TWI836240B - Optical fingerprint recognition method and touch display and information processing device utilizing the same - Google Patents

Abstract

An optical fingerprint identification method, including: performing a binarization operation on one of the first optical fingerprint images of sample fingerprints collected from a collectible range of a touch screen to generate a first binarized image, and Perform a linear regression calculation on the first binarized image to obtain a first fingerprint inclination angle of the sample fingerprint on a two-dimensional coordinate plane in the collectible image range; for the collectible image on the touch screen A second optical fingerprint image of a fingerprint to be identified is collected from the range and is subjected to the binarization operation to generate a second binarized image, and the linear regression calculation is performed on the second binarized image. To obtain a second fingerprint inclination angle of the fingerprint to be identified on the two-dimensional coordinate plane of the collectible image range; and to make the first optical fiber optic according to the angle difference between the second fingerprint inclination angle and the first fingerprint inclination angle. The fingerprint image performs a relative rotation on the second optical fingerprint image to obtain two frames of fingerprint images with the same axial direction, and the two frames of fingerprint images are matched and compared.

Description

Optical fingerprint recognition method and touch display and information processing device using the same

The present invention relates to fingerprint recognition, and more particularly to an optical fingerprint recognition method capable of improving recognition accuracy.

Optical fingerprint recognition is often set up in the display module of a smart phone in the form of under-screen detection to collect the reflected images of the valleys and ridges of the fingerprint for comparison with a frame of recorded fingerprint image. Please refer to Figure 1, which is a schematic diagram of a conventional under-screen optical fingerprint recognition device. As shown in Figure 1, a display screen 10 has a sampleable range 11, and a finger 20 presses a local range 11a in the sampleable range 11 to perform fingerprint recognition.

The operation of the known under-screen optical fingerprint recognition device includes: recording a user's fingerprint in advance to obtain a template fingerprint image and save it; and during recognition, collecting a fingerprint image of a fingerprint to be verified and matching it with the template fingerprint image. Generally speaking, if the same finger is recorded and then verified, the result should be a match; if it is not the same finger, it is not a match.

However, in extreme environments (strong light, low temperature, etc.), the collected fingerprint images, as shown in Figure 2, will be severely interfered, the effective area of the fingerprint image will become smaller, and the fingerprints will no longer be continuous, resulting in The number of incorrect matches increases, the recognition rate decreases, and the risk of use increases. In addition, the general fingerprint recognition algorithm gradually rotates the entire frame of the fingerprint image by an angle and then calculates the matching error with the entire frame of the template fingerprint image to obtain multiple error values, that is, blindly testing each rotation angle. The entire frame of the fingerprint image is matched with the entire frame of the template fingerprint image, and then the smallest of the error values is compared with a threshold to determine whether there is a match. However, this angle blind test method cannot ensure that the fingerprint and the sample fingerprint are compared at the same tilt angle, which may lead to misjudgment, especially when the effective area of a frame of fingerprint image to be identified becomes smaller and the frame is to be When there is an inclination angle difference between the fingerprint and the template fingerprint, the effective area for identifying the fingerprint image may match a local texture of the template fingerprint, resulting in misjudgment.

To solve the above problems, a novel optical fingerprint recognition method is urgently needed in this field.

The main purpose of the present invention is to disclose an optical fingerprint recognition method, which can improve the accuracy of fingerprint recognition by finding the axial angle difference between a fingerprint to be recognized and a sample fingerprint, performing a rotation operation according to the axial angle difference to align the axes of the two fingerprints and then performing matching and comparison.

Another object of the present invention is to disclose a touch display that can provide a fingerprint recognition function with a high accuracy rate through the above-mentioned optical fingerprint recognition method.

Another object of the present invention is to disclose an information processing device that can provide a fingerprint recognition function with high accuracy through the above-mentioned touch display.

To achieve the aforementioned purpose, an optical fingerprint recognition method is proposed, which includes:

A first optical fingerprint image of a sample fingerprint acquired within a scannable range of a touch screen is binarized to generate a first binarized image, and a linear regression calculation is performed on the first binarized image to obtain a first fingerprint tilt angle of the sample fingerprint on a two-dimensional coordinate plane within the scannable range;

Perform the binarization operation on a second optical fingerprint image of a fingerprint to be identified that is collected in the collectible image range of the touch screen to generate a second binarized image, and the second binary value Perform the linear regression calculation on the image to obtain a second fingerprint inclination angle of the fingerprint to be identified on the two-dimensional coordinate plane of the collectible image range; and

According to the angle difference between the second fingerprint tilt angle and the first fingerprint tilt angle, the first optical fingerprint image is rotated relative to the second optical fingerprint image to obtain two fingerprint images with the same axis, and the two fingerprint images are matched and compared.

In one embodiment, the binarization operation includes the following steps: setting a threshold _GT , _GT is a positive number, when the pixel grayscale value is greater than _GT , its binarization value is 1, when the pixel grayscale value is less than _GT , its binarization value is 0.

In one embodiment, the linear regression calculation includes the following steps: Substituting the coordinates of n pixels with a binary value of 1 into a linear regression tilt angle calculation formula to generate a tilt angle value, n is greater than An integer of 1, the formula for calculating the linear regression tilt angle is:

,in, is the tilt angle value, ( _xi , _yi ) represents the coordinates of any pixel point, is the average X coordinate of n pixels, is the average Y coordinate of n pixels.

In order to achieve the above object, the present invention further proposes a touch display, which has a touch screen and a control unit. The control unit is used to control the screen display, touch scanning and fingerprint collection of the touch screen, wherein the control unit The touch screen has a fingerprint acquisition area, and the control unit has a firmware program to execute the optical fingerprint recognition method as mentioned above.

In a possible embodiment, the touch display may be a micro-LED touch display, a mini-LED touch display, a quantum dot LED touch display or an organic light-emitting diode. Polar body touch display.

In order to achieve the above object, the present invention further proposes an information processing device, which has a central processing unit and a touch display as mentioned above, wherein the central processing unit is used to communicate with the touch display.

In a possible embodiment, the information processing device may be a smart phone, a portable computer, a car computer, a wearable electronic device or an access control device.

In order to enable the Review Committee to further understand the structure, features and purpose of the present invention, the following are attached with drawings and detailed descriptions of preferred specific embodiments.

The principle of the present invention is:

(1) performing a binarization operation on a frame of a first optical image including a sample fingerprint and a frame of a second optical image including a fingerprint to be identified acquired within an optical fingerprint acquisition range of a touch screen to generate two frames of binarized images, and performing a linear regression calculation on the two frames of binarized images to obtain two fingerprint axes on a two-dimensional coordinate plane within the acquisition range; and

(2) Perform a relative rotation on the first optical image and the second optical image according to the angle difference between the two fingerprint axes to obtain two frames of fingerprint images with the same axial direction, and perform a relative rotation on the two frames Fingerprint images are matched and compared.

Based on the above principle, the present invention can improve the accuracy of fingerprint identification to effectively avoid misjudgments.

Please refer to FIG. 3 , which is a flow chart of an embodiment of the optical fingerprint recognition method of the present invention. As shown in Figure 3, the method includes: performing a binarization operation on one of the first optical fingerprint images of the template fingerprint collected in a collectible range of a touch screen to generate a first binarized image. , and perform a linear regression calculation on the first binarized image to obtain a first fingerprint inclination angle of the sample fingerprint on a two-dimensional coordinate plane in the collectible image range (step a); A second optical fingerprint image of a fingerprint to be identified is collected from the image capture range of the touch screen and is subjected to the binarization operation to generate a second binarized image, and the second binarized image is Perform the linear regression calculation to obtain a second fingerprint inclination angle of the fingerprint to be identified on the two-dimensional coordinate plane of the collectible image range (step b); and based on the second fingerprint inclination angle and the first The angle difference between the fingerprint inclination angle causes the first optical fingerprint image to perform a relative rotation on the second optical fingerprint image to obtain two frames of fingerprint images with the same axial direction, and perform a matching comparison on the two frames of fingerprint images. Right (step c).

Please refer to FIGS. 4a-4b together, wherein FIG. 4a is a schematic diagram of an optical fingerprint image of a pressed fingerprint captured within a captureable range of a touch screen; and FIG. 4b is a binarized image generated after performing the binarization operation on the optical fingerprint image.

In addition, the method in Figure 3 will simultaneously collect images around the fingerprint when collecting the fingerprint image. As shown in Figure 4a, the optical fingerprint image includes images of the 1-9 area, where the finger press position is 5.

In addition, when performing the binarization operation, it includes the following steps: setting a threshold _GT , _GT is a positive number, when the pixel grayscale value is greater than _GT , its binarization value is 1, when the pixel grayscale value is not greater than (that is, less than or equal to) _GT , its binarization value is 0. In the binarized image shown in FIG4b, the white area includes n pixels whose binarization value is 1, and n is an integer greater than 1.

，其中，θ為所述傾斜角數值，(x_i，y_i)為像素 點的座標，

為n個像素點的X座標平均值，

為n個像素點的Y座標平均值。 In addition, when performing the linear regression calculation, it includes the following steps: Substituting the coordinates of the n pixel points in the white area into a linear regression tilt angle calculation formula to generate a tilt angle value. The linear regression tilt angle calculation formula is as follows:

, where θ is the tilt angle value, (x _i , y _i ) is the coordinate of the pixel point,

is the average value of the X coordinates of n pixels,

is the average Y coordinate of n pixels.

After obtaining the first fingerprint tilt angle and the second fingerprint tilt angle, the first optical fingerprint image can be relatively rotated with respect to the second optical fingerprint image according to the angle difference between the second fingerprint tilt angle and the first fingerprint tilt angle to obtain two fingerprint images with the same axis, and the two fingerprint images can be matched and compared, wherein the relative rotation can be a first rotation of the first optical fingerprint image with the angle difference, or a second rotation of the second optical fingerprint image with the angle difference which is opposite to the first rotation.

Please refer to FIG. 5 , which is a block diagram of an embodiment of the touch display of the present invention. As shown in Figure 5, a touch display 100 has a touch screen 110 and a control unit 120. The control unit 120 is used to control the screen display, touch scanning and fingerprint collection of the touch screen 110. The touch screen 110 has a fingerprint acquisition area 111, and the control unit 120 has a firmware program to execute the optical fingerprint identification method, specifically including:

(1) The control unit 120 performs the binarization operation on one of the first optical fingerprint images of the sample fingerprints collected in the fingerprint sampling area 111 of the touch screen 110 to generate a binarized image, and The binary image is subjected to the linear regression calculation to obtain a first fingerprint inclination angle of the sample fingerprint on a two-dimensional coordinate plane of the fingerprint acquisition area 111 .

(2) The control unit 120 performs the binary operation on a second optical fingerprint image of a fingerprint to be identified collected in the fingerprint acquisition area 111 of the touch screen 110 to generate a binary image, and The binary image is subjected to the linear regression calculation to obtain a second fingerprint inclination angle of the fingerprint to be identified on the two-dimensional coordinate plane of the fingerprint acquisition area 111; and

(iii) the control unit 120 rotates the first optical fingerprint image relative to the second optical fingerprint image according to the angle difference between the second fingerprint tilt angle and the first fingerprint tilt angle to obtain two fingerprint images with the same axis, and performs matching comparison on the two fingerprint images.

In addition, the touch display 100 may be a micro LED touch display, a mini LED touch display, a quantum dot LED touch display or an organic light emitting diode touch display.

According to the above description, the present invention further proposes an information processing device. Please refer to FIG. 6, which shows a block diagram of an embodiment of the information processing device of the present invention. As shown in FIG. 6, an information processing device 200 has a central processing unit 210 and a touch display 220, wherein the touch display 220 is implemented by the touch display 100, and the central processing unit 210 is used to communicate with the touch display 220. In addition, the information processing device 200 can be a smart phone, a portable computer, a car computer, a wearable electronic device or an access control device.

Through the above disclosed design, the present invention has the following advantages:

1. The optical fingerprint recognition method of the present invention can improve the accuracy of fingerprint recognition by finding the axial angle difference between a fingerprint to be recognized and a sample fingerprint, performing a rotation operation according to the axial angle difference to align the axes of the two fingerprints and then performing matching and comparison.

2. The touch display of the present invention can provide a fingerprint recognition function with a high accuracy rate through the above-mentioned optical fingerprint recognition method.

3. The information processing device of the present invention can provide a fingerprint recognition function with a high accuracy rate through the above-mentioned touch display.

What is disclosed in this case is a preferred embodiment. Any partial changes or modifications derived from the technical ideas of this case and easily inferred by those familiar with the art will not deviate from the scope of the patent rights of this case.

In summary, this case shows that its purpose, means and effects are all different from the known technology, and it is the first invention that is practical and indeed meets the patent requirements for invention. We sincerely request the review committee to examine this carefully and grant a patent as soon as possible to benefit the society. This is our utmost prayer.

10: display screen 11: imageable range 11a: local range 20: finger 100: touch display 110: touch screen 111: fingerprint image acquisition area 120: control unit 200: information processing device 210: central processing unit 220: touch display Step a: perform a binarization operation on a first optical fingerprint image of a sample fingerprint acquired within an imageable range of a touch screen to generate a first binarized image, and perform a linear regression calculation on the first binarized image to obtain a first fingerprint tilt angle of the sample fingerprint on a two-dimensional coordinate plane within the imageable range. Step b: performing the binarization operation on a second optical fingerprint image of a fingerprint to be identified captured in the imageable range of the touch screen to generate a second binarized image, and performing the linear regression calculation on the second binarized image to obtain a second fingerprint tilt angle of the fingerprint to be identified on the two-dimensional coordinate plane of the imageable range. Step c: performing a relative rotation of the first optical fingerprint image to the second optical fingerprint image according to the angle difference between the second fingerprint tilt angle and the first fingerprint tilt angle to obtain two fingerprint images with the same axis, and matching and comparing the two fingerprint images.

FIG. 1 is a schematic diagram of a conventional under-screen optical fingerprint recognition device. Figure 2 is a schematic diagram of a fingerprint image collected in an extreme environment. FIG. 3 is a flow chart of an embodiment of the optical fingerprint recognition method of the present invention. FIG. 4a is a schematic diagram of an optical fingerprint image of a pressed fingerprint captured in a captureable range of a touch screen. Figure 4b is a binarized image generated after performing a binarization operation on the optical fingerprint image in Figure 4a. FIG. 5 is a block diagram of an embodiment of the touch display of the present invention. FIG. 6 is a block diagram of an embodiment of the information processing device of the present invention.

Step a: Perform a binarization operation on a first optical fingerprint image of a sample fingerprint acquired within a sampleable range of a touch screen to generate a first binarized image, and perform a linear regression calculation on the first binarized image to obtain a first fingerprint tilt angle of the sample fingerprint on a two-dimensional coordinate plane within the sampleable range.

Step b: Perform the binarization operation on a second optical fingerprint image of a fingerprint to be identified captured in the imageable range of the touch screen to generate a second binarized image, and perform the linear regression calculation on the second binarized image to obtain a second fingerprint tilt angle of the fingerprint to be identified on the two-dimensional coordinate plane of the imageable range

Step c: Perform a relative rotation on the first optical fingerprint image and the second optical fingerprint image according to the angle difference between the second fingerprint inclination angle and the first fingerprint inclination angle to obtain two frames of fingerprint images with the same axial direction. image, and perform matching and comparison on the two frames of fingerprint images.

Claims (12)

An optical fingerprint recognition method includes: performing a binarization operation on a first optical fingerprint image of a sample fingerprint acquired within a sampleable range of a touch screen to generate a first binarized image, and performing a linear regression calculation on the first binarized image to obtain a first fingerprint tilt angle of the sample fingerprint on a two-dimensional coordinate plane within the sampleable range; performing the binarization operation on a second optical fingerprint image of a fingerprint to be recognized acquired within the sampleable range of the touch screen to generate a second binarized image, and performing the linear regression calculation on the second binarized image. Calculate to obtain a second fingerprint tilt angle of the fingerprint to be identified on the two-dimensional coordinate plane of the imageable range; and perform a relative rotation of the first optical fingerprint image to the second optical fingerprint image according to the angle difference between the second fingerprint tilt angle and the first fingerprint tilt angle to obtain two fingerprint images with the same axis, and match and compare the two fingerprint images; wherein the relative rotation is to make the first optical fingerprint image perform a first rotation with the angle difference, or to make the second optical fingerprint image perform a second rotation opposite to the first rotation with the angle difference. As described in the optical fingerprint recognition method of item 1 of the patent application scope, the binarization operation includes the following steps: setting a threshold _GT , _GT is a positive number, when the pixel grayscale value is greater than _GT , its binarization value is 1, when the pixel grayscale value is not greater than _GT , its binarization value is 0. The optical fingerprint recognition method as described in item 2 of the patent application, wherein the linear regression calculation includes the following steps: substituting the coordinates of the n pixels with a binary value of 1 into a linear regression tilt angle calculation formula In order to generate an inclination angle value, n is an integer greater than 1. The linear regression inclination angle calculation formula is as follows:

, where θ is the tilt angle value, ( _xi , y _i ) represents the coordinates of any pixel point,

is the average X coordinate of n pixels,

is the average Y coordinate of n pixels. A touch display has a touch screen and a control unit. The control unit is used to control the screen display, touch scanning and fingerprint collection of the touch screen. The touch screen has a fingerprint collection area. And the control unit has a firmware program to execute an optical fingerprint identification method. The method includes: performing a binarization operation on one of the first optical fingerprint images of the template fingerprints collected in the fingerprint sampling area. To generate a first binarized image, and perform a linear regression calculation on the first binarized image to obtain a first fingerprint of the sample fingerprint on a two-dimensional coordinate plane in the fingerprint sampling area Tilt angle; perform the binarization operation on a second optical fingerprint image of a fingerprint to be identified collected in the fingerprint sampling area to generate a second binarized image, and the second binary value The linear regression calculation is performed on the image to obtain a second fingerprint inclination angle of the fingerprint to be identified on the two-dimensional coordinate plane of the fingerprint sampling area; and based on the second fingerprint inclination angle and the first fingerprint The difference in tilt angle causes the first optical fingerprint image to perform a relative rotation on the second optical fingerprint image to obtain two frames of fingerprint images with the same axial direction, and perform matching and comparison on the two frames of fingerprint images. ; Wherein, the relative rotation is to allow the first optical fingerprint image to perform a first rotation with the angle difference, or to allow the second optical fingerprint image to perform one of the opposite directions with the first rotation using the angle difference. Second spin. The touch display as described in item 4 of the patent application, wherein the binarization operation includes the following steps: setting a threshold G _T , G _T is a positive number, and when the pixel grayscale value is greater than G _T , the second The binarized value is 1. When the pixel grayscale value is not greater than G _T , its binarized value is 0. The touch display as described in item 5 of the patent application, wherein the linear regression calculation comprises the following steps: Substituting the coordinates of the n pixel points whose binarized values are 1 into a linear regression tilt angle calculation formula to generate a tilt angle value, where n is an integer greater than 1, and the linear regression tilt angle calculation formula is as follows:

, where θ is the tilt angle value, ( _xi , _yi ) is the coordinate of any pixel point,

is the average value of the X coordinates of the n pixels,

is the average value of the Y coordinates of the n pixels. The touch display described in item 4 of the patent application scope is a touch display selected from a group consisting of a micro-LED touch display, a mini LED touch display, a quantum dot LED touch display and an organic LED touch display. An information processing device has a central processing unit and a touch display, wherein the central processing unit is used to communicate with the touch display, the touch display has a touch screen and a control unit, the control unit is used to control the screen display, touch scanning and fingerprint collection of the touch screen, the touch screen has a fingerprint collection area, and the control unit has a firmware program An optical fingerprint recognition method is implemented based on the optical fingerprint recognition method, the method comprising: performing a binarization operation on a first optical fingerprint image of a sample fingerprint acquired in the fingerprint acquisition area to generate a first binarized image, and performing a linear regression calculation on the first binarized image to obtain a first fingerprint tilt angle of the sample fingerprint on a two-dimensional coordinate plane of the fingerprint acquisition area; ... A second optical fingerprint image of a fingerprint to be identified is subjected to the binarization operation to generate a second binarized image, and the second binarized image is subjected to the linear regression calculation to obtain a second fingerprint tilt angle of the fingerprint to be identified on the two-dimensional coordinate plane of the fingerprint sampling area; and the first fingerprint tilt angle is adjusted according to the angle difference between the second fingerprint tilt angle and the first fingerprint tilt angle. An optical fingerprint image performs a relative rotation on the second optical fingerprint image to obtain two fingerprint images with the same axis, and the two fingerprint images are matched and compared; wherein the relative rotation is to make the first optical fingerprint image perform a first rotation with the angle difference, or to make the second optical fingerprint image perform a second rotation opposite to the first rotation with the angle difference. The information processing device as described in item 8 of the patent application, wherein the binarization operation includes the following steps: setting a threshold G _T , G _T is a positive number, and when the pixel grayscale value is greater than G _T , the second The binarized value is 1. When the pixel grayscale value is not greater than G _T , its binarized value is 0. The information processing device as described in Item 9 of the patent application, wherein the linear regression calculation includes the following steps: substituting the coordinates of the n pixels with a binary value of 1 into a linear regression tilt angle calculation formula To generate an inclination angle value, n is an integer greater than 1. The linear regression inclination angle calculation formula is as follows:

, where θ is the tilt angle value, ( _xi , _yi ) is the coordinate of any pixel point,

is the average X coordinate of n pixels,

is the average Y coordinate of n pixels. For example, the information processing device described in item 8 of the patent application scope, wherein the touch display is composed of a micro-light emitting diode touch display, a mini-light emitting diode touch display, and a quantum dot light-emitting diode touch display. A touch display selected from a group consisting of a control display and an organic light emitting diode touch display. The information processing device described in Item 8 of the patent application scope is a device selected from the group consisting of a smart phone, a portable computer, a car computer, a wearable electronic device and an access control device.

Images