WO2021077903A1

WO2021077903A1 - Finger image collection method and apparatus, and storage medium

Info

Publication number: WO2021077903A1
Application number: PCT/CN2020/112697
Authority: WO
Inventors: 王鹏飞; 邓家璧; 罗晓宇; 陈向文
Original assignee: 珠海格力电器股份有限公司; 珠海联云科技有限公司
Priority date: 2019-10-22
Filing date: 2020-08-31
Publication date: 2021-04-29
Also published as: CN110751105A; CN110751105B

Abstract

Disclosed are a finger image collection method and apparatus, and a storage medium. The method comprises: when an image collection instruction sent by a requester is received, adjusting a camera to a plurality of preset positions to obtain finger images photographed by the camera at the plurality of preset positions; respectively calculating a finger region area in each finger image, and setting the finger image with the largest finger region area to be a target finger image; and determining whether the target finger image includes a finger pulp or finger dorsum by means of a trained convolutional neural network (CNN) algorithm, and if a finger pulp is included, sending the target finger image to the requester.

Description

Finger image collection method, device and storage medium

This application claims the priority of a patent application filed to the State Intellectual Property Office of China on October 22, 2019 with the application number 201911008514.0 and the invention title "A method, device and storage medium for finger image acquisition".

Technical field

This application relates to the field of image recognition technology, and in particular to a method, device and storage medium for acquiring finger images.

Background technique

With the continuous advancement of national consumption upgrades, more and more attention has been paid to the security field; finger vein recognition is currently a relatively cutting-edge biometric technology, which has the advantages of fast recognition speed, good performance, and features that are not easy to forge; however, finger vein recognition To a large extent, the recognition of finger veins depends on the quality of the finger vein image collection; in the prior art finger vein collection equipment, in order to improve the quality of the collected finger images, the user’s finger placement and posture are mandatory. The experience is poor; if the user does not place the finger position according to the regulations during the finger image collection process, the quality of the obtained finger image will be poor, which will seriously affect the accuracy of finger vein recognition.

Summary of the invention

The purpose of this application is to provide a finger image collection method, device and storage medium, which solves the problem of poor finger image quality caused by incorrect placement of the user’s finger in current finger vein collection equipment, which affects the finger The problem of the accuracy of vein recognition.

According to one aspect of the present application, there is provided a finger image acquisition method, which specifically includes the following steps:

Adjusting the camera to a plurality of preset positions when receiving an image acquisition instruction sent by the requesting party, so as to obtain finger images taken by the camera at the plurality of preset positions;

Calculate the area of the finger area in each finger image separately, and set the finger image with the largest area of the finger area as the target finger image;

The trained CNN convolutional neural network algorithm is used to determine whether the finger pad or the back of the finger is included in the target finger image. If the finger pad is included, the target finger image is sent to the requesting party.

Optionally, the step of adjusting the camera to a plurality of preset positions to obtain finger images taken by the camera at the plurality of preset positions specifically includes:

After the camera captures an image at a preset position, it cuts out the finger image from the captured image according to the preset size.

Optionally, the separately calculating the area of the finger area in each finger image, and setting the finger image with the largest area of the finger area as the target finger image, specifically includes:

Binarize each finger image according to the gray value of each pixel in each finger image; among them, the gray value of the pixel in the finger image after binarization only includes 0 and 255 ；

Count the total number of pixels with a pixel value of 255 in each finger image after binarization processing, and set the finger image with the largest total count as the target finger image.

Optionally, the performing binarization processing on each finger image according to the gray value of each pixel in each finger image, specifically includes:

For a finger image, determine in turn whether the gray value of each pixel in the finger image is greater than or equal to the preset gray threshold, if yes, adjust the gray value of the corresponding pixel to 255, if not, change Adjust the gray value of the corresponding pixel to 0.

Optionally, the judging whether the finger belly or the back of the finger is included in the target finger image by using the trained CNN convolutional neural network algorithm specifically includes:

Using the trained CNN convolutional neural network algorithm, calculate the probability value of the back of the finger contained in the target finger image;

It is determined whether the probability value is greater than or equal to a preset probability threshold; if it is, it is determined that the back of the finger included in the target finger image is determined; if not, it is determined that the pad of the finger is included in the target finger image.

Optionally, the method further includes:

If the target finger image includes the back of the finger, the camera is adjusted based on the preset position corresponding to the target finger image, and the finger image including the finger pad is re-photographed through the adjusted camera, and the The re-photographed finger image is sent to the requesting party.

Optionally, if the back of the finger is included in the target finger image, adjusting the camera based on the preset position corresponding to the shooting of the target finger image specifically includes:

If the target finger image includes the back of the finger, the rotation angle of the finger image including the finger pad is calculated based on the preset position corresponding to the target finger image, and the rotation angle is adjusted according to the rotation angle. The camera.

According to another aspect of the present application, a finger image acquisition device is also provided, which specifically includes the following components:

webcam;

A rotatable housing configured to house the camera;

A drive motor configured to drive the rotatable housing to rotate;

A controller configured to control the drive motor to drive the rotatable housing to rotate and to control the camera to shoot. The controller includes a memory and a processor, and the memory stores a controller that can run on the processor. A program, when the processor executes the program, the steps of the finger image acquisition method described above can be realized.

Optionally, the camera is installed on the inner wall of the rotatable housing.

According to another aspect of the present application, there is also provided a computer-readable storage medium on which a program is stored, and the program is executed by a processor to implement the steps of the finger image acquisition method described above.

The finger image collection method, device and storage medium provided in this application can be applied to various types of finger vein collection and recognition systems, and solve the problem of the current finger vein collection equipment caused by incorrect placement of the user’s finger. The quality of the obtained finger image is poor, which affects the accuracy of finger vein recognition; this application can not only increase the user’s finger placement freedom when using it, but also ensure the stability of the collected finger image and improve the user Experience of use.

Description of the drawings

By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only used for the purpose of illustrating the preferred embodiments, and are not considered as a limitation to the application. Also, throughout the drawings, the same reference symbols are used to denote the same components. In the attached picture:

FIG. 1 is a schematic diagram of an optional flow chart of the finger image acquisition method provided in the first embodiment;

2 is a schematic diagram of an optional program module of the finger image acquisition device provided in the second embodiment;

3 is a schematic diagram of an optional hardware architecture of the controller provided in the second embodiment;

Fig. 4 is a schematic diagram of an optional program module of the finger image acquisition device provided in the third embodiment.

Detailed ways

In order to make the purpose, technical solutions, and advantages of this application clearer, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the application, and are not used to limit the application. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of this application.

Example one

The embodiment of the present application provides a finger image acquisition method, which is applied to a finger vein acquisition device. As shown in FIG. 1, the method specifically includes the following steps:

Step S101: When an image acquisition instruction sent by the requesting party is received, the camera is adjusted to a plurality of preset positions to obtain finger images taken by the camera at the plurality of preset positions.

Specifically, the adjusting the camera to a plurality of preset positions to obtain finger images taken by the camera at the plurality of preset positions specifically includes:

In this embodiment, in order to reduce the amount of calculation in the subsequent calculations, a more appropriate part of the image will be selected as the finger image from the captured image. For example, the size of the image captured by the camera is 1080×810, according to The size is preset, 300 to 800 are captured from the leftmost side of the captured image, and the area from 200 to 600 is captured from the uppermost edge of the captured image to obtain a finger image.

Step S102: Calculate the area of the finger area in each finger image, and set the finger image with the largest area of the finger area as the target finger image.

Specifically, calculating the area of the finger area in each finger image separately and setting the finger image with the largest area of the finger area as the target finger image specifically includes:

Step A1: Binarize each finger image according to the gray value of each pixel in each finger image; wherein, the gray value of the pixel in the finger image after the binarization process only includes 0 and 255;

Step A2: Count the total number of pixels with a pixel value of 255 in each finger image after binarization processing, and set the finger image with the largest total count as the target finger image.

Further, the performing binarization processing on each finger image according to the gray value of each pixel in each finger image, specifically includes:

For a finger image, sequentially determine whether the gray value of each pixel in the finger image is greater than or equal to the preset gray threshold, and if the gray value of the first pixel is greater than or equal to the preset gray threshold, then The gray value of the first pixel is adjusted to 255, and if the gray value of the second pixel is less than the preset gray threshold, the gray value of the second pixel is adjusted to 0.

Step S103: Use the trained CNN convolutional neural network algorithm to determine whether the finger pad or the back of the target finger image is included, and if it is a finger pad, send the target finger image to the requesting party.

Specifically, before step S103, the method further includes:

Obtain a set number of sample finger images; where each sample finger image is marked whether it contains the finger belly or the back of the finger;

According to the set number of sample finger images, the CNN convolutional neural network algorithm is trained to obtain a recognition model that can be used to determine the probability value of the back of the finger contained in the finger image.

Further, the use of the trained CNN convolutional neural network algorithm to determine whether the target finger image contains the finger belly or the finger back specifically includes:

Using the trained CNN convolutional neural network algorithm to calculate the probability value of the back of the finger contained in the target finger image;

Determine whether the probability value is greater than or equal to the preset probability threshold; if the probability value is greater than or equal to the preset probability threshold, determine that the finger back is included in the target finger image; if the probability value is less than the If a preset probability threshold is set, it is determined that the finger pad is included in the target finger image.

Further, the method further includes:

Furthermore, if the back of the finger is included in the target finger image, adjusting the camera based on the preset position corresponding to the shooting of the target finger image specifically includes:

Preferably, the camera is rotated 180 degrees around the finger at a preset position corresponding to the image of the target finger to obtain a finger image including the finger pads.

In this embodiment, when the requesting party needs to collect an image of the user's finger, the user can place the finger in any posture, and the camera will take finger images of the user's finger at different angles according to the preset position, and filter from multiple finger images The finger image with the largest finger area is displayed; since the finger image with the largest finger area may include the finger pad or the back of the finger, in this embodiment, the CNN convolutional neural network algorithm is also used to identify the finger image; If it is the back of the finger, the image of the finger containing the finger pad is sent to the requesting party, so that the requesting party can perform the follow-up finger vein recognition operation based on the finger image containing the finger pad; if it is the back of the finger, adjust the camera's Angle so that the camera captures the image of the finger including the fingertips.

Example two

The embodiment of the present application provides a finger image acquisition device. As shown in FIG. 2, the device specifically includes the following components:

Camera 201;

A rotatable housing 202 configured to house the camera 201; specifically, the camera 201 is installed on the inner wall of the rotatable housing 202;

A drive motor 203 configured to drive the rotatable housing 202 to rotate;

A controller 204 configured to control the driving motor 203 to drive the rotatable housing 202 to rotate and to control the camera 201 to shoot.

Specifically, as shown in FIG. 3, the controller 204 at least includes but is not limited to: a memory and a processor that can communicate with each other through a system bus; it should be pointed out that FIG. 3 only shows a memory and a processor. The controller 204, however, it should be understood that it is not required to implement all the illustrated components, and more or fewer components may be implemented instead.

In this embodiment, the memory (ie, readable storage medium) includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), Read only memory (ROM), electrically erasable programmable read only memory (EEPROM), programmable read only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory may be an internal storage unit of the controller 204, such as a hard disk or a memory of the controller 204. In other embodiments, the memory may also be an external storage device of the controller 204, such as a plug-in hard disk equipped on the controller 204, a smart memory card (Smart Media Card, SMC), and a secure digital (Secure Digital, SD) ) Card, Flash Card, etc. Of course, the memory can also include both

The internal storage unit of the controller 204 also includes its external storage device. In this embodiment, the memory is usually configured to store the operating system and various application software installed in the controller 204. In addition, the memory can also be set to temporarily store various types of data that have been output or will be output.

In some embodiments, the processor may be a central processing unit (Central Processing Unit, CPU), a microcontroller, a microprocessor, or other data processing chips. The processor is usually configured to control the overall operation of the controller 204.

Further, in this embodiment, the memory stores a program that can run on the processor, and the processor can implement the following steps when executing the program:

The trained CNN convolutional neural network algorithm is used to determine whether the finger pad or the back of the finger is included in the target finger image. If it is a finger pad, the target finger image is sent to the requesting party.

For the specific embodiment process of the above method steps, please refer to the first embodiment, and this embodiment will not be repeated here.

Furthermore, in practical applications, when the controller 204 receives the image acquisition instruction sent by the requester, it sends a rotation instruction to the drive motor 203; the drive motor 203 drives the rotatable housing 202 to rotate according to the rotation instruction to The camera 201 on the inner wall of the rotatable housing 202 is adjusted to a plurality of preset positions; the camera 201 captures a finger image at each preset position, and sends the captured finger image to the controller 204.

Example three

The embodiment of the application provides a finger image acquisition device. As shown in FIG. 4, the device specifically includes the following components: a fixing member 401, a connecting member 402, a drive motor 403, a rotatable housing 404, and an image acquisition module 405 , Fingertip slot 406;

Wherein, the fixing member 401 is configured to fix the finger image acquisition device;

One end of the connecting piece 402 is connected with the fixing piece 401, the other end is provided with a fingertip groove 406, and the connecting piece 402 is connected with the rotatable housing 404;

The driving motor 403 is arranged inside the connecting piece 402, and is connected to the rotatable housing 404, and is configured to drive the rotatable housing 404 to rotate;

The rotatable housing 404 is cylindrical, and the inner wall of the rotatable housing 404 is black;

The image acquisition module 405 includes: a camera, a fill light, and is configured to capture images, and the image acquisition module 405 is installed on the inner wall of the rotatable housing 404;

The fingertip groove 406 is configured to place the fingertip of the user's finger, and the user can place the finger in any posture.

In addition, the finger image acquisition device further includes: a controller (not shown in FIG. 4), which is electrically connected to the driving motor 403 and the image acquisition module 405, based on the finger image acquisition device, the implementation of this application The example also provides a finger image acquisition method, which specifically includes the following steps:

Step S1: After the user's fingertip is placed in the fingertip slot, the direction controller is requested to send an image acquisition instruction.

Step S2: After receiving the image acquisition instruction, the controller sends a rotation instruction to the drive motor;

Step S3: After receiving the rotation instruction, the driving motor drives the rotatable housing to rotate according to a preset angle, so that the image acquisition module on the inner wall of the rotatable housing is adjusted to a plurality of preset positions.

Among them, the initial position of the image acquisition module is at the lowest position of the rotatable housing; after receiving the rotation instruction, the drive motor first drives the rotatable housing to rotate clockwise by a set number of preset angles in order to enable image acquisition The module reaches the preset position of the set number; for example, the rotatable shell rotates 9 10 degrees clockwise in turn; before, the drive motor drives the rotatable shell to rotate to the initial position, and then drives the rotatable shell counterclockwise in turn Rotate a set number of set angles so that the image capture module reaches a set number of preset positions; for example, the rotatable housing can be rotated counterclockwise by nine 10 degrees in sequence.

Step S4: The image acquisition module photographs the user's finger at each preset position to obtain finger images of different angles, and sends the photographed finger images to the controller.

Step S5: The controller determines the gray value of each pixel in each finger image, and performs binarization processing on each finger image according to the gray value of each pixel in each finger image.

Specifically, the performing binarization processing on each finger image according to the gray value of each pixel in each finger image includes:

For a finger image, the controller sequentially determines whether the gray value of each pixel in the finger image is greater than or equal to the preset gray threshold, and if so, adjusts the gray value of the corresponding pixel to 255, if not, Then adjust the gray value of the corresponding pixel to 0.

Step S6: The controller separately counts the total number of pixel points with a pixel value of 255 in each finger image after binarization processing, and sets the finger image with the most counted total number as the target finger image.

Through the above steps S5 and S6, the controller can determine the target finger image containing the largest area of the finger from the multiple finger images captured by the image acquisition module. Since the fingers are flat, no matter what posture the user inserts the finger into the fingertip groove, the target finger image containing the largest area of the finger can be determined from multiple finger images taken from different angles.

Step S7: The controller uses the trained CNN convolutional neural network algorithm to determine whether the target finger image contains the finger belly or the finger back.

Since the target finger image may include the finger belly or the finger back, in this embodiment, the CNN convolutional neural network algorithm is also used to perform image recognition on the finger image.

Specifically, step S7 includes:

Step S8: If the finger pad is included in the target finger image, the controller sends the target finger image to the requesting party for the requesting party to perform a finger vein recognition operation based on the target finger image.

Step S9: If the back of the finger is included in the target finger image, the controller calculates the rotation angle at which the finger image including the finger pad can be captured based on the preset position corresponding to the captured target finger image, and the calculated rotation The angle is sent to the drive motor so that the drive motor drives the rotatable housing to rotate according to the rotation angle.

Step S10: The image acquisition module re-photographs the finger image including the finger pad, and forwards the re-photographed finger image to the requesting party through the controller.

In this embodiment, the user can place the finger in the fingertip groove in any posture, find the finger pad of the user's finger by adjusting the position of the image acquisition module, and take an image of the finger directly under the finger pad, thereby Get the best finger image for finger vein recognition.

Example four

This embodiment also provides a computer-readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), only Readable memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disks, optical disks, servers, App application malls, etc., on which computer programs are stored, When the computer program is executed by the processor, the following method steps are implemented:

When receiving the image collection instruction sent by the requesting party, adjust the camera to a plurality of preset positions to obtain finger images taken by the camera at the plurality of preset positions;

It should be noted that in this article, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, It also includes other elements not explicitly listed, or elements inherent to the process, method, article, or device. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, article or device that includes the element.

The serial numbers of the foregoing embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

Through the description of the above embodiments, those skilled in the art can clearly understand that the method of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better.的实施方式。

The above are only the preferred embodiments of the application, and do not limit the scope of the patent for this application. Any equivalent structure or equivalent process transformation made using the content of the description and drawings of the application, or directly or indirectly applied to other related technical fields , The same reason is included in the scope of patent protection of this application.

Industrial applicability

The finger image acquisition method, device and storage medium adopt the camera to take finger images of the user's finger at different angles according to the preset position, and filter the finger image with the largest finger area from multiple finger images, and use CNN convolution The neural network algorithm recognizes the finger image; if it is recognized as a finger pad, the finger image containing the finger pad is sent to the requesting party, so that the requesting party can perform subsequent finger vein recognition operations based on the finger image containing the finger pad; The back of the finger is output, and the angle of the camera is adjusted so that the camera captures the image of the finger including the finger pad. Compared with related technologies, it can not only increase the freedom of the user’s finger placement when using it, but also ensure the captured image. The stability of the finger image improves the user experience.

Claims

A finger image acquisition method, the method includes:

Adjusting the camera to a plurality of preset positions when receiving an image acquisition instruction sent by the requesting party, so as to obtain finger images taken by the camera at the plurality of preset positions;

Calculate the area of the finger area in each finger image separately, and set the finger image with the largest area of the finger area as the target finger image;

The trained CNN convolutional neural network algorithm is used to determine whether the finger pad or the back of the finger is included in the target finger image. If it is a finger pad, the target finger image is sent to the requesting party.
The finger image collection method according to claim 1, wherein said adjusting the camera to a plurality of preset positions to obtain finger images taken by the camera at the plurality of preset positions comprises:

After the camera captures an image at a preset position, it cuts out the finger image from the captured image according to the preset size.
The finger image acquisition method according to claim 1, wherein the calculating the area of the finger area in each finger image separately and setting the finger image with the largest area of the finger area as the target finger image comprises:

Binarize each finger image according to the gray value of each pixel in each finger image; among them, the gray value of the pixel in the finger image after binarization only includes 0 and 255 ；

Count the total number of pixels with a pixel value of 255 in each finger image after binarization processing, and set the finger image with the largest total count as the target finger image.
The finger image acquisition method according to claim 3, wherein the binarization processing on each finger image according to the gray value of each pixel in each finger image respectively comprises:

For a finger image, sequentially determine whether the gray value of each pixel in the finger image is greater than or equal to the preset gray threshold, and if the gray value of the first pixel is greater than or equal to the preset gray threshold, then The gray value of the first pixel is adjusted to 255, and if the gray value of the second pixel is less than the preset gray threshold, the gray value of the second pixel is adjusted to 0.
The finger image acquisition method according to claim 1, wherein, before the trained CNN convolutional neural network algorithm is used to determine whether the target finger image contains the finger belly or the finger back, the method further comprises :

Obtain a set number of sample finger images; where each sample finger image is marked whether it contains the finger belly or the back of the finger;

According to the set number of sample finger images, the CNN convolutional neural network algorithm is trained to obtain the trained CNN convolutional neural network algorithm.
The finger image acquisition method according to claim 1, wherein the judging whether the target finger image contains a finger belly or a finger back using a trained CNN convolutional neural network algorithm comprises:

Using the trained CNN convolutional neural network algorithm to calculate the probability value of the back of the finger contained in the target finger image;

Determine whether the probability value is greater than or equal to the preset probability threshold; if the probability value is greater than or equal to the preset probability threshold, determine that the finger back is included in the target finger image; if the probability value is less than the If a preset probability threshold is set, it is determined that the finger pad is included in the target finger image.
The finger image acquisition method according to claim 1, wherein the method further comprises:

If the target finger image includes the back of the finger, the camera is adjusted based on the preset position corresponding to the target finger image, and the finger image including the finger pad is re-photographed through the adjusted camera, and the The re-photographed finger image is sent to the requesting party.
7. The finger image acquisition method according to claim 7, wherein if the back of the finger is included in the target finger image, adjusting the camera based on a preset position corresponding to the target finger image, comprising :

If the target finger image includes the back of the finger, the rotation angle of the finger image including the finger pad is calculated based on the preset position corresponding to the target finger image, and the rotation angle is adjusted according to the rotation angle. The camera.
A finger image acquisition device, the device includes:

webcam;

A rotatable housing configured to house the camera;

A drive motor configured to drive the rotatable housing to rotate;

A controller configured to control the drive motor to drive the rotatable housing to rotate and to control the camera to shoot. The controller includes a memory and a processor, and the memory stores a controller that can run on the processor. A program, which can implement the steps of the method according to any one of claims 1 to 8 when the processor executes the program.
The finger image acquisition device according to claim 9, wherein:

The camera is installed on the inner wall of the rotatable housing.
A computer-readable storage medium having a program stored thereon, wherein the program is executed by a processor to implement the steps of the method according to any one of claims 1 to 8.
A finger image acquisition device, the device includes:

Set as a fixing member for fixing the finger image acquisition device;

A connecting piece, one end of the connecting piece is connected with the fixing piece, the other end of the connecting piece is provided with a fingertip groove, and the connecting piece is connected with the rotatable housing;

A drive motor configured to drive the rotatable housing to rotate, the drive motor is arranged inside the connecting piece and connected to the rotatable housing;

The image acquisition module set for image shooting, including: camera, fill light;

The controller is configured to perform the steps of the method according to any one of claims 1 to 8, and the controller is electrically connected to the drive motor and the image acquisition module, respectively.
The finger image acquisition device according to claim 12, wherein:

The rotatable shell is cylindrical, and the inner wall of the rotatable shell is black.
The finger image acquisition device according to claim 12, wherein:

The image acquisition module is installed on the inner wall of the rotatable housing.
The finger image acquisition device according to claim 12, wherein:

The initial position of the image acquisition module is located at the lowest position of the rotatable housing.
The finger image acquisition device according to claim 12, wherein:

The driving motor is configured to drive the rotatable housing to rotate a set number of preset angles clockwise after receiving a rotation instruction, so that the image capture module reaches a set number of preset positions.
A finger image acquisition method, the method includes:

When receiving the image acquisition instruction sent by the requesting party, adjust the image acquisition module to a plurality of preset positions to obtain finger images taken by the image acquisition module at the plurality of preset positions;

Calculate the area of the finger area in each finger image separately, and set the finger image with the largest area of the finger area as the target finger image;

The trained CNN convolutional neural network algorithm is used to determine whether the finger pad or the back of the finger is included in the target finger image. If it is a finger pad, the target finger image is sent to the requesting party.