A face recognition method, system, electronic device, and storage medium

Abstract

The application provides a face recognition method, system, electronic device and storage medium. The face recognition method comprises the following steps: obtaining three-dimensional face information and three-dimensional vein information of a face to be recognized; determining to-be-recognized information based on the three-dimensional face information and the three-dimensional vein information; processing the to-be-recognized information by using a target recognition algorithm to obtain a recognition result. The method improves the accuracy, security and stability of face recognition.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, and in particular to a face recognition method and system, an electronic device, and a storage medium. BACKGROUND

[0002] Portrait recognition technology is widely used in the fields of finance, public security, access control, consumer electronics, etc., but has certain defects, such as face recognition technology being susceptible to interference from factors such as light, makeup, and expression, and the current difficulty in distinguishing similar individuals such as identical twins. SUMMARY

[0003] The present application provides a face recognition method, system, electronic device, and storage medium, which improves the accuracy, security, and stability of face recognition.

[0004] To solve the above technical problems, the first technical solution provided by the present application is to provide a face recognition method, comprising: acquiring three-dimensional face information and three-dimensional vein information of a face to be recognized; determining to-be-recognized information based on the three-dimensional face information and the three-dimensional vein information; and processing the to-be-recognized information using a target recognition algorithm to obtain a recognition result.

[0005] The step of acquiring the three-dimensional face information and the three-dimensional vein information of the face to be recognized comprises: acquiring a visible light image and an infrared light image of the face to be recognized from multiple angles; determining a three-dimensional face point cloud of the face to be recognized based on the visible light image to obtain the three-dimensional face information, and determining a three-dimensional vein point cloud of the face to be recognized based on the infrared light image to obtain the three-dimensional vein information.

[0006] The step of determining the to-be-recognized information based on the three-dimensional face information and the three-dimensional vein information comprises: fusing the visible light image, the three-dimensional face point cloud, and the three-dimensional vein point cloud to obtain the to-be-recognized information.

[0007] The step of determining the three-dimensional vein point cloud of the face to be recognized based on the infrared light image comprises: pre-processing the infrared light image; the pre-processing comprises at least one of scatter medium imaging reconstruction, image enhancement, and image segmentation; and determining the three-dimensional vein point cloud based on the pre-processed infrared light image.

[0008] The step of determining the three-dimensional face point cloud of the face to be recognized based on the visible light image comprises: determining a first parallax map based on the visible light image, the first parallax map representing distance information between the imaging device and the face to be recognized; and determining the three-dimensional face point cloud of the face to be recognized based on the first parallax map. The step of determining the three-dimensional vein point cloud of the face to be recognized based on the infrared light image comprises: determining a second parallax map based on the infrared light image, the second parallax map representing distance information between the imaging device and the face to be recognized; and determining the three-dimensional vein point cloud of the face to be recognized based on the second parallax map.

[0009] Before the step of obtaining the visible light image and the infrared light image of the face to be recognized, the method comprises: calibrating a plurality of imaging devices so that the plurality of imaging devices are in the same world coordinate system. The step of fusing the visible light image, the three-dimensional face point cloud, and the three-dimensional vein point cloud comprises: performing coordinate system registration on the visible light image and the three-dimensional face point cloud; fusing the visible light image and the three-dimensional face point cloud based on the registered coordinate system to obtain a first fused image; performing coordinate system registration on the first fused image and the three-dimensional vein point cloud; and fusing the first fused image and the three-dimensional vein point cloud based on the registered coordinate system.

[0010] The target recognition algorithm is PointNet++, the channel of the PointNet++ is 7, and the size of a rotation matrix is 7x7, and the rotation matrix is used for rotating and calibrating a point cloud.

[0011] The step of processing the information to be recognized by using the target recognition algorithm to obtain a recognition result comprises: processing the information to be recognized by using a plurality of point set abstraction modules respectively; performing feature fusion on outputs of the plurality of point set abstraction modules; processing the feature-fused result by using a fully connected layer to obtain an N-dimensional feature vector, the output channel number of the fully connected layer being N; and searching in a database based on the N-dimensional feature vector to obtain the recognition result.

[0012] To solve the above technical problems, the second technical solution provided by the present application is to provide a face recognition system, comprising: a first imaging device, a second imaging device, a first filter, a second filter; the first filter is arranged corresponding to the first imaging device, the second filter is arranged corresponding to the second imaging device, in response to the first filter and the second filter away from the first imaging device and the second imaging device, the first imaging device and the second imaging device acquire three-dimensional face information of the face to be identified; in response to the first filter and the second filter close to the first imaging device and the second imaging device, the first imaging device and the second imaging device acquire three-dimensional vein information of the face to be identified; a processing component is used to process the to-be-identified information by using a target recognition algorithm to obtain a recognition result.

[0013] Among them, it also includes: a projector, in response to the first filter and the second filter away from the first imaging device and the second imaging device, the projector projects a coded structured light pattern to the face to be identified; the infrared LED is annular, including a plurality of LED bulbs of different wavebands, in response to the first filter and the second filter close to the first imaging device and the second imaging device, the infrared LED projects infrared light to the face to be identified.

[0014] To solve the above technical problems, the third technical solution provided by the present application is to provide an electronic device, comprising: a memory and a processor, wherein the memory stores program instructions, and the processor retrieves program instructions from the memory to execute any one of the above methods.

[0015] To solve the above technical problems, the fourth technical solution provided by the present application is to provide a computer readable storage medium, which stores a program file, and the program file can be executed to implement any one of the above methods.

[0016] The beneficial effects of the present application, different from the prior art, the present application determines the to-be-identified information based on three-dimensional face information and three-dimensional vein information; the to-be-identified information is processed by using a target recognition algorithm to obtain a recognition result. This method improves the accuracy, security and stability of face recognition. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings, wherein:

[0018] Figure 1A structural schematic diagram of an embodiment of the face recognition system of the present application;

[0019] Figure 2 A flowchart of an embodiment of the face recognition method of the present application;

[0020] Figure 3 A structural schematic diagram of an embodiment of the electronic device of the present application;

[0021] Figure 4 A structural schematic diagram of an embodiment of the computer readable storage medium of the present application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0023] Specifically, please refer to Figure 1 , Figure 1 A structural schematic diagram of an embodiment of the face recognition system of the present application, specifically comprising: a first imaging device 101, a second imaging device 102, a first filter 103, a second filter 104, a projector 105, an infrared LED 106, and a processing component 107. The first filter 103 is arranged corresponding to the first imaging device 101, the second filter 104 is arranged corresponding to the second imaging device 102, in response to the first filter 103 and the second filter 104 being away from the first imaging device 101 and the second imaging device 102, the first imaging device 101 and the second imaging device 102 acquire three-dimensional face information of a face to be recognized. In response to the first filter 103 and the second filter 104 being close to the first imaging device 101 and the second imaging device 102, the first imaging device 101 and the second imaging device 102 acquire three-dimensional vein information of the face to be recognized. The processing component 107 is configured to process the to-be-recognized information by using a target recognition algorithm to obtain a recognition result.

[0024] In an embodiment, when the first filter 103 and the second filter 104 are away from the first imaging device 101 and the second imaging device 102, the projector 105 projects a coded structured light pattern to the face to be recognized. The infrared LED 106 is annular and includes a plurality of LED bulbs of different wavebands, and when the first filter 103 and the second filter 104 are close to the first imaging device 101 and the second imaging device 102, the infrared LED 106 projects infrared light to the face to be recognized.

[0025] Specifically, the first filter 103 and the second filter 104 are arranged at positions corresponding to the first imaging device 101 and the second imaging device 102 by the dial respectively. The first filter 103 and the second filter 104 are infrared filters. When the dial drives the first filter 103 and the second filter 104 to rotate in front of the lenses of the first imaging device 101 and the second imaging device 102, the first imaging device 101 and the second imaging device 102 are in an infrared waveband acquisition mode, and at this time, the first imaging device 101 and the second imaging device 102 acquire three-dimensional vein information of the face to be recognized. When the dial drives the first filter 103 and the second filter 104 to move away from the lenses of the first imaging device 101 and the second imaging device 102, the first imaging device 101 and the second imaging device 102 are in a visible light acquisition mode, and at this time, the first imaging device 101 and the second imaging device 102 acquire three-dimensional face information of the face to be recognized.

[0026] In an embodiment, in the infrared waveband acquisition mode, the infrared LED 106 projects infrared light to the face to be recognized. In the visible light acquisition mode, the projector projects a coded structured light pattern to the face to be recognized.

[0027] Specifically, please refer to Figure 2 , Figure 2 is a flowchart of an embodiment of the face recognition method of the present application, and specifically includes:

[0028] Step S11: acquiring three-dimensional face information and three-dimensional vein information of the face to be recognized.

[0029] Specifically, in the first stage, the first filter 103 and the second filter 104 are away from the lenses of the first imaging device 101 and the second imaging device 102, the projector 105 projects a coded light pattern to the face to be recognized, and the first imaging device 101 and the second imaging device 102 collect the pattern projected to the face to be recognized from different angles, thereby realizing acquisition of three-dimensional face information of the face to be recognized. In the second stage, the first filter 103 and the second filter 104 are located in front of the lenses of the first imaging device 101 and the second imaging device 102, the infrared LED 106 projects infrared light to the face to be recognized, and the first imaging device 101 and the second imaging device 102 collect three-dimensional vein information of the face to be recognized under the infrared waveband from different angles through the first filter 103 and the second filter 104, thereby realizing acquisition of three-dimensional vein information of the face to be recognized.

[0030] In an embodiment, the position switching of the first filter 103 and the second filter 104 can be realized by driving the dial by the motor through the processing assembly 107.

[0031] Specifically, in an embodiment, the imaging device obtains a visible light image and an infrared light image of a face to be recognized from multiple angles. That is, the imaging device 101 and the imaging device 102 first obtain visible light images of the face to be recognized from different angles in a visible light mode, and obtain infrared light images from different angles in an infrared waveband acquisition mode. The three-dimensional face point cloud of the face to be recognized is determined based on the visible light images, and the three-dimensional face information is obtained, and the three-dimensional vein point cloud of the face to be recognized is determined based on the infrared light images, and the three-dimensional vein information is obtained.

[0032] Specifically, a first disparity map is determined based on the visible light images, and the first disparity map represents distance information of the imaging device and the face to be recognized. The three-dimensional face point cloud of the face to be recognized is determined based on the first disparity map.

[0033] A second disparity map is determined based on the infrared light images, and the second disparity map represents distance information of the imaging device and the face to be recognized. The three-dimensional vein point cloud of the face to be recognized is determined based on the second disparity map.

[0034] In an embodiment, the disparity map of the visible light image or the infrared light image can be obtained by binocular matching, and the disparity map is converted into a point cloud. Specifically, the conversion of the disparity map into the point cloud can utilize a triangulation method. The depth of a pixel point can be calculated according to the disparity map, and the x and y coordinates can be determined according to the imaging coordinates of the imaging device, and the obtained coordinates are further converted into world coordinates to obtain the point cloud.

[0035] In an embodiment, since the skin is a scattering medium, if the infrared light image is directly processed, the effect is poor. Therefore, in the embodiment of the present application, the infrared light image needs to be preprocessed. The preprocessing includes at least one of scattering medium imaging reconstruction, image enhancement, and image segmentation. The three-dimensional vein point cloud is determined based on the preprocessed infrared light image.

[0036] In an embodiment, the convolutional neural network is used to perform scattering medium imaging reconstruction on the infrared light image to remove the interference of the skin on the imaging of the vein. Further, the image enhancement technology is used to improve the definition and sharpness of the vein image. In another embodiment, the image segmentation technology can also be used to remove the invalid area outside the vein.

[0037] Step S12: determining the to-be-recognized information based on the three-dimensional face information and the three-dimensional vein information.

[0038] In an embodiment, the visible light image, the three-dimensional face point cloud and the three-dimensional vein point cloud are fused to obtain the to-be-identified information. In a specific embodiment, in order to improve the accuracy of fusion, further registration of the coordinate system is required. For example, the plurality of imaging devices are calibrated so that the plurality of imaging devices are in the same world coordinate system. Specifically, the first imaging device 101 and the second imaging device 102 are calibrated so that the first imaging device 101 and the second imaging device 102 are in the same world coordinate system. In this way, the coordinate system does not need to be aligned again when fusion is performed.

[0039] The visible light image, the three-dimensional face point cloud and the three-dimensional vein point cloud are fused, including: first, the visible light image and the three-dimensional face point cloud are registered in the coordinate system; based on the registered coordinate system, the visible light image and the three-dimensional face point cloud are fused to obtain a first fused image. In this embodiment, if the coordinate systems of the visible light image and the three-dimensional face point cloud are consistent, the visible light image and the three-dimensional face point cloud can also be fused without registration of the coordinate system. The first fused image and the three-dimensional vein point cloud are registered in the coordinate system; based on the registered coordinate system, the first fused image and the three-dimensional vein point cloud are fused to obtain the to-be-identified information.

[0040] In another embodiment, the visible light image and the three-dimensional vein point cloud can also be fused first to obtain a first fused image, and then the first fused image and the three-dimensional face point cloud are fused.

[0041] In a specific embodiment, the visible light image and the three-dimensional face point cloud are fused to obtain a first fused image (X, Y, Z, R, G, B), wherein X, Y and Z are three-dimensional space coordinates, and R, G and B represent pixel values of red, green and blue three channels. The first fused image and the three-dimensional vein point cloud are fused to obtain the to-be-identified information (X, Y, Z, R, G, B, V). When the point is a vein point, V = 1, and when the point is not a vein point, V = 0.

[0042] Step S13: processing the to-be-identified information by using a target recognition algorithm to obtain a recognition result.

[0043] Specifically, the fused result is processed by using a target recognition algorithm, and the fused result is the to-be-identified information.

[0044] In an embodiment, the target recognition algorithm is PointNet++, and the to-be-identified information is input into the PointNet++ algorithm. The input channel of the PointNet++ algorithm is expanded to 7, and the size of the rotation matrix used for rotation and calibration of the point cloud is 7x7.

[0045] In another embodiment, the information to be identified is processed by a plurality of point set abstraction modules respectively; the outputs of the plurality of point set abstraction modules are fused in features; the result after the feature fusion is processed by a full connection layer to obtain an N-dimensional feature vector, the output channel number of the full connection layer is N; the N-dimensional feature vector is searched in a database to obtain the identification result. The point set abstraction module of PointNet++ is expanded to M groups in the application, M can be 3, in the case of M being 3, the information to be identified is processed by 3 groups of point set abstraction modules respectively, the outputs of the 3 groups of point set abstraction modules are fused in features once, and then the full connection layer is processed, the output channel number of the full connection layer is N, so as to output an N-dimensional feature vector. The N-dimensional feature vector is used for searching and matching in the database, if the matching result is greater than or equal to a threshold, the identification is passed, and if the matching result is less than the threshold, the identification fails.

[0046] The application uses Figure 1 The face three-dimensional data is obtained by the optical system shown in the figure, not a TOF camera, and the precision is higher. The application uses an infrared binocular system to obtain the three-dimensional vein, and through the scattering medium imaging reconstruction technology, the good precision is maintained on the premise of greatly reducing the cost. The feature vector is constructed by the improved PointNet++, and the optimal feature extraction scheme can be obtained through iterative training. The face point cloud acquisition device and the vein acquisition device share two cameras, so they can be easily calibrated to the same world coordinate system, and there is no need for later sensor alignment.

[0047] The application builds a device that can synchronously obtain high-precision color three-dimensional face and three-dimensional vein information in view of the inherent defects of portrait recognition technology, such as being easily disturbed by light, expression and makeup, being difficult to distinguish similar individuals, being easily stolen, copied and destroyed by exposing features to the outside world, and needing live detection. The infrared vein image after the scattering medium imaging reconstruction has high imaging quality and low cost, the color three-dimensional face information and the three-dimensional vein information after the fusion are used for biometric recognition, the accuracy, security and stability of the recognition are effectively improved, and the live detection is avoided.

[0048] Please refer to Figure 3 , which is a structural schematic diagram of an embodiment of the electronic device, and the electronic device includes a memory 202 and a processor 201 connected to each other.

[0049] The memory 202 is used for storing program instructions for implementing any one of the above methods.

[0050] The processor 201 is used for executing the program instructions stored in the memory 202.

[0051] The processor 201 can also be called a CPU (Central Processing Unit). The processor 201 can be an integrated circuit chip having a processing capability of signals. The processor 201 can also be a general processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general processor can be a microprocessor or the processor can also be any conventional processor.

[0052] The memory 202 can be a memory bar, a TF card, etc., and can store all information in the electronic device of the device, including input raw data, computer programs, intermediate motion results and final motion results. It is saved in the memory. It is stored and retrieved according to the location specified by the controller. With the memory, the electronic device has a memory function and can work normally. The memory of the electronic device can be divided into main memory (memory) and auxiliary memory (external storage) according to the purpose, and there is also a classification method of dividing external memory and internal memory. The external storage is usually a magnetic medium or an optical disc, etc., which can store information for a long time. The memory refers to the storage component on the motherboard, which is used to store the data and programs currently being executed, but only for temporarily storing programs and data. When the power is off or the power is off, the data will be lost.

[0053] In several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only schematic, for example, the division of modules or units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0054] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the present embodiment scheme.

[0055] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0056] The integrated unit, if implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in part, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the methods in the embodiments of the present application.

[0057] Please refer to Figure 4 , the structural schematic diagram of the computer readable storage medium of the present application. The storage medium of the present application stores a program file 203 capable of realizing all the above methods, wherein the program file 203 can be stored in the above storage medium in the form of a software product, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the methods in the embodiments of the present application. The foregoing storage device includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media capable of storing program codes, or a computer, a server, a mobile phone, a tablet terminal device, etc.

[0058] The above is only the embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent flow transformation, or direct or indirect application in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A face recognition method, characterized in that, The face recognition method is applied to a face recognition system, which includes: a first imaging device, a second imaging device, a first filter, and a second filter; the first filter is configured corresponding to the first imaging device, and the second filter is configured corresponding to the second imaging device; the method includes: The first imaging device and the second imaging device are used to acquire three-dimensional facial information and three-dimensional vein information of the face to be identified from different angles; The information to be identified is determined based on the three-dimensional face information and the three-dimensional vein information; the information to be identified is obtained by fusing the visible light image, the three-dimensional face point cloud, and the three-dimensional vein point cloud; the three-dimensional face point cloud is the three-dimensional face information, and the three-dimensional vein point cloud is the three-dimensional vein information. The target recognition algorithm is used to process the information to be recognized to obtain the recognition result; the target recognition algorithm is PointNet++, the channels of PointNet++ are determined based on the channels of the information to be recognized, and the size of the rotation matrix is ​​determined based on the channels of the information to be recognized. The rotation matrix is ​​used to rotate and calibrate the point cloud. The steps of acquiring the three-dimensional facial information and three-dimensional vein information of the face to be identified include: Multiple imaging devices are calibrated so that they are in the same world coordinate system; Acquire visible light and infrared images of the face to be identified from different angles; Based on the visible light image, a three-dimensional face point cloud of the face to be identified is determined, thereby obtaining the three-dimensional face information; and based on the infrared light image, a three-dimensional vein point cloud of the face to be identified is determined, thereby obtaining the three-dimensional vein information. The step of determining the information to be identified based on the three-dimensional face information and the three-dimensional vein information includes: The visible light image is registered with the three-dimensional face point cloud in a coordinate system. The visible light image and the three-dimensional face point cloud are fused based on the registered coordinate system to obtain a first fused image; Register the first fused image with the three-dimensional vein point cloud in the coordinate system. The first fused image is fused with the three-dimensional vein point cloud based on the registered coordinate system to obtain the information to be identified.

2. The identification method according to claim 1, characterized in that, The step of determining the three-dimensional vein point cloud of the face to be identified based on the infrared light image includes: The infrared image is preprocessed; the preprocessing includes at least one of the following: scattering medium imaging reconstruction, image enhancement, and image segmentation. The three-dimensional vein point cloud is determined based on the preprocessed infrared image.

3. The identification method according to claim 1, characterized in that, The step of determining the 3D face point cloud of the face to be identified based on the visible light image includes: A first disparity map is determined based on the visible light image, and the first disparity map represents the distance information between the imaging device and the face to be identified; The three-dimensional face point cloud of the face to be identified is determined based on the first disparity map; The step of determining the three-dimensional vein point cloud of the face to be identified based on the infrared light image includes: A second disparity map is determined based on the infrared light image, and the second disparity map represents the distance information between the imaging device and the face to be identified; The three-dimensional vein point cloud of the face to be identified is determined based on the second disparity map.

4. The identification method according to any one of claims 1 to 3, characterized in that, The target recognition algorithm is PointNet++, which has 7 channels and a rotation matrix of size 7×7. The rotation matrix is ​​used to rotate and calibrate the point cloud.

5. The identification method according to claim 4, characterized in that, The step of processing the information to be identified using a target recognition algorithm to obtain the recognition result includes: The information to be identified is processed using multiple point set abstraction modules respectively; The outputs of the multiple point set abstraction modules are fused using features. The result of feature fusion is processed by a fully connected layer to obtain an N-dimensional feature vector, and the number of output channels of the fully connected layer is N. The identification result is obtained by searching the database based on the N-dimensional feature vector.

6. A face recognition system, characterized in that, include: First imaging device, second imaging device, first filter, second filter; The first filter is configured corresponding to the first imaging device, and the second filter is configured corresponding to the second imaging device. In response to the first and second filters moving away from the first and second imaging devices, the first and second imaging devices acquire a visible light image of the face to be identified, and determine a three-dimensional face point cloud based on the visible light image, thereby obtaining three-dimensional face information. In response to the first and second filters moving closer to the first and second imaging devices, the first and second imaging devices acquire an infrared light image of the face to be identified, and determine a three-dimensional vein point cloud based on the infrared light image, thereby obtaining three-dimensional vein information. A processing component is used to process the information to be identified using a target recognition algorithm to obtain a recognition result. The information to be identified is obtained by fusing a visible light image, a 3D face point cloud, and a 3D vein point cloud. The 3D face point cloud is the 3D face information, and the 3D vein point cloud is the 3D vein information. The target recognition algorithm is PointNet++, where the channels of PointNet++ are determined based on the channels of the information to be identified, and the size of the rotation matrix is ​​determined based on the channels of the information to be identified. The rotation matrix is ​​used to rotate and calibrate the point cloud. The processing component is also used to calibrate the first imaging device and the second imaging device so that multiple first imaging devices and second imaging devices are in the same world coordinate system. The processing component is further configured to register the visible light image with the three-dimensional face point cloud in a coordinate system; and to fuse the visible light image with the three-dimensional face point cloud based on the registered coordinate system to obtain a first fused image; The first fused image is registered with the three-dimensional vein point cloud in a coordinate system; based on the registered coordinate system, the first fused image and the three-dimensional vein point cloud are fused to obtain the information to be identified.

7. The identification system according to claim 6, characterized in that, Also includes: The projector projects an coded structured light pattern onto the face to be identified in response to the first filter and the second filter moving away from the first imaging device and the second imaging device. An infrared LED, which is ring-shaped and includes multiple LED bulbs of different wavelengths, projects infrared light onto the face to be identified in response to the first filter and the second filter being close to the first imaging device and the second imaging device.

8. An electronic device, characterized in that, include: A memory and a processor, wherein the memory stores program instructions, and the processor retrieves the program instructions from the memory to perform the method as claimed in any one of claims 1-5.

9. A computer-readable storage medium, characterized in that, The system contains a program file that can be executed to implement the method as described in any one of claims 1-5.

Images