Stereoscopic vision fingerprint signature method and device, electronic equipment and storage medium

Abstract

The invention relates to a stereoscopic vision fingerprint signature method and device, electronic equipment and a storage medium. The stereoscopic vision fingerprint signature method comprises the following steps: S1, acquiring a target image containing a hand and an image of signature handwriting; s2, inputting the target image into a plurality of pre-designed algorithms to obtain a normal graph of the target image; and S3, importing the normal graph into a 3D engine to realize stereoscopic vision fingerprint signature. According to the stereoscopic vision fingerprint signature method, multimedia stereoscopic vision fingerprint signature interaction experience is adopted, under the conditions that fingerprints and palmprint of a user are not collected and user privacy is protected, multimedia stereoscopic vision fingerprint signature can be well achieved, user experience is improved through fingerprint signature interaction, and user experience is improved. The attention of users is attracted in a unique and effective mode, the brand popularity and influence are improved, visitors can participate in exhibition or activities in a more visual mode, interactivity and interestingness are enhanced, and the experience feeling of participants is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer image processing, and in particular to a method, device, electronic device and storage medium for stereoscopic visual handprint signature. Background Art

[0002] Multimedia three-dimensional visual handprint signature interactive experience, an art that is full of creativity and technological integration.

[0003] However, in design, how to skillfully use various multimedia elements, such as dynamic images, interactive interfaces, etc., to create an engaging atmosphere, and through innovative interactive methods, allow users to actively participate and feel the charm and value of the brand has become an urgent problem to be solved. Summary of the invention

[0004] The technical problem to be solved by the present invention is how to use multimedia elements to achieve a multimedia stereoscopic visual handprint signature interactive experience.

[0005] In order to solve the above technical problems, according to one aspect of the present invention, a method for stereoscopic visual handprint signature is provided, which includes the following steps: step S1, obtaining a target image containing a hand and an image of a signature handwriting; step S2, inputting the target image into a plurality of pre-designed algorithms to obtain a normal map of the target image; step S3, importing the normal map into a 3D engine to realize stereoscopic visual handprint signature.

[0006] According to an embodiment of the present invention, step S1 may include the following steps: step S11, detecting the hand pressing the display screen through the infrared frame set on the display screen, and controlling the back of the hand top view camera to capture the appropriate back of the hand picture, completing the back of the hand picture acquisition; step S12, uploading the back of the hand picture to Baidu Cloud hand key point recognition, and obtaining the coordinate data of the hand key points in the back of the hand picture, wherein the hand key points include: the key point of the index finger root of the palm, the key point of the middle finger root, the center point of the wrist, and the key point of the index finger joint adjacent to the root of the index finger.

[0007] According to an embodiment of the present invention, step S2 may include the following steps: step S21, using a first algorithm to analyze a back of hand picture to obtain a black and white picture of the back of hand picture, wherein the back of hand pixels are white and the non-back of hand pixels are black, wherein the first algorithm includes: taking out the color values ​​of all pixel points of the back of hand picture to be processed, first converting the color from the RGB color space to the HSV color space, and then performing a dot multiplication with the HSV color space of the skin color, and judging whether the colors are similar based on the dot multiplication result, wherein similar pixels return to white, otherwise return to black; step S22, further processing the back of hand black and white picture through a second algorithm to make the white in the back of hand black and white picture The colored area is more like a palm, and a black and white picture of the palm is obtained, wherein the second algorithm includes: drawing a circle with the key point at the root of the middle finger of the palm as the center and the distance from the center point of the wrist to the key point at the root of the middle finger of the palm as the radius, and changing the pixels of the arm area to the left of the center point of the wrist and below the circle to black; drawing a circle with the key point at the root of the index finger as the center and the distance from the center point of the wrist to the key point at the root of the index finger as the radius, and changing the pixels of the arm area to the right of the center point of the wrist and below the circle to black, wherein the functions involved include: the function expression of the circle: (xa)+(yb)>r2, wherein there are three parameters a, b, r, that is, the coordinates of the center of the circle are (a, b), and r is the radius. This function is used to determine whether points outside the circular area will be excluded.

[0008] According to an embodiment of the present invention, step S2 may further include the following steps: step S23, performing black and white gradient transition processing on the edge of the palm in the black and white palm picture by a third algorithm to obtain a first palm grayscale image, wherein the third algorithm includes: obtaining the colors of all pixel points of the image to be processed, averaging the colors of the pixels connected to the surrounding pixels, and then returning the average value; step S24, further processing the first palm grayscale image by the third algorithm to increase the black and white transition width of the edge in the white area of ​​the palm, to obtain a second palm grayscale image; step S25, merging the first palm grayscale image and the second palm grayscale image by a fourth algorithm, according to the black and white palm picture, the white area takes the pixel color of the second palm grayscale image, and the black area takes the pixel color of the first palm grayscale image, to obtain a third palm grayscale image, wherein the fourth algorithm includes: palm black and white picture, using color array c Color_0 is used to store the binary number of each pixel in the black and white palm picture, where black is 0 and white is 1; the first palm grayscale image is represented by the color array color_1, which is used to store the RGBA value of each pixel in the first palm grayscale image; the second palm grayscale image is represented by the color array color_2, which is used to store the RGBA value of each pixel in the second palm grayscale image; the third palm grayscale image is represented by the color array color_3, which is used to store the RGBA value of each pixel in the new third palm grayscale image after calculation; the value in the color_0 array is judged, if it is black, the RGBA value of the corresponding pixel in color_1 is assigned to color_3, otherwise the RGBA value of the corresponding pixel in color_2 is assigned to color_3, and finally color_3 is output.

[0009] According to an embodiment of the present invention, step S2 may also include the following steps: step S26, based on the coordinate data of the hand key points in the back of the hand picture, the joint texture grayscale image made by the artist is fused to obtain a third palm grayscale image through the fifth algorithm to obtain a fourth palm grayscale image, wherein the fifth algorithm includes: adding the joint texture grayscale image at the position of the key point at the root of the index finger of the palm, and determining the position and direction of the added joint texture grayscale image based on the direction from the key point at the root of the index finger to the key point of the index finger joint according to the position coordinates of the key point at the root of the index finger of the palm.

[0010] According to an embodiment of the present invention, step S2 may further include the following steps: step 27, using the sixth algorithm to generate a black-and-white image of handwriting based on the trajectory of the signature pen on the touch screen; using the third algorithm to convert the black-and-white image of handwriting into a grayscale image to obtain a grayscale image of handwriting; using the fourth algorithm to merge the grayscale image of handwriting with the fourth palm grayscale image into a final grayscale image, wherein the sixth algorithm includes: the movement of the signature pen on the touch screen is essentially sending the real-time two-dimensional coordinate position of the signature pen on the touch screen to the signature program, and the handwriting is formed by connecting the real-time coordinate positions in sequence, which results in the handwriting having sharp corners and not being smooth. Therefore, in order to improve the smoothness of the handwriting and imitate the effect of real handwriting, the smooth effect of the handwriting is achieved by using a second-order Bezier curve and a third-order Bezier curve, wherein the path of the second-order Bezier curve is tracked by the function B(t) of the given points P0, P1, and P2:

[0011] B(t)=(1-t) 2 P0+2t(1-t)P1+t 2 P2, t∈[0, 1]

[0012] Among them, P0, P1, and P2 are three points in the given plane, P0 is the starting point of the curve, P2 is the end point of the curve, and P1 is the control point; the value range of t is 0 to 1, and B is a point on the smooth curve;

[0013] The third-order Bezier curve starts at P0 and goes to P1, and then goes to P3 from P2. It usually does not go through P1 or P2. These two points are just there to provide directional information. The distance between P0 and P1 determines how long the curve goes to P2 before turning to P3. The parameter form of the third-order Bezier curve is:

[0014] B(t)=P0(1-t) 3 +3P1t(1-t) 2 +3P2t 2 (1-t)+P3t 3 , t∈[0,1]

[0015] According to an embodiment of the present invention, step S2 may further include the following steps: step S28, converting the final grayscale image into a normal map through the seventh algorithm, wherein the seventh algorithm includes: converting the grayscale image into a height map, converting each pixel value in the grayscale image into a corresponding height value, which can be achieved by mapping the grayscale value to a certain height range; calculating the normal direction of each pixel, for each pixel in the height map, estimating the normal direction of the pixel by calculating the height difference of the surrounding pixels; using the central difference method or the forward difference method to calculate the partial derivative, and then according to the partial derivative, The derivative calculates the normal direction; normalizes the normal direction. When the calculated normal direction is not a unit vector, it needs to be normalized so that its length is 1; converts the normal direction to a normal map. The normalized normal direction is converted to an RGB color value. Usually, three channels in the RGB color space are used to represent the three components of the normal; this is achieved by mapping the components of the normal direction to the range of RGB color values; generates a normal map, and saves the converted RGB color values ​​as a normal map; the normal map is used to simulate the concave and convex effects of the surface in rendering to make the object look more realistic.

[0016] According to a second aspect of the present invention, there is provided a device for stereoscopic handprint signature, comprising: a display screen, a camera, and a touch screen, wherein the display screen has an infrared frame for detecting a hand pressing the display screen, and the camera is arranged at a top-view position on the back of the hand, for acquiring a target image containing the hand; a signature pen and a touch screen for acquiring an image of the signature handwriting; a computing unit, wherein the computing unit is used to adopt the stereoscopic handprint signature method as described above, input the target image, and acquire a normal map of the target image; and import the normal map into a 3D engine to realize stereoscopic handprint signature.

[0017] According to the third aspect of the present invention, there is provided an electronic device, comprising: a memory, a processor, and a stereoscopic vision handprint signature program stored in the memory and executable on the processor, wherein the steps of the above-mentioned stereoscopic vision handprint signature method are implemented when the stereoscopic vision handprint signature program is executed by the processor.

[0018] According to a fourth aspect of the present invention, a computer storage medium is provided, wherein a stereoscopic vision handprint signature program is stored on the computer storage medium, and when the stereoscopic vision handprint signature program is executed by a processor, the steps of the above-mentioned stereoscopic vision handprint signature method are implemented.

[0019] Compared with the prior art, the technical solution provided by the embodiments of the present invention can at least achieve the following beneficial effects:

[0020] The method for stereoscopic vision handprint signature according to the present invention adopts a multimedia stereoscopic vision handprint signature interactive experience, which does not involve collecting users' fingerprints and palm prints and protects users' privacy. It can well realize multimedia stereoscopic vision handprint signature, improve user experience through handprint signature interaction, attract users' attention in a unique and effective way, enhance brand awareness and influence, and allow visitors to participate in exhibitions or activities in a more intuitive way, enhance interactivity and fun, and improve the participants' sense of experience.

[0021] The method of stereoscopic handprint signature according to the present invention is environmentally friendly and energy-saving. Compared with the traditional paper signature method, electronic signature interaction does not require consumables such as paper and ink, which reduces resource consumption and waste generation, and meets the requirements of modern society for energy conservation and environmental protection. Highly customized, the system supports highly customized settings, and can be personalized according to the theme and content of the exhibition. Whether it is the style of the signature interface, the choice of background pattern or the arrangement of interactive links, it can be adjusted and optimized according to actual needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments will be briefly introduced below. Obviously, the drawings in the following description only relate to some embodiments of the present invention, but are not intended to limit the present invention.

[0023] Figure 1 is a flow chart showing a method for stereoscopic visual handprint signature according to an embodiment of the present invention;

[0024] Figure 2 is a palm diagram illustrating an embodiment of the present invention;

[0025] Figure 3 is a schematic diagram showing a second-order Bezier curve;

[0026] Figure 4 is a schematic diagram showing a third-order Bezier curve. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] Unless otherwise defined, the technical or scientific terms used herein shall have the common meanings understood by persons with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the patent application specification and claims of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "a" do not indicate a quantity limitation, but indicate the existence of at least one.

[0029] Figure 1 is a flow chart showing a method for stereoscopic visual handprint signature according to an embodiment of the present invention; Figure 2 is a diagram illustrating a palm according to an embodiment of the present invention.

[0030] like Figure 1 and Figure 2 As shown, the method of stereoscopic visual handprint signature includes the following steps: step S1, obtaining a target image containing a hand and an image of a signature handwriting; step S2, inputting the target image into a plurality of pre-designed algorithms to obtain a normal map of the target image; step S3, importing the normal map into a 3D engine to realize stereoscopic visual handprint signature.

[0031] According to one or some embodiments of the present invention, step S1 includes the following steps:

[0032] Step S11, the infrared frame set on the display screen detects that the hand is pressing the display screen, and controls the hand back top view camera to capture the appropriate picture of the hand back, thereby completing the picture collection of the hand back.

[0033] Step S12, upload the back of the hand image to Baidu Cloud hand key point recognition, and obtain the coordinate data of the hand key points in the back of the hand image, wherein the hand key points include: key point 1 at the base of the index finger of the palm, key point 2 at the base of the middle finger, center point 3 of the wrist, and key point 4 of the index finger joint adjacent to the base of the index finger.

[0034] According to one or some embodiments of the present invention, step S2 includes the following steps:

[0035] Step S21, using the first algorithm to analyze the back of the hand image, and obtain a black and white image of the back of the hand, wherein the back of the hand pixels are white and the non-back of the hand pixels are black, wherein the first algorithm includes: taking out the color values ​​of all pixels of the back of the hand image to be processed, first converting the color from the RGB color space to the HSV color space, and then performing a dot product with the HSV color space of the skin color, and judging whether the colors are similar based on the dot product result, wherein similar pixels are returned to white, otherwise they are returned to black.

[0036] Step S22, further processing the black-and-white image of the back of the hand through a second algorithm to make the white area in the black-and-white image of the back of the hand look more like a palm, and obtaining a black-and-white image of the palm, wherein the second algorithm includes: drawing a circle with the key point 2 at the root of the middle finger of the palm as the center and the distance from the center point 3 of the wrist to the key point 2 at the root of the middle finger of the palm as the radius, and changing the 5 pixels of the arm area to the left of the center point 3 of the wrist and below the circle to black; drawing a circle with the key point 1 at the root of the index finger as the center and the distance from the center point 3 of the wrist to the key point 1 at the root of the index finger as the radius, and changing the 5 pixels of the arm area to the right of the center point 3 and below the circle to black.

[0037] Among them, the second algorithm involves functions including:

[0038] Function expression of circle: (xa)+(yb)>r2,

[0039] Among them, there are three parameters a, b, and r, that is, the coordinates of the center of the circle are (a, b), and r is the radius.

[0040] This function is used to exclude points outside the circular area.

[0041] According to one or some embodiments of the present invention, step S2 further includes the following steps:

[0042] Step S23, performing a black-and-white gradient transition process on the edge of the palm in the black-and-white palm image by a third algorithm to obtain a first palm grayscale image, wherein the third algorithm includes: obtaining the colors of all pixels of the image to be processed, averaging the colors of the pixels connected to the surrounding pixels, and then returning the average value:

[0043] Color(0,0)Color(1,0)Color(2,0)

[0044] Color(0,1)Color(1,1)Color(2,1)

[0045] Color(0,2)Color(1,2)Color(2,2).

[0046] Step S24: further process the first palm grayscale image by a third algorithm to increase the black-white transition width of the edge in the white area of ​​the palm, thereby obtaining a second palm grayscale image.

[0047] Step S25, using the fourth algorithm, the first palm grayscale image and the second palm grayscale image are merged, and according to the black and white palm image, the white area takes the pixel color of the second palm grayscale image, and the black area takes the pixel color of the first palm grayscale image, to obtain a third palm grayscale image.

[0048] Wherein, the fourth algorithm includes:

[0049] The black and white palm image is represented by the color array color_0, which is used to store the binary number of each pixel in the black and white palm image, where black is 0 and white is 1.

[0050] The first palm grayscale image is represented by the color array color_1, which is used to store the RGBA value of each pixel in the first palm grayscale image.

[0051] The second palm grayscale image is represented by the color array color_2, which is used to store the RGBA value of each pixel in the second palm grayscale image.

[0052] The third palm grayscale image is represented by the color array color_3, which is used to store the RGBA value of each pixel of the new third palm grayscale image after the calculation.

[0053] Judge the value in the color_0 array. If it is black, assign the RGBA value of the corresponding pixel in color_1 to color_3. Otherwise, assign the RGBA value of the corresponding pixel in color_2 to color_3, and finally output color_3.

[0054] The code is as follows:

[0055]

[0056] According to one or some embodiments of the present invention, step S2 further includes the following steps:

[0057] Step S26, according to the coordinate data of the hand key points in the back of the hand picture, the joint texture grayscale image made by the artist is merged into the third palm grayscale image through the fifth algorithm to obtain the fourth palm grayscale image, wherein the fifth algorithm includes: adding the joint texture grayscale image at the position of the key point 1 at the root of the index finger of the palm, and according to the position coordinates of the key point 1 at the root of the index finger of the palm, the position and direction of the added joint texture grayscale image are determined from the direction from the key point 1 at the root of the index finger to the key point 4 of the index finger joint adjacent to the root of the index finger.

[0058] According to one or some embodiments of the present invention, step S2 further includes the following steps:

[0059] Step 27, using the sixth algorithm, generate a black-and-white image of the handwriting along the motion trajectory of the signature pen on the touch screen; using the third algorithm, convert the black-and-white image of the handwriting into a grayscale image to obtain a handwriting grayscale image; using the fourth algorithm, merge the handwriting grayscale image with the fourth palm grayscale image into a final grayscale image.

[0060] Among them, the sixth algorithm includes:

[0061] The movement of the signature pen on the touch screen essentially sends the real-time two-dimensional coordinate position of the signature pen on the touch screen to the signature program. The handwriting is formed by connecting the real-time coordinate positions in sequence. This results in the handwriting being angular and not smooth. Therefore, in order to improve the smoothness of the handwriting and imitate the effect of real handwriting, the second-order Bezier curve and the third-order Bezier curve are used to achieve the smooth effect of the handwriting.

[0062] Figure 3 is a schematic diagram showing a second-order Bezier curve; Figure 4 is a schematic diagram showing a third-order Bezier curve.

[0063] The path of the second-order Bezier curve is traced by the function B(t) at given points P0, P1, and P2:

[0064] B(t)=(1-t) 2 P0+2t(1-t)P1+t 2 P2, t∈[0, 1]

[0065] like Figure 3 As shown, P0, P1, and P2 are three points in a given plane, P0 is the starting point of the curve, P2 is the end point of the curve, and P1 is the control point; the value range of t is 0 to 1, and B is a point on the smooth curve.

[0066] like Figure 4 As shown, the third-order Bezier curve starts at P0 and goes to P1, and from P2 to P3; it generally does not pass through P1 or P2; these two points are just there to provide directional information; the distance between P0 and P1 determines how long the curve is in the direction of P2 before turning to P3.

[0067] The parametric form of the third-order Bezier curve is:

[0068] B(t)=P0(1-t) 3 +3P1t(1-t) 2 +3P2t 2 (1-t)+P3t 3 , t∈[0,1]

[0069] According to one or some embodiments of the present invention, step S2 further includes the following steps:

[0070] Step S28: convert the final grayscale image into a normal image through the seventh algorithm, wherein the seventh algorithm includes:

[0071] Converting a grayscale image to a height map, converting each pixel value in the grayscale image to a corresponding height value, can be achieved by mapping the grayscale value to a certain height range.

[0072] Calculate the normal direction of each pixel. For each pixel in the height map, estimate the normal direction of the pixel by calculating the height difference of its surrounding pixels; use the central difference method or the forward difference method to calculate the partial derivatives, and then calculate the normal direction based on the partial derivatives.

[0073] Normalize the normal direction. When the calculated normal direction is not a unit vector, it needs to be normalized so that its length is 1.

[0074] The normal direction is converted into a normal map, and the normalized normal direction is converted into an RGB color value. The three channels in the RGB color space are usually used to represent the three components of the normal; this is achieved by mapping the components of the normal direction to the range of RGB color values.

[0075] Generate a normal map and save the converted RGB color values ​​as a normal map; the normal map is used to simulate the bumpy effect of the surface in rendering to make the object look more realistic.

[0076] The method for stereoscopic vision handprint signature according to the present invention adopts a multimedia stereoscopic vision handprint signature interactive experience, which does not involve collecting users' fingerprints and palm prints and protects users' privacy. It can well realize multimedia stereoscopic vision handprint signature, improve user experience through handprint signature interaction, attract users' attention in a unique and effective way, enhance brand awareness and influence, and allow visitors to participate in exhibitions or activities in a more intuitive way, enhance interactivity and fun, and improve the participants' sense of experience.

[0077] According to a second aspect of the present invention, there is provided a device for stereoscopic handprint signature, comprising: a display screen, a camera, and a touch screen, wherein the display screen has an infrared frame for detecting a hand pressing the display screen, and the camera is arranged at a top-view position on the back of the hand, for acquiring a target image containing the hand; a signature pen and a touch screen for acquiring an image of the signature handwriting; a computing unit, wherein the computing unit is used to adopt the stereoscopic handprint signature method as described above, input the target image, and acquire a normal map of the target image; and import the normal map into a 3D engine to realize stereoscopic handprint signature.

[0078] The method of stereoscopic handprint signature according to the present invention is environmentally friendly and energy-saving. Compared with the traditional paper signature method, electronic signature interaction does not require consumables such as paper and ink, which reduces resource consumption and waste generation, and meets the requirements of modern society for energy conservation and environmental protection. Highly customized, the system supports highly customized settings, and can be personalized according to the theme and content of the exhibition. Whether it is the style of the signature interface, the choice of background pattern or the arrangement of interactive links, it can be adjusted and optimized according to actual needs.

[0079] According to another aspect of the present invention, there is provided a device for stereoscopic visual handprint signature, comprising: a memory, a processor, and a stereoscopic visual handprint signature program stored in the memory and executable on the processor, wherein the steps of the above-mentioned stereoscopic visual handprint signature method are implemented when the stereoscopic visual handprint signature program is executed by the processor.

[0080] According to the present invention, a computer storage medium is also provided.

[0081] A stereoscopic vision handprint signature program is stored on the computer storage medium, and when the stereoscopic vision handprint signature program is executed by the processor, the steps of the above-mentioned stereoscopic vision handprint signature method are implemented.

[0082] Among them, the method implemented when the stereoscopic vision handprint signature program running on the processor is executed can refer to the various embodiments of the stereoscopic vision handprint signature method of the present invention, and will not be repeated here.

[0083] The present invention also provides a computer program product.

[0084] The computer program product of the present invention includes a stereoscopic vision handprint signature program, which implements the steps of the stereoscopic vision handprint signature method described above when executed by a processor.

[0085] Among them, the method implemented when the stereoscopic vision handprint signature program running on the processor is executed can refer to the various embodiments of the stereoscopic vision handprint signature method of the present invention, and will not be repeated here.

[0086] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0087] The above description is merely an exemplary embodiment of the present invention and is not intended to limit the protection scope of the present invention. The protection scope of the present invention is determined by the appended claims.

Claims

1. A method for three-dimensional visual handprint signature, comprising the following steps: Step S1, obtaining a target image containing a hand and an image of a signature handwriting; Step S2, inputting the target image into a plurality of pre-designed algorithms to obtain a normal map of the target image; Step S3, importing the normal map into the 3D engine to realize the stereoscopic visual handprint signature.

2. The method for three-dimensional visual handprint signature as claimed in claim 1, wherein: Step S1 includes the following steps: Step S11, detecting that a hand is pressing the display screen through an infrared frame set on the display screen, and controlling the hand back top view camera to capture an appropriate picture of the hand back, thereby completing the collection of the hand back picture; Step S12, upload the back of the hand picture to Baidu Cloud hand key point recognition, and obtain the coordinate data of the hand key points in the back of the hand picture, wherein the hand key points include: the key point at the base of the index finger, the key point at the base of the middle finger, the center point of the wrist, and the key point of the index finger joint adjacent to the base of the index finger.

3. The method of stereoscopic handprint signature as claimed in claim 2, wherein: Step S2 includes the following steps: Step S21, using the first algorithm to analyze the back of the hand picture, and obtaining a black and white picture of the back of the hand picture, wherein the back of the hand pixels are white and the non-back of the hand pixels are black, wherein the first algorithm includes: taking out the color values ​​of all pixel points of the back of the hand picture to be processed, first converting the color from the RGB color space to the HSV color space, and then performing a dot product with the HSV color space of the skin color, and judging whether the colors are similar according to the dot product result, wherein similar pixel points return white, otherwise return black; Step S22, further processing the back of hand black and white picture by a second algorithm, so that the white area in the back of hand black and white picture is more like a palm, and obtaining a palm black and white picture, wherein the second algorithm includes: taking the key point at the root of the middle finger of the palm as the center of the circle and the distance from the center of the wrist to the key point at the root of the middle finger of the palm as the radius to draw a circle, and changing the pixels of the arm area to the left of the center of the wrist and below the circle to black; taking the key point at the root of the index finger as the center of the circle and the distance from the center of the wrist to the key point at the root of the index finger as the radius to draw a circle, and changing the pixels of the arm area to the right of the center of the wrist and below the circle to black, wherein the functions involved include: Function expression of circle: (xa)+(yb)>r2, Among them, there are three parameters a, b, and r, that is, the coordinates of the center of the circle are (a, b), and r is the radius. This function is used to exclude points outside the circular area.

4. The method for three-dimensional visual handprint signature as claimed in claim 3, wherein: Step S2 also includes the following steps: Step S23, performing a black-white gradient transition process on the edge of the palm in the black-and-white palm image by a third algorithm to obtain a first palm grayscale image, wherein the third algorithm includes: obtaining the colors of all pixels of the image to be processed, averaging the colors of the pixels connected to the surrounding pixels, and then returning the average value; Step S24: further process the first palm grayscale image by a third algorithm to increase the black-white transition width of the edge of the white area of ​​the palm, thereby obtaining a second palm grayscale image; Step S25: The first palm grayscale image and the second palm grayscale image are fused by a fourth algorithm. According to the black and white palm image, the white area takes the pixel color of the second palm grayscale image, and the black area takes the pixel color of the first palm grayscale image, so as to obtain a third palm grayscale image. The fourth algorithm includes: The black and white palm image is represented by the color array color_0, which is used to store the binary number of each pixel in the black and white palm image, where black is 0 and white is 1; The first palm grayscale image is represented by the color array color_1, which is used to store the RGBA value of each pixel in the first palm grayscale image; The second palm grayscale image is represented by the color array color_2, which is used to store the RGBA value of each pixel in the second palm grayscale image; The third palm grayscale image is represented by the color array color_3, which is used to store the RGBA value of each pixel of the new third palm grayscale image after the calculation; Judge the value in the color_0 array. If it is black, assign the RGBA value of the corresponding pixel in color_1 to color_3. Otherwise, assign the RGBA value of the corresponding pixel in color_2 to color_3, and finally output color_3.

5. The method of stereoscopic visual handprint signature as claimed in claim 4, wherein: Step S2 also includes the following steps: Step S26, according to the coordinate data of the hand key points in the back of the hand picture, the joint texture grayscale image made by the artist is merged into the third palm grayscale image through the fifth algorithm to obtain the fourth palm grayscale image, wherein the fifth algorithm includes: adding the joint texture grayscale image at the position of the key point at the root of the index finger of the palm, and determining the position and direction of the added joint texture grayscale image according to the position coordinates of the key point at the root of the index finger of the palm and the direction from the key point at the root of the index finger to the key point of the index finger joint.

6. The method for stereoscopic handprint signature as claimed in claim 5, wherein: Step S2 also includes the following steps: Step 27: Generate a black-and-white handwriting image by using the sixth algorithm to track the motion of the signature pen on the touch screen; and convert the black-and-white handwriting image into a grayscale image by using the third algorithm to obtain a grayscale handwriting image. The handwriting grayscale image and the fourth palm grayscale image are fused into a final grayscale image through the fourth algorithm. Among them, the sixth algorithm includes: The smooth effect of handwriting is achieved through the second-order Bezier curve and the third-order Bezier curve. The path of the second-order Bezier curve is traced by the function B(t) at given points P0, P1, and P2: B(i)=(1-t) 2 P0+2t(1-t)P1+t 2 P2,t∈[0、1] Among them, P0, P1, and P2 are three points in the given plane, P0 is the starting point of the curve, P2 is the end point of the curve, and P1 is the control point; the value range of t is 0 to 1, and B is a point on the smooth curve; A third-order Bezier curve starts at P0 and goes to P1, and from P2 to P3; it does not usually go through P1 or P2; these two points are just there to provide directional information; the distance between P0 and P1 determines how long the curve is in the direction of P2 before turning to P3. The parametric form of the third-order Bezier curve is: B(t)=P0(1-t) 3 +3P1t(1-t) 2 +3P2t 2 (1-t)+P3t 3 ,t∈[0,1]。 7. The method for stereoscopic handprint signature as claimed in claim 6, wherein: Step S2 also includes the following steps: Step S28: convert the final grayscale image into a normal image through the seventh algorithm, wherein the seventh algorithm includes: Convert the grayscale image to a height map, convert each pixel value in the grayscale image to a corresponding height value, and achieve this by mapping the grayscale value to the height; Calculate the normal direction of each pixel. For each pixel in the height map, estimate the normal direction of the pixel by calculating the height difference of its surrounding pixels; use the central difference method or the forward difference method to calculate the partial derivative, and then calculate the normal direction based on the partial derivative; Normalize the normal direction. When the calculated normal direction is not a unit vector, it needs to be normalized so that its length is 1. Convert the normal direction to a normal map, convert the normalized normal direction to an RGB color value, usually using three channels in the RGB color space to represent the three components of the normal; this is achieved by mapping the components of the normal direction to the range of RGB color values; Generate a normal map and save the converted RGB color values ​​as a normal map; the normal map is used to simulate the bumpy effect of the surface in rendering to make the object look more realistic.

8. A device for three-dimensional visual handprint signature, comprising: A display screen, a camera, and a touch screen. The display screen has an infrared frame for detecting a hand pressing the display screen. The camera is arranged at a top-view position of the back of the hand to obtain a target image containing the hand. Signature pen and touch screen for acquiring an image of the signature handwriting; A computing unit, wherein the computing unit is used to adopt the stereoscopic visual handprint signature method as described in any one of claims 1 to 7, input the target image, obtain the normal map of the target image; and import the normal map into a 3D engine to realize stereoscopic visual handprint signature.

9. An electronic device, comprising: A memory, a processor, and a stereoscopic vision handprint signature program stored in the memory and executable on the processor, wherein the stereoscopic vision handprint signature program, when executed by the processor, implements the steps of the stereoscopic vision handprint signature method as described in any one of claims 1 to 7.

10. A computer storage medium, wherein: The computer storage medium stores a stereoscopic vision handprint signature program, which, when executed by a processor, implements the steps of the stereoscopic vision handprint signature method as described in any one of claims 1 to 7.

Images