queues; Constructing a training sample corresponding to each of the fingerprint image queues, wherein the training sample includes the fingerprint image queue, the timing information corresponding to the fingerprint image queue, and the carrier state information of the last sample fingerprint image in the fingerprint image queue; The recurrent neural network model is obtained by training the model to be trained through each of the training samples.

6. The method according to claim 5, characterized in that The step of acquiring multiple fingerprint image queues includes: Collecting sample fingerprint images corresponding to fingerprints carried by at least two fingerprint carriers respectively to obtain at least two sample fingerprint image libraries; Extracting a plurality of sample fingerprint images from the at least two sample fingerprint image libraries at target time intervals to obtain a plurality of groups of sample fingerprint image queues, wherein the sample fingerprint image queues include a plurality of sample fingerprint images extracted from at least one sample fingerprint library; Each group of sample fingerprint image queues is sorted according to the extraction order to obtain multiple fingerprint image queues.

7. The method according to claim 5, characterized in that The training of the model to be trained by each of the training samples to obtain the recurrent neural network model includes: Input the fingerprint image queue and the time sequence information corresponding to the fingerprint image queue included in each training sample into the to-be-trained model respectively, and obtain the recognition result corresponding to each training sample, wherein the recognition result is used to indicate the carrier state of the last sample fingerprint image in the fingerprint image queue in the training sample; Determine the matching result corresponding to each training sample according to the recognition result corresponding to each training sample and the carrier state information of the last sample fingerprint image in the training sample; According to the matching results corresponding to each training sample, the parameters of the model to be trained are modified until the matching results corresponding to each training sample meet the training completion conditions, thereby obtaining the recurrent neural network model.

8. The method according to claim 7, characterized in that The step of inputting the fingerprint image queues and the time sequence information corresponding to the fingerprint image queues included in each of the training samples into the to-be-trained model to obtain the recognition results corresponding to each of the training samples comprises: The following operations are performed on each fingerprint image queue in the training sample: Get the initial variables; Inputting the initial variable, the sample fingerprint image ranked first in the fingerprint image queue, and the time information corresponding to the sample fingerprint image ranked first in the fingerprint image queue into the model to be trained to obtain the first variable; Input the n-1th variable, the sample fingerprint image ranked nth in the fingerprint image queue, and the time information corresponding to the sample fingerprint image ranked nth in the fingerprint image queue into the to-be-trained model to obtain the nth variable; The m-1th variable, the last sample fingerprint image in the fingerprint image queue, and the time information corresponding to the last sample fingerprint image in the fingerprint image queue are input into the model to be trained to obtain the recognition result corresponding to the fingerprint image queue, where m is the number of sample fingerprint images included in the fingerprint image queue, and n is an integer less than or equal to m-1 and greater than or equal to 2.

9. The method according to claim 1, characterized in that: The recurrent neural network model is obtained by the following method: Get model parameters; The parameters in the model to be trained are modified according to the model parameters to obtain the recurrent neural network model.

10. A fingerprint carrier state recognition device, characterized in that: include: An acquisition unit, used for acquiring a fingerprint image of a current frame; An input unit, used to input feature information of the current frame fingerprint image, time information corresponding to the current frame fingerprint image, and hidden variables corresponding to the current frame fingerprint image into a recurrent neural network model, to obtain carrier state information output by the recurrent neural network model and hidden variables corresponding to the next frame fingerprint image, wherein the fingerprint carrier is a medium that carries the fingerprint corresponding to the fingerprint image; A determination unit is used to determine the state of the fingerprint carrier according to the carrier state information.

11. An electronic device, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute the fingerprint carrier state recognition method as described in any one of claims 1-9.

12. A computer storage medium, characterized in that: A computer program is stored thereon, and when the program is executed by a processor, the fingerprint carrier state recognition method as described in any one of claims 1 to 9 is implemented.

Images