TW202133041A - A method, a device for extracting features of fingerprint images and computer-readable storage medium - Google Patents

Abstract

The present disclosure relates to a method, a device for extracting features of fingerprint images, and a computer-readable storage medium. The method include: obtaining fingerprint images, pre-processing the fingerprint images to remove noise information in the fingerprint images, and extracting two-value characteristics of the fingerprint images by using a trained preset convolution neural network model, in which the preset convolution neural network includes a continuous activation function that simulates a step function, and matching the two-value feature of the fingerprint images with a two-value feature of a pre-stored fingerprint image to get a matching result. The present disclosure makes two-value processing on the characteristics of the output of the preset convolution neural network model by using the activation function of the preset convolution neural network model to, which improves the matching efficiency of the two-value features of the fingerprint image.

Description

Fingerprint image feature extraction method, device and computer readable storage medium

The invention relates to the field of fingerprint image recognition, in particular to a fingerprint image feature extraction method, device and computer readable storage medium.

The current neural networks used for feature extraction of fingerprint images all output floating-point descriptors, and their feature comparison speed is slow, which cannot meet the requirements of high-speed platform applications. For example, image matching algorithms based on feature points such as sift and orb are used to extract features from fingerprint images. There are many secondary/third-level feature points of the image. If floating-point feature matching is still used, the matching efficiency of the image Will be extremely low.

In view of the above, it is necessary to propose a fingerprint image feature extraction method, device, and computer-readable storage medium to extract the binary features of the fingerprint image, so as to improve the matching efficiency of the fingerprint image.

The first aspect of the present application provides a fingerprint image feature extraction method. The method includes the steps:

Get fingerprint image;

Preprocessing the fingerprint image to remove noise in the fingerprint image;

Extracting the binary features of the fingerprint image by using a trained preset convolutional neural network model, wherein the preset convolutional neural network includes a continuous activation function with an analogous step function function; and

The binary feature of the fingerprint image is matched with the binary feature of the pre-stored fingerprint image to obtain a matching result.

Preferably, the preset convolutional neural network includes an input layer, a plurality of convolutional layers for feature extraction, a fully connected layer, an activation function, and an output layer. The input layer, convolutional layer, fully connected layer, and activation The function and output layer are connected in turn.

Preferably, the extracting the binary features of the fingerprint image by using the trained preset convolutional neural network model includes:

Provide an input channel for the fingerprint image through the input layer; train and extract the features of the fingerprint image through the convolutional layer; integrate the features trained and extracted by each convolutional layer through the fully connected layer Binarize the features output by the fully connected layer through the activation function and segment the results of the binarization process according to a preset threshold to obtain the binary features of the fingerprint image; and through the output The layer outputs binary features.

，所述預設閾值為1/2，

為使用者設定的常數，所述啟動函數對所述全連接層輸出的特徵進行二值化處理及將二值化處理的結果依據預設閾值進行分割得到所述指紋圖像的二值特徵包括：Preferably, the start function

, The preset threshold is 1/2,

It is a constant set by the user, the activation function performs binarization processing on the features output by the fully connected layer, and divides the result of the binarization processing according to a preset threshold to obtain the binary features of the fingerprint image including :

對所述全連接層輸出的特徵進行二值化處理及將二值化處理的結果依據1/2進行分割得到所述指紋圖像的二值特徵。pass through

Binarization is performed on the features output by the fully connected layer, and the result of the binarization is divided according to 1/2 to obtain the binary features of the fingerprint image.

，所述預設閾值為0，

為使用者設定的常數，所述啟動函數對所述全連接層輸出的特徵進行二值化處理及將二值化處理的結果依據預設閾值進行分割得到所述指紋圖像的二值特徵包括：Preferably, the start function is

, The preset threshold is 0,

It is a constant set by the user, the activation function performs binarization processing on the features output by the fully connected layer, and divides the result of the binarization processing according to a preset threshold to obtain the binary features of the fingerprint image including :

對所述全連接層輸出的特徵進行二值化處理及將二值化處理的結果依據0進行分割得到所述指紋圖像的二值特徵。By the start function

Binarization is performed on the features output by the fully connected layer, and the result of the binarization is divided according to 0 to obtain the binary features of the fingerprint image.

Preferably, the matching the binary feature of the fingerprint image with the binary feature of the pre-stored fingerprint image to obtain the matching result includes:

Calculating the Hamming distance between the binary feature of the fingerprint image and the feature value of the pre-stored fingerprint image; and

When the Hamming distance value is less than or equal to the preset distance value, the fingerprint image is successfully matched with the pre-stored fingerprint image.

Preferably, the calculating the Hamming distance value between the binary feature of the fingerprint image and the feature value of the pre-stored fingerprint image includes:

Perform an XOR operation on the binary feature of the fingerprint image and the feature value of the prestored fingerprint image to obtain an XOR result, and the number of the XOR results is the fingerprint image and the prestored fingerprint image The Hamming distance value.

Preferably, preprocessing the fingerprint image to remove noise in the fingerprint image includes:

Performing fingerprint segmentation on the fingerprint image to remove the background area in the fingerprint image;

Performing fingerprint enhancement on the fingerprint image to remove cross-connections, interruptions, and blurred parts in the fingerprint image; and

Refinement processing is performed on the fingerprint image to delete edge primitives of Chinese lines in the fingerprint image.

A second aspect of the present application provides a fingerprint image feature extraction device, the device includes:

The acquisition module is used to acquire fingerprint images;

A preprocessing module for preprocessing the fingerprint image to remove noise in the fingerprint image;

The feature extraction module is used to extract the binary features of the fingerprint image by using a trained preset convolutional neural network model, wherein the preset convolutional neural network includes a function with an analogous step function And continuous activation function; and

The matching module is used to match the binary feature of the fingerprint image with the binary feature of the pre-stored fingerprint image to obtain a matching result.

The third party of this application provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the fingerprint image feature extraction method is realized.

In this case, binarization processing is performed on the features output by the preset convolutional neural network model through the activation function of the preset convolutional neural network model to obtain the binary features of the fingerprint image, and the fingerprint image The binary feature is matched with the binary feature of the pre-stored fingerprint image to obtain the matching result, which improves the matching efficiency of the binary feature of the fingerprint image.

In order to be able to understand the above objectives, features and advantages of the present invention more clearly, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments of the application and the features in the embodiments can be combined with each other if there is no conflict.

In the following description, many specific details are explained in order to fully understand the present invention. The described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments, and are not intended to limit the present invention.

Preferably, the fingerprint image feature extraction method of the present invention is applied to one or more electronic devices. The electronic device is a device that can automatically perform numerical calculation and/or information processing in accordance with pre-set or stored instructions. Its hardware includes, but is not limited to, a microprocessor and a dedicated integrated circuit (ASIC) , Field-Programmable Gate Array (FPGA), Digital Signal Processor (DSP), embedded devices, etc.

The electronic device may be a computing device such as a desktop computer, a notebook computer, a tablet computer, and a cloud server. The device can interact with the user through a keyboard, a mouse, a remote control, a touch pad, or a voice control device.

Fig. 1 is a flowchart of a feature extraction method of a fingerprint image in an embodiment of the present invention. According to different needs, the order of the steps in the flowchart can be changed, and some steps can be omitted.

Referring to FIG. 1, the feature extraction method of the fingerprint image specifically includes the following steps:

Step S101: Obtain a fingerprint image.

In this embodiment, the fingerprint image is collected by the fingerprint collector. In this embodiment, the fingerprint collector is an optical fingerprint collector, a thermal fingerprint reader or a biometric radio frequency fingerprint reader.

Step S102: Perform preprocessing on the fingerprint image to remove noise in the fingerprint image.

In this embodiment, the preprocessing the fingerprint image to remove noise in the fingerprint image includes: performing fingerprint segmentation on the fingerprint image to remove the background area in the fingerprint image. In a specific implementation, a threshold-based image segmentation algorithm may be used to perform fingerprint segmentation on the fingerprint image to remove the background area in the fingerprint image.

In one embodiment, the preprocessing the fingerprint image to remove noise in the fingerprint image includes: performing fingerprint enhancement on the fingerprint image to remove cross-connections in the fingerprint image. , Interruption points and obscure parts. In a specific implementation, the fingerprint image is segmented by the fingerprint enhancement algorithm of STFT (Short Time Fourier Transform), so as to remove the background area in the fingerprint image.

In one embodiment, the preprocessing of the fingerprint image to remove noise in the fingerprint image includes: thinning the fingerprint image to delete edge primitives of Chinese lines in the fingerprint image. In a specific embodiment, the fingerprint image is refined by the OPTA refinement algorithm to delete the edge primitives of the Chinese line of the fingerprint image.

In an embodiment, the preprocessing the fingerprint image to remove noise in the fingerprint image includes: performing fingerprint segmentation on the fingerprint image to remove the background area in the fingerprint image ; Perform fingerprint enhancement on the fingerprint image to remove the cross-connections, interruptions and fuzzy parts in the fingerprint image; and refine the fingerprint image to delete the edge map of the Chinese line of the fingerprint image Yuan.

Step S103: Extract the binary features of the fingerprint image by using a trained preset convolutional neural network model, wherein the preset convolutional neural network includes a function of an analogous step function and continuous activation function.

In this embodiment, the preset convolutional neural network includes an input layer, a plurality of convolutional layers for feature extraction, a fully connected layer, an activation function, and an output layer. The input layer, convolutional layer, fully connected layer, activation function and output layer are connected in sequence. In this embodiment, the "extracting the binary features of the fingerprint image using the trained preset convolutional neural network model" includes: providing an input channel for the fingerprint image through the input layer; The buildup layer trains and extracts the features of the fingerprint image; the fully connected layer integrates the features extracted by the training of each convolutional layer; the binarization process is performed on the features output by the fully connected layer through the start function, and The result of the binarization process is segmented according to a preset threshold to obtain the binary feature of the fingerprint image; the binary feature is output through the output layer.

，所述預設閾值為1/2，

為用戶設定的常數。所述“啟動函數對全連接層輸出的特徵進行二值化處理及將二值化處理的結果依據預設閾值進行分割得到指紋圖像的二值特徵”包括：通過所述啟動函數

全連接層輸出的特徵進行二值化處理及將二值化處理的結果依據1/2進行分割得到所述指紋圖像的二值特徵。本實施方式中，當逐漸增大設定的常數

時，所述啟動函數

，接近於階躍函數

。In this embodiment, the startup function is

, The preset threshold is 1/2,

A constant set by the user. The “starting function binarizes the features output by the fully connected layer and segmenting the result of the binarization processing according to a preset threshold to obtain the binary features of the fingerprint image” includes: through the start function

The feature output by the fully connected layer is binarized, and the result of the binarization process is divided according to 1/2 to obtain the binary feature of the fingerprint image. In this embodiment, when the set constant is gradually increased

When the start function

, Close to the step function

.

，所述預設閾值為0，

為用戶設定的常數。所述“啟動函數對全連接層輸出的特徵進行二值化處理及將二值化處理的結果依據預設閾值進行分割得到指紋圖像的二值特徵”包括：通過所述啟動函數

對全連接層輸出的特徵進行二值化處理及將二值化處理的結果依據0進行分割得到所述指紋圖像的二值特徵。本實施方式中，當逐漸增大設定的常數

時，所述啟動函數

接近於階躍函數

。In another embodiment, the start function is

, The preset threshold is 0,

A constant set by the user. The “starting function binarizes the features output by the fully connected layer and segmenting the result of the binarization processing according to a preset threshold to obtain the binary features of the fingerprint image” includes: through the start function

Binarization is performed on the features output by the fully connected layer, and the result of the binarization is divided according to 0 to obtain the binary features of the fingerprint image. In this embodiment, when the set constant is gradually increased

When the start function

Close to step function

.

In this embodiment, the method further includes: training the preset convolutional neural network model. Specifically, when training the preset convolutional neural network model, a preset number of fingerprint image samples can be stored in the server 2, and the fingerprint image samples can be classified by the user; for example, Prepare 1000 fingerprint image samples, and then classify these 1000 fingerprint image samples according to the users to whom these fingerprint image samples belong, and perform binary feature calibration on each classified two fingerprint image sample. When the classification of the prepared preset number of fingerprint image samples is completed, the preset convolutional neural network model can be used as the classification model at this time, and the face fingerprint image samples can be input to the preset as training samples. Set the convolutional neural network model for training, and adjust the weight parameters of the connections between nodes on each base layer of the preset convolutional neural network model according to the classification results output by the preset convolutional neural network model . After the preset convolutional neural network model is trained based on the input training samples after each adjustment, the accuracy of the output classification result will be gradually improved compared with the classification result calibrated by the user. At the same time, the user can set an accuracy threshold in advance. In the continuous adjustment process, if the classification result output by the preset convolutional neural network model is compared with the classification result calibrated by the user, the accuracy reaches the preset accuracy. After the accuracy threshold, at this time, the weight parameters connected between the base nodes in the preset convolutional neural network model are all optimal weight parameters, and it can be considered that the preset convolutional neural network model has been trained. In this embodiment, after the training of the preset convolutional neural network model is completed, the fingerprint image input into the preset convolutional neural network model is performed through the trained preset convolutional neural network model. Extraction of binary features.

Step S104: Match the binary feature of the fingerprint image with the binary feature of the pre-stored fingerprint image to obtain a matching result.

In this embodiment, the "matching the binary feature of the fingerprint image with the binary feature of the pre-stored fingerprint image to obtain a matching result" includes: calculating the binary feature of the fingerprint image and the pre-stored fingerprint image When the Hamming distance value is less than or equal to the preset distance value, the fingerprint image and the pre-stored fingerprint image are matched successfully.

In this embodiment, the "calculating the Hamming distance value between the binary feature of the fingerprint image and the feature value of the pre-stored fingerprint image" includes: comparing the binary feature of the fingerprint image with the pre-stored fingerprint image The feature value is XORed to obtain XOR results, and the number of XOR results is the Hamming distance value between the fingerprint image and the pre-stored fingerprint image.

In this embodiment, the feature output by the fully connected layer is binarized by the activation function of the preset convolutional neural network model, and the result of the binarization process is segmented according to a preset threshold to obtain a fingerprint image. For the binary feature of the image, the binary feature of the fingerprint image is matched with the binary feature of the pre-stored fingerprint image to obtain a matching result, which improves the matching efficiency of the binary feature of the fingerprint image.

FIG. 2 is a structural diagram of a fingerprint image feature extraction device 40 in an embodiment of the present invention.

In some embodiments, the feature extraction and display device 40 of the fingerprint image runs in an electronic device. The fingerprint image feature extraction device 40 may include a plurality of functional modules composed of code segments. The code of each program segment in the fingerprint image feature extraction device 40 can be stored in the memory and executed by at least one processor to perform the feature extraction function of the fingerprint image.

In this embodiment, the fingerprint image feature extraction device 40 can be divided into multiple functional modules according to the functions it performs. Referring to FIG. 2, the fingerprint image feature extraction device 40 may include an acquisition module 401, a preprocessing module 402, a feature extraction module 403 and a matching module 404. The module referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, which are stored in the memory. In some embodiments, the functions of each module will be described in detail in subsequent embodiments.

The acquiring module 401 acquires a fingerprint image.

In this embodiment, the acquisition module 401 collects fingerprint images through a fingerprint collector. In this embodiment, the fingerprint collector is an optical fingerprint collector, a thermal fingerprint reader or a biometric radio frequency fingerprint reader.

The preprocessing module 402 preprocesses the fingerprint image to remove noise in the fingerprint image.

In this embodiment, the preprocessing module 402 performs fingerprint segmentation on the fingerprint image to remove the background area in the fingerprint image. In a specific embodiment, the preprocessing module 402 may use a threshold-based image segmentation algorithm to perform fingerprint segmentation on the fingerprint image to remove the background area in the fingerprint image.

In one embodiment, the pre-processing module 402 performs fingerprint enhancement on the fingerprint image to remove cross-connections, interruptions, and blurred parts in the fingerprint image. In a specific embodiment, the pre-processing module 402 performs fingerprint segmentation on the fingerprint image through the STFT (Short Time Fourier Transform) fingerprint enhancement algorithm, so as to remove the background area in the fingerprint image.

In one embodiment, the preprocessing module 402 performs a thinning process on the fingerprint image to delete the edge primitives of the Chinese line of the fingerprint image. In a specific embodiment, the pre-processing module 402 performs refinement processing on the fingerprint image through the OPTA refinement algorithm to delete the edge primitives of the Chinese line of the fingerprint image.

In one embodiment, the preprocessing module 402 performs fingerprint segmentation on the fingerprint image to remove the background area in the fingerprint image; performs fingerprint enhancement on the fingerprint image to remove the fingerprint image Cross-connects, interrupted points and fuzzy parts in the middle; and refine the fingerprint image to delete the edge primitives of the Chinese line of the fingerprint image.

The feature extraction module 403 uses a trained preset convolutional neural network model to extract the binary features of the fingerprint image, wherein the preset convolutional neural network includes a function with an analogous step function And continuous start function.

In this embodiment, the preset convolutional neural network includes an input layer, a plurality of convolutional layers for feature extraction, a fully connected layer, an activation function, and an output layer. The input layer, convolutional layer, fully connected layer, activation function and output layer are connected in sequence. In this embodiment, the feature extraction module 403 provides an input channel for the fingerprint image through the input layer; trains and extracts the features of the fingerprint image through the convolutional layer; Integrating the features extracted by training; binarizing the features output by the fully connected layer through the activation function and segmenting the results of the binarization processing according to a preset threshold to obtain the binary features of the fingerprint image; by The output layer outputs binary features.

，所述預設閾值為1/2，

為用戶設定的常數。所述特徵提取模組403通過所述啟動函數

對全連接層輸出的特徵進行二值化處理及將二值化處理的結果依據1/2進行分割得到所述指紋圖像的二特徵。本實施方式中，當逐漸增大設定的常數

時，所述

接近於階躍函數

。In this embodiment, the startup function is

, The preset threshold is 1/2,

A constant set by the user. The feature extraction module 403 uses the activation function

Binarization is performed on the features output by the fully connected layer, and the result of the binarization is divided according to 1/2 to obtain the binary features of the fingerprint image. In this embodiment, when the set constant is gradually increased

When the

Close to step function

.

，所述預設閾值為0，

為用戶設定的常數,例如

為大於1的整數。所述特徵提取模組403通過所述啟動函數

對全連接層輸出的特徵進行二值化處理及將二值化處理的結果依據0進行分割得到所述指紋圖像的二值特徵。本實施方式中，當逐漸增大設定的常數

時，所述

接近於階躍函數

。In another embodiment, the start function is

, The preset threshold is 0,

Constants set by the user, for example

Is an integer greater than 1. The feature extraction module 403 uses the activation function

Binarization is performed on the features output by the fully connected layer, and the result of the binarization is divided according to 0 to obtain the binary features of the fingerprint image. In this embodiment, when the set constant is gradually increased

When the

Close to step function

.

In this embodiment, the feature extraction module 403 is also used to train the preset convolutional neural network model. Specifically, when training the preset convolutional neural network model, a preset number of fingerprint image samples can be stored in the server 2, and the fingerprint image samples can be classified by the user; for example, Prepare 1000 fingerprint image samples, and then classify these 1000 fingerprint image samples according to the users to whom these fingerprint image samples belong, and perform binary feature calibration on each classified two fingerprint image sample. When the classification of the prepared preset number of fingerprint image samples is completed, the preset convolutional neural network model can be used as the classification model at this time, and the face fingerprint image samples can be input to the preset as training samples. Set the convolutional neural network model for training, and adjust the weight parameters of the connections between nodes on each base layer of the preset convolutional neural network model according to the classification results output by the preset convolutional neural network model . After the preset convolutional neural network model is trained based on the input training samples after each adjustment, the accuracy of the output classification result will be gradually improved compared with the classification result calibrated by the user. At the same time, the user can set an accuracy threshold in advance. In the continuous adjustment process, if the classification result output by the preset convolutional neural network model is compared with the classification result calibrated by the user, the accuracy reaches the preset accuracy. After the accuracy threshold, at this time, the weight parameters connected between the base nodes in the preset convolutional neural network model are all optimal weight parameters, and it can be considered that the preset convolutional neural network model has been trained. In this embodiment, after the training of the preset convolutional neural network model is completed, the fingerprint image input into the preset convolutional neural network model is performed through the trained preset convolutional neural network model. Extraction of binary features.

The matching module 404 matches the binary feature of the fingerprint image with the binary feature of the pre-stored fingerprint image to obtain a matching result.

In this embodiment, the matching module 404 calculates the Hamming distance value between the binary feature of the fingerprint image and the feature value of the pre-stored fingerprint image, and when the Hamming distance value is less than or equal to the preset distance value, Then the fingerprint image is successfully matched with the pre-stored fingerprint image.

In this embodiment, the matching module 404 is used to perform an XOR operation on the binary feature of the fingerprint image and the feature value of the prestored fingerprint image to obtain an XOR result, and the number of XOR results is determined by the number of XOR results. The Hamming distance between the fingerprint image and the pre-stored fingerprint image.

In this embodiment, the matching module 404 binarizes the features output by the fully connected layer through the startup function of the preset convolutional neural network model, and performs binarization processing on the result of the binarization process according to a preset threshold. The binary feature of the fingerprint image can be obtained by segmentation, and the binary feature of the fingerprint image is matched with the binary feature of the pre-stored fingerprint image to obtain a matching result, thus improving the matching efficiency of the binary feature of the fingerprint image.

FIG. 3 is a schematic diagram of the electronic device 6 in an embodiment of the present invention.

The electronic device 6 includes a memory 61, a processor 62, and a computer program 63 stored in the memory 61 and running on the processor 62. The processor 62 implements the steps in the embodiment of the fingerprint image feature extraction method when the computer program 63 is executed, for example, steps S101 to S104 shown in FIG. 1. Alternatively, when the processor 62 executes the computer program 63, the functions of the modules/units in the embodiment of the fingerprint image feature extraction device are realized, such as the modules 401 to 404 in FIG. 2.

Exemplarily, the computer program 63 may be divided into one or more modules/units, and the one or more modules/units are stored in the memory 61 and executed by the processor 62 , To complete the present invention. The one or more modules/units may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program 63 in the electronic device 6. For example, the computer program 63 can be divided into the acquisition module 401, the preprocessing module 402, the feature extraction module 403, and the matching module 404 in FIG. 2.

The electronic device 6 may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server. Those skilled in the art can understand that the schematic diagram is only an example of the electronic device 6 and does not constitute a limitation on the electronic device 6. It may include more or less components than those shown in the figure, or a combination of certain components, or different components. Components, for example, the electronic device 6 may also include input and output devices, network access devices, bus bars, and the like.

The so-called processor 62 may be a central processing unit (Central Processing Unit, CPU), other general-purpose processors, digital signal processors (Digital Signal Processors, DSP), and dedicated integrated circuits (Application Specific Integrated Circuits, ASICs). ), off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor 62 can also be any conventional processor, etc. The processor 62 is the control center of the electronic device 6, which uses various interfaces and lines to connect the entire electronic device 6 Various parts.

The memory 61 can be used to store the computer programs 63 and/or modules/units, and the processor 62 runs or executes the computer programs and/or modules/units stored in the memory 61, and The data stored in the memory 61 is called to realize various functions of the computer and electronic device 6. The memory 61 may mainly include a storage program area and a storage data area, where the storage program area can store an operating system, an application program required by at least one function (such as a sound playback function, an image playback function, etc.), etc.; The area can store data (such as audio data, phone book, etc.) created based on the use of the electronic device 6 and so on. In addition, the memory 61 may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart memory card (Smart Media Card, SMC), and secure digital (Secure Digital, SD) card, flash memory card (Flash Card), at least one magnetic disk memory device, flash memory device, or other volatile solid-state memory device.

If the integrated module/unit of the electronic device 6 is realized in the form of a software function module and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiments and methods, and can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the foregoing method embodiments can be realized. Wherein, the computer program includes computer program code, and the computer program code may be in the form of original program code, object code, executable file, or some intermediate forms. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read- Only Memory), Random Access Memory (RAM, Random Access Memory), electrical carrier signal, telecommunications signal, and software distribution media, etc. It should be noted that the content contained in the computer-readable medium can be appropriately added or deleted according to the requirements of the legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to the legislation and patent practice, the computer-readable medium Does not include electrical carrier signals and telecommunication signals.

In the several embodiments provided by the present invention, it should be understood that the disclosed electronic device and method may be implemented in other ways. For example, the electronic device embodiments described above are only illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

In addition, the functional modules in the various embodiments of the present invention may be integrated in the same processing module, or each module may exist alone physically, or two or more modules may be integrated in the same module. The above-mentioned integrated modules can be implemented either in the form of hardware or in the form of hardware plus software functional modules.

For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description, and therefore it is intended to fall within the claims. All changes within the meaning and scope of the equivalent elements of are included in the present invention. Any reference signs in the claims should not be regarded as limiting the claims involved. In addition, it is obvious that the word "include" does not exclude other modules or steps, and the singular does not exclude the plural. Multiple modules or electronic devices stated in the claims of an electronic device can also be implemented by the same module or electronic device through software or hardware. Words such as first and second are used to denote names, but do not denote any specific order.

The series of detailed descriptions listed above are only specific descriptions of feasible implementations of the present invention. They are not intended to limit the scope of protection of the present invention. Any equivalent implementations or implementations made without departing from the technical spirit of the present invention All changes should be included in the protection scope of the present invention.

40: Feature extraction device for fingerprint images 401: Get Module 402: preprocessing module 403: Feature Extraction Module 404: matching module 6: Electronic equipment 61: Memory 62: processor 63: computer program S101~S104: steps

Fig. 1 is a flowchart of a feature extraction method of a fingerprint image in an embodiment of the present invention.

Fig. 2 is a structural diagram of a fingerprint image feature extraction device in an embodiment of the present invention.

Fig. 3 is a schematic diagram of the electronic device of the present invention.

S101~S104: steps

Claims (10)

A fingerprint image feature extraction method, wherein the method includes the steps: Get fingerprint image; Preprocessing the fingerprint image to remove noise in the fingerprint image; Using a trained preset convolutional neural network model to extract the binary features of the fingerprint image, wherein the preset convolutional neural network includes a continuous starting function with an analogous step function function; and The binary feature of the fingerprint image is matched with the binary feature of the pre-stored fingerprint image to obtain a matching result. The fingerprint image feature extraction method according to claim 1, wherein the preset convolutional neural network includes an input layer, a plurality of convolutional layers for feature extraction, a fully connected layer, an activation function, and an output layer , The input layer, convolutional layer, fully connected layer, activation function and output layer are connected in sequence. The fingerprint image feature extraction method according to claim 2, wherein the extracting the binary features of the fingerprint image by using a trained preset convolutional neural network model includes: Provide an input channel for the fingerprint image through the input layer; train and extract the features of the fingerprint image through the convolutional layer; integrate the features trained and extracted by each convolutional layer through the fully connected layer Binarize the features output by the fully connected layer through the activation function and segment the results of the binarization process according to a preset threshold to obtain the binary features of the fingerprint image; and through the output The layer outputs the binary feature. The fingerprint image feature extraction method according to claim 3, wherein the activation function is

, The preset threshold is 1/2,

It is a constant set by the user, the activation function performs binarization processing on the features output by the fully connected layer, and divides the result of the binarization processing according to a preset threshold to obtain the binary features of the fingerprint image including : pass through

Binarization is performed on the features output by the fully connected layer, and the result of the binarization is divided according to 1/2 to obtain the binary features of the fingerprint image. The fingerprint image feature extraction method according to claim 3, wherein the activation function is

, The preset threshold is 0,

It is a constant set by the user, the activation function performs binarization processing on the features output by the fully connected layer, and divides the result of the binarization processing according to a preset threshold to obtain the binary features of the fingerprint image including : Through the start function

Binarization is performed on the features output by the fully connected layer, and the result of the binarization is divided according to 0 to obtain the binary features of the fingerprint image. The fingerprint image feature extraction method according to claim 1, wherein the matching the binary feature of the fingerprint image with the binary feature of the pre-stored fingerprint image to obtain the matching result includes: Calculating the Hamming distance between the binary feature of the fingerprint image and the feature value of the pre-stored fingerprint image; and When the Hamming distance value is less than or equal to the preset distance value, the fingerprint image is successfully matched with the pre-stored fingerprint image. The fingerprint image feature extraction method according to claim 6, wherein the calculating the Hamming distance value between the binary feature of the fingerprint image and the feature value of the prestored fingerprint image includes: Perform an XOR operation on the binary feature of the fingerprint image and the feature value of the prestored fingerprint image to obtain an XOR result, and the number of the XOR results is the fingerprint image and the prestored fingerprint image The Hamming distance value. The fingerprint image feature extraction method according to claim 1, wherein, preprocessing the fingerprint image to remove noise in the fingerprint image includes: Performing fingerprint segmentation on the fingerprint image to remove the background area in the fingerprint image; Performing fingerprint enhancement on the fingerprint image to remove cross-connections, interruptions, and blurred parts in the fingerprint image; and Refinement processing is performed on the fingerprint image to delete edge primitives of Chinese lines in the fingerprint image. A fingerprint image feature extraction device, wherein the device includes: The acquisition module is used to acquire fingerprint images; A preprocessing module for preprocessing the fingerprint image to remove noise in the fingerprint image; The feature extraction module is used to extract the binary features of the fingerprint image by using a trained preset convolutional neural network model, wherein the preset convolutional neural network includes a function with an analogous step function And continuous activation function; and The matching module is used to match the binary feature of the fingerprint image with the binary feature of the pre-stored fingerprint image to obtain a matching result. A computer-readable storage medium having a computer program stored thereon, wherein: the computer program is executed by a processor to implement the fingerprint image feature extraction method according to any one of claim items 1-8.

Images