Fingerprint image non-contact acquisition device and method based on multi-camera array

Abstract

The embodiment of the invention provides a fingerprint image non-contact acquisition device and method based on a multi-camera array, and the device comprises a main camera which is connected with a central processing unit and is used for obtaining hand information, transmitting the hand information to the central processing unit, carrying out the photographing after receiving a photographing instruction, and transmitting an obtained palm line image to the central processing unit; the central processing unit is connected with the plurality of slave cameras and the user side, and is used for determining an adjustment instruction according to the hand information, sending the adjustment instruction to the plurality of slave cameras, sending a shooting instruction to the main camera and the plurality of slave cameras at the same time, obtaining palm grain images fed back by each camera, and sending the palm grain images to the user side; obtaining a required fingerprint image based on the palm line image, and transmitting the fingerprint image to the user side; the subordinate cameras are used for adjusting according to the adjusting instruction, performing multi-angle shooting after receiving the shooting instruction, and transmitting obtained multi-angle palm grain images to the central processing unit; and the user side is used for providing an interactive interface according to user requirements after acquiring the fingerprint image.

Description

Technical Field

[0001] The present invention relates to the field of biometric identification technology, and in particular to a non-contact acquisition device and method for handprint images based on a multi-camera array. Background Art

[0002] With the widespread application of biometric technology, fingerprint recognition has been widely used in identity authentication, security checks, criminal investigation and other fields due to its uniqueness and difficulty in forging. However, traditional fingerprint collection equipment usually requires the person being collected to place his hands in a specific posture multiple times. In particular, when completing three-sided fingerprint collection, others are needed to assist in rolling the person's fingers on the lens surface. This not only increases the collection time, but also reduces the user experience and easily leads to the risk of cross-transmission of pathogens. Summary of the invention

[0003] The object of the present invention is to provide a non-contact handprint image acquisition device and method based on a multi-camera array, aiming to solve the above-mentioned problems in the prior art.

[0004] The embodiment of the present invention provides a non-contact handprint image acquisition device based on a multi-camera array, comprising:

[0005] The main camera is connected to the central processor and is arranged in front of the collection area, and is used to obtain hand information in real time, and transmit the hand information to the central processor, and take pictures of the palm lines after receiving the shooting instruction sent by the central processor, and transmit the obtained palm line image to the central processor; wherein the hand information includes whether the hand enters the collection area and the position and posture of the collected hand;

[0006] a central processing unit connected to the main camera, the plurality of slave cameras and the user end, and configured to obtain the hand information, determine the shooting angles and distances of the plurality of slave cameras according to the hand information and send the shooting angles and distances to the plurality of slave cameras as adjustment instructions, send the shooting instructions to the main camera and the plurality of slave cameras at the same time after the adjustment of the plurality of slave cameras is completed, obtain the palm line images fed back by the main camera and the plurality of slave cameras, obtain the required handprint images based on the palm line images, and transmit the handprint images to the user end;

[0007] A plurality of slave cameras are connected to the central processor and are arranged around the acquisition area, and are used to obtain the adjustment instruction, adjust the camera angle and focal length according to the adjustment instruction, perform multi-angle palm pattern shooting after receiving the shooting instruction sent by the central processor, and transmit the obtained multi-angle palm pattern images to the central processor;

[0008] The user end is connected to the central processor and is used to obtain the handprint image and provide a user interaction interface of the real-time acquisition status of the handprint image, image preview and related operation guidance according to user needs.

[0009] The embodiment of the present invention provides a method for non-contact acquisition of handprint images based on a multi-camera array, comprising:

[0010] The hand information is acquired in real time through the main camera, and the hand information is transmitted to the central processor, and the palm lines are photographed after receiving the shooting instruction sent by the central processor, and the obtained palm lines image is transmitted to the central processor; wherein the hand information includes whether the hand enters the collection area and the position and posture of the collected hand;

[0011] The hand information is obtained through a central processor, and shooting angles and distances of multiple slave cameras are determined according to the hand information and sent to the multiple slave cameras as adjustment instructions. After the adjustment of the multiple slave cameras is completed, the shooting instructions are sent to the master camera and the multiple slave cameras at the same time, and the palm line images fed back by the master camera and the multiple slave cameras are obtained, and the required handprint image is obtained based on the palm line image, and the handprint image is transmitted to the user end;

[0012] Acquire the adjustment instruction through multiple slave cameras, adjust the camera angle and focal length according to the adjustment instruction, perform multi-angle palm pattern shooting after receiving the shooting instruction sent by the central processor, and transmit the obtained multi-angle palm pattern image to the central processor;

[0013] The handprint image is acquired through the user terminal, and a user interaction interface is provided according to user needs, including real-time acquisition status of the handprint image, image preview and related operation guidance.

[0014] The use of the embodiments of the present invention can include the following beneficial effects: the embodiments of the present invention provide a multi-angle non-contact automatic fingerprint collection system, which can quickly and accurately complete the collection of all-round fingerprint images including flat fingerprints, three-sided fingerprints, side palm prints, half palm prints and four-finger images in a single operation, significantly improving the fingerprint collection efficiency and the quality of the collected fingerprint images, and optimizing the fingerprint collection process, realizing true non-contact collection, and improving the user experience. The device is compactly designed, easy to install and use, and suitable for a variety of occasions. At the same time, the device also has the ability to automatically decontaminate and disinfect the lens surface. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate one or more embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0016] Figure 1 is a schematic diagram of a non-contact acquisition device for handprint images based on a multi-camera array according to an embodiment of the present invention;

[0017] Figure 2 is a schematic diagram of the three-dimensional structure of the device according to an embodiment of the present invention;

[0018] Figure 3 is a schematic diagram of the shooting effect of the main camera of an embodiment of the present invention;

[0019] Figure 4 is a schematic diagram of the shooting effect of a slave camera according to an embodiment of the present invention;

[0020] Figure 5 It is a flow chart of a method for non-contact acquisition of handprint images based on a multi-camera array according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the following will be combined with the drawings in one or more embodiments of this specification to clearly and completely describe the technical solutions in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this document.

[0022] Device Embodiment

[0023] According to an embodiment of the present invention, a non-contact handprint image acquisition device based on a multi-camera array is provided. Figure 1 is a schematic diagram of a non-contact handprint image acquisition device based on a multi-camera array according to an embodiment of the present invention. Figure 1 As shown, the non-contact handprint image acquisition device based on a multi-camera array according to an embodiment of the present invention specifically includes:

[0024] The main camera 10 is connected to the central processor and is arranged in front of the collection area, and is used to obtain hand information in real time, and transmit the hand information to the central processor, and take a picture of the palm lines after receiving a shooting instruction sent by the central processor, and transmit the obtained palm lines image to the central processor; wherein the hand information includes whether the hand enters the collection area and the position and posture of the collected hand;

[0025] The central processor 12 is connected to the main camera, the plurality of slave cameras and the user end, and is used to obtain the hand information, determine the shooting angles and distances of the plurality of slave cameras according to the hand information and send them as adjustment instructions to the plurality of slave cameras, and after the adjustment of the plurality of slave cameras is completed, send the shooting instructions to the main camera and the plurality of slave cameras at the same time, obtain the palm line images fed back by the main camera and the plurality of slave cameras, obtain the required handprint image based on the palm line image, and transmit the handprint image to the user end, specifically for:

[0026] A continuous three-dimensional handprint model is obtained by a 3D reconstruction algorithm based on the palm pattern image, and a desired handprint image is obtained by using an image processing algorithm on the three-dimensional handprint model;

[0027] The handprint images include ten flat fingerprint images, ten three-side fingerprint images, one half palm print image, one side palm print image, one four-finger fingerprint image and one four-finger knuckle print image;

[0028] The three-dimensional handprint model is converted into a plurality of two-dimensional handprint plane images at different angles, the handprint plane images are denoised and enhanced, the processed images are combined using a splicing algorithm to obtain a preliminary handprint image, a neural network model is used to identify key points of the preliminary handprint image, and the preliminary handprint image is segmented by a deep learning segmentation network according to the identified key points, texture features are extracted from the segmented image using image processing technology, the extracted features are matched with different types of fingerprints or palm prints by a machine learning classification algorithm to obtain a matched handprint image, and morphological operations are used to optimize the details of the matched handprint image to obtain the desired handprint image;

[0029] Wherein, the image processing technology is Gabor filter or deep learning model;

[0030] A plurality of slave cameras 14 are connected to the central processor and are arranged around the acquisition area, and are used to obtain the adjustment instruction, adjust the camera angle and focal length according to the adjustment instruction, perform multi-angle palm pattern shooting after receiving the shooting instruction sent by the central processor, and transmit the obtained multi-angle palm pattern images to the central processor;

[0031] The user terminal 16 is connected to the central processor and is used to obtain the handprint image and provide a user interaction interface of the real-time acquisition status of the handprint image, image preview and related operation guidance according to user needs;

[0032] The device further comprises:

[0033] A cleaning device, connected to the central processor, disposed around the master camera and the plurality of slave cameras, comprising a wiping tool and a cleaning fluid, and used to clean the lenses of the master camera and the plurality of slave cameras according to a trigger mechanism preset by the central processor;

[0034] The trigger mechanism includes a time trigger mechanism and an event trigger mechanism;

[0035] The disinfection equipment is connected to the central processor and is arranged around the collection area, and is used to disinfect the collection area through a disinfection tool according to the disinfection time preset by the central processor; wherein the disinfection tool is an ultraviolet lamp or a chemical spray.

[0036] The following is a specific example of a non-contact handprint image acquisition device based on a multi-camera array according to an embodiment of the present invention. Figure 2 As shown, the above technical solution of the embodiment of the present invention is described in detail.

[0037] The embodiment of the present invention proposes a one-time non-contact handprint image acquisition device using a multi-camera array, aiming to solve the problems of low efficiency and low acquisition accuracy in existing handprint acquisition technology. The device adopts advanced multi-camera array technology to automatically complete the acquisition of handprint images in one time without any sense. Compared with the traditional step-by-step acquisition method, the embodiment of the present invention significantly improves the acquisition process. It is no longer necessary to separately collect ten plane fingerprints, ten three-sided fingerprints, one half palm print, one side palm print and one four-finger image. The person being collected only needs to put his hand into the acquisition device to complete the acquisition of all handprint images including plane fingerprints, three-sided fingerprints, side palm prints, half palm prints and four-finger images at one time without any perception or special operation.

[0038] 1. Device composition

[0039] It is mainly composed of multiple high-definition cameras, structured light illumination sources, hand positioning devices, as well as data processing, lens decontamination and disinfection units.

[0040] 2. Working Principle

[0041] The person being collected places his hand in the hand positioning device, and the system automatically adjusts the position of the camera and lighting source. The structured light lighting source projects a specific pattern onto the surface of the hand. Multiple cameras simultaneously capture the light pattern reflected by the surface of the hand from different angles. The data processing unit reconstructs a high-precision three-dimensional model of the handprint based on the captured image. The three-dimensional model image is converted into a 2D grayscale handprint image through an image processing algorithm, and then the image segmentation algorithm is used to segment the entire handprint image into plane fingerprints, three-sided fingerprints, half palm prints, side palm prints, and four-connected fingerprints.

[0042] The collection device proposed in the embodiment of the present invention is specifically as follows:

[0043] 1. Main components of the device

[0044] (1) Main camera

[0045] Function: Such as Figure 3 As shown, whether the hand enters the acquisition area is monitored in real time. If it enters, the system is activated to notify the slave camera to prepare for acquisition.

[0046] Technical implementation: Use highly sensitive sensors or image recognition technology to ensure accurate hand detection.

[0047] (2) Slave Camera

[0048] Layout: Multiple angles are arranged around the main acquisition area to cover different viewing angles.

[0049] Function: Such as Figure 4 As shown, the angle and posture are automatically adjusted according to the hand data provided by the main camera to capture multi-angle images.

[0050] 2. Construction of 3D images

[0051] (1) Image capture: Collaboratively capture hand images from multiple angles to obtain perspective differences.

[0052] (2) Data integration: Use 3D reconstruction algorithms to process multi-angle image data and synthesize a 3D image of the hand.

[0053] 3. Image processing algorithm

[0054] (1) Handprint image generation: The 3D image of the hand is processed by an image processing algorithm to extract high-quality handprint images.

[0055] (2) Stitching algorithm: A stitching algorithm is used to ensure the continuity and integrity of fingerprints in multiple images.

[0056] (3) Automatic image segmentation algorithm: The collected handprint image is segmented into ten flat fingerprints, ten rolled fingerprints, one half palm print each for the left and right hands, side palm prints for the left and right hands, and four-finger fingerprint images for the left and right hands.

[0057] 4. System control and software

[0058] (1) Control logic: The software system automatically schedules the slave cameras based on the master camera signal and processes the image data.

[0059] (2) User interface: A user-friendly interface that provides real-time acquisition status, image preview, and operation guidance.

[0060] 5. Security and Privacy

[0061] (1) Data protection: Comply with data protection regulations and use encrypted storage and transmission technology.

[0062] (2) System security: protection against any external illegal access, including physical and network security measures.

[0063] The embodiment of the present invention proposes a handprint image acquisition device using a multi-camera array, which can collect hand images without any sense at one time, wherein the camera array is configured to simultaneously capture multi-angle images of the hand, including but not limited to planar fingerprints, three-sided fingerprints, half palm prints, side palm prints and full palm prints, so as to achieve efficient and comprehensive handprint data acquisition. The specific implementation process is as follows:

[0064] 1. Hand detection and triggering

[0065] When the person being collected puts his hand into the collection device, the main camera first detects the presence of the hand through a built-in sensor (such as an infrared sensor or a motion sensor). The main camera is then activated and determines the approximate position, size, and intrusion depth of the hand through a preset software algorithm to ensure that the slave camera can collect images from the best angle.

[0066] 2. Slave Camera Layout

[0067] The collection device is also equipped with four slave cameras, located above, below, on the left and right sides of the handprint collection device, to fully cover different sides of the hand. Each slave camera is equipped with an automatic adjustment function, which can fine-tune the angle and focal length according to the specific position and posture of the hand.

[0068] 3. Collaborative collection and angle adjustment

[0069] The main camera sends the hand detection data to the central processing unit, which calculates the optimal shooting angle and distance for each slave camera based on the data, and then sends the data to each slave camera. The slave camera adjusts its camera angle and focal length according to the received instructions to ensure that the hand images at other angles can be captured from the best angle.

[0070] 4. Image acquisition and synthesis

[0071] Once all slave cameras are adjusted into position, they will take high-resolution images of the hand simultaneously or in a predetermined sequence. These images are transmitted back to the central processing unit in real time, where a specialized image processing algorithm synthesizes the images from different angles into a continuous 3D handprint model.

[0072] 5. Image processing and optimization

[0073] The 3D handprint model undergoes further image processing, such as denoising, sharpening and contrast adjustment, to enhance image quality and details. Stitching algorithms are used to ensure seamless transitions between images, especially the fingerprint and handprint parts, to ensure that the collected handprint images are complete and detailed.

[0074] In summary, the acquisition device proposed in the embodiment of the present invention uses a multi-camera array to achieve non-contact detection and image capture of the hand through infrared, laser scanning, high-frequency camera and other technologies. The one-time non-contact automatic acquisition includes ten plane fingerprints, ten three-side fingerprints, one half palm print, one side palm print, one full palm print, one four-finger fingerprint and one four-finger knuckle print handprint images, thereby significantly improving the acquisition efficiency and user experience.

[0075] The device uses an image processing algorithm that can automatically segment ten flat fingerprints, ten three-sided fingerprints, one half palm print, one side palm print, one four-finger fingerprint and one four-finger knuckle print from the entire handprint image collected, ensuring that each part of the image can be identified and stored with the highest accuracy, including:

[0076] 1. Image preprocessing

[0077] Purpose: To improve image quality and reduce errors in subsequent processing.

[0078] Technical implementation: 1) Denoising: Apply filtering algorithms, such as median filtering or Gaussian filtering, to remove image noise; 2) Enhancement: Use histogram equalization or local contrast enhancement technology to improve the contrast and detail visibility of the image.

[0079] 2. Hand detection and positioning

[0080] Purpose: To precisely locate the position of the hand and its components.

[0081] Technical implementation: 1) Hand recognition: Use deep learning network models, such as convolutional neural networks (CNN), to detect and identify hands and fingers; 2) Key point recognition: Identify key points such as finger joints and fingertips to provide accurate reference for subsequent image segmentation.

[0082] 3. Image Segmentation

[0083] Purpose: To segment the handprint image into required parts.

[0084] Technical implementation: 1) Key point-based segmentation: According to the identified key points, use image processing algorithms (such as edge detection and region growing) to segment ten planar fingerprints, ten three-sided fingerprints, etc.; 2) Deep learning segmentation: Use a deep learning segmentation network such as U-Net to further refine the segmentation results to ensure the accuracy and consistency of the segmentation.

[0085] 4. Feature extraction and classification

[0086] Purpose: To extract effective biometric features from segmented images.

[0087] Technical implementation: 1) Feature extraction: Use image processing technology such as Gabor filter or deep learning network model to extract the texture features of fingerprints and palmprints; 2) Classification: Use machine learning classification algorithm (such as support vector machine or random forest) to match the extracted features with different types of fingerprints and palmprints.

[0088] 5. Post-processing and optimization

[0089] Purpose: To optimize image processing results and improve the quality of segmented images.

[0090] Technical implementation: 1) Image post-processing: Apply morphological operations such as corrosion and dilation to process details and improve edge clarity; 2) Optimization algorithm: Combine actual application feedback to continuously adjust and optimize the parameters and processing flow of the AI ​​model.

[0091] Preferably, the acquisition device provided in the embodiment of the present invention further includes an automatic lens surface decontamination system, which adopts an automatic cleaning mechanism for automatically cleaning the camera lens surface regularly or as needed to ensure the image acquisition quality and avoid image quality degradation caused by lens contamination, specifically including:

[0092] 1. Cleaning and disinfection unit design

[0093] Cleaning mechanism: Uses a micro motorized brush or soft wiping material to physically remove dust and smudges from the lens.

[0094] Layout: The cleaning and disinfection unit should be compactly mounted around the camera array, ensuring that every lens can be covered.

[0095] 2. Cleaning and disinfection process

[0096] Trigger modes: (1) Time trigger: set periodic automatic cleaning and disinfection, for example, twice a day; (2) Event trigger: automatically start cleaning and disinfection after detecting lens contamination or completing a certain number of acquisition cycles.

[0097] Cleaning process (physical cleaning): The electric brush or wiping cloth is automatically extended under the drive of the control system and gently wipes the lens surface. For stains that are difficult to remove, a small amount of cleaning liquid can be sprayed to enhance the cleaning effect.

[0098] Preferably, the acquisition device provided in the embodiment of the present invention further comprises an automatic disinfection system for automatically disinfecting the image acquisition area after each acquisition to ensure the hygiene and safety of the acquisition environment and prevent cross-infection of microorganisms, specifically including:

[0099] 1. Disinfection mechanism: Use ultraviolet (UV) light or chemical spray for disinfection. UV light can effectively kill bacteria and viruses, while chemical spray can use alcohol or other disinfectants.

[0100] 2. Disinfection process: After physical cleaning, the ultraviolet lamp is turned on to cover the entire lens surface and continuously irradiated for a certain period of time (for example, 30 seconds) to ensure that all microorganisms are killed; or a chemical spray is used, and the nozzle will evenly spray the disinfectant until the lens surface is covered, and then dry naturally or blow dry with a fan.

[0101] The device proposed in the embodiment of the present invention has an integrated temperature measurement function, which is used to automatically measure the body temperature of the person being collected during the handprint image collection process to implement health monitoring and safety management. It uses a multi-camera array and / or other sensors to collect 3D images of the hands of the person being collected, providing richer spatial information for the handprint image and enhancing the accuracy of the recognition and verification process. The device also adopts a set of image quality evaluation and control mechanisms to evaluate the quality of the collected handprint images in real time, and automatically re-collect when necessary to ensure the high quality standards of the collected images to meet the needs of subsequent processing and analysis.

[0102] Method Embodiment

[0103] According to an embodiment of the present invention, a method for non-contact acquisition of handprint images based on a multi-camera array is provided. Figure 5 FIG. 1 is a flow chart of a method for non-contact acquisition of handprint images based on a multi-camera array according to an embodiment of the present invention. Figure 5 As shown, the non-contact acquisition method of handprint images based on a multi-camera array according to an embodiment of the present invention specifically includes:

[0104] Step S501, acquiring hand information in real time through the main camera, transmitting the hand information to the central processor, photographing the palm lines after receiving the shooting instruction sent by the central processor, and transmitting the obtained palm line image to the central processor; wherein the hand information includes whether the hand enters the collection area and the position and posture of the collected hand;

[0105] Step S502, obtaining the hand information through a central processor, determining the shooting angles and distances of multiple slave cameras according to the hand information and sending them as adjustment instructions to the multiple slave cameras, sending the shooting instructions to the master camera and the multiple slave cameras at the same time after the adjustment of the multiple slave cameras is completed, obtaining the palm line images fed back by the master camera and the multiple slave cameras, obtaining the required handprint image based on the palm line image, and transmitting the handprint image to the user end, specifically including:

[0106] A continuous three-dimensional handprint model is obtained by a 3D reconstruction algorithm based on the palm pattern image, and a desired handprint image is obtained by using an image processing algorithm on the three-dimensional handprint model;

[0107] The handprint images include ten flat fingerprint images, ten three-side fingerprint images, one half palm print image, one side palm print image, one four-finger fingerprint image and one four-finger knuckle print image;

[0108] The method of using an image processing algorithm to obtain the desired handprint image on the three-dimensional handprint model specifically includes:

[0109] The three-dimensional handprint model is converted into a plurality of two-dimensional handprint plane images at different angles, the handprint plane images are denoised and enhanced, the processed images are combined using a splicing algorithm to obtain a preliminary handprint image, a neural network model is used to identify key points of the preliminary handprint image, and the preliminary handprint image is segmented by a deep learning segmentation network according to the identified key points, texture features are extracted from the segmented image using image processing technology, the extracted features are matched with different types of fingerprints or palm prints by a machine learning classification algorithm to obtain a matched handprint image, and morphological operations are used to optimize the details of the matched handprint image to obtain the desired handprint image;

[0110] Wherein, the image processing technology is Gabor filter or deep learning model;

[0111] Step S503, obtaining the adjustment instruction through multiple slave cameras, adjusting the camera angle and focal length according to the adjustment instruction, performing multi-angle palm pattern shooting after receiving the shooting instruction sent by the central processor, and transmitting the obtained multi-angle palm pattern image to the central processor;

[0112] Step S504, acquiring the handprint image through the user terminal, and providing a user interaction interface of the real-time acquisition status of the handprint image, image preview and related operation guidance according to user needs;

[0113] The method further comprises:

[0114] Cleaning the lenses of the master camera and the plurality of slave cameras by a cleaning device according to a trigger mechanism preset by the central processor;

[0115] The trigger mechanism includes a time trigger mechanism and an event trigger mechanism;

[0116] The collection area is disinfected by a disinfection tool according to the disinfection time preset by the central processor through the disinfection device; wherein the disinfection tool is an ultraviolet lamp or a chemical spray.

[0117] The embodiment of the present invention is a method embodiment corresponding to the above-mentioned device embodiment. The specific operation of each step can be understood by referring to the description of the device embodiment, which will not be repeated here.

[0118] In summary, the embodiment of the present invention discloses a device for automatically collecting fingerprint images in one go using a multi-camera array, which changes the traditional method of collecting ten plane fingerprints, ten three-sided fingerprints, one half palm print, one side palm print, and one four-finger image and then finally completing the fingerprint collection. The person being collected only needs to put his hand into the collection device to collect fingerprint images including plane fingerprints, three-sided fingerprints, side palm prints, half palm prints, and four-finger images in one go without feeling. It not only greatly improves the collection efficiency of fingerprint images, but also significantly improves the accuracy of the collected images by accurately synchronizing and analyzing the data obtained by the multi-camera array. An automatic stain removal and disinfection device is also added to the surface of the imaging system lens to reduce the risk of cross infection when different people collect fingerprint images. The embodiment of the present invention can provide a more efficient and accurate fingerprint collection solution for the fields of law, security, and personal identity verification.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A non-contact handprint image acquisition device based on a multi-camera array, characterized in that: include: The main camera is connected to the central processor and is arranged in front of the collection area, and is used to obtain hand information in real time, and transmit the hand information to the central processor, and take pictures of the palm lines after receiving the shooting instruction sent by the central processor, and transmit the obtained palm line image to the central processor; wherein the hand information includes whether the hand enters the collection area and the position and posture of the collected hand; a central processing unit connected to the main camera, the plurality of slave cameras and the user end, and configured to obtain the hand information, determine the shooting angles and distances of the plurality of slave cameras according to the hand information and send the shooting angles and distances to the plurality of slave cameras as adjustment instructions, send the shooting instructions to the main camera and the plurality of slave cameras at the same time after the adjustment of the plurality of slave cameras is completed, obtain the palm line images fed back by the main camera and the plurality of slave cameras, obtain the required handprint images based on the palm line images, and transmit the handprint images to the user end; A plurality of slave cameras are connected to the central processor and are arranged around the acquisition area, and are used to obtain the adjustment instruction, adjust the camera angle and focal length according to the adjustment instruction, perform multi-angle palm pattern shooting after receiving the shooting instruction sent by the central processor, and transmit the obtained multi-angle palm pattern images to the central processor; The user end is connected to the central processor and is used to obtain the handprint image and provide a user interaction interface of the real-time acquisition status of the handprint image, image preview and related operation guidance according to user needs.

2. The device according to claim 1, characterized in that The device further comprises: A cleaning device, connected to the central processor, disposed around the master camera and the plurality of slave cameras, comprising a wiping tool and a cleaning fluid, and used to clean the lenses of the master camera and the plurality of slave cameras according to a trigger mechanism preset by the central processor; The disinfection equipment is connected to the central processor and is arranged around the collection area, and is used to disinfect the collection area through a disinfection tool according to the disinfection time preset by the central processor; wherein the disinfection tool is an ultraviolet lamp or a chemical spray.

3. The device according to claim 1, characterized in that The central processing unit is specifically used for: A continuous three-dimensional handprint model is obtained by a 3D reconstruction algorithm based on the palm pattern image, and a desired handprint image is obtained by using an image processing algorithm on the three-dimensional handprint model; The handprint images include ten plane fingerprint images, ten three-side fingerprint images, one half palm print image, one side palm print image, one four-finger fingerprint image and one four-finger knuckle print image.

4. The device according to claim 3, characterized in that The central processing unit is specifically used for: The three-dimensional handprint model is converted into a plurality of two-dimensional handprint plane images at different angles, the handprint plane images are denoised and enhanced, the processed images are combined using a splicing algorithm to obtain a preliminary handprint image, a neural network model is used to identify key points of the preliminary handprint image, and the preliminary handprint image is segmented by a deep learning segmentation network according to the identified key points, texture features are extracted from the segmented image using image processing technology, the extracted features are matched with different types of fingerprints or palm prints by a machine learning classification algorithm to obtain a matched handprint image, and morphological operations are used to optimize the details of the matched handprint image to obtain the desired handprint image; Wherein, the image processing technology is Gabor filter or deep learning model.

5. The device according to claim 2, characterized in that The trigger mechanism includes a time trigger mechanism and an event trigger mechanism.

6. A non-contact acquisition method of handprint images based on a multi-camera array, characterized in that: include: The hand information is acquired in real time through the main camera, and the hand information is transmitted to the central processor, and the palm lines are photographed after receiving the shooting instruction sent by the central processor, and the obtained palm lines image is transmitted to the central processor; wherein the hand information includes whether the hand enters the collection area and the position and posture of the collected hand; The hand information is obtained through a central processor, and shooting angles and distances of multiple slave cameras are determined according to the hand information and sent to the multiple slave cameras as adjustment instructions. After the adjustment of the multiple slave cameras is completed, the shooting instructions are sent to the master camera and the multiple slave cameras at the same time, and the palm line images fed back by the master camera and the multiple slave cameras are obtained, and the required handprint image is obtained based on the palm line image, and the handprint image is transmitted to the user end; Acquire the adjustment instruction through multiple slave cameras, adjust the camera angle and focal length according to the adjustment instruction, perform multi-angle palm pattern shooting after receiving the shooting instruction sent by the central processor, and transmit the obtained multi-angle palm pattern image to the central processor; The handprint image is acquired through the user terminal, and a user interaction interface is provided according to user needs, including real-time acquisition status of the handprint image, image preview and related operation guidance.

7. The method according to claim 6, characterized in that The method further comprises: Cleaning the lenses of the master camera and the plurality of slave cameras by a cleaning device according to a trigger mechanism preset by the central processor; The collection area is disinfected by a disinfection tool according to the disinfection time preset by the central processor through the disinfection device; wherein the disinfection tool is an ultraviolet lamp or a chemical spray.

8. The method according to claim 6, characterized in that The required handprint image is obtained based on the palm pattern image, specifically including: A continuous three-dimensional handprint model is obtained by a 3D reconstruction algorithm based on the palm pattern image, and a desired handprint image is obtained by using an image processing algorithm on the three-dimensional handprint model; The handprint images include ten plane fingerprint images, ten three-side fingerprint images, one half palm print image, one side palm print image, one four-finger fingerprint image and one four-finger knuckle print image.

9. The method according to claim 8, characterized in that The method of using an image processing algorithm to obtain the desired handprint image on the three-dimensional handprint model specifically includes: The three-dimensional handprint model is converted into a plurality of two-dimensional handprint plane images at different angles, the handprint plane images are denoised and enhanced, the processed images are combined using a splicing algorithm to obtain a preliminary handprint image, a neural network model is used to identify key points of the preliminary handprint image, and the preliminary handprint image is segmented by a deep learning segmentation network according to the identified key points, texture features are extracted from the segmented image using image processing technology, the extracted features are matched with different types of fingerprints or palm prints by a machine learning classification algorithm to obtain a matched handprint image, and morphological operations are used to optimize the details of the matched handprint image to obtain the desired handprint image; Wherein, the image processing technology is Gabor filter or deep learning model.

10. The method according to claim 7, characterized in that The trigger mechanism includes a time trigger mechanism and an event trigger mechanism.

Images