Fingerprint image expansion method, system, electronic device and storage medium

Abstract

The present application relates to a fingerprint image expansion method, system, electronic device and storage medium, wherein the fingerprint image expansion method includes the following steps: obtaining a first fingerprint image; extracting the vector field and frequency field of the first fingerprint image; expanding the first fingerprint image based on the frequency field to obtain an expanded image; correcting the expanded image based on the vector field to obtain a corrected image; splicing the first fingerprint image and the corrected image to obtain a second fingerprint image, and outputting the second fingerprint image. The present application has the effects of fast expansion speed, good continuity of the expanded image relative to the original image, does not increase the false recognition rate, and increases the size of the effective area for extracting fingerprint feature points, thereby improving the pass rate during fingerprint authentication. It is achieved without upgrading the CPU speed and memory, and effectively improving the pass rate applied to embedded fingerprint processors.

Description

Technical Field

[0001] The present application relates to the field of fingerprint recognition, and in particular to a fingerprint image expansion method, system, electronic device and storage medium. Background Art

[0002] Nowadays, fingerprint recognition has been widely used in various fields. Fingerprint recognition is performed by comparing the feature points of the fingerprint image. Feature points usually refer to points with unique shapes and positions in the fingerprint image, such as bifurcation points, end points, short lines, ring points, etc. When using the fingerprint algorithm to extract feature points, an area of ​​8*8 pixels around the feature point is required as the feature information of the feature point, so 8 pixels must be subtracted from the top, bottom, left, and right sides of the fingerprint sensor.

[0003] For fingerprint sensors in consumer electronics, the information collection area is becoming smaller and smaller to adapt to the continuous innovation of products. For example, for fingerprint sensors used on the side of mobile phones, the size of the fingerprint sensor usually matches the thickness of the mobile phone, and the size is 32*100 pixels. The effective area that can actually extract fingerprint feature points is only 16*84 pixels in size. For another example, the size of the fingerprint unlocking button on a computer is 64*80 pixels in size, and the effective area that can actually extract fingerprint feature points is only 48*64 pixels in size. The fingerprint recognition rate is related to the fingerprint information that can be collected. The smaller the size of the collected fingerprint image, the less available information, so the fewer feature points can be extracted, and the fingerprint recognition pass rate will be relatively lower.

[0004] At present, there are two main solutions in the related technology to improve the recognition pass rate by expanding the useful information of the fingerprint image. One is to set a certain weight for each edge pixel directly according to the edge information of the fingerprint image, and then calculate the value of the expanded pixel according to the edge pixel and its corresponding weight. However, the expanded image obtained by the weight method is very different from the texture direction and the width of the fingerprint in the fingerprint image, resulting in too large an error in the extracted feature point information, which will greatly increase the false recognition rate of the fingerprint, and because most of the expanded information is invalid, the improvement in the pass rate is also very limited; another way is to conduct deep learning through a large number of fingerprint libraries, train models such as convolutional neural networks (CNN), so that the fingerprint posture of the expanded area can be automatically estimated through a large number of convolution operations based on the fingerprint image. However, if image expansion is to be achieved through deep learning, the processor needs to be equipped with an NPU and a large memory to support a large number of convolution operations, which cannot be applied in embedded applications. Summary of the invention

[0005] Without increasing the CPU speed and memory, in order to improve the pass rate of existing fingerprint processors used in embedded systems and without increasing the false recognition rate, the present application provides a fingerprint image expansion method, system, electronic device and storage medium. The direction and shape of fingerprint images have certain rules, which we summarize into two important indicators: image vector field and frequency field. Through these indicators, the four sides of the image are expanded to achieve the purpose of increasing the effective area size for extracting fingerprint feature points.

[0006] In the first aspect, the present application provides a fingerprint image expansion method, which adopts the following technical solution:

[0007] A fingerprint image expansion method specifically comprises the following steps:

[0008] Acquire a first fingerprint image;

[0009] Extracting a vector field and a frequency field of the first fingerprint image;

[0010] Expanding the first fingerprint image based on the frequency field to obtain an expanded image;

[0011] Correcting the extended image based on the vector field to obtain a corrected image;

[0012] A second fingerprint image is obtained by splicing the first fingerprint image and the corrected image, and the second fingerprint image is output.

[0013] The frequency field refers to the number of ridges per unit length in the first fingerprint image, and the frequency field can reflect the density of the ridges in the fingerprint image. The vector field describes the direction of the ridges in the first fingerprint image. The vector on each pixel points to the direction of the ridge at that point. The size of the vector can indicate the strength or confidence of the ridge. The vector field can reflect the global and local directions of the fingerprint image. By adopting the above technical solution, the image is expanded at the edge of the first fingerprint image according to the ridge density at the corresponding position in the frequency field and the ridge direction at the corresponding position in the vector field, and the expanded fingerprint pattern is obtained. The expansion speed is fast, and it is ensured that the expanded image has good continuity with the original image, and the false recognition rate will not be increased. At the same time, the size of the effective area for extracting fingerprint feature points is increased, thereby improving the pass rate during fingerprint authentication. It is achieved without upgrading the CPU speed and memory and without increasing the false recognition rate of fingerprints, and effectively improving the pass rate of embedded fingerprint processors.

[0014] Optionally, extracting the vector field and frequency field of the first fingerprint image specifically includes the following steps:

[0015] Dividing the first fingerprint image into a plurality of window units;

[0016] Calculating the local vector and the local frequency of each of the window units;

[0017] The local vectors constitute the vector field, and the local frequencies constitute the frequency field.

[0018] By adopting the above technical solution, the first fingerprint image is partitioned, and the local vectors and local frequencies are calculated for the pixels in each window unit as samples, representing the vector information and frequency information of the corresponding window unit. The local vectors of all window units are combined together to form the vector field of the first fingerprint image, and the local frequencies of all window units are combined together to form the frequency field of the first fingerprint image. The size of the window unit is defined according to the needs of specific applications and image analysis, and it is necessary to balance the ability to capture image features and computational efficiency. A larger window provides richer local information, but increases computational complexity, while a smaller window does the opposite. The amount of calculation of the frequency field and the vector field is greatly reduced, thereby increasing the expansion speed of the fingerprint image.

[0019] Optionally, the calculating of the local vector and the local frequency of each window unit specifically comprises the following steps:

[0020] Calculate the directional gradient of the central pixel point of the window unit;

[0021] Calculating the local vector based on the directional gradient;

[0022] Normalizing the first fingerprint image;

[0023] A frequency wave is obtained based on image modeling in a local direction window along the direction of the local vector, wherein the local direction window is characterized as a rectangular area with the central pixel point as the center and the normal direction of the local vector as the length direction;

[0024] The local frequency is calculated based on the frequency wave.

[0025] By adopting the above technical solution, a local direction window is set according to the normal direction of the central pixel and the local vector. The local direction window contains multiple alternating fingerprint ridges and valleys, and the corresponding frequency wave is modeled. The number of ridges per unit length is the local frequency of the window unit. The size of the local direction window is also defined according to the specific application and image analysis requirements.

[0026] Optionally, the central pixel point of the window unit is (x, y), the side length of the window unit is w, the gradient of the central pixel along the x direction is ∇x, and the gradient along the y direction is ∇y; the local vector is calculated based on the directional gradient, and the specific calculation formula is:

[0027]

[0028] Wherein, the θ(x, y) is the angle of the local vector of the window unit.

[0029] Optionally, the step of obtaining a frequency wave based on image modeling in a local direction window along the direction of the local vector specifically includes the following steps:

[0030] Along the angle of the local vector corresponding to the window unit, the amplitude X[k] of the frequency wave corresponding to each column of pixels is calculated. The specific calculation formula is:

[0031]

[0032] The frequency wave is fitted based on the amplitude X[k].

[0033] By adopting the above technical solution, the values ​​of each column of pixels in the local direction window are calculated along the direction of the local vector, a scatter plot is formed in the normal direction of the local vector, and further fitting is formed to form a frequency wave. The crest of the frequency wave corresponds to the fingerprint ridge, the trough corresponds to the fingerprint valley, and the number of crests corresponds to the number of ridges.

[0034] Optionally: the first fingerprint image is expanded based on the frequency field to obtain an extended image, specifically: a number of extended areas are created at the edge of the first fingerprint image, and fingerprint peak and valley pixels of adjacent extended areas are filled based on the local frequency of the window unit to obtain extended image units, and each of the extended image units constitutes the extended image.

[0035] By adopting the above technical solution, an expansion area is created at the edge of the first fingerprint image. For the convenience of calculation, the size of a single expansion area is preferably consistent with the size of the window unit. According to the local frequency of the adjacent window unit, the fingerprint peak and valley pixels are filled in the expansion area. The peak of the local frequency sine wave corresponds to the ridge line of the fingerprint, and the trough corresponds to the valley line of the fingerprint, so that the image is initially expanded, which has the characteristics of fast expansion speed.

[0036] Optionally, the extended image is corrected based on the vector field to obtain a corrected image, specifically:

[0037] Directional filtering is performed on the adjacent extended image units based on the local vector of the window unit to obtain corrected image units, and each of the corrected image units constitutes the corrected image.

[0038] By adopting the above technical solution, the extended image is directionally filtered according to the local vector of the corresponding window unit, and the filled peak and valley pixels are corrected, so that the extended image has good continuity with the ridges and valleys of the corresponding window unit, ensuring the accuracy of the feature point information corresponding to the extracted feature points, thereby improving the pass rate of fingerprint authentication.

[0039] In the second aspect, the present application provides a fingerprint image expansion system, which adopts the following technical solution:

[0040] A fingerprint image expansion system, comprising:

[0041] An image acquisition module, used to acquire a first fingerprint image;

[0042] An information extraction module, used to extract the vector field and frequency field of the first fingerprint image;

[0043] An image expansion module, used for expanding the first fingerprint image according to the frequency field to obtain an expanded image;

[0044] An image correction module, used for correcting the extended image according to the vector field to obtain a corrected image;

[0045] The merging output module is used to obtain a second fingerprint image by splicing the first fingerprint image and the corrected image, and output the second fingerprint image.

[0046] In a third aspect, the present application provides an electronic device, which adopts the following technical solution:

[0047] An electronic device, comprising:

[0048] at least one processor;

[0049] Memory; at least one application, wherein the at least one application is stored in the memory and is configured to be executed by at least one processor, and the at least one application is configured to: execute the above-mentioned fingerprint image expansion method.

[0050] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

[0051] A computer-readable storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the above-mentioned fingerprint image expansion method.

[0052] In summary, the present application includes at least one of the following beneficial technical effects:

[0053] The method of the present invention is used to expand the image, the expansion speed is fast, and the expanded fingerprint pattern is more accurate, which can greatly increase the effective area of ​​the extracted feature points, thereby improving the comparison pass rate;

[0054] The image is expanded by calculating the vector field and frequency field. The expansion speed is fast enough and the expanded image has good continuity with the original image. The pass rate can be improved without increasing the CPU speed and memory of the embedded fingerprint processor and avoiding increasing the false recognition rate.

[0055] For example, for the first fingerprint image with a pixel size of 64*80, assuming that by expanding 4 pixels on each of the four sides, a second fingerprint image with a size of 72*88 can be obtained. In this way, the effective pixels increase from 48*64 to 56*72, and the effective area increases by 31.25%, which greatly improves the pass rate of fingerprint recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a flowchart of the fingerprint image expansion method in the embodiment of the present application;

[0057] Figure 2 yes Figure 1 The expanded flow chart of step S2;

[0058] Figure 3 yes Figure 2 The expanded flow chart of step S22;

[0059] Figure 4 is a schematic diagram of extracting local frequency, showing the setting method of the local direction window and the obtained frequency wave;

[0060] Figure 5 is an example fingerprint image, where (1) is the first fingerprint image (64*80), (2) is the fingerprint image after frequency field expansion (72*88), and (3) is the second fingerprint image (72*88) obtained by splicing the expanded image after vector field correction with the first fingerprint image;

[0061] Figure 6 The following are examples of extracting feature points from fingerprint images. (1) shows the locations of feature points extracted from the first fingerprint image (64*80), and (2) shows the locations of feature points extracted from the second fingerprint image (72*88).

[0062] Figure 7 It is a structural block diagram of the fingerprint image expansion system in the embodiment of the present application;

[0063] Figure 8 It is a structural schematic diagram of an electronic device in an embodiment of the present application.

[0064] Explanation of the accompanying drawings: 1. Image acquisition module; 2. Information extraction module; 3. Image expansion module; 4. Image correction module; 5. Merging output module; 6. Processor; 7. Memory; 8. I / O interface; 9. Communication component; 10. Communication bus. DETAILED DESCRIPTION

[0065] To more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. However, it should be understood by those of ordinary skill in the art that the present application can be implemented without these details. In some cases, in order to avoid unnecessary descriptions that make various aspects of the present application obscure, well-known methods, processes, systems, components and / or circuits that have been described at a higher level will not be described in detail. For those of ordinary skill in the art, it is obvious that various changes can be made to the embodiments disclosed in the present application, and without departing from the principles and scope of the present application, the general principles defined in the present application can be applied to other embodiments and application scenarios. Therefore, the present application is not limited to the embodiments shown, but conforms to the broadest scope consistent with the scope claimed for protection of the present application.

[0066] It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention. In addition, the technical features involved in each embodiment of the present invention described below can be combined with each other as long as there is no conflict between them.

[0067] In the description of this application, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed", etc. are understood to exclude the number itself, and "above", "below", "within", etc. are understood to include the number itself. If there is a description of "first" or "second", it is only used to distinguish the technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0068] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples.

[0069] In order to enable those skilled in the art to understand the technical solution provided by the present application in detail, the working principle of general fingerprint recognition is briefly described first, and the steps are as follows:

[0070] Fingerprint collection: The fingerprint sensor obtains an image of the surface of the user's finger and captures the pattern formed by the ridges and valleys of the fingerprint.

[0071] Image preprocessing: Perform noise reduction, grayscale, binarization and other operations on the original fingerprint image to improve the accuracy of subsequent processing. It may include image filtering (noise removal), enhancement (contrast improvement) and normalization (size and orientation adjustment).

[0072] Feature extraction: Extract fingerprint feature points such as endpoints, bifurcation points, isolated points, short ridges and ridges from the preprocessed image.

[0073] Feature matching: Compare the extracted fingerprint features with the fingerprint templates stored in the database, and use various algorithms to determine the similarity between the fingerprint to be identified and the fingerprints in the database.

[0074] Decision: Based on the similarity score generated by the matching algorithm, the system will make a decision: if the score is higher than a preset threshold, the fingerprint match is considered successful and the system passes the authentication; if the score is lower than the threshold, the match is considered a failure and the system refuses authentication.

[0075] The pass rate of fingerprint recognition indicates the proportion of registered users' fingerprints that are correctly recognized and accepted.

[0076] The false recognition rate of fingerprint recognition indicates the proportion of fingerprints of unregistered users that are incorrectly accepted.

[0077] Among them, in the feature extraction stage, when the fingerprint algorithm extracts feature points, it needs to use an area of ​​8*8 pixels around the feature points as feature point information. Therefore, 8 pixels are subtracted from the area on each side of the top, bottom, left, and right sides where fingerprint features can be extracted. For example, the fingerprint pixels on the side of a mobile phone are 32*100, and the corresponding effective area is only 16*84. The fingerprint unlocking button on a computer is 64*80, and the corresponding effective area is only 48*64. In this way, too few feature points are extracted, resulting in a relatively low finger pass rate. The system will frequently deny access to legitimate users, resulting in a poor user experience.

[0078] The direction and shape of fingerprint images have certain rules, which we summarize into two important indicators: image vector field and frequency field. We use these indicators to expand the four sides of the image (up, down, left, and right) to increase the effective area. Based on this:

[0079] The embodiment of the present application discloses a fingerprint image expansion method.

[0080] Example 1

[0081] like Figure 1 As shown, the fingerprint image expansion method includes, specifically, the following steps:

[0082] S1: Acquire a first fingerprint image.

[0083] Explanatory, depending on the type of fingerprint sensor, the physical features of the ridges and valleys are converted into electrical signals, which are then converted into digital data to obtain a first fingerprint image.

[0084] Among them, the types of fingerprint sensors generally include optical sensors that illuminate the finger with a light source and use an optical lens to capture images of ridges and valleys, or capacitive sensors that use changes in capacitance between the finger and the sensor surface to detect ridges.

[0085] S2: Extracting the vector field and frequency field of the first fingerprint image.

[0086] like Figure 2 As shown, the specific steps for extracting the vector field and frequency field are as follows:

[0087] S21: Divide the first fingerprint image into a number of window units.

[0088] Explanatory, the first fingerprint image is partitioned to obtain a number of window units, and the local vectors and local frequencies are calculated for the pixels in each window unit as samples, representing the vector information and frequency information of the corresponding window unit. The local vectors of all window units are combined together to form the vector field of the first fingerprint image, and the local frequencies of all window units are combined together to form the frequency field of the first fingerprint image.

[0089] Illustratively, the size of the window unit is defined according to the requirements of specific applications and image analysis, and it is necessary to balance the ability to capture image features with computational efficiency, that is, the larger the window, the lower the accuracy but the faster the speed, while the smaller the window, the greater the accuracy but the slower the speed.

[0090] Preferably, the pixel size of the window unit is 4*4.

[0091] For example, if the pixels of the first fingerprint image are 64*80, and the first fingerprint image is partitioned into 4*4 window units, 320 window units can be obtained.

[0092] S22: Calculate the local vector and local frequency of each of the window units.

[0093] like Figure 3 As shown, the specific steps for calculating the local vector and local frequency are as follows:

[0094] S221: Calculate the directional gradient of the central pixel point of the window unit.

[0095] The central pixel point of the window unit is (x, y), and the side length of the window unit is w.

[0096] For example, the gradient ∇x in the x direction and the gradient ∇y in the y direction of the central pixel point (x, y) may be calculated by using a 3*3 Sobel operator.

[0097] S222: Calculate the local vector based on the directional gradient, and the specific calculation formula is:

[0098]

[0099] The direction function of the local vector of the window unit is calculated according to the gradient function.

[0100] S223: Normalize the first fingerprint image.

[0101] Explanatory, the frequency field of the fingerprint image reflects the intervals between the ridges and valleys of the fingerprint. Along the direction of the ridges and valleys, it can be modeled into a sine-like wave according to the grayscale of the fingerprint image.

[0102] S224: A frequency wave is obtained along the direction of the local vector based on image modeling in a local direction window, where the local direction window is represented by a rectangular area centered on the central pixel point and with the normal direction of the local vector as its length direction.

[0103] Combined with Figure 4 In the figure, the size of the window unit is w*w, the size of the local direction window is l*w, the long side l of the local direction window is perpendicular to the ridge direction, and the long side l of the local direction window should be greater than or at least equal to the diagonal length of the window unit to include all the ridges in the window unit.

[0104] Illustratively, the size of the local direction window is defined according to the needs of specific applications and image analysis, and it is necessary to balance the ability to capture image features with computational efficiency. A larger window provides richer local information but increases computational complexity, while a smaller window has the opposite effect.

[0105] Along the angle θ of the local vector corresponding to the window unit, the amplitude X[k] of the frequency wave corresponding to each column of pixels is calculated. The specific calculation formula is:

[0106]

[0107] The frequency wave is fitted based on the amplitude X[k].

[0108] Illustratively, the sinusoidal wave curve fitted according to the values ​​of X[0], X[1], X[2], . . . , X[l-1] is the frequency wave.

[0109] S225: Calculate the local frequency based on the frequency wave.

[0110] Explanatory, reference Figure 4 , extract the maximum point in the frequency wave, that is, the peak vertex β, and calculate the average value of the distance between adjacent maximum points, which is the pixel value α of the adjacent ridge line interval. The reciprocal of the average value is the number of ridge lines per unit length in the window unit, that is, the local frequency.

[0111] S23: The local vectors constitute the vector field, and the local frequencies constitute the frequency field.

[0112] The local vectors calculated based on the central pixel point (x, y) of each window unit together constitute the vector field, and the local frequencies calculated based on the central pixel point (x, y) of each window unit together constitute the frequency field.

[0113] Illustratively, there are many methods for extracting vector fields and frequency fields, not limited to the methods mentioned in this article, but also include other similar extraction methods, such as using the Sobel operator or vector field extraction based on integral images. The key point is that through the combination of these two methods, the accuracy of fingerprint expansion can be greatly improved.

[0114] S3: Expanding the first fingerprint image based on the frequency field to obtain an expanded image.

[0115] Specifically, the fingerprint peak and valley pixels of the adjacent extended areas are filled based on the local frequency of the window unit to obtain extended image units, and each of the extended image units constitutes the extended image.

[0116] Explanatory, several expansion areas are created at the edge of the first fingerprint image. For the convenience of calculation, the size of the expansion area can be consistent with the size of the window unit. In the example, the size of the window unit is 4*4, so the size of the expansion area is also set to 4*4, that is, each side of the first fingerprint image is expanded by 4 pixels.

[0117] Reference Figure 5 , (1) is the first fingerprint image (64*80), (2) is the fingerprint image after frequency field expansion (72*88).

[0118] S4: Correcting the extended image based on the vector field to obtain a corrected image.

[0119] Specifically, based on the local vector of the window unit, directional filtering is performed on the adjacent extended image units to obtain corrected image units, and each of the corrected image units constitutes the corrected image.

[0120] S5: obtaining a second fingerprint image based on the first fingerprint image and the corrected image by stitching, and outputting the second fingerprint image.

[0121] Reference Figure 5, (3) Figure shows the second fingerprint image (72*88) obtained by splicing the extended image after vector field correction with the first fingerprint image.

[0122] Reference for the effect after fingerprint image expansion Figure 6 , (1) shows the locations of feature points extracted from the first fingerprint image (64*80), and (2) shows the locations of feature points extracted from the second fingerprint image (72*88). It can be seen that the number of extracted feature points has increased significantly.

[0123] The change in pass rate after fingerprint image expansion is as follows:

[0124] Fingerprint Type First fingerprint image (64*80) pass rate Second fingerprint image (72*88) pass rate Children's fingers 90.1% 93.2% Old man's finger 88.3% 91.6% Normal finger 93.2% 96.7% Wet fingers 87.5% 90.6% Dry fingers 91.2% 94.4%

[0125] Note: The false recognition rate is one in one hundred thousand.

[0126] Explanatory, in actual image expansion, the local vectors and frequency vectors of the window units at the edge positions on both sides of the first fingerprint image can be calculated first, and the two sides of the first fingerprint image (64*80) can be expanded, corrected and merged based on these local vectors and frequency vectors, and then the local vectors and frequency vectors of the window units at the edge positions on the other two sides are calculated based on the expanded image (72*80), and then the local vectors and frequency vectors are expanded, corrected and merged again based on these local vectors and frequency vectors, and finally the second fingerprint image (72*88) is obtained. Similarly, it can be continuously expanded to obtain a larger second fingerprint image.

[0127] Example 2

[0128] The embodiment of the present application also discloses a fingerprint image expansion system.

[0129] Reference Figure 7 The fingerprint image expansion system includes an image acquisition module 1, an information extraction module 2, an image expansion module 3, an image correction module 4 and a merging output module 5.

[0130] The image acquisition module 1 is used to acquire a first fingerprint image.

[0131] The information extraction module 2 is used to extract the vector field and frequency field of the first fingerprint image.

[0132] Specifically, the first fingerprint image is divided into several window units. The local vector and local frequency of each window unit are calculated, more specifically, the directional gradient of the central pixel of the window unit is calculated; the local vector is calculated based on the directional gradient, where the central pixel of the window unit is (x, y), the side length of the window unit is w, the gradient of the central pixel along the x direction is ∇x, and the gradient along the y direction is ∇y. The specific calculation formula is:

[0133]

[0134] Among them, θ (x, y) is the angle of the local vector of the window unit. The first fingerprint image is normalized; along the direction of the local vector, the frequency wave is obtained based on the image modeling in the local direction window. The local direction window is characterized as a rectangular area with the center pixel as the center and the normal direction of the local vector as the length direction. More specifically: along the angle of the local vector corresponding to the window unit, the amplitude X[k] of the frequency wave corresponding to each column of pixels is calculated. The specific calculation formula is:

[0135]

[0136] Among them, θ(x,y) is the direction function of the central pixel (x,y); the frequency wave is fitted based on the amplitude X[k]; the local frequency is calculated based on the frequency wave. Each local vector constitutes a vector field, and each local frequency constitutes a frequency field.

[0137] The image expansion module 3 is used to expand the first fingerprint image according to the frequency field to obtain an expanded image.

[0138] Specifically, several extended areas are created at the edge of the first fingerprint image, and the fingerprint peak and valley pixels of adjacent extended areas are filled based on the local frequency of the window unit to obtain extended image units, and each extended image unit constitutes an extended image.

[0139] The image correction module 4 is used to correct the extended image according to the vector field to obtain a corrected image.

[0140] Specifically, based on the local vector of the window unit, directional filtering is performed on adjacent extended image units to obtain corrected image units, and each corrected image unit constitutes a corrected image.

[0141] The merging and outputting module 5 is used for obtaining a second fingerprint image by splicing the first fingerprint image and the corrected image, and outputting the second fingerprint image.

[0142] The other functions performed in the above-mentioned image acquisition module 1, information extraction module 2, image expansion module 3, image correction module 4 and merge output module 5 and the technical details of each function are the same or similar to the corresponding features in the fingerprint image expansion method described above, so they will not be repeated here.

[0143] The embodiment of the present application also discloses an electronic device.

[0144] like Figure 8 As shown, the electronic device includes a processor 6 and a memory 7 , and may further include one or more of an I / O interface 8 and a communication component 9 .

[0145] The processor 6 is used to control the overall operation of the electronic device to complete all or part of the steps in the above-mentioned fingerprint image expansion method; the memory 7 is used to store various types of data to support the operation of the electronic device, and these data may include instructions for any application or method used to operate on the electronic device, as well as application-related data. The memory 7 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0146] The I / O interface 8 provides an interface between the processor 6 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 9 is used to test the wired or wireless communication between the electronic device and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 9 can include: Wi-Fi components, Bluetooth components, NFC components.

[0147] The communication bus 10 may include a path to transmit information between the above components. The communication bus 10 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 10 may be divided into an address bus, a data bus, a control bus, etc.

[0148] The electronic device can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the fingerprint image expansion method given in the above embodiment.

[0149] The electronic device may include, but is not limited to, mobile terminals such as digital broadcast receivers, PDAs (Personal Digital Assistants), PMPs (Portable Multimedia Players), etc., and fixed terminals such as digital TVs, desktop computers, etc., and may also be servers, etc.

[0150] The embodiment of the present application also discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor 6, the steps of the fingerprint image expansion method described above are implemented.

[0151] The computer-readable storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0152] It should be understood that although the steps in the flowchart of the accompanying drawings are shown in sequence according to the instructions of the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise clearly stated in this document, the execution of these steps is not strictly limited in order and can be performed in other orders.

[0153] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A fingerprint image expansion method, characterized in that: The specific steps include: Acquire a first fingerprint image; Extracting a vector field and a frequency field of the first fingerprint image; Expanding the first fingerprint image based on the frequency field to obtain an expanded image; Correcting the extended image based on the vector field to obtain a corrected image; obtaining a second fingerprint image based on the first fingerprint image and the corrected image by stitching, and outputting the second fingerprint image; Extracting the vector field and the frequency field of the first fingerprint image specifically includes the following steps: Dividing the first fingerprint image into a plurality of window units; Calculating the local vector and the local frequency of each of the window units; The local vectors constitute the vector field, and the local frequencies constitute the frequency field; The expanding the first fingerprint image based on the frequency field to obtain the expanded image is specifically: Creating a plurality of extended areas at the edge of the first fingerprint image, and performing fingerprint peak and valley pixel filling on adjacent extended areas based on the local frequency of the window unit to obtain extended image units, wherein each of the extended image units constitutes the extended image; The extended image is corrected based on the vector field to obtain a corrected image, specifically: based on the local vector of the window unit, directional filtering is performed on adjacent extended image units to obtain corrected image units, and each of the corrected image units constitutes the corrected image.

2. The fingerprint image expansion method according to claim 1, characterized in that: Calculating the local vector and the local frequency of each window unit specifically includes the following steps: Calculate the directional gradient of the central pixel point of the window unit; Calculating the local vector based on the directional gradient; Normalizing the first fingerprint image; A frequency wave is obtained based on image modeling in a local direction window along the direction of the local vector, wherein the local direction window is characterized as a rectangular area with the central pixel point as the center and the normal direction of the local vector as the length direction; The local frequency is calculated based on the frequency wave.

3. The fingerprint image expansion method according to claim 2, characterized in that: The central pixel point of the window unit is , the side length of the window unit is , the gradient of the central pixel along the x direction is , the gradient along the y direction is ; Calculate the local vector based on the directional gradient, the specific calculation formula is: in, is the angle of the local vector of the window unit.

4. The fingerprint image expansion method according to claim 3, characterized in that: The frequency wave is obtained based on the image modeling in the local direction window along the direction of the local vector, which specifically includes the following steps: Along the angle of the local vector corresponding to the window unit, the amplitude X[k] of the frequency wave corresponding to each column of pixels is calculated. The specific calculation formula is: The frequency wave is fitted based on the amplitude X[k].

5. A fingerprint image expansion system, characterized in that: include: An image acquisition module (1), used to acquire a first fingerprint image; The information extraction module (2) is used to extract the vector field and frequency field of the first fingerprint image, and specifically comprises the following steps: Dividing the first fingerprint image into a plurality of window units; Calculating the local vector and the local frequency of each of the window units; The local vectors constitute the vector field, and the local frequencies constitute the frequency field; An image expansion module (3) is used to expand the first fingerprint image according to the frequency field to obtain an expanded image, specifically: Creating a plurality of extended areas at the edge of the first fingerprint image, and performing fingerprint peak and valley pixel filling on adjacent extended areas based on the local frequency of the window unit to obtain extended image units, wherein each of the extended image units constitutes the extended image; An image correction module (4) is used to correct the extended image according to the vector field to obtain a corrected image, specifically: based on the local vector of the window unit, directional filtering is performed on the adjacent extended image units to obtain corrected image units, and each of the corrected image units constitutes the corrected image; A merging and outputting module (5) is used to obtain a second fingerprint image by splicing the first fingerprint image and the corrected image, and output the second fingerprint image.

6. An electronic device, characterized in that: include: at least one processor (6); Memory (7); At least one application, wherein the at least one application is stored in a memory (7) and is configured to be executed by at least one processor (6), and the at least one application is configured to: execute the fingerprint image expansion method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor (6) to implement the fingerprint image expansion method according to any one of claims 1 to 4.

Images