Fingerprint identification device and fingerprint identification method

Abstract

The invention discloses a fingerprint identification device and a fingerprint identification method. The fingerprint identification device comprises a memory and a processor. The memory is used for storing at least one instruction. The processor is used for reading at least one instruction to execute the following steps: dividing the fingerprint image into a plurality of sub-regions; dividing the plurality of sub-regions into a plurality of high-quality regions and a plurality of low-quality regions according to the gray-scale value proportions of the plurality of sub-regions; obtaining a first feature sample of the plurality of high-quality areas; obtaining second feature samples of the plurality of low-quality areas, wherein the number of the first feature samples is greater than the number of the second feature samples; and outputting the first feature sample and the second feature sample.

Description

Technical Field

[0001] The present disclosure relates to a fingerprint recognition device and a fingerprint recognition method, and more particularly to a fingerprint recognition device and a fingerprint recognition method that divide a same fingerprint image into regions of different qualities for adaptive sampling. Background Art

[0002] With the advancement of technology, fingerprint recognition has become one of the most commonly used unlocking methods for electronic products. Fingerprint recognition can be divided into different unlocking modes based on the state of the fingerprint, such as dry fingerprint unlocking mode and wet fingerprint unlocking mode.

[0003] However, in the case where the same fingerprint may exist in both dry and wet states, if only a single fingerprint unlocking mode (for example, only using dry fingerprint unlocking mode or wet fingerprint unlocking mode) is used to unlock the phone, the recognition rate of fingerprint recognition will be affected, and even fingerprint recognition may fail. Summary of the Invention

[0004] In view of the deficiencies in the prior art, one of the objectives of the present disclosure is (but not limited to) to provide a fingerprint recognition device and a fingerprint recognition method to improve the deficiencies in the prior art.

[0005] In some embodiments, a fingerprint recognition device includes a memory and a processor. The memory is configured to store at least one instruction. The processor is configured to read the at least one instruction to execute the following steps: dividing a fingerprint image into a plurality of sub-regions; dividing the plurality of sub-regions into a plurality of high-quality regions and a plurality of low-quality regions based on grayscale value ratios of the plurality of sub-regions; obtaining first feature samples of the plurality of high-quality regions; obtaining second feature samples of the plurality of low-quality regions, wherein the number of the first feature samples is greater than the number of the second feature samples; and outputting the first feature samples and the second feature samples.

[0006] In some embodiments, a processor reads at least one instruction to execute a fingerprint recognition method, which includes: dividing a fingerprint image into multiple sub-regions; dividing the multiple sub-regions into multiple high-quality regions and multiple low-quality regions based on the grayscale value ratios of the multiple sub-regions; obtaining first feature samples of the multiple high-quality regions; obtaining second feature samples of the multiple low-quality regions, wherein the number of the first feature samples is greater than the number of the second feature samples; and outputting the first feature samples and the second feature samples.

[0007] The technical means embodied in the embodiments of this disclosure can alleviate at least one of the shortcomings of the existing technology. The fingerprint recognition device and fingerprint recognition method disclosed herein can divide the same fingerprint image into different quality areas and adaptively sample each quality area to facilitate subsequent fingerprint recognition. Therefore, this disclosure effectively improves the problem of using only a single fingerprint recognition mode for the same fingerprint image, which can affect the recognition rate and even cause fingerprint recognition failure.

[0008] The features, implementation and technical effects of the present disclosure are described in detail below with reference to the accompanying drawings for preferred embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic diagram of a fingerprint recognition device according to some embodiments of the present disclosure;

[0010] Figure 2 A flowchart of a fingerprint recognition method according to some embodiments of the present disclosure;

[0011] Figure 3 A schematic diagram of a fingerprint image drawn according to some embodiments of the present disclosure;

[0012] Figure 4 A schematic diagram of a fingerprint image drawn according to some embodiments of the present disclosure;

[0013] Figure 5 A schematic diagram of feature sampling according to some embodiments of the present disclosure; and

[0014] Figure 6 A schematic diagram of feature sampling according to some embodiments of the present disclosure is shown.

[0015] Explanation of symbols

[0016] 100: Fingerprint recognition device

[0017] 110: Processor

[0018] 120: Memory

[0019] 200: Method

[0020] 210-250: Steps

[0021] 300: Fingerprint image

[0022] 400: Fingerprint image

[0023] 410, 420: Block

[0024] A1~A16: Sub-areas

[0025] F1~F10:feature samples DETAILED DESCRIPTION

[0026] All terms used herein have their ordinary meanings. The definitions of the aforementioned terms in commonly used dictionaries, including any usage examples of any term discussed herein, are provided for illustrative purposes only and should not limit the scope and meaning of the present disclosure. Similarly, the present disclosure is not limited to the various embodiments described herein.

[0027] As used herein, "coupled" or "connected" may refer to two or more components making direct physical or electrical contact with each other, or making indirect physical or electrical contact with each other, or may refer to two or more components operating or acting on each other. As used herein, the term "circuit" may refer to a device comprised of at least one transistor and / or at least one active or passive component connected in a specific manner to process signals.

[0028] As used herein, the term "and / or" includes any combination of one or more of the listed associated items. Terms such as first, second, and third are used herein to describe and identify individual elements. Thus, a first element herein could also be referred to as a second element without departing from the spirit of the present disclosure. For ease of understanding, similar elements in the drawings will be designated with the same reference numerals.

[0029] To improve the existing problem of using only a single fingerprint unlocking mode to unlock, which will affect the recognition rate of fingerprint recognition and even cause fingerprint recognition failure, the present disclosure proposes a fingerprint recognition device and a fingerprint recognition method, which are described in detail below.

[0030] Figure 1 This is a schematic diagram of a fingerprint recognition device 100 drawn according to some embodiments of the present disclosure. As shown in the figure, the fingerprint recognition device 100 includes a processor 110 and a memory 120. The memory 120 is used to store at least one instruction. The processor 110 is used to read at least one instruction to perform fingerprint recognition. To make the operation of the fingerprint recognition device 100 easier to understand, please refer to Figure 2 , Figure 2 FIG. 2 is a flow chart of a fingerprint recognition method 200 according to some embodiments of the present disclosure.

[0031] First, in some embodiments, fingerprint images can be acquired by sensors such as capacitive sensors, optical sensors, and ultrasonic sensors. The acquired fingerprint images are typically grayscale two-dimensional images that accurately depict the fingerprint's ridges and valleys. Furthermore, before executing fingerprint recognition method 200, the present disclosure may perform pre-processing on the fingerprint image to obtain a processed fingerprint image. For example, the fingerprint image may be subjected to noise reduction, smoothing, and boundary enhancement. These processes may be performed using a mean filter, a sharpening filter, or other methods, or may be processed using normalization methods such as histogram equalization.

[0032] See also Figure 1 and Figure 2 In step 210, the fingerprint image is divided into a plurality of sub-regions. For example, see Figure 3 , Figure 3 FIG. 3 is a schematic diagram of a fingerprint image 300 drawn according to some embodiments of the present disclosure. The present disclosure may divide the fingerprint image 300 into a plurality of sub-areas A1 - A16 .

[0033] In step 220 , the plurality of sub-regions are divided into a plurality of high-quality regions and a plurality of low-quality regions according to the grayscale value ratios of the plurality of sub-regions.

[0034] First, we will explain how to obtain grayscale values, and then explain how to calculate the grayscale value ratios mentioned above using grayscale values. Figure 3 The fingerprint recognition device 100 can obtain multiple grayscale values of multiple pixels in each of the sub-areas A1 to A16, for example, multiple grayscale values of multiple pixels in the sub-area A1, multiple grayscale values of multiple pixels in the sub-area A2, ..., multiple grayscale values of multiple pixels in the sub-area A16.

[0035] Next, for each of the multiple sub-areas, the fingerprint recognition device 100 determines whether each of the multiple grayscale values is greater than a grayscale threshold. For example, it determines whether the grayscale value of each pixel in sub-area A1 is greater than the grayscale threshold. If one of the multiple grayscale values (e.g., the grayscale value of a pixel in sub-area A1) is greater than the grayscale threshold, then the one of the multiple grayscale values (e.g., the grayscale value of the pixel in sub-area A1) is determined to be a high grayscale value. Conversely, if one of the multiple grayscale values (e.g., the grayscale value of a pixel in sub-area A1) is not greater than the grayscale threshold, then the one of the multiple grayscale values (e.g., the grayscale value of the pixel in sub-area A1) is determined to be a low grayscale value.

[0036] In some embodiments, for a region of the plurality of sub-regions, the grayscale value ratio in step 220 is the ratio of the number of high grayscale values in the region to the number of low grayscale values in the region. In some embodiments, the grayscale threshold is between grayscale 107 and grayscale 147. In some embodiments, the grayscale threshold may be grayscale 127. In some embodiments, the grayscale threshold is between grayscale 100 and grayscale 140. In some embodiments, the grayscale threshold is between grayscale 110 and grayscale 150.

[0037] Next, we will explain how the grayscale value ratio is used. Figure 4 , Figure 4 This is a schematic diagram of a fingerprint image 400 drawn according to some embodiments of the present disclosure. As shown, the fingerprint in block 410 of fingerprint image 400 is relatively complete, while the fingerprint in block 420 of fingerprint image 400 is relatively blurred. This shows that even within the same fingerprint image 400, fingerprints in different blocks may have different conditions. Therefore, the fingerprint recognition device 100 of the present disclosure can divide the same fingerprint image 400 into different quality areas and adaptively sample each quality area to facilitate subsequent fingerprint recognition. Details are described below.

[0038] See also Figure 4 The fingerprint recognition device 100 of the present disclosure can determine whether the grayscale value ratio of high grayscale values to low grayscale values of one of the multiple sub-regions (e.g., one of the sub-regions A1-A16) is within a grayscale value ratio range. If the grayscale value ratio of one of the multiple sub-regions (e.g., one of the sub-regions A1-A16) is within a grayscale value ratio range, then the one of the multiple sub-regions (e.g., one of the sub-regions A1-A16) is determined to be a high-quality region.

[0039] For example, the fingerprint in block 410 of fingerprint image 400 is relatively complete. As can be seen in the figure, the ratio of high grayscale values to low grayscale values in the pixels of sub-regions A13, A14, and A15 corresponding to block 410 is close to 1:1, i.e., the grayscale value ratio is close to 1. The fingerprint recognition device 100 can determine that the grayscale value ratio of high grayscale values to low grayscale values in sub-regions A13, A14, and A15 (e.g., 1) is within a grayscale value ratio range (e.g., 0.6 to 1.4). In this case, sub-regions A13, A14, and A15 are determined to be high-quality regions. In another embodiment, the sub-regions corresponding to block 410 may further include sub-regions A9 to A11, but are not limited thereto.

[0040] Furthermore, if the grayscale value ratio of one of the sub-regions (eg, one of the sub-regions A1 - A16 ) is not within the grayscale value ratio range, the one of the sub-regions (eg, one of the sub-regions A1 - A16 ) is determined to be a low-quality region.

[0041] For example, the fingerprint in block 420 of the fingerprint image 400 is blurred. Therefore, the grayscale value ratio (e.g., 2) of the corresponding sub-regions A2-A4 and A6-A8 of block 420 is not within the grayscale value ratio range (e.g., 0.6-1.4). In this case, sub-regions A2-A4 and A6-A8 are determined to be low-quality regions.

[0042] In some embodiments, the grayscale value ratio range is 0.6 to 1.4. For example, the grayscale value ratio range formula of the present disclosure is as follows:

[0043] (1-Bias_TH)~(1+Bias_TH)…Formula 1

[0044] As shown in Formula 1, Bias_TH is a bias threshold. The bias threshold of the present disclosure may be 0.4. Substituting the bias threshold 0.4 into Formula 1 yields a grayscale value ratio range of 0.6 to 1.4.

[0045] In some embodiments, the bias threshold of the present disclosure may be 0.3. Substituting the bias threshold of 0.3 into Formula 1 yields a grayscale value ratio range of 0.7 to 1.3. In some embodiments, the bias threshold of the present disclosure may be 0.5. Substituting the bias threshold of 0.5 into Formula 1 yields a grayscale value ratio range of 0.5 to 1.5.

[0046] In some embodiments, in addition to using grayscale value ratios to determine quality regions, the present disclosure may also utilize grayscale standard deviation within a region, grayscale contrast within a region, and other values as a basis for determination, depending on actual needs.

[0047] In step 230, a plurality of first feature samples of the plurality of high-quality regions are obtained. In step 240, a plurality of second feature samples of the plurality of low-quality regions are obtained, wherein the number of the plurality of first feature samples is greater than the number of the plurality of second feature samples.

[0048] To illustrate how features are sampled, see Figure 5 and Figure 6 , Figure 5 Schematic diagram of a feature sampling method according to the prior art, wherein the white sub-areas A4 and A7-A16 are high-quality areas, and the gray sub-areas A1-A3 and A5-A6 are low-quality areas. Figure 5 As shown, five feature samples F1 to F5 are sampled for high-quality regions A4 and A7 to A16, and five feature samples F6 to F10 are sampled for low-quality regions A1 to A3 and A5 to A6. In other words, the feature samples are evenly distributed between high-quality and low-quality regions.

[0049] In contrast, Figure 6As shown, the fingerprint recognition device 100 of the present disclosure can adaptively sample the high-quality areas A4 and A7-A16 and the low-quality areas A1-A3 and A5-A6. For example, five feature samples F1-F5 are sampled for the high-quality areas A4 and A7-A16, and three feature samples F6, F8, and F10 are sampled for the low-quality areas A1-A3 and A5-A6. It can be seen that the number of feature samples sampled for the high-quality areas A4 and A7-A16 (e.g., 5) is greater than the number of feature samples sampled for the low-quality areas A1-A3 and A5-A6 (e.g., 3).

[0050] The fingerprint recognition device 100 of the present disclosure performs adaptive sampling for different quality areas based on the characteristics of fingerprints of different qualities. For example, the fingerprint images of the high-quality areas A4 and A7~A16 are clearer, so the feature points obtained from the high-quality areas A4 and A7~A16 are more likely to be valid feature points rather than noise. In addition, the fingerprint images of the low-quality areas A1~A3 and A5~A6 are more blurred, which means that the feature samples sampled therefrom are more difficult to distinguish between fingerprints and noise, and are more likely to cause errors. Therefore, the fingerprint recognition device 100 of the present disclosure concentrates the sampling of feature points on the high-quality areas A4 and A7~A16, and reduces the sampling of the low-quality areas A1~A3 and A5~A6, thereby improving the recognition rate of the fingerprint recognition device 100 of the present disclosure.

[0051] In some embodiments, the plurality of first feature samples include a plurality of first feature points, and the plurality of second feature samples include a plurality of second feature points, wherein the number of the plurality of first feature points is greater than the number of the plurality of second feature points. For example, see Figure 6 , the number of feature points sampled in the high-quality areas A4 and A7-A16 (eg, 5) is greater than the number of feature points sampled in the low-quality areas A1-A3 and A5-A6 (eg, 3).

[0052] In some embodiments, the present disclosure may predefine a maximum number of feature points for (each) fingerprint image, and then calculate the intensity of each feature point based on the grayscale value difference between each feature point and its area, or other intensity calculation methods (such as directional intensity), and sort them according to the intensity of each feature point. Feature points with stronger intensity are selected as feature points first.

[0053] In some embodiments, for the high-quality regions of the fingerprint image, the minimum values within a small region or a large region can be selected as candidate feature points, and the definition of the above-mentioned minimum value in the present disclosure can be relatively loose to increase the number of effective feature points in the high-quality regions of the fingerprint image. In some embodiments, for the low-quality regions of the fingerprint image, the minimum values within a small region or a large region can be selected as candidate feature points, and the definition of the above-mentioned minimum value in the present disclosure can be relatively strict to reduce the number of invalid feature points in the low-quality regions of the fingerprint image.

[0054] For example, for the high-quality regions of the fingerprint image, the definition of the minimum value within a small region or a large region is as shown in Formula 2 below. For the low-quality regions of the fingerprint image, the definition of the minimum value within a small region or a large region is as shown in Formula 3 below:

[0055] I(x,y)≤I(x - 2:x + 2,y - 2:y + 2)…Formula 2

[0056] I(x,y)<I(x - 2:x + 2,y - 2:y + 2)…Formula 3

[0057] As shown in Formula 2, the definition of the minimum value in the high-quality region is relatively loose (including both less than and equal conditions). As shown in Formula 3, the definition of the minimum value in the low-quality region is relatively strict (only including the less than condition). Therefore, the present disclosure can reduce the number of invalid feature points in the low-quality regions of the fingerprint image.

[0058] In some embodiments, multiple first feature samples include first feature confidence values of multiple high-quality regions, and multiple second feature samples include second feature confidence values of multiple low-quality regions, where the first feature confidence value is greater than the second feature confidence value. For example, please refer to Figure 6 , the feature confidence values of the feature points sampled in the high-quality regions A4 and A7 - A16 are greater than the feature confidence values of the feature points sampled in the low-quality regions A1 - A3 and A5 - A6. In some embodiments, the feature confidence value of the above-mentioned feature point represents the degree of reliability of the feature point.

[0059] In some embodiments, the present disclosure can reduce the number of second feature samples in multiple low-quality regions. For example, if the recognition rate of subsequent fingerprint recognition is still low, the present disclosure can further reduce the sampling of the low-quality regions A1 - A3 and A5 - A6 to avoid the influence of unclear fingerprint feature samples on the fingerprint recognition rate.

[0060] In step 250, a plurality of first feature samples and a plurality of second feature samples are output. For example, after obtaining the first feature samples and the second feature samples, the fingerprint recognition device 100 of the present disclosure may output the first feature samples and the second feature samples for subsequent fingerprint recognition. In some embodiments, the first feature samples and the second feature samples output by the present disclosure include the coordinates of the feature samples (e.g., feature points) in the fingerprint image, such as the x-coordinate and the y-coordinate of a Cartesian coordinate system.

[0061] It should be noted that this disclosure does not Figures 1 to 6 The embodiments shown are limited to the examples shown and are merely illustrative of one embodiment of the present disclosure to facilitate understanding of the technology of the present disclosure. The scope of the present disclosure shall be determined by the claims. Those skilled in the art may modify or embellish the embodiments of the present disclosure without departing from the scope of the present disclosure and still fall within the scope of the claims of the present disclosure.

[0062] In summary, the fingerprint recognition device 100 and fingerprint recognition method 200 disclosed herein can divide a single fingerprint image into different quality regions and adaptively sample each quality region to facilitate subsequent fingerprint recognition. Therefore, the present disclosure effectively improves the problem of using only a single fingerprint recognition mode for a single fingerprint image, which can affect the recognition rate and even cause fingerprint recognition failure.

[0063] Although the embodiments of the present disclosure are described above, these multiple embodiments are not intended to limit the present disclosure. Those skilled in the art may make changes to the technical features of the present disclosure based on the explicit or implicit content of the present disclosure. All such changes may fall within the scope of patent protection sought by the present disclosure. In other words, the scope of patent protection of the present disclosure shall be based on the definition of the claims of this specification.

Claims

1. A fingerprint recognition device, comprising: a memory for storing at least one instruction; and A processor is configured to read the at least one instruction to execute the following steps: Dividing a fingerprint image into multiple sub-regions; dividing the plurality of sub-regions into a plurality of high-quality regions and a plurality of low-quality regions according to a grayscale value ratio of the plurality of sub-regions; Obtaining a plurality of first feature samples of the plurality of high-quality areas; Obtaining a plurality of second feature samples of the plurality of low-quality areas, wherein the number of the plurality of first feature samples is greater than the number of the plurality of second feature samples; as well as The plurality of first feature samples and the plurality of second feature samples are output.

2. The fingerprint recognition device as claimed in claim 1 , wherein the processor is further configured to read the at least one instruction to execute the following steps: The number of the second feature samples in the low-quality regions is reduced.

3. The fingerprint recognition device as claimed in claim 1 , wherein the processor is further configured to read the at least one instruction to execute the following steps: For a sub-region of the plurality of sub-regions, obtaining a plurality of grayscale values of a plurality of pixels in the sub-region; determining whether one of the plurality of grayscale values is greater than a grayscale threshold; If the one of the plurality of grayscale values is greater than the grayscale threshold, determining that the one of the plurality of grayscale values is a high grayscale value; and If one of the multiple grayscale values is not greater than the grayscale threshold, then the one of the multiple grayscale values is determined to be a low grayscale value, wherein the grayscale value ratio is the ratio of the number of the high grayscale values of the multiple sub-areas to the number of the low grayscale values of the multiple sub-areas. 4 . The fingerprint recognition device as claimed in claim 3 , wherein the grayscale threshold is between grayscale 107 and grayscale 147.

5. The fingerprint recognition device as claimed in claim 3 , wherein the processor is further configured to read the at least one instruction to execute the following steps: determining whether a grayscale value ratio of the high grayscale value to the low grayscale value of one of the plurality of sub-regions is within the grayscale value ratio range; If the grayscale value ratio of the one of the plurality of sub-regions is within the grayscale value ratio range, determining that the one of the plurality of sub-regions is the high-quality region; and If the grayscale value ratio of one of the plurality of sub-regions is not within the grayscale value ratio range, then the one of the plurality of sub-regions is determined to be the low-quality region. 6 . The fingerprint recognition device as claimed in claim 5 , wherein the grayscale value ratio ranges from 0.6 to 1.

4.

7. The fingerprint recognition device according to any one of claims 1 to 6, wherein the plurality of first feature samples include a plurality of first feature points, the plurality of second feature samples include a plurality of second feature points, and the number of the plurality of first feature points is greater than the number of the plurality of second feature points.

8. The fingerprint recognition device as claimed in any one of claims 1 to 6, wherein the plurality of first feature samples include a first feature confidence value for the plurality of high-quality regions, and the plurality of second feature samples include a second feature confidence value for the plurality of low-quality regions, wherein the first feature confidence value is greater than the second feature confidence value.

9. A fingerprint recognition method, comprising: a processor reading at least one instruction to execute the fingerprint recognition method; Dividing a fingerprint image into multiple sub-regions; dividing the plurality of sub-regions into a plurality of high-quality regions and a plurality of low-quality regions according to a grayscale value ratio of the plurality of sub-regions; Obtaining a plurality of first feature samples of the plurality of high-quality areas; Obtaining a plurality of second feature samples of the plurality of low-quality areas, wherein the number of the plurality of first feature samples is greater than the number of the plurality of second feature samples; as well as The plurality of first feature samples and the plurality of second feature samples are output.

10. The fingerprint recognition method according to claim 9, further comprising: The number of the second feature samples in the low-quality regions is reduced.

Images