Under-screen fingerprint sensor

By using image sensors and controllers to process fingerprint images in the under-screen fingerprint sensor, the problem of insufficient security in identifying real fingers and fake fingers is solved, and effective recognition of fake fingers is achieved and improve the security of personal data.

CN112287903BActive Publication Date: 2024-11-12EGIS TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202011379861.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2020-06-08
Filing Date
2020-11-30
Publication Date
2024-11-12
Estimated Expiration
2040-11-30

AI Technical Summary

Technical Problem

The existing under-screen fingerprint sensors have insufficient security in identifying real and fake fingers, which makes it difficult to effectively prevent theft of personal data.

Method used

Using an image sensor and controller combination, the greyscale value of the fingerprint image and the Hale feature are calculated internally, and combined with the grayscale average value of the central area and the surrounding area, it is determined whether the fingerprint image comes from a real finger or a fake finger.

Benefits of technology

Accurate identification of real and fake fingers is achieved, improving the security of personal data and preventing data theft.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112287903B_ABST
    Figure CN112287903B_ABST
Patent Text Reader

Abstract

The present invention provides an under-screen fingerprint sensing device, comprising an image sensor and a controller. The image sensor is arranged below a display, wherein the display emits an illumination light, which is transmitted to a finger pressed on the display, and the finger reflects the illumination light into a signal light carrying fingerprint information of the finger, and the signal light penetrates the display to form a fingerprint image on the image sensor. The controller is electrically connected to the image sensor and is used to process the fingerprint image sensed by the image sensor. The controller is used to perform an inner product between the grayscale value on two axes in different directions of the fingerprint image and the Haar feature, and thereby determine whether the fingerprint image is from a real finger or a fake finger.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a sensing device, and in particular to an under-screen fingerprint sensing device. Background Art

[0002] As portable electronic devices develop towards a larger screen-to-body ratio, the capacitive fingerprint sensor originally placed on the front of the electronic device is no longer applicable and needs to be placed on the side or back of the electronic device. However, fingerprint sensors placed on the side or back are inconvenient to use, so under-screen fingerprint sensors were developed.

[0003] Under-screen fingerprint sensors can be roughly divided into optical fingerprint sensors and ultrasonic fingerprint sensors, among which optical fingerprint sensors are lower in cost and more suitable for mass production. Since fingerprint sensing is related to the security of users' personal data, anti-spoofing functions need to be developed. For example, someone who wants to steal personal data may collect the user's fingerprint in the environment, make a fake finger with this fingerprint, and then use the fake finger to press the electronic device to achieve successful fingerprint recognition and unlock. Therefore, it is necessary to develop a fingerprint sensor that can identify fake fingers instead of real fingers to further enhance the security of users' personal data. Summary of the invention

[0004] The present invention is directed to an under-screen fingerprint sensing device, which can identify real fingers and fake fingers.

[0005] One embodiment of the present invention proposes an under-screen fingerprint sensing device for use with a display. The display emits an illumination light, which is transmitted to a finger pressed on the display. The finger reflects the illumination light into a signal light carrying the fingerprint information of the finger. The under-screen fingerprint sensing device includes an image sensor and a controller. The image sensor is arranged below the display, wherein the signal light penetrates the display and forms a fingerprint image on the image sensor. The controller is electrically connected to the image sensor, processes the fingerprint image sensed by the image sensor, performs an inner product of the grayscale values ​​on two axes in different directions of the fingerprint image and the Haar-like feature, and thereby determines whether the fingerprint image is from a real finger or a fake finger.

[0006] One embodiment of the present invention provides an under-screen fingerprint sensing device for use with a display. The display emits an illumination light, which is transmitted to a finger pressed on the display. The finger reflects the illumination light into a signal light carrying the fingerprint information of the finger. The under-screen fingerprint sensing device includes an image sensor and a controller. The image sensor is arranged below the display, wherein the signal light penetrates the display and forms a fingerprint image on the image sensor. The controller is electrically connected to the image sensor and is used to process the fingerprint image sensed by the image sensor. The controller is used to calculate the grayscale average value of the central area of ​​the fingerprint image and the grayscale average values ​​of the two peripheral areas on both sides of the central area, and thereby determine whether the fingerprint image is from a real finger or a fake finger.

[0007] One embodiment of the present invention provides an under-screen fingerprint sensing device for use with a display. The display emits an illumination light, which is transmitted to a finger pressed on the display. The finger reflects the illumination light into a signal light carrying the fingerprint information of the finger. The under-screen fingerprint sensing device includes an image sensor and a controller. The image sensor is arranged below the display, wherein the signal light penetrates the display and forms a fingerprint image on the image sensor. The controller is electrically connected to the image sensor and is used to process the fingerprint image sensed by the image sensor. The controller is used to perform an inner product of the grayscale value on two axes in different directions of the fingerprint image and the Haar feature to obtain a first result. The controller is used to calculate the grayscale average value of the central area of ​​the fingerprint image and the grayscale average values ​​of the two peripheral areas on both sides of the central area to obtain a second result. The controller is used to combine the first result and the second result to determine whether the fingerprint image is from a real finger or a fake finger.

[0008] In the under-screen fingerprint sensing device of the embodiment of the present invention, the controller performs an inner product between the grayscale values ​​on two axes in different directions of the fingerprint image and the Haar feature, and / or calculates the grayscale average value of the central area of ​​the fingerprint image and the grayscale average values ​​of the two peripheral areas on both sides of the central area, and thereby determines whether the fingerprint image is from a real finger or a fake finger. Therefore, the under-screen fingerprint sensing device of the embodiment of the present invention can achieve an anti-spoofing effect in fingerprint sensing, thereby enhancing the security of the user's personal data. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 is a cross-sectional schematic diagram of an electronic device according to an embodiment of the present invention;

[0010] Figure 2A Show Figure 1 The case where the illumination light with P polarization direction is reflected by the finger;

[0011] Figure 2B Show Figure 1 The case where the illumination light with S polarization direction is reflected by the finger;

[0012] Figure 3 for Figure 1 The distribution curve of the reflectivity and transmittance of the illumination light at the interface between the finger and the glass cover relative to the incident angle of the illumination light on the interface;

[0013] Figure 4A for Figure 1 A regional comparison diagram of a fingerprint image sensed by an image sensor and the P polarized light and S polarized light received by the image sensor;

[0014] Figure 4B for Figure 1 The image sensor senses the fingerprint image of the fake finger;

[0015] Figure 5A for Figure 1 The average grayscale distribution diagram of the fingerprint image of the real finger sensed by the image sensor on the diagonal line MN;

[0016] Figure 5B for Figure 1 An average grayscale distribution diagram of the fingerprint image of the fake finger sensed by the image sensor on the diagonal line MN;

[0017] Figure 6 for Figure 1 A schematic diagram of a controller in processing a fingerprint image;

[0018] Figure 7 for Figure 1 Another schematic diagram of the controller processing a fingerprint image;

[0019] Figure 8 for Figure 1 Schematic diagram of a controller in processing a first eigenvalue and a second eigenvalue of a fingerprint image. DETAILED DESCRIPTION

[0020] Reference will now be made in detail to exemplary embodiments of the present invention, examples of which are illustrated in the accompanying drawings. Whenever possible, the same reference numerals are used in the drawings and the description to refer to the same or like parts.

[0021] Figure 1 FIG. 1 is a cross-sectional diagram of an electronic device according to an embodiment of the present invention. Figure 1The electronic device 100 of this embodiment includes a display 105 and an under-screen fingerprint sensing device 200. The electronic device 100 is, for example, a smart phone, a tablet computer, a laptop computer, a personal digital assistant (PDA), or other appropriate electronic devices. The under-screen fingerprint sensing device 200 is used to work with the display 105, and includes an image sensor 210 and a controller 230. The image sensor 210 is disposed below the display 105. The display 105 emits an illumination light 60, which is transmitted to a finger 50 pressed on the display 105, and the finger 50 reflects the illumination light 60 into a signal light 70 carrying the fingerprint information of the finger. The signal light 70 penetrates the display 105 and forms a fingerprint image on the image sensor 210. In this way, the image sensor 210 can sense the fingerprint image of the finger 50. The controller 230 is electrically connected to the image sensor 210 and is used to process the fingerprint image sensed by the image sensor 210.

[0022] In the present embodiment, the display 105 includes a cover glass 120, a linear polarizer 116, a phase retarder film 114, and an organic light emitting diode display panel 112. A finger is pressed on the cover glass 120. The linear polarizer 116 is disposed between the cover glass 120 and the image sensor 210. The phase retarder film 114 is disposed between the linear polarizer 116 and the image sensor 210, wherein the phase retarder film 114 is, for example, a wave plate, which can be formed of the material of the display itself or an additional wave plate. The organic light emitting diode display panel 112 is disposed between the phase retarder film 114 and the image sensor 210. In other embodiments, the organic light emitting diode display panel 112 can also be replaced by a liquid crystal display panel, a micro-light emitting diode display panel, an electrophoretic display panel, or other appropriate display panels.

[0023] In this embodiment, the under-screen fingerprint sensing device 200 further includes a lens 220, which is disposed on the path of the signal light 70 and is located between the display 105 and the image sensor 210. The lens 220 may include one or more lenses, which may image the signal light 70 on the image sensor 210 to form a fingerprint image on the image sensor 210.

[0024] Specifically, the illumination light 60 emitted by the organic light emitting diode display panel 112 is not polarized, and after passing through the phase delay film 114, it still does not have polarization. However, after passing through the linear polarizer 116, the illumination light 60 has a polarization direction K0, wherein the polarization direction K0 can be decomposed into two components, a first polarization direction K1 and a second polarization direction K2, as shown in FIG. Figure 2A and Figure 2B As shown, the first polarization direction K1 is the P polarization direction, and the second polarization direction K2 is the S polarization direction. In other words, the illumination light 60 with the polarization direction K0 can be regarded as the combination of the illumination light 62 with the first polarization direction K1 and the illumination light 64 with the second polarization direction K2.

[0025] Figure 2A Show Figure 1 In the case where the illumination light with P polarization direction is reflected by the finger, Figure 2B Show Figure 1 The illumination light with S polarization direction is reflected by the finger. Figure 1 , Figure 2A and Figure 2B The illumination light 62 having the first polarization direction K1 travels in the glass cover 120 (eg Figure 2A ), and is partially reflected by the finger 50 at the upper surface 122 of the glass cover 120 to form the signal light 72, while another part of the illumination light 82 is transmitted in the finger 50. On the other hand, the illumination light 64 having the second polarization direction K2 travels in the glass cover 120 (such as Figure 2B ), and is partially reflected by the finger 50 at the upper surface 122 of the glass cover 120 to form the signal light 74, while the other part of the illumination light 84 is transmitted in the finger 50. The signal light 72 and the signal light 74 are combined into the signal light 70, and transmitted in the direction of the image sensor 210.

[0026] The light intensities and ratios of the signal light 72 and the signal light 74 sensed by the image sensor 210 are related to the reflectivity caused by the refractive indexes of the finger 50 and the glass cover 120 . Figure 3 for Figure 1 The distribution curve of the reflectivity and transmittance of the interface between the finger and the glass cover to the illumination light relative to the incident angle of the interface, where R p The distribution curve of the reflectivity of the interface to the illumination light 62 with respect to the incident angle is: s The distribution curve of the reflectivity of the interface to the illumination light 64 with respect to the incident angle is: p The distribution curve of the transmittance of the interface to the illumination light 62 relative to the incident angle is shown as follows: s The distribution curve of the transmittance of the interface to the illumination light 64 relative to the incident angle is shown as follows. p =1-R p , and T s =1-R s .Depend on Figure 3It can be seen that the difference in light intensity between the signal light 72 (P polarized light) and the signal light 74 (S polarized light) corresponding to a larger incident angle is greater. Since the image sensor 210 receives the signal light 70 reflected by the illumination light 60 with a larger incident angle at the outer circle, the difference in light intensity between the signal light 72 (P polarized light) and the signal light 74 (S polarized light) sensed at the outer circle is greater.

[0027] Figure 4A for Figure 1 The fingerprint image sensed by the image sensor and the area comparison diagram of the P polarized light and S polarized light received by it. Please refer to Figure 1 , Figure 2A , Figure 2B and Figure 4A , Figure 4A In the figure, the image brightness in the area where P is located is mainly contributed by the signal light 72 (P polarized light), the image brightness in the area where S is located is mainly contributed by the signal light 74 (S polarized light), and the image brightness in the area where S+P is located is mainly contributed by both the signal light 72 (P polarized light) and the signal light 74 (S polarized light). The contrast of the fingerprint image is stronger in the area where P is located because there is more penetrating light; the contrast of the fingerprint image is weaker in the area where S is located because there is less penetrating light; the DC value of the fingerprint image is low in the area where P is located because there is less reflected light, that is, the average brightness is low; the DC value of the fingerprint image is high in the area where S is located because there is more reflected light, that is, the average brightness is high.

[0028] Figure 4B for Figure 1 The image sensor senses the fingerprint image of the fake finger. Figure 5A for Figure 1 The average grayscale distribution diagram of the fingerprint image of the real finger sensed by the image sensor on the diagonal MN, and Figure 5B for Figure 1 The average grayscale distribution diagram of the fingerprint image of the fake finger sensed by the image sensor on the diagonal MN. Figure 5A and Figure 5B The average grayscale of a pixel on the diagonal line MN is obtained by taking the average grayscale value of several pixels near the pixel and the grayscale value of the pixel itself. Figure 4A and Figure 4B ,as well as Figure 5A and Figure 5B, it can be found that the average brightness of the fingerprint image of the real finger is lower in the center, while the average brightness near the two ends on the diagonal line MN is higher. On the contrary, the average brightness of the fingerprint image of the fake finger is higher in the center, while the average brightness near the two ends on the diagonal line MN is lower. One of the main reasons for this phenomenon is that the refractive index of the fake finger is different from that of the real finger, which causes the reflectivity of the P polarized light and the S polarized light to be different, and then causes the distribution curve of the average brightness at different positions to be different. Therefore, the embodiment of the present invention can take advantage of this phenomenon to produce a solution for identifying real fingers and fake fingers, which will be described in detail in the following content.

[0029] Figure 6 for Figure 1 Schematic diagram of the controller processing fingerprint images in . Figure 1 and Figure 6 In this embodiment, the controller 230 is used to perform an inner product between the grayscale values ​​on two axes L1 and L2 in different directions of the fingerprint image and the Haar feature to obtain a first result. Specifically, in this embodiment, the controller first performs a blurring process on the fingerprint image, and then performs an inner product between the grayscale values ​​on two axes L1 and L2 in different directions of the blurred fingerprint image and the Haar feature. The blurring here is, for example, mean blurring or Gaussian blurring, for example, taking the average of the grayscale value of each pixel and the grayscale values ​​of several pixels around the pixel, and using the average as the new grayscale value of the pixel.

[0030] In the present embodiment, the Haar feature is a function that the middle area C1 has a first value and the two side areas C2 and C3 have a second value, wherein the first value is greater than the second value. For example, the first value is +1 and the second value is -1. The blurred fingerprint image has, for example, 200×200 pixels, and the Haar feature has values ​​of, for example, -1 for the 1st to 75th pixels from the left end, +1 for the 76th to 125th pixels, and -1 for the 126th to 200th pixels. The axis L1 of the blurred fingerprint image also has 200 pixels, and each pixel has its own grayscale value, and this grayscale value represents the brightness of the blurred fingerprint image. Next, the grayscale values ​​of the 200 pixels on the axis L1 are sequentially multiplied with the values ​​of the 200 pixels of the Haar feature (ie -1 or +1) to obtain 200 products, and then the 200 products are added together to obtain the value obtained after the inner product operation.

[0031] Then, the controller 230 adds the two inner product values ​​corresponding to the axis L1 and the axis L2 respectively to obtain the first eigenvalue. Figure 5A, Figure 5B and Figure 6 It can be seen that the grayscale values ​​on the axes L1 and L2 are calculated by inner product with the Haar feature. Figure 5A The inner product value or the first eigenvalue of the true finger is smaller, and Figure 5B The larger the inner product value or the first eigenvalue of the fake finger, the calculated first eigenvalue can be used to distinguish whether the sensed finger is a real finger or a fake finger.

[0032] Figure 7 for Figure 1 Another schematic diagram of the controller processing fingerprint images in FIG. Figure 1 and Figure 7 , the controller 230 is used to calculate the grayscale average of the central area D2 of the fingerprint image and the grayscale averages of the two peripheral areas D1 and D3 on both sides of the central area D2 to obtain a second result. In addition, the controller 230 is used to combine the first result and the second result to determine whether the fingerprint image is from a real finger or a fake finger.

[0033] Specifically, the grayscale average value of the central area D2 is R2, and the grayscale average values ​​of the two peripheral areas D1 and D3 are R1 and R3 respectively. The controller 230 is used to perform arithmetic operations on R1, R2, and R3 to obtain the second eigenvalue. In the present embodiment, the central area D2 and the two peripheral areas D1 and D3 are arranged on the diagonal line of the fingerprint image, and the direction of the diagonal line can be determined according to the direction of the penetration axis of the linear polarizer 116, that is, the direction in which the difference between the second eigenvalues ​​of the real finger and the fake finger is greater is selected as the direction of the diagonal line. In one embodiment, the above-mentioned arithmetic operation is, for example, (R2-R1)+(R2-R3), or the above-mentioned arithmetic operation is, for example, R2-R1-R3. In the comparison Figure 5A , Figure 5B and Figure 7 It can be seen that for the second eigenvalues ​​obtained by these two arithmetic operations, the second eigenvalue of the real finger is smaller, while the second eigenvalue of the fake finger is larger. Therefore, the calculated second eigenvalue can be used to distinguish whether the sensed finger is a real finger or a fake finger.

[0034] Figure 8 for Figure 1 Schematic diagram of the controller processing the first characteristic value and the second characteristic value of the fingerprint image. Please refer to Figure 1 and Figure 8 , the controller 230 is used to determine the two-dimensional coordinates formed by the first characteristic value and the second characteristic value (such as Figure 8The controller 230 determines whether the fingerprint image is from a real finger. If the two-dimensional coordinates fall within the predetermined area F1 or F2, the controller 230 determines that the fingerprint image is from a fake finger. In this way, the under-screen fingerprint sensing device 200 of this embodiment can achieve an anti-spoofing effect in fingerprint sensing, thereby enhancing the security of the user's personal data.

[0035] The above-mentioned embodiment judges the real finger and the fake finger by combining the results of the first eigenvalue and the second eigenvalue. However, in other embodiments, the real finger and the fake finger can be judged based on the first eigenvalue alone, or the real finger and the fake finger can be judged based on the second eigenvalue alone. For example, the controller 230 can only calculate the first eigenvalue and determine whether the first eigenvalue falls within the preset range. If the first eigenvalue falls within the preset range, it is determined that the fingerprint image is from a real finger; if the first eigenvalue falls outside the preset range, it is determined that the fingerprint image is from a fake finger. Alternatively, the controller 230 can only calculate the second eigenvalue and determine whether the second eigenvalue falls within the preset range. If the second eigenvalue falls within the preset range, it is determined that the fingerprint image is from a real finger; if the second eigenvalue falls outside the preset range, it is determined that the fingerprint image is from a fake finger.

[0036] In one embodiment, the controller 230 is, for example, a central processing unit (CPU), a microprocessor (microprocessor), a digital signal processor (DSP), a programmable controller, a programmable logic device (PLD), or other similar devices or a combination of these devices, and the present invention is not limited thereto. In addition, in one embodiment, each function of the controller 230 can be implemented as a plurality of program codes. These program codes are stored in a memory and executed by the controller 230. Alternatively, in one embodiment, each function of the controller 230 can be implemented as one or more circuits. The present invention is not limited to implementing each function of the controller 230 in software or hardware.

[0037] In summary, in the under-screen fingerprint sensing device of the embodiment of the present invention, the controller performs an inner product between the grayscale values ​​on two axes in different directions of the fingerprint image and the Haar feature, and / or calculates the grayscale average value of the central area of ​​the fingerprint image and the grayscale average values ​​of the two peripheral areas on both sides of the central area, and thereby determines whether the fingerprint image is from a real finger or a fake finger. Therefore, the under-screen fingerprint sensing device of the embodiment of the present invention can achieve an anti-spoofing effect in fingerprint sensing, thereby improving the security of the user's personal data.

[0038] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An under-screen fingerprint sensing device, characterized in that: Used in conjunction with a display, the display emits illumination light, the illumination light is transmitted to a finger pressed on the display, and the finger reflects the illumination light into signal light carrying fingerprint information of the finger. The under-screen fingerprint sensing device includes: an image sensor disposed below the display, wherein the signal light penetrates the display to form a fingerprint image on the image sensor; and A controller is electrically connected to the image sensor, processes the fingerprint image sensed by the image sensor, performs an inner product between the grayscale values ​​on two axes in different directions of the fingerprint image and the Haar feature, and thereby determines whether the fingerprint image is from a real finger or a fake finger. The controller first performs a fuzzy process on the fingerprint image, and then performs an inner product between the grayscale values ​​on two axes in different directions of the blurred fingerprint image and the Haar feature. The controller adds two inner product values ​​corresponding to two axes in different directions and determines whether the added value falls within a preset range. The controller responds to the fact that the added value falls within the preset range and determines that the fingerprint image is from a real finger, and responds to the fact that the added value falls outside the preset range and determines that the fingerprint image is from a fake finger.

2. The under-screen fingerprint sensing device according to claim 1, characterized in that: The Haar characteristic is a function in which a middle region has a first value and two side regions have a second value, wherein the first value is greater than the second value.

3. The under-screen fingerprint sensing device according to claim 1, characterized in that: The display comprises: a glass cover, wherein the finger is pressed on the glass cover; A linear polarizer, disposed between the glass cover and the image sensor; a phase retardation film disposed between the linear polarizer and the image sensor; and The organic light emitting diode display panel is arranged between the phase delay film and the image sensor.

4. The under-screen fingerprint sensing device according to claim 1, characterized in that: The device also includes a lens, which is arranged on the path of the signal light and located between the display and the image sensor.

5. An under-screen fingerprint sensing device, characterized in that: Used in conjunction with a display, the display emits illumination light, the illumination light is transmitted to a finger pressed on the display, and the finger reflects the illumination light into signal light carrying fingerprint information of the finger. The under-screen fingerprint sensing device includes: an image sensor disposed below the display, wherein the signal light penetrates the display to form a fingerprint image on the image sensor; and A controller is electrically connected to the image sensor, processes the fingerprint image sensed by the image sensor, calculates the grayscale average value of the central area of ​​the fingerprint image and the grayscale average values ​​of the two peripheral areas on both sides of the central area, and thereby determines whether the fingerprint image is from a real finger or a fake finger. The grayscale average value of the central area is R2, and the grayscale average values ​​of the two peripheral areas are R1 and R3 respectively. The controller is used to perform arithmetic operations on R1, R2, and R3 to obtain characteristic values, and to determine whether the characteristic values ​​fall within a preset range. The controller responds to the characteristic value falling within the preset range and determines that the fingerprint image is from a real finger, and responds to the characteristic value falling outside the preset range and determines that the fingerprint image is from a fake finger.

6. The under-screen fingerprint sensing device according to claim 5, characterized in that: The central area and the two peripheral areas are arranged on a diagonal line of the fingerprint image.

7. The under-screen fingerprint sensing device according to claim 5, characterized in that: The arithmetic operation is (R2-R1)+(R2-R3), or the arithmetic operation is R2-R1-R3.

8. The under-screen fingerprint sensing device according to claim 5, characterized in that: The display comprises: a glass cover, wherein the finger is pressed on the glass cover; A linear polarizer, disposed between the glass cover and the image sensor; a phase retardation film disposed between the linear polarizer and the image sensor; and The organic light emitting diode display panel is arranged between the phase delay film and the image sensor.

9. The under-screen fingerprint sensing device according to claim 5, characterized in that: The device also includes a lens, which is arranged on the path of the signal light and located between the display and the image sensor.

10. An under-screen fingerprint sensing device, characterized in that: Used in conjunction with a display, the display emits illumination light, the illumination light is transmitted to a finger pressed on the display, and the finger reflects the illumination light into signal light carrying fingerprint information of the finger. The under-screen fingerprint sensing device includes: an image sensor disposed below the display, wherein the signal light penetrates the display to form a fingerprint image on the image sensor; and A controller is electrically connected to the image sensor, processes the fingerprint image sensed by the image sensor, performs inner product between the grayscale values ​​on two axes in different directions of the fingerprint image and the Haar feature to obtain a first result, the controller is used to calculate the grayscale average value of the central area of ​​the fingerprint image and the grayscale average values ​​of two peripheral areas on both sides of the central area to obtain a second result, and the controller is used to combine the first result and the second result to determine whether the fingerprint image is from a real finger or a fake finger, The controller is used to first perform blurring on the fingerprint image, and then perform inner product between the grayscale values ​​on two axes in different directions of the blurred fingerprint image and the Haar feature. The controller adds two inner product values ​​corresponding to two axes in different directions to obtain a first eigenvalue. The grayscale average value of the central area is R2, and the grayscale average values ​​of the two peripheral areas are R1 and R3 respectively. The controller is used to perform arithmetic operations on R1, R2, and R3 to obtain a second eigenvalue. The controller is used to determine whether the two-dimensional coordinates formed by the first eigenvalue and the second eigenvalue fall within a predetermined area. The controller responds to the two-dimensional coordinates falling within the predetermined area and determines that the fingerprint image is from a real finger, and the controller responds to the two-dimensional coordinates falling outside the predetermined area and determines that the fingerprint image is from a fake finger.

11. The under-screen fingerprint sensing device according to claim 10, characterized in that: The Haar characteristic is a function in which a middle region has a first value and two side regions have a second value, wherein the first value is greater than the second value.

12. The under-screen fingerprint sensing device according to claim 10, characterized in that: The arithmetic operation is (R2-R1)+(R2-R3), or the arithmetic operation is R2-R1-R3.

13. The under-screen fingerprint sensing device according to claim 10, characterized in that: The central area and the two peripheral areas are arranged on a diagonal line of the fingerprint image.

14. The under-screen fingerprint sensing device according to claim 10, characterized in that: The display comprises: a glass cover, wherein the finger is pressed on the glass cover; A linear polarizer, disposed between the glass cover and the image sensor; a phase retardation film disposed between the linear polarizer and the image sensor; and The organic light emitting diode display panel is arranged between the phase delay film and the image sensor.

15. The under-screen fingerprint sensing device according to claim 10, characterized in that: The device also includes a lens, which is arranged on the path of the signal light and located between the display and the image sensor.