Fingerprint detection method and device, program product, and storage medium

Abstract

This application provides a fingerprint detection method, apparatus, program product, and storage medium. The method includes: acquiring the native domain information of a fingerprint to be detected, wherein the native domain information is the raw data collected during fingerprint verification; performing fingerprint feature extraction on the native domain information to obtain a fingerprint feature matrix of the fingerprint to be detected; performing a classification operation on the fingerprint feature matrix through a trained media classification network to obtain a fingerprint media classification vector of the fingerprint to be detected, wherein the fingerprint media classification vector is used to represent the probability that the fingerprint to be detected belongs to multiple media types in vector form; and determining the media type of the fingerprint to be detected based on the fingerprint media classification vector. This application solves the problem in related technologies of the inability to effectively detect the type of fingerprint media.

Description

Technical Field

[0001] This application relates to the field of fingerprint detection, specifically to a fingerprint detection method and apparatus, a program product, and a storage medium. Background Technology

[0002] Currently, most existing technologies rely on explicit calculations and manual fitting of real-time physical characteristics such as acoustic impedance and surface distance using ultrasonic echo signals to identify fingerprint patterns and distinguish between genuine and fake fingerprints. However, this method can only identify whether a fingerprint is genuine or fake; it cannot detect the type of medium on which the fingerprint is located, thus hindering further analysis of the fingerprint.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a fingerprint detection method, apparatus, program product, and storage medium to at least solve the problem in related technologies that cannot effectively detect the type of fingerprint medium.

[0005] According to one embodiment of this application, a fingerprint detection method is provided, comprising: acquiring native domain information of a fingerprint to be detected, wherein the native domain information is raw data collected during fingerprint verification of the fingerprint to be detected; performing fingerprint feature extraction on the native domain information to obtain a fingerprint feature matrix of the fingerprint to be detected; performing a classification operation on the fingerprint feature matrix through a trained media classification network to obtain a fingerprint media classification vector of the fingerprint to be detected, wherein the fingerprint media classification vector is used to represent, in vector form, the probability that the fingerprint to be detected belongs to multiple media types; and determining the media type of the fingerprint to be detected based on the fingerprint media classification vector.

[0006] In an exemplary embodiment, before performing a classification operation on the fingerprint feature matrix using a trained media classification network to obtain the fingerprint media classification vector of the fingerprint to be detected, the method further includes: collecting native domain information of simulated fingerprints simulated on multiple media to obtain multiple sets of simulated fingerprint native domain information, wherein the multiple sets of simulated fingerprint native domain information include: multiple sets of native domain information of simulated fingerprints on the same medium obtained after multiple information collections on simulated fingerprints on the same medium; performing fingerprint feature extraction on each of the simulated fingerprint native domain information included in the multiple sets of simulated fingerprint native domain information to obtain multiple sets of simulated fingerprint feature matrices; and training the media classification network using the sets of simulated fingerprint feature matrices and the classification labels of the multiple media to obtain the media classification network, wherein the classification labels are used to represent the media types of the multiple media.

[0007] In an exemplary embodiment, a media classification network is trained using the simulated fingerprint feature matrix set and the classification labels of the various media to obtain the trained media classification network. This includes: inputting the simulated fingerprint feature matrix set into the media classification network to perform a training operation, wherein the training operation includes: inputting the simulated fingerprint feature matrix into a feature extraction layer of the media classification network; outputting a feature vector of the simulated fingerprint feature matrix through the feature extraction layer; inputting the feature vector of the simulated fingerprint feature matrix into a classification layer of the media classification network; calculating the probability that the feature vector of the simulated fingerprint feature matrix belongs to the classification labels of the various media through the classification layer; and outputting a media classification vector; determining the loss value output by the loss function of the media classification network based on the media classification vector and the media classification labels; ending the training when the loss value meets a preset training termination condition to obtain the trained media classification network; and adjusting the parameter values ​​of the media classification network when the loss value does not meet the preset training termination condition to reduce the loss value output by the loss function of the media classification network in the next training iteration.

[0008] In an exemplary embodiment, after inputting the set of simulated fingerprint feature matrices into the media classification network for training, the method further includes: extracting feature vectors of each simulated fingerprint feature matrix included in the set of simulated fingerprint feature matrices output by the feature extraction layer; mapping the feature vectors of each simulated fingerprint feature matrix to a coordinate space of a preset dimension, and performing dimensionality reduction processing on the feature vectors of each simulated fingerprint feature matrix in the coordinate space to obtain feature vectors of a target dimension; and performing visualization operations on the feature vectors of the target dimension of each simulated fingerprint feature matrix to display them on a target interface.

[0009] In an exemplary embodiment, performing fingerprint feature extraction on the native domain information to obtain a fingerprint feature matrix of the fingerprint to be detected includes: constructing a native domain information set, wherein the native domain information set includes multiple native domain information of the fingerprint to be detected acquired within the current time period, or the native domain information set includes multiple historical native domain information acquired within a historical time period and the native domain information acquired within the current time period; converting all the native domain information included in the native domain information set into digital signals using a signal converter to obtain multiple sets of digital signals; extracting digital features from each set of digital signals to obtain fingerprint features of each set of native domain information; generating a matrix including fingerprint features of each set of native domain information according to the signal acquisition channel of the target sensor acquiring the native domain information included in the native domain information set to obtain multiple sets of fingerprint feature matrices; and combining the multiple sets of fingerprint feature matrices to obtain the fingerprint feature matrix to be detected.

[0010] In an exemplary embodiment, a classification operation is performed on the fingerprint feature matrix using a trained media classification network to obtain a fingerprint media classification vector for the fingerprint to be detected. This includes: inputting the fingerprint feature matrix to be detected into a feature extraction layer included in the media classification network; extracting feature vectors for each group of fingerprint feature matrices included in the fingerprint feature matrix to be detected through the feature extraction layer; inputting the feature vectors of each group of fingerprint feature matrices output by the feature extraction layer into a classification layer included in the media classification network; calculating the probability that the feature vectors of each group of fingerprint feature matrices belong to the classification labels of multiple media through the classification layer; and combining the multiple classification vectors output by the classification layer to obtain the fingerprint media classification vector.

[0011] According to another embodiment of this application, a fingerprint detection device is provided, comprising: a first acquisition module, configured to acquire native domain information of a fingerprint to be detected, wherein the native domain information is raw data collected during fingerprint verification of the fingerprint to be detected; a first extraction module, configured to perform fingerprint feature extraction on the native domain information to obtain a fingerprint feature matrix of the fingerprint to be detected; a first classification module, configured to perform a classification operation on the fingerprint feature matrix through a trained media classification network to obtain a fingerprint media classification vector of the fingerprint to be detected, wherein the fingerprint media classification vector is used to represent the probability that the fingerprint to be detected belongs to multiple media types in vector form; and a first determination module, configured to determine the media type of the fingerprint to be detected based on the fingerprint media classification vector.

[0012] According to yet another embodiment of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0013] According to yet another embodiment of this application, a computer-readable storage medium is also provided, wherein a computer program is stored therein, and the computer program is configured to perform the steps in any of the above method embodiments when it is run.

[0014] According to yet another embodiment of this application, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0015] In this embodiment, the native domain information of the fingerprint to be detected is obtained, fingerprint feature extraction is performed on the native domain information to obtain a fingerprint feature matrix, and a trained media classification network is used to perform a classification operation on the fingerprint feature matrix to obtain a media classification vector. Finally, the media type of the fingerprint to be detected is analyzed based on the media classification vector. By using a trained media classification network to perform a classification operation on the fingerprint feature matrix, a more accurate and high-quality media classification vector can be generated, thereby improving the accuracy and efficiency of identifying the media type of the fingerprint to be detected and effectively detecting the type of fingerprint media. This achieves the technical effect of accurately and quickly identifying the type of fingerprint media, thus solving the problem of ineffective detection of fingerprint media type in related technologies. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the application environment of the fingerprint detection method according to the embodiments of this application;

[0017] Figure 2 This is a flowchart of a fingerprint detection method according to an embodiment of this application;

[0018] Figure 3 This is a schematic diagram of the structure of a media classification network according to an embodiment of this application;

[0019] Figure 4 This is a schematic diagram of the specific structure of the media classification network according to an embodiment of this application;

[0020] Figure 5 This is a visualization scatter plot according to an embodiment of this application;

[0021] Figure 6 This is a flowchart of constructing a CNN model according to an embodiment of this application;

[0022] Figure 7 This is a structural block diagram of a fingerprint detection device according to an embodiment of this application;

[0023] Figure 8This is a schematic diagram of the structure of a product for determining the fingerprint medium type according to an embodiment of this application;

[0024] Figure 9 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0025] The embodiments of this application will be described in detail below with reference to the accompanying drawings and examples.

[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0027] According to one aspect of the embodiments of this application, a fingerprint detection method is provided. Optionally, in this embodiment, the fingerprint detection method can be applied to, for example... Figure 1 The hardware environment shown consists of server 101 and terminal device 103. For example... Figure 1 As shown, server 101 is connected to terminal 103 via a network and can be used to provide services to terminal devices or applications installed on terminal devices. The applications can be video applications, instant messaging applications, browser applications, educational applications, game applications, etc. Database 105 can be set up on the server or independently of the server to provide data storage services for server 101, such as a game data storage server. The network mentioned above can include, but is not limited to, wired networks and wireless networks. The wired network includes local area networks, metropolitan area networks, and wide area networks. The wireless network includes Bluetooth, WIFI, and other networks that enable wireless communication. Terminal device 103 can be a terminal configured with an application, and can include, but is not limited to, at least one of the following: mobile phones (such as Android phones, iOS phones, etc.), laptops, tablets, handheld computers, MID (Mobile Internet Devices), PADs, desktop computers, smart TVs, smart voice interaction devices, smart home appliances, vehicle terminals, aircraft, virtual reality (VR) terminals, augmented reality (AR) terminals, mixed reality (MR) terminals, and other computer devices. The server mentioned above can be a single server, a server cluster composed of multiple servers, or a cloud server.

[0028] Combination Figure 1 As shown, the above fingerprint detection method can be implemented in terminal device 103 through the following steps:

[0029] S1, Obtain the native domain information of the fingerprint to be detected, wherein the native domain information is the raw data collected when performing fingerprint verification on the fingerprint to be detected;

[0030] S2, perform fingerprint feature extraction on the native domain information to obtain the fingerprint feature matrix of the fingerprint to be detected;

[0031] S3, the fingerprint feature matrix is ​​classified by the trained media classification network to obtain the fingerprint media classification vector of the fingerprint to be detected. The fingerprint media classification vector is used to represent the probability of the fingerprint to be detected belonging to multiple media types in the form of a vector.

[0032] S4. Determine the media type of the fingerprint to be detected based on the fingerprint media classification vector.

[0033] Optionally, in this embodiment, the above fingerprint detection method can also be implemented via a server, for example, in Figure 1 It is implemented in server 101 shown; or it is implemented jointly by the terminal device and the server.

[0034] It should be noted that the aforementioned servers can be independent physical servers, server clusters or distributed systems composed of multiple physical servers, or cloud servers providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. Terminals can be smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, etc., but are not limited to these. Terminals and servers can be directly or indirectly connected via wired or wireless communication, and this application does not impose any restrictions on this connection.

[0035] The above is merely an example, and this embodiment does not impose any specific limitations.

[0036] This embodiment provides a fingerprint detection method. Figure 2 This is a flowchart of a fingerprint detection method according to an embodiment of this application, such as... Figure 2 As shown, the process includes the following steps:

[0037] Step S202: Obtain the native domain information of the fingerprint to be detected, wherein the native domain information is the raw data collected when performing fingerprint verification on the fingerprint to be detected;

[0038] Optionally, the fingerprint to be detected in this embodiment can be a simulated fingerprint on the target medium. This embodiment can acquire the native domain information of the simulated fingerprint on the target medium using at least one of the following methods: Optical scanning: Scanning the simulated fingerprint on the target medium using an optical scanner to convert the fingerprint image into a digital signal. Capacitive scanning: Acquiring fingerprint information by detecting the capacitance change between the simulated fingerprint on the target medium and the sensor. Piezoelectric scanning: Acquiring fingerprint information by detecting the pressure change applied to the sensor by the target medium. Ultrasonic scanning: Acquiring fingerprint information of the simulated fingerprint on the target medium by emitting and receiving ultrasonic waves. Thermal imaging scanning: Acquiring fingerprint information by detecting temperature changes on the surface of the target medium. Radio frequency scanning: Acquiring fingerprint information of the simulated fingerprint on the target medium by emitting radio frequency signals and receiving the reflected signals. Multispectral scanning: Acquiring fingerprint information of the simulated fingerprint on the target medium by using light of different wavelengths.

[0039] Optionally, the target medium in this embodiment may be made of materials including, but not limited to, gel, polyurethane (PU), latex, silicone, acrylic, and resin. Simulating the fingerprint to be detected on the target medium refers to creating a copy on the target medium that has similar characteristics to the original fingerprint. For example, under the same environment, using different mediums such as gel, PU, ​​latex, silicone, acrylic, and resin, a forged fingerprint corresponding to a real finger can be manufactured.

[0040] Optionally, during the fingerprint acquisition process, raw data refers to the unprocessed fingerprint information obtained by the fingerprint acquisition device. Raw data includes, but is not limited to, the following information: fingerprint ridges, fingerprint orientation, fingerprint size, fingerprint image quality, acquisition time, acquisition device information, and other biometric information (e.g., finger thickness, temperature, etc.). For example, using an optical fingerprint scanner, an image of the fingerprint is acquired by scanning a fingerprint mounted on silicone. This image data constitutes the raw domain information. It captures features such as fingerprint texture, ridges, and sweat pores. Capacitive sensors acquire fingerprints by detecting changes in capacitance between the silicone finger and the sensor surface. These capacitance changes reflect the distribution of ridges and valleys in the fingerprint, constituting the raw domain information.

[0041] Step S204: Perform fingerprint feature extraction on the native domain information to obtain the fingerprint feature matrix of the fingerprint to be detected;

[0042] Optionally, this embodiment can perform fingerprint feature extraction on the native domain information in the following way to obtain a fingerprint feature matrix: perform signal conversion operation on the native domain information through a signal converter to obtain a set of digital signals of the native domain information; extract digital features from the set of digital signals to obtain a set of digital features of the native domain information; and determine the digital matrix of the native domain information as a set of digital features including the native domain information.

[0043] In this embodiment, a signal converter is a device that converts different types of signals into another type, such as conversion between analog and digital signals, or conversion between signals of different frequencies or encoding methods. Signal conversion of native domain information can be performed as follows: First, determine the type of native domain information—whether it is an analog signal, a digital signal, or other form of information. Then, select a suitable signal converter based on the type of native domain information. For example, if converting an analog signal to a digital signal, an analog-to-digital converter is needed. The relevant parameters of the converter, such as sampling rate, resolution, and filtering, are configured according to the characteristics and requirements of the analog signal. Next, the native domain information is input into the signal converter for signal conversion. After conversion, the signal converter outputs a set of digital signals, which are digital representations of the native domain information.

[0044] For example, a capacitive sensor is used to acquire an image of the fingerprint to be detected. This image contains detailed information such as the ridges and valleys of the fingerprint. A signal converter is used to perform signal conversion on the fingerprint image, including mathematical methods such as Fourier transform and wavelet transform, converting the image information into a set of digital signals. These digital signals contain the feature information of the fingerprint to be detected. Digital features are extracted from the converted digital signals. These features can be the ridge direction, ridge frequency, ridge curvature, etc. The extracted digital features are arranged in a certain order to form a digital matrix. This matrix contains all the key information of the fingerprint to be detected. Through the above methods, signal conversion, digital feature extraction, and digital matrix construction of the fingerprint to be detected can be achieved, providing a crucial information foundation for target medium type identification.

[0045] In this embodiment, the fingerprint feature matrix is ​​a digital matrix data obtained after signal-to-analog conversion using a sensor. Since it has not undergone preprocessing, the corresponding fingerprint path information may not be visually observable. However, it contains all the original, unfiltered information, making it more suitable for automatic feature extraction using deep learning techniques.

[0046] Step S206: Perform a classification operation on the fingerprint feature matrix through the trained media classification network to obtain the fingerprint media classification vector of the fingerprint to be detected. The fingerprint media classification vector is used to represent the probability of the fingerprint to be detected belonging to multiple media types in the form of a vector.

[0047] Optionally, the media classification network includes, but is not limited to, various types of neural networks, such as Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), and Graph Neural Networks (GNN). For example, when the media classification network is a CNN, the structure of the media classification network includes, but is not limited to, as shown in the example below. Figure 3 The network structure shown includes a feature extraction layer and a classification layer. The fingerprint feature matrix is ​​input into the feature extraction layer via a CNN network. The feature extraction layer outputs a feature vector of the fingerprint feature matrix. This feature vector is then input into the classification layer. The classification layer calculates the probability that the feature vector of the fingerprint feature matrix belongs to the classification label of multiple media and outputs a fingerprint media classification vector.

[0048] Optionally, the feature extraction layer may include multiple sub-network layers. For example, when the media classification network is a CNN network, the feature extraction layer consists of Conv layers (convolutional layers), ReLU activation layers, MaxPool max pooling layers, and GlobalAvgPool layers (global pooling layers). The Conv layers are used to extract the image features. The ReLU activation layer is a non-linear activation function used to introduce non-linearity into the neural network, enabling the network to learn and simulate more complex functions. The MaxPool max pooling layer reduces the spatial size of the features output by the convolutional layers, thereby reducing the number of parameters and computational complexity, while also making feature detection more robust. The GlobalAvgPool global pooling layer is a special pooling operation that performs average pooling on the entire feature set. The number and stacking of the Conv layers (convolutional layers), ReLU activation layers, MaxPool max pooling layers, and GlobalAvgPool layers in the feature extraction layer can be adjusted. For example, the feature extraction layer consists of Conv layers (convolutional layers), ReLU activation layers, MaxPool max pooling layers, and GlobalAvgPool layers (global pooling layers) in terms of quantity and stacking method, as follows: Figure 4As shown, this method is used to extract a one-dimensional depth feature vector from a fingerprint feature matrix (e.g., an ultrasonic fingerprint input image). For example, the fingerprint feature matrix is ​​input into a Conv layer, convolved using a 3x3 kernel, and the output matrix is ​​processed. A ReLU activation layer is applied to the output matrix of the convolutional layer. Then, a max-pooling operation is performed on the output of the ReLU activation layer using a max-pooling kernel in a MaxPool layer. Finally, a global average pooling operation is performed on the output of the max-pooling layer to obtain the feature vector of the fingerprint feature matrix.

[0049] Optionally, the classification layer may include multiple sub-network layers, for example, when the media classification network is a CNN network, such as... Figure 4 As shown, the classification layer consists of fully connected (FC) layers and softmax activation function layers, used to predict the probability scores of seven media from the deep feature vector. The FC layers linearly transform the features of the input data before providing them to subsequent layers for further processing. By stacking multiple fully connected layers, the neural network can learn more complex features and patterns. The softmax activation function is a function that transforms input data into a probability distribution. In multi-class classification problems, the softmax function is used in the output layer of the neural network, converting the model's output into the probability of each class.

[0050] Optionally, the number of elements in the output fingerprint medium classification vector is related to the number of media preset in the medium classification network. For example, when the classification layer includes FC, the FC output is a 1*N vector, where each bit corresponds to the probability of belonging to this category. The medium classification network outputs guide labels as 0-unknown, 1-gel, 2-PU, 3-latex, 4-silicone, 5-acrylic, 6-resin, where only the element value of the corresponding category is 1, and the element values ​​of other categories are 0. For example, the output vector corresponding to the gel category is [0, 1, 0, 0, 0, 0, 0].

[0051] Step S208: Determine the media type of the fingerprint to be detected based on the fingerprint media classification vector.

[0052] Optionally, the medium type of the fingerprint to be detected can be determined from the value corresponding to each element in the fingerprint medium classification vector. For example, when the fingerprint medium classification vector is [0, 1, 0, 0, 0, 0, 0], the medium type of the fingerprint to be detected can be determined to be gel.

[0053] Optionally, the fingerprint detection method provided in this embodiment can be applied to various scenarios that require fingerprint media recognition, including but not limited to scenarios of fingerprint media testing and scenarios of recognizing real and fake fingerprints.

[0054] Through the above steps, the native domain information of the fingerprint to be detected, simulated on the target medium, is obtained. Fingerprint feature extraction is performed on the native domain information to obtain a fingerprint feature matrix. Then, a trained medium classification network is used to perform a classification operation on the fingerprint feature matrix to obtain a fingerprint medium classification vector. Finally, the medium type of the fingerprint to be detected is analyzed based on the fingerprint medium classification vector. By using the trained medium classification network to perform classification operations on the fingerprint feature matrix, more accurate and high-quality fingerprint medium classification vectors can be generated, thereby improving the accuracy and efficiency of medium type identification and effectively detecting the type of fingerprint medium. This achieves the technical effect of accurately and quickly identifying the type of fingerprint medium, thus solving the problem of ineffective fingerprint medium type detection in related technologies.

[0055] In an exemplary embodiment, before performing a classification operation on the fingerprint feature matrix using a trained media classification network to obtain the fingerprint media classification vector of the fingerprint to be detected, the method further includes:

[0056] Step S302: Collect the native domain information of simulated fingerprints simulated on multiple media respectively to obtain a set of multiple simulated fingerprint native domain information. The set of multiple simulated fingerprint native domain information includes: multiple simulated fingerprint native domain information of the same medium obtained after collecting information multiple times from the simulated fingerprint of the same medium.

[0057] Optionally, this embodiment mainly focuses on the training process of the media classification network. Various media include, but are not limited to, media made of materials such as gel, PU, ​​latex, silicone, acrylic, and resin. For example, under the same environment, multiple forged fingerprints corresponding to genuine fingers can be manufactured using different media processes such as gel, PU, ​​latex, silicone, acrylic, and resin. The method for collecting the native domain information of the simulated fingerprints simulated on various media is the same as the method for collecting the native domain information of the fingerprint to be detected, and will not be repeated here.

[0058] Optionally, to enable the media classification network to learn more generalized feature representations, the training data should be as abundant as possible, and the amount of data for each medium should be as consistent as possible. To enhance the advantages of the classification scheme over previous manual feature modeling and extraction schemes, and to improve the media classification system's ability to handle differences in fingers, temperature, and system environment, the sampling for each medium should cover at least more fingers (e.g., no less than 20), temperature (e.g., temperature range covering -25°C to 55°C), and machine range (using at least 3 different sampling machines for sampling). If training from scratch is required, the training data specification should be more than 100,000.

[0059] Step S304: Perform fingerprint feature extraction on each of the simulated fingerprint native domain information included in the set of multiple simulated fingerprint native domain information to obtain a set of multiple simulated fingerprint feature matrices;

[0060] Step S306: Using the simulated fingerprint feature matrix set and the classification labels of the various media, train the media classification network to obtain the media classification network, wherein the classification labels are used to represent the media types of the various media.

[0061] Optionally, when collecting the fingerprint native domain data corresponding to each medium, the corresponding medium can be labeled with a classification tag. For example, gel is labeled as 1, PU as 2, latex as 3, silicone as 4, acrylic as 5, resin as 6, and unknown type as 0. If there are additional media to be classified, the tag value can be incremented and accumulated.

[0062] This embodiment, by collecting simulated fingerprints on multiple media and performing digital conversion, can collect more comprehensive information, thereby improving the accuracy of the recognition system. Training with multiple media helps enhance the model's generalization ability to different media features, enabling it to maintain high recognition accuracy even when facing unknown media. By training the media classification network using a set of digit matrices and classification labels, automated learning and recognition of media types are achieved, reducing manual intervention and improving work efficiency. Because the training process involves multiple media, the media classification network can adapt to different media types, enhancing the algorithm's applicability and flexibility. By collecting information multiple times from the same simulated fingerprint on the same medium, the resulting native domain information of multiple simulated fingerprints increases data richness, helping to improve the model's ability to recognize subtle differences.

[0063] In an exemplary embodiment, step S304 above involves performing fingerprint feature extraction on each of the simulated fingerprint native domain information included in the simulated fingerprint native domain information set to obtain a simulated fingerprint feature matrix set, including:

[0064] Step S3041: For each simulated fingerprint native domain information included in the simulated fingerprint native domain information set, perform the following operations to obtain the simulated fingerprint feature matrix of each simulated fingerprint native domain information, wherein the simulated fingerprint feature matrix set includes the simulated fingerprint feature matrix of each simulated fingerprint native domain information: perform signal conversion operation on the simulated fingerprint native domain information through a signal converter to obtain a set of digital signals of the simulated fingerprint native domain information; extract digital features from the set of digital signals to obtain a set of fingerprint features of the simulated fingerprint native domain information; determine the simulated fingerprint feature matrix of the simulated fingerprint native domain information as a set of fingerprint features including the simulated fingerprint native domain information.

[0065] Optionally, the signal conversion of the simulated fingerprint native domain information can be performed as follows: Determine the type of the simulated fingerprint native domain information—whether it is an analog signal, a digital signal, or other form of information. Select a suitable signal converter based on the type of the simulated fingerprint native domain information. For example, converting an analog signal to a digital signal might require an ADC. Configure the relevant parameters of the converter, such as sampling rate, resolution, and filtering, according to the characteristics and requirements of the signal. Input the simulated fingerprint native domain information into the signal converter for signal conversion. After conversion, the signal converter will output a set of digital signals, which are the digital representations of the simulated fingerprint native domain information.

[0066] For example, a capacitive sensor is used to acquire an image of the fingerprint to be detected. This image contains detailed information such as the ridges and valleys of the fingerprint. A signal converter is used to perform signal conversion on the fingerprint image, including mathematical methods such as Fourier transform and wavelet transform, converting the image information into a set of digital signals. These digital signals contain the feature information of the fingerprint to be detected. Digital features are extracted from the converted digital signals. These features can be the ridge direction, ridge frequency, ridge curvature, etc. The extracted digital features are arranged in a certain order to form a digital matrix. This matrix contains all the key information of the fingerprint to be detected.

[0067] This embodiment uses a signal converter to perform signal conversion and digital feature extraction on the native domain information, which can improve the efficiency of signal processing, enhance the readability of the signal, describe the characteristics of fingerprints more accurately, and facilitate machine learning.

[0068] In an exemplary embodiment, step S306 above, which uses the simulated fingerprint feature matrix set and classification labels for multiple media to train the pre-training media classification network to obtain the trained media classification network, includes:

[0069] Step S3061: Input the set of simulated fingerprint feature matrices into the media classification network to perform a training operation. The training operation includes: inputting the simulated fingerprint feature matrix into the feature extraction layer of the media classification network, outputting the feature vector of the simulated fingerprint feature matrix through the feature extraction layer, inputting the feature vector of the digital matrix into the classification layer of the media classification network, calculating the probability that the feature vector of the simulated fingerprint feature matrix belongs to the classification labels of multiple media through the classification layer, and outputting the media classification vector. The simulated fingerprint feature matrix is ​​the fingerprint feature matrix included in the set of digital matrices.

[0070] Optionally, in step S3061, the medium classification network includes, but is not limited to, various types of neural networks, such as CNN, RNN, LSTM, and GNN. For example, when the medium classification network is a CNN network, the structure of the CNN network includes, but is not limited to, as shown in the image below. Figure 3 The network structure shown includes a feature extraction layer and a classification layer. Each digit matrix in the digit matrix set is input into the feature extraction layer via a CNN network. The feature extraction layer outputs a feature vector for each digit matrix. This feature vector is then input into the classification layer, which calculates the probability that each digit matrix's feature vector belongs to a classification label for multiple media and outputs a media classification vector.

[0071] Optionally, the feature extraction layer may include multiple sub-network layers. For example, when the media classification network is a CNN network, the feature extraction layer consists of Conv layers (convolutional layers), ReLU activation layers, MaxPool max pooling layers, and GlobalAvgPool layers (global pooling layers). The Conv layers are used to extract the image features. The ReLU activation layer is a non-linear activation function used to introduce non-linearity into the neural network, enabling the network to learn and simulate more complex functions. The MaxPool max pooling layer reduces the spatial size of the features output by the convolutional layers, thereby reducing the number of parameters and computational complexity, while also making feature detection more robust. The GlobalAvgPool global pooling layer is a special pooling operation that performs average pooling on the entire feature set. The number and stacking of the Conv layers (convolutional layers), ReLU activation layers, MaxPool max pooling layers, and GlobalAvgPool layers in the feature extraction layer can be adjusted. For example, the feature extraction layer consists of Conv layers (convolutional layers), ReLU activation layers, MaxPool max pooling layers, and GlobalAvgPool layers (global pooling layers) in terms of quantity and stacking method, as follows: Figure 4 As shown, it is used to extract a one-dimensional depth feature vector from a digital matrix (e.g., ultrasonic fingerprint Raw domain whole map data).

[0072] Optionally, the classification layer may include multiple sub-network layers, for example, when the media classification network is a CNN network, such as... Figure 4As shown, the classification layer consists of fully connected (FC) layers and softmax activation function layers, used to predict the probability scores of seven media from the deep feature vector. The FC layers linearly transform the features of the input data before providing them to subsequent layers for further processing. By stacking multiple fully connected layers, the neural network can learn more complex features and patterns. The softmax activation function is a function that transforms input data into a probability distribution. In multi-class classification problems, the softmax function is typically used in the output layer of the neural network, converting the model's output into the probability of each class.

[0073] Step S3062: Determine the loss value output by the loss function of the media classification network based on the media classification vector and the classification label of the target media, wherein the target media is the media corresponding to the digital matrix.

[0074] Optionally, the number of elements in the output medium classification vector is related to the number of media in the medium classification network. For example, when the classification layer includes a fully connected (FC) layer, the FC output is a 1*N vector, where each bit corresponds to the probability of belonging to that category. The medium classification network outputs the following labels: 0-unknown, 1-gel, 2-PU, 3-latex, 4-silicone, 5-acrylic, 6-resin. For ease of training, these are converted to one-hot encoding, where each category corresponds to a one-hot encoded vector, where only the elements of the corresponding category have a value of 1, and the elements of other categories have a value of 0. For example, the gel category corresponds to [0, 1, 0, 0, 0, 0, 0].

[0075] Step S3064: If the above loss value meets the preset training termination condition, the training is terminated, and the above media classification network is obtained.

[0076] Optionally, before training the media classification network, it is also necessary to configure the training settings for the media classification network. For example, the optimizer is selected as AdamW, and the initial learning rate is 1×10⁻⁶. -3 The learning rate scheduling strategy uses cosine annealing to dynamically adjust the learning rate, and training is performed for 100 epochs. During training, parameters need to be continuously updated. For example, ultrasonic fingerprint images are randomly selected from the digit matrix set and input into the media classification network to obtain the probability distribution of each media category. The predicted probability distribution is then... With respect to the true distribution Y={y i For each i ∈ [0, 6], calculate the Cross Entropy (CE) loss, update the network parameters using the backpropagation algorithm, and gradually reduce the loss function. The CE loss calculation method in step S3062 includes:

[0077] Step S3066: If the above loss value does not meet the preset training termination condition, adjust the values ​​of the parameters of the above media classification network to reduce the loss value output by the above loss function of the above media classification network in the next training.

[0078] Optionally, the preset training termination condition in step S3066 includes: whether the maximum number of iterations (epochs) has been reached; if yes, optimization stops and network training is complete; otherwise, the above steps are repeated until the maximum number of iterations (epochs) is reached. Optionally, after the media classification network is trained, the obtained CNN classification network can be directly applied to the fingerprint media detection task. In practical applications, the press frame data is divided into... Figure 4 The input is a CNN classification network, and the output is the prediction score for each of the five media. The range is (0, 1), and the maximum value is... The category corresponding to the value subscript is the media category predicted by the CNN classification network for that input.

[0079] Optionally, in step S3066, when the loss value does not meet the preset training termination condition, the parameters of the media classification network can be adjusted using the following methods to reduce the output value of the loss function: adjusting the learning rate, increasing the network depth or width, adjusting the optimizer, using regularization techniques, using data augmentation, adjusting the loss function, and adjusting hyperparameters. It should be noted that adjusting the parameters of the media classification network requires trial and error based on the specific task and dataset. By comprehensively considering multiple methods, the output value of the loss function can be effectively reduced, and classification performance improved.

[0080] This embodiment trains a media classification network by inputting a set of digit matrices, effectively extracting features from the digit matrices and improving classification accuracy. By inputting the feature vectors of the digit matrices into the classification layer, the probability of the media's classification label can be calculated, providing a basis for subsequent classification decisions. Based on the media classification vector and the target media's classification label, the loss value output by the media classification network's loss function can be determined, providing feedback information for subsequent training. When the loss value meets the preset training termination condition, training can end, resulting in a trained media classification network, improving training efficiency. When the loss value does not meet the preset training termination condition, adjusting the parameters of the media classification network can reduce the loss value in the next training iteration, improving training flexibility and adaptability. Through multiple iterative training iterations, the performance of the media classification network can be continuously optimized, improving classification accuracy and robustness. In summary, the above steps, including inputting a set of digit matrices, extracting features, calculating loss values, and adjusting parameters, can effectively train a high-performance media classification network, improving the accuracy and efficiency of classification tasks.

[0081] In an exemplary embodiment, after inputting the set of simulated fingerprint feature matrices into the media classification network to perform a training operation, the method further includes:

[0082] Step S402: Extract the feature vector of each simulated fingerprint feature matrix included in the set of simulated fingerprint feature matrices output by the feature extraction layer;

[0083] Step S404: Map the feature vectors of each simulated fingerprint feature matrix to a coordinate space of a preset dimension, and perform dimensionality reduction processing on the feature vectors of each simulated fingerprint feature matrix in the coordinate space to obtain feature vectors of the target dimension.

[0084] Step S406: Perform a visualization operation on the feature vectors of the target dimension of each simulated fingerprint feature matrix to display them on the target interface.

[0085] Optionally, this embodiment is applicable to scenarios where the effectiveness of a media classification network is verified during its training. In this embodiment, to effectively demonstrate the effectiveness of the constructed media classification network in fingerprint media classification, nonlinear dimensionality reduction technology (T-distributed stochastic neighbor embedding, or T-SNE for short) can be used to visualize the deep learning features of different media. The T-SNE-based dimensionality reduction visualization is then used for material feature analysis. T-SNE is used to embed high-dimensional data into a low-dimensional space (the preset dimension coordinate space can be two-dimensional or three-dimensional) to facilitate the visualization of deep learning features, thereby revealing the similarities and differences between data points.

[0086] For example, in the qualitative analysis phase, the TSNE algorithm maps the eigenvectors of each numerical matrix to a two-dimensional coordinate space, allowing for visualization of the TSNE algorithm's output. For instance, as... Figure 5 As shown, the visualization distribution of the 7-dimensional deep learning features after dimensionality reduction can be achieved through methods such as scatter plots. The position of each data point represents its coordinates in the low-dimensional space, and different colors or shapes can be used to distinguish different media. The deep learning features of fingerprint data from six different media, including gel, silicone, and resin, exhibit strong separability and aggregability based on their physical properties, demonstrating the feasibility of using deep learning technology to guide the learning of the implicit physical properties in the native domains of different media.

[0087] For example, in the quantitative analysis phase, the quantitative results of media detection are compared in Table 1. The classification accuracy confusion matrix for five different pressable media is a classic statistical method for evaluating the performance and accuracy of multi-classification tasks. The values ​​on the diagonal of the confusion matrix represent the probability that each media category is accurately classified into the correct result. These values ​​can be used to calculate classification accuracy and other relevant metrics. The classification accuracy confusion matrix is ​​used to evaluate the performance of the classification network. For a classification task with five different pressable media, the confusion matrix will be a 5x5 matrix, where each row represents the actual category and each column represents the predicted category. For example, there are five pressable media: A, B, C, D, and E. The confusion matrix will include the following cells: True Positives (TP): The number of samples predicted as belonging to this category and actually belonging to this category. False Positives (FP): The number of samples predicted as belonging to this category but not actually belonging to this category. False Negatives (FN): The number of samples predicted as not belonging to this category but actually belonging to this category. True Negatives (TN): For binary classification problems, this can be the number of samples predicted as not belonging to this category and actually belonging to this category. Actual categories: AB CDE; Predicted categories: AB ED A. The confusion matrix can look like this:

[0088] “AB CDE

[0089] A TP1 0 0 0 0FN1

[0090] B 0TP2 0FN2 0

[0091] C 0 0TP3 0 0

[0092] D 0FN3 0TP4 0

[0093] E FN4 0 0 0TP5”;

[0094] Here, `TP1` represents the number of samples that are actually class A and are correctly predicted as class A, and `FN1` represents the number of samples that are actually class A but are incorrectly predicted as other classes.

[0095] In this embodiment, the confusion matrix provides a comprehensive view of classification performance, including details of correct and incorrect classifications. The confusion matrix allows for a more complete understanding of the classifier's performance, enabling model improvement through adjusting model parameters or using different classification algorithms.

[0096] Table 1:

[0097]

[0098] This embodiment analyzes how the features of different fingerprint media are distributed in low-dimensional space by observing the visualization results. If the features of fingerprint media form different clusters in the visualization, it means that the data of different media are distinguishable in the feature space. Based on the visualization results, it may be necessary to return to the network training or feature extraction steps to improve the network performance or select more suitable features. Qualitative analysis of the correlation and differences of deep learning features of different media is achieved. It can also effectively realize deep anti-counterfeiting feature extraction based on native domain features. Materials of the same medium achieve effective aggregation in the dimensionality reduction space, while different media exhibit strong separability even in two-dimensional space.

[0099] In an exemplary embodiment, step S204 involves performing fingerprint feature extraction on the native domain information to obtain the fingerprint feature matrix of the fingerprint to be detected, including:

[0100] Step S2041: Construct a native domain information set, wherein the native domain information set includes multiple native domain information of the fingerprint to be detected acquired within the current time period, or the native domain information set includes multiple historical native domain information acquired within a historical time period and the native domain information acquired within the current time period;

[0101] Optionally, the multiple native domain information of the fingerprint to be detected acquired within the current time period refers to pressing the same fingerprint medium multiple times on the fingerprint scanner to obtain more comprehensive and clearer fingerprint information. This increases the number of fingerprint feature points collected, thereby improving recognition accuracy. Continuous pressing reduces the number of times data needs to be repeated due to acquisition failure, thus improving overall acquisition efficiency. The fingerprints corresponding to multiple historical native domain information acquired within a historical time period can be the same fingerprint or different fingerprints corresponding to the native domain information acquired within the current time period.

[0102] Step S2042: Convert all native domain information included in the native domain information set into digital signals using a signal converter to obtain multiple sets of digital signals;

[0103] Optionally, in this embodiment, the process of converting the native domain information into digital signals and extracting digital feature signals through a signal converter is the same as the above-described processing method for the native domain information, and will not be repeated here.

[0104] Step S2043: Extract digital features from each group of digital signals to obtain fingerprint features of each group of native domain information;

[0105] Step S2044: According to the signal acquisition channel of the target sensor that acquires the native domain information included in the native domain information set, generate a matrix including fingerprint features of each set of native domain information to obtain multiple sets of fingerprint feature matrices.

[0106] Optionally, generating a digital matrix including the digital features of each set of digital signals according to the signal acquisition channels of the target sensor that acquires the native domain information included in the aforementioned native domain information set is achieved, but is not limited to, through the following methods: Defining matrix dimensions: Determining the number of rows and columns of the generated digital matrix. Typically, the number of rows represents the time series or the number of samples, and the number of columns represents different signal channels. Initializing the matrix: Creating an empty matrix with appropriate dimensions to store the signal data acquired from each channel. Signal acquisition: Acquiring signals from each channel using the sensor. Each channel may correspond to a different type of signal, such as temperature, pressure, fingerprint patterns, etc. Data formatting: Formatting the acquired signal data into numerical values, which can be digital sample values ​​of analog signals. Filling the matrix: Filling the signal data of each channel sequentially into the corresponding columns of the matrix. For example, if the data of the first channel is temperature, it will be filled into the first column of the matrix; if the second channel is pressure, it will be filled into the second column, and so on.

[0107] Step S2045: Combine the multiple sets of fingerprint feature matrices to obtain the fingerprint feature matrix to be detected.

[0108] Optionally, combining the above-mentioned multiple sets of numerical matrices can be done in ways including, but not limited to, merging multiple sets of numerical matrices into a single numerical matrix. For example, merging two matrices into a larger matrix. Specifically, this includes: vertical merging (row-wise merging): placing two matrices vertically to form a new matrix whose number of rows is the sum of the number of rows in the two matrices, while the number of columns remains unchanged. Horizontal merging (column-wise merging): placing two matrices side-by-side to form a new matrix whose number of columns is the sum of the number of columns in the two matrices, while the number of rows remains unchanged. If the dimensions of the two matrices allow, they can also be placed in specific positions within a larger matrix to form a block matrix.

[0109] This embodiment identifies fingerprint media by collecting multimodal fingerprint data, which can capture fingerprint features from different angles. This reduces the uncertainty that may exist with single fingerprint features, thereby improving the accuracy of fingerprint media identification.

[0110] In an exemplary embodiment, step S204 above, performing a classification operation on the fingerprint feature matrix using a trained media classification network to obtain a fingerprint media classification vector for the fingerprint to be detected, includes:

[0111] Step S2041: Input the fingerprint feature matrix to be detected into the feature extraction layer included in the medium classification network, and extract the feature vector of each fingerprint feature matrix included in the fingerprint feature matrix to be detected through the feature extraction layer.

[0112] Step S2042: The feature vector of each fingerprint feature matrix output by the feature extraction layer is input into the classification layer included in the medium classification network, and the probability that the feature vector of each fingerprint feature matrix belongs to the classification label of multiple media is calculated by the classification layer.

[0113] Step S2043: Combine the multiple classification vectors output by the above classification layer to obtain the above fingerprint medium classification vector.

[0114] Optionally, the fingerprint feature matrix is ​​input into the feature extraction layer included in the aforementioned media classification network. The feature extraction layer extracts the digit matrix from each layer of the fingerprint feature matrix, obtaining the feature vector for each layer. Similarly, the classification layer calculates the probability of each feature vector belonging to each category. The probability of each category is converted into a vector distribution, ultimately resulting in multiple sets of media vectors. Finally, the media vector corresponding to the current time period at the bottom layer is determined as the media vector of the fingerprint medium to be detected.

[0115] This embodiment uses combined data modalities as input to the media classification network. Different data modalities may capture different features of the fingerprint medium to be detected. Combining these modalities can improve the performance of the classification layer. Furthermore, combining multiple modalities can increase the diversity of fingerprint features, thereby reducing the risk of overfitting. In summary, this can significantly improve the performance and adaptability of the media classification network in media detection tasks.

[0116] The present application will be further described below with reference to specific embodiments:

[0117] This specific embodiment, based on the differences in the physical characteristics of different media in simulated fingerprints (including but not limited to capacitance impedance, relative depth of valleys and ridges, acoustic impedance, optical polarization characteristics, etc.), uses a deep learning network to build a media detection model, collects data for training and optimization, and constructs a deep learning network CNN model. This model is used as an example to illustrate the detection and classification of fingerprint media in an under-display fingerprint system. Figure 6 The specific embodiment described herein includes the following steps:

[0118] S601 collects training data, which is a set of simulated fingerprint native domain information. For example, under the same environment, using different media processes such as gel, PU, ​​latex, silicone, acrylic, and resin, a forged fingerprint corresponding to a real finger is created. The native fingerprint domain data corresponding to each medium is collected.

[0119] S602, Build a CNN model. For example, build a CNN model like... Figure 4 The CNN model shown takes ultrasonic fingerprint raw image data as input and outputs the medium classification result. The network architecture mainly consists of a feature extraction layer and a classification layer. The feature extraction layer is composed of stacked Conv layers, ReLU activation layers, MaxPool layers, and GlobalAvgPool layers, used to extract a one-dimensional depth feature vector from the ultrasonic fingerprint input image. The classification layer consists of FC and Softmax activation function layers, used to predict the probability scores of seven media from the depth feature vector. The CNN model outputs guide labels: 0-unknown, 1-gel, 2-PU, 3-latex, 4-silicone, 5-acrylic, 6-resin. For ease of training, these are converted to one-hot encoding, with each category corresponding to a one-hot encoded vector where only elements of the corresponding category have a value of 1, and elements of other categories have a value of 0.

[0120] S603 trains and optimizes the CNN model. It uses the training set built by S601 to train and optimize the CNN model built by S602, making it suitable for media detection tasks.

[0121] S604, forged fingerprint detection. After training, the S603 CNN model can be directly applied to the fingerprint medium detection task. In practical applications, the pressed frame data is input into the CNN model, and the CNN model outputs the prediction scores for each of the five mediums. The range is (0,1), and the maximum value is... The subscript value corresponds to the media category predicted by the model for the input image. The detection range includes, but is not limited to, the detection of simulated fingerprints in materials such as gel, PU, ​​latex, silicone, acrylic, resin, and unknown materials.

[0122] This specific embodiment takes an ultrasonic under-display fingerprint system as an example to demonstrate the differences in physical characteristics hidden in the features extracted by deep learning technology for different media. A deep learning convolutional neural network model is constructed for media detection, which reduces the computational complexity of the algorithm without relying on additional complex sensors, and ensures high accuracy of media detection.

[0123] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0124] This embodiment also provides a fingerprint detection device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0125] Figure 7 This is a structural block diagram of a fingerprint detection device according to an embodiment of this application, such as... Figure 7 As shown, the device includes:

[0126] The first acquisition module 72 is used to acquire the native domain information of the fingerprint to be detected, wherein the native domain information is the raw data collected when performing fingerprint verification on the fingerprint to be detected.

[0127] The first extraction module 74 is used to perform fingerprint feature extraction on the native domain information to obtain the fingerprint feature matrix of the fingerprint to be detected;

[0128] The first classification module 76 is used to perform a classification operation on the fingerprint feature matrix through the trained media classification network to obtain the fingerprint media classification vector of the fingerprint to be detected, wherein the fingerprint media classification vector is used to represent the probability that the fingerprint to be detected belongs to multiple media types in the form of a vector.

[0129] The first determining module 78 is used to determine the media type of the fingerprint to be detected based on the fingerprint media classification vector.

[0130] In an exemplary embodiment, the above apparatus further includes: a first acquisition unit, configured to acquire simulated fingerprint native domain information of simulated fingerprints simulated on multiple media before performing a classification operation on the fingerprint feature matrix through a trained media classification network to obtain the fingerprint media classification vector of the fingerprint to be detected, thereby obtaining a set of simulated fingerprint native domain information, wherein the simulated fingerprint native domain information set includes a portion of the simulated fingerprint native domain information obtained by acquiring multiple information from the same simulated fingerprint simulated on the same media; a first execution unit, configured to perform fingerprint feature extraction on each of the simulated fingerprint native domain information included in the set of simulated fingerprint native domain information to obtain a set of simulated fingerprint feature matrices; and a first training unit, configured to train the media classification network using the set of simulated fingerprint feature matrices and the classification labels of the multiple media to obtain the media classification network, wherein the classification labels are used to represent the media types of the multiple media.

[0131] In an exemplary embodiment, the first execution unit includes: a first execution submodule, configured to perform the following operations for each simulated fingerprint native domain information included in the simulated fingerprint native domain information set: obtaining a simulated fingerprint feature matrix for each simulated fingerprint native domain information, wherein the simulated fingerprint feature matrix set includes the simulated fingerprint feature matrices of each simulated fingerprint native domain information; performing a signal conversion operation on the simulated fingerprint native domain information using a signal converter to obtain a set of digital signals for the simulated fingerprint native domain information; extracting digital features from the set of digital signals to obtain a set of fingerprint features for the simulated fingerprint native domain information; and determining the simulated fingerprint feature matrix of the simulated fingerprint native domain information as a set of fingerprint features including the simulated fingerprint native domain information.

[0132] In an exemplary embodiment, the first training unit includes: a first input submodule, configured to input the set of simulated fingerprint feature matrices into the media classification network to perform a training operation, wherein the training operation includes: inputting the simulated fingerprint feature matrix into a feature extraction layer of the media classification network, outputting a feature vector of the simulated fingerprint feature matrix through the feature extraction layer, inputting the feature vector of the digit matrix into a classification layer of the media classification network, calculating the probability that the feature vector of the simulated fingerprint feature matrix belongs to the classification labels of multiple media through the classification layer, and outputting a media classification vector, wherein the simulated fingerprint feature matrix is ​​the... The digital matrix set includes a fingerprint feature matrix; a first determining submodule is used to determine the loss value output by the loss function of the media classification network based on the media classification vector and the classification label of the target media, wherein the target media is the media corresponding to the digital matrix; a first processing submodule is used to end the training and obtain the media classification network when the loss value meets a preset training termination condition; a first adjusting submodule is used to adjust the parameter values ​​of the media classification network to reduce the loss value output by the loss function of the media classification network in the next training if the loss value does not meet the preset training termination condition.

[0133] In an exemplary embodiment, the above apparatus further includes: a first extraction module, configured to extract feature vectors of each simulated fingerprint feature matrix included in the simulated fingerprint feature matrix set output by the feature extraction layer after inputting the simulated fingerprint feature matrix set into the medium classification network for training operations; a first mapping module, configured to map the feature vectors of each simulated fingerprint feature matrix to a coordinate space of a preset dimension, and perform dimensionality reduction processing on the feature vectors of each simulated fingerprint feature matrix in the coordinate space to obtain feature vectors of a target dimension; and a first operation module, configured to perform visualization operations on the feature vectors of the target dimension of each simulated fingerprint feature matrix to display them on a target interface.

[0134] In an exemplary embodiment, the first extraction module includes: a first construction unit, configured to construct a native domain information set, wherein the native domain information set includes multiple native domain information of the fingerprint to be detected acquired within the current time period, or the native domain information set includes multiple historical native domain information acquired within a historical time period and the native domain information acquired within the current time period; a first conversion unit, configured to convert all the native domain information included in the native domain information set into digital signals using a signal converter to obtain multiple sets of digital signals; a first extraction unit, configured to extract digital features from each set of digital signals to obtain fingerprint features of each set of native domain information; a first generation unit, configured to generate a matrix including fingerprint features of each set of native domain information according to the signal acquisition channel of the target sensor that acquires the native domain information included in the native domain information set, to obtain multiple sets of fingerprint feature matrices; and a first combination unit, configured to combine the multiple sets of fingerprint feature matrices to obtain the fingerprint feature matrix to be detected.

[0135] In an exemplary embodiment, the first classification module includes: a first input unit, configured to input the fingerprint feature matrix to be detected into a feature extraction layer included in the media classification network, and extract feature vectors of each group of fingerprint feature matrices included in the fingerprint feature matrix to be detected through the feature extraction layer; a second input unit, configured to input the feature vectors of each group of fingerprint feature matrices output by the feature extraction layer into a classification layer included in the media classification network, and calculate the probability that the feature vectors of each group of fingerprint feature matrices belong to the classification labels of multiple media through the classification layer; and a second combination unit, configured to combine multiple classification vectors output by the classification layer to obtain the fingerprint media classification vector.

[0136] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0137] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0138] According to one aspect of this application, a computer program product is provided, the computer program product comprising a computer program / instructions, the computer program / instructions including instructions for performing a process. Figure 8The program code for the method shown. In such an embodiment, the computer program can be downloaded and installed from a network via communication section 809, and / or installed from removable medium 811. When the computer program is executed by central processing unit 801, it performs various functions provided in the embodiments of this application.

[0139] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0140] Figure 8 A schematic block diagram of a computer system architecture for implementing an electronic device according to embodiments of the present application is shown.

[0141] It should be noted that, Figure 8 The computer system 800 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0142] like Figure 8 As shown, the computer system 800 includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 802 or programs loaded from storage section 808 into random access memory (RAM). The random access memory 803 also stores various programs and data required for system operation. The CPU 801, ROM 802, and RAM 803 are interconnected via a bus 804. An input / output interface 805 (I / O interface) is also connected to the bus 804.

[0143] The following components are connected to the input / output interface 805: an input section 1006 including a keyboard, mouse, etc.; an output section 807 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a local area network card, modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the input / output interface 805 as needed. A removable medium 811, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 810 as needed so that computer programs read from it can be installed into the storage section 808 as needed.

[0144] Specifically, according to embodiments of this application, the processes described in the various method flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 809, and / or installed from removable medium 811. When the computer program is executed by central processing unit 801, it performs various functions defined in the system of this application.

[0145] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described fingerprint detection method is also provided. This electronic device may be... Figure 1 The terminal device or server shown. This embodiment uses this electronic device as an example for illustration. Figure 9 As shown, the electronic device includes a memory 902 and a processor 904. The memory 902 stores a computer program, and the processor 904 is configured to execute the steps of any of the above method embodiments through the computer program.

[0146] Optionally, in this embodiment, the aforementioned electronic device may be located in at least one of a plurality of network devices in a computer network.

[0147] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0148] S1, Obtain the native domain information of the fingerprint to be detected, wherein the native domain information is the raw data collected when performing fingerprint verification on the fingerprint to be detected;

[0149] S2, perform fingerprint feature extraction on the native domain information to obtain the fingerprint feature matrix of the fingerprint to be detected;

[0150] S3, the fingerprint feature matrix is ​​classified by the trained media classification network to obtain the fingerprint media classification vector of the fingerprint to be detected. The fingerprint media classification vector is used to represent the probability of the fingerprint to be detected belonging to multiple media types in the form of a vector.

[0151] S4. Determine the media type of the fingerprint to be detected based on the fingerprint media classification vector.

[0152] Alternatively, as those skilled in the art will understand, Figure 9 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones (such as Android phones, iOS phones, etc.), tablets, PDAs, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 9 This does not limit the structure of the aforementioned electronic devices or electronic equipment. For example, electronic devices or electronic equipment may also include components that are more... Figure 9 The more or fewer components shown (such as network interfaces, etc.), or having the same Figure 9 The different configurations shown.

[0153] The memory 902 can be used to store software programs and modules, such as the program instructions / modules corresponding to the fingerprint detection method and device in this embodiment. The processor 904 executes various functional applications and data processing by running the software programs and modules stored in the memory 902, thereby realizing the fingerprint detection method described above. The memory 902 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 902 may further include memory remotely located relative to the processor 904, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. Specifically, the memory 902 may be used, but is not limited to, to store information such as the code of an item. As an example, such as... Figure 9 As shown, the memory 902 may include, but is not limited to, the first acquisition module 72, the first extraction module 74, the first classification module 76, and the first determination module 78 of the fingerprint detection device. Furthermore, it may include, but is not limited to, other module units of the fingerprint detection device, which will not be elaborated in this example.

[0154] Optionally, the transmission device 906 described above is used to receive or send data via a network. Specific examples of the network described above may include wired networks and wireless networks. In one example, the transmission device 906 includes a Network Interface Controller (NIC), which can be connected to other network devices and routers via a network cable to communicate with the Internet or a local area network. In another example, the transmission device 906 is a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0155] In addition, the aforementioned electronic device also includes: a display 908 for displaying the aforementioned initial code; and a connection bus 910 for connecting the various module components in the aforementioned electronic device.

[0156] In other embodiments, the aforementioned terminal device or server can be a node in a distributed system, wherein the distributed system can be a blockchain system, which is a distributed system formed by connecting multiple nodes through network communication. The nodes can form a peer-to-peer network, and any form of computing device, such as a server, terminal, or other electronic device, can become a node in the blockchain system by joining this peer-to-peer network.

[0157] According to one aspect of this application, a computer-readable storage medium is provided, from which a processor of a computer device reads computer instructions, and the processor executes the computer instructions, causing the computer device to perform the fingerprint detection method provided in various alternative implementations of the fingerprint medium type determination aspect described above.

[0158] Optionally, in this embodiment, the computer-readable storage medium may be configured to store a computer program for performing the following steps:

[0159] S1, Obtain the native domain information of the fingerprint to be detected, wherein the native domain information is the raw data collected when performing fingerprint verification on the fingerprint to be detected;

[0160] S2, perform fingerprint feature extraction on the native domain information to obtain the fingerprint feature matrix of the fingerprint to be detected;

[0161] S3, the fingerprint feature matrix is ​​classified by the trained media classification network to obtain the fingerprint media classification vector of the fingerprint to be detected. The fingerprint media classification vector is used to represent the probability of the fingerprint to be detected belonging to multiple media types in the form of a vector.

[0162] S4. Determine the media type of the fingerprint to be detected based on the fingerprint media classification vector.

[0163] Optionally, in this embodiment, those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0164] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0165] If the integrated units in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in the aforementioned computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0166] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0167] In the several embodiments provided in this application, it should be understood that the disclosed client can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between units or modules, and may be electrical or other forms.

[0168] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0169] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0170] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A fingerprint detection method, characterized in that, include: Obtain the native domain information of the fingerprint to be detected, wherein the native domain information is the raw data collected when performing fingerprint verification on the fingerprint to be detected; Fingerprint feature extraction is performed on the native domain information to obtain the fingerprint feature matrix of the fingerprint to be detected; The fingerprint feature matrix is ​​classified by the trained media classification network to obtain the fingerprint media classification vector of the fingerprint to be detected. The fingerprint media classification vector is used to represent the probability that the fingerprint to be detected belongs to multiple media types in the form of a vector. The media type of the fingerprint to be detected is determined based on the fingerprint media classification vector; The step of performing fingerprint feature extraction on the native domain information to obtain the fingerprint feature matrix of the fingerprint to be detected includes: Construct a native domain information set, wherein the native domain information set includes multiple historical native domain information obtained within a historical time period and the native domain information obtained within the current time period; The native domain information included in the native domain information set is converted into digital signals by a signal converter to obtain multiple sets of digital signals; Digital features are extracted from each set of digital signals to obtain fingerprint features of each set of native domain information, wherein the fingerprint features include at least one of the ridge direction, ridge frequency, and ridge curvature of the fingerprint. According to the signal acquisition channel of the target sensor that acquires the native domain information included in the native domain information set, a matrix including fingerprint features of each set of native domain information is generated to obtain multiple sets of fingerprint feature matrices. The multiple sets of fingerprint feature matrices are combined to obtain the fingerprint feature matrix to be detected.

2. The method according to claim 1, characterized in that, Before performing a classification operation on the fingerprint feature matrix using a trained media classification network to obtain the fingerprint media classification vector of the fingerprint to be detected, the method further includes: The native domain information of simulated fingerprints simulated on multiple media is collected to obtain multiple sets of native domain information of simulated fingerprints. The multiple sets of native domain information of simulated fingerprints include: the native domain information of multiple simulated fingerprints on the same medium obtained after multiple information collections of simulated fingerprints on the same medium. Fingerprint feature extraction is performed on each of the simulated fingerprint native domain information included in the set of multiple simulated fingerprint native domain information to obtain a set of multiple simulated fingerprint feature matrices; The media classification network is trained using the simulated fingerprint feature matrix set and the classification labels of the various media to obtain the media classification network, wherein the classification labels are used to represent the media types of the various media.

3. The method according to claim 2, characterized in that, Using the simulated fingerprint feature matrix set and the classification labels of the various media, the media classification network before training is trained to obtain the trained media classification network, including: The simulated fingerprint feature matrix set is input into the media classification network to perform a training operation, wherein the training operation includes: inputting the simulated fingerprint feature matrix into the feature extraction layer of the media classification network, outputting the feature vector of the simulated fingerprint feature matrix through the feature extraction layer, inputting the feature vector of the simulated fingerprint feature matrix into the classification layer of the media classification network, calculating the probability that the feature vector of the simulated fingerprint feature matrix belongs to the classification labels of multiple media through the classification layer, and outputting the media classification vector; The loss value output by the loss function of the media classification network is determined based on the media classification vector and the media classification label. If the loss value meets the preset training termination condition, the training ends, and the trained media classification network is obtained. If the loss value does not meet the preset training termination condition, the parameter values ​​of the media classification network are adjusted to reduce the loss value output by the loss function of the media classification network in the next training.

4. The method according to claim 3, characterized in that, After inputting the simulated fingerprint feature matrix set into the media classification network to perform training operations, the method further includes: Extract the feature vector of each simulated fingerprint feature matrix included in the set of simulated fingerprint feature matrices output by the feature extraction layer; The feature vectors of each simulated fingerprint feature matrix are mapped to a coordinate space of a preset dimension, and the feature vectors of each simulated fingerprint feature matrix are subjected to dimensionality reduction processing in the coordinate space to obtain feature vectors of the target dimension. Visualize the feature vectors of the target dimension of each simulated fingerprint feature matrix to display them on the target interface.

5. The method according to claim 1, characterized in that, The fingerprint feature matrix is ​​classified using a trained media classification network to obtain a fingerprint media classification vector for the fingerprint to be detected, including: The fingerprint feature matrix to be detected is input into the feature extraction layer included in the medium classification network, and the feature vector of each group of fingerprint feature matrices included in the fingerprint feature matrix to be detected is extracted by the feature extraction layer. The feature extraction layer inputs the feature vector of each fingerprint feature matrix into the classification layer in the medium classification network, and the classification layer calculates the probability that the feature vector of each fingerprint feature matrix belongs to the classification label of multiple media. The fingerprint medium classification vector is obtained by combining the multiple classification vectors output by the classification layer.

6. A fingerprint detection device, characterized in that, include: The first acquisition module is used to acquire the native domain information of the fingerprint to be detected, wherein the native domain information is the raw data collected when performing fingerprint verification on the fingerprint to be detected; The first extraction module is used to perform fingerprint feature extraction on the native domain information to obtain the fingerprint feature matrix of the fingerprint to be detected; The first classification module is used to perform a classification operation on the fingerprint feature matrix through the trained media classification network to obtain the fingerprint media classification vector of the fingerprint to be detected, wherein the fingerprint media classification vector is used to represent the probability that the fingerprint to be detected belongs to multiple media types in the form of a vector. The first determining module is used to determine the media type of the fingerprint to be detected based on the fingerprint media classification vector; The step of performing fingerprint feature extraction on the native domain information to obtain the fingerprint feature matrix of the fingerprint to be detected includes: Construct a native domain information set, wherein the native domain information set includes multiple historical native domain information obtained within a historical time period and the native domain information obtained within the current time period; The native domain information included in the native domain information set is converted into digital signals by a signal converter to obtain multiple sets of digital signals; Digital features are extracted from each set of digital signals to obtain fingerprint features of each set of native domain information, wherein the fingerprint features include at least one of the ridge direction, ridge frequency, and ridge curvature of the fingerprint. According to the signal acquisition channel of the target sensor that acquires the native domain information included in the native domain information set, a matrix including fingerprint features of each set of native domain information is generated to obtain multiple sets of fingerprint feature matrices. The multiple sets of fingerprint feature matrices are combined to obtain the fingerprint feature matrix to be detected.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the method described in any one of claims 1 to 5.

9. An electronic 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, it implements the steps of the method described in any one of claims 1 to 5.

Images