Target recognition method, storage medium and device

Abstract

The present application provides a target recognition method, storage medium and device, the target recognition method comprising: acquiring a first optical image and a first depth image containing a target to be identified, the first depth image being an image containing distance information of the target to be identified; extracting a first marker image containing a first identification marker from the first optical image, and correcting characteristic parameters of the first identification marker in the first marker image according to the first depth image; extracting target recognition parameters of the first identification marker based on the corrected first marker image; comparing the target recognition parameters with pre-stored target authentication parameters to generate a target recognition result. The present application combines optical images and TOF depth images to realize the recognition of the identities of different targets and the recognition of specific scenes.

Description

Technical Field

[0001] The present application belongs to the technical field of image analysis and relates to a target recognition method, and in particular to a target recognition method, storage medium, and device. Background Art

[0002] At present, with the popularization and development of wearable devices or related electronic devices such as AR (Augmented Reality), VR (Virtual Reality), and smart watches, the user's identity recognition method is also constantly updated. In the existing technology, the face recognition solution of mobile phones has been relatively popular. However, wearable devices such as AR glasses and VR glasses can often only be worn on the head and cannot achieve comprehensive scanning of facial biometric information like mobile phones.

[0003] However, as users' demand for various wearable electronic devices continues to increase, identity and scene recognition have become particularly important. However, in existing image recognition methods, when recognition is based on an image obtained by the device itself, the information that the image can provide is limited, resulting in low accuracy and reliability of operations such as identity recognition.

[0004] Therefore, how to solve the defects of existing technologies, such as the inability to combine multiple image information to improve the accuracy and reliability of target recognition, has become a technical problem that needs to be urgently solved by those skilled in the art. Summary of the Invention

[0005] In view of the shortcomings of the prior art described above, the purpose of this application is to provide a target recognition method, storage medium and device to solve the problem that the prior art cannot combine multiple image information to improve the accuracy and reliability of target recognition.

[0006] To achieve the above-mentioned purpose and other related purposes, the present application provides a target recognition method on the one hand, which includes: respectively obtaining a first optical image and a first depth image containing a target to be identified, wherein the first depth image refers to an image containing distance information of the target to be identified; extracting a first marker image containing a first identification marker from the first optical image, and correcting the characteristic parameters of the first identification marker in the first marker image according to the first depth image, wherein the first identification marker includes a first part of the target to be identified; extracting target recognition parameters of the first identification marker based on the corrected first marker image; comparing the target recognition parameters with pre-stored target authentication parameters to generate a target recognition result.

[0007] In one embodiment of the present application, the step of extracting a first marker image containing a first identification marker from the first optical image includes: inputting the first optical image into a pre-trained marker image model, and obtaining an image containing the first part of the target to be identified as the first marker image.

[0008] In one embodiment of the present application, the step of correcting the characteristic parameters of the first identification marker in the first marker image based on the first depth image includes: determining the characteristic parameters of the first part in the first marker image as a first distance from the first depth image; obtaining a second depth image of the same first part of the target to be identified in the feature recognition library, and correcting the first distance in the first marker image to a second distance corresponding to the second image; wherein, the first distance refers to the distance information of the first part in the first depth image, and the second distance refers to the distance information of the first part in the second depth image.

[0009] In one embodiment of the present application, the step of comparing the target identification parameters with the pre-stored target authentication parameters to generate a target identification result includes: in response to the target identification parameters being consistent with the pre-stored target authentication parameters, the target identification result is a target identification success; in response to the target identification parameters being inconsistent with the pre-stored target authentication parameters, the target identification result is a target identification failure; wherein, the target identification result includes an identity recognition result.

[0010] In one embodiment of the present application, the target identification parameters include biometric parameters of the target to be identified; the step of comparing the target identification parameters with pre-stored target authentication parameters to generate a target identification result includes: in response to the biometric parameters being consistent with the pre-stored biometric parameters, the target identification result is a target identification success; in response to the biometric parameters not being consistent with the pre-stored biometric parameters, the target identification result is a target identification failure.

[0011] In one embodiment of the present application, the target identification parameters also include posture characteristic parameters of the target to be identified; the step of comparing the target identification parameters with pre-stored target authentication parameters to generate a target identification result includes: in response to the biometric parameters being consistent with the pre-stored biometric parameters and the posture characteristic parameters being consistent with the pre-stored posture characteristic parameters, the target identification result is a successful target identification; in response to the biometric parameters not being consistent with the pre-stored biometric parameters or the posture characteristic parameters not being consistent with the pre-stored posture characteristic parameters, the target identification result is a failed target identification.

[0012] In one embodiment of the present application, the method further includes: extracting a second marker image containing a second identification marker from the first optical image, and correcting the characteristic parameters of the second identification marker in the second marker image according to the first depth image, wherein the second identification marker includes a first object associated with the target to be identified or a second object associated with the scene; based on the corrected second marker image, extracting target identification parameters of the second identification marker; based on the target identification parameters of the first identification marker and the target parameters of the second identification marker, comparing them with pre-stored target authentication parameters to generate a target identification result.

[0013] In one embodiment of the present application, the target recognition parameters of the second recognition marker are physical feature parameters and posture feature parameters.

[0014] In one embodiment of the present application, before the step of extracting the first identification marker from the first optical image, the method also includes: preprocessing the first optical image and the first depth image, and the preprocessing at least includes: image fusion and denoising of the first optical image and the first depth image.

[0015] To achieve the above-mentioned purpose and other related purposes, the present application provides, on the other hand, a computer-readable storage medium having a computer program stored thereon, which implements the target recognition method when executed by a processor.

[0016] To achieve the above-mentioned purpose and other related purposes, the present application provides an electronic device on the other hand, including: a processor and a memory; the memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory, so that the electronic device performs the target recognition method.

[0017] To achieve the above-mentioned objectives and other related objectives, the last aspect of the present application provides a smart wearable device, which includes: an optical camera, which captures an optical image containing a test target; a depth camera, which captures a depth image containing a test target; one or more processors, which communicate with the optical camera and the depth camera; and a memory, which is configured to store instructions. When the stored instructions are executed by the one or more processors, the one or more processors execute steps, which include: respectively acquiring a first optical image and a first depth image containing a target to be identified, wherein the first depth image refers to an image containing distance information of the target to be identified; extracting a first marker image containing a first identification marker from the first optical image, and correcting the characteristic parameters of the first identification marker in the first marker image according to the first depth image, wherein the first identification marker includes a first part of the target to be identified; extracting target identification parameters of the first identification marker based on the corrected first marker image; comparing the target identification parameters with pre-stored target authentication parameters to generate a target recognition result.

[0018] As described above, the target recognition method, storage medium, and device described in this application have the following beneficial effects:

[0019] This application uses a TOF (Time of Flight) camera combined with an optical camera to capture static images or multiple frames of dynamic images, and then processes and recognizes the data to identify the wearer or user. This technology can be applied to wearable devices such as AR and VR, and has potential applications in smart homes, intelligent transportation, assisted driving, and other fields.

[0020] The TOF camera in this application can provide distance data, which can serve as a powerful supplement to the image data of traditional optical cameras. The TOF camera combined with a high-resolution optical camera can not only measure the size of the target through the TOF camera, but also obtain more details of the target through the optical camera. This application combines the two to provide a flexible identity recognition and scene recognition method for wearable devices.

[0021] This application can realize contactless recognition; improve the accuracy and robustness of recognition by combining biometrics with posture features such as gestures; and can flexibly set various recognition features, with high playability. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Shown is a principle flow chart of the target recognition method of the present application in one embodiment.

[0023] Figure 2Shown is a calibration flow chart of the target recognition method of the present application in one embodiment.

[0024] Figure 3 Shown is a calibration diagram of an embodiment of the target recognition method of the present application.

[0025] Figure 4 Shown is a comparison and recognition flow chart of the target recognition method of the present application in one embodiment.

[0026] Figure 5 Shown is a target recognition flow chart of a target recognition method according to an embodiment of the present application.

[0027] Figure 6 Shown is a schematic diagram of the structural connection of an electronic device in one embodiment of the present application.

[0028] Component number description

[0029] 6 Electronic devices

[0030] 61 processors

[0031] 62 Memory

[0032] Steps S11 to S14

[0033] Steps S121 to S123

[0034] Steps S141-S142

[0035] Steps S51 to S55 DETAILED DESCRIPTION

[0036] The following describes the embodiments of the present application through specific examples. Those skilled in the art can easily understand the other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0037] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. Therefore, the illustrations only show components related to the present application and are not drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component can be changed at will, and the component layout type may also be more complicated.

[0038] The target recognition method, storage medium, and device described in this application can combine optical images and TOF depth images to realize the identification of different targets and the recognition of specific scenes. Furthermore, on the one hand, this application can realize the identity recognition of the wearer of the smart wearable device based on the wearer's biometrics or posture characteristics, and on the other hand, it can realize the recognition of specific scenes by any electronic device.

[0039] The target recognition method provided in the embodiments of the present application can be applied to electronic devices, including but not limited to mobile phones, smart wearable devices, vehicle-mounted devices, augmented reality (AR) / virtual reality (VR) devices, laptops, ultra-mobile personal computers (UMPCs), netbooks, personal digital assistants (PDAs), smart speakers, set-top boxes (STBs), or televisions. The embodiments of the present application do not impose any restrictions on the specific types of electronic devices.

[0040] The following will be combined Figures 1 to 6 The principles and implementation methods of a target recognition method, storage medium, and device of this embodiment are described in detail so that those skilled in the art can understand the target recognition method, storage medium, and device of this embodiment without creative work.

[0041] See also Figure 1 , which is a flow chart showing the principle of the target recognition method of the present application in one embodiment. Figure 1 As shown, the target recognition method specifically includes the following steps:

[0042] S11 , respectively acquiring a first optical image and a first depth image containing a target to be identified, where the first depth image refers to an image depth image containing distance information of the target to be identified.

[0043] Specifically, the optical image can be a static image or a multi-frame dynamic image. The depth image can also be a static image or a multi-frame dynamic image. Among them, using a multi-frame dynamic image for recognition has higher accuracy.

[0044] S12. Extract a first marker image containing a first identification marker from the first optical image, and correct characteristic parameters of the first identification marker in the first marker image according to the first depth image, wherein the first identification marker includes an optical image, an optical image, and a depth image of a first part of the target to be identified.

[0045] In one embodiment, the identification marker includes a body part of the target person or an object associated with an identity or scene. Specifically, the identification marker can be a body part of the device user, such as a hand, arm, or leg, or a specific object, such as an office pen holder or a pen in a hand. The characteristics of a body part of the target person can be used to identify a person's identity, the characteristics of an object associated with an identity can be used to identify a person's identity, and the characteristics of an object associated with a scene can be used to identify a specific scene.

[0046] See also Figure 2 , which is a correction flow chart of the target recognition method of the present application in one embodiment. Figure 2 As shown, S12 specifically includes the following steps:

[0047] S121: Input the first optical image into a pre-trained marker image model to obtain an image containing a first part of the target to be identified as the first marker image. Specifically, the optical image is input into the pre-trained marker image model to determine that the identification marker is at least one of a body part of the target, an object associated with an identity, or an object associated with a scene. In one embodiment, the body part image of the target is an image of a palm.

[0048] S122: Determine, from the first depth image, characteristic parameters of the first part in the first marker image as a first range depth image.

[0049] S123: Obtain a second depth image of the same first part of the target to be identified in the feature recognition library, and correct the first distance in the first marker image to a second distance corresponding to the second image; wherein the first distance refers to the distance information of the first part in the first depth image, and the second distance refers to the distance information of the first part in the second depth image. Specifically, based on the same palm image, the first distance is corrected to the second distance corresponding to the palm image in the feature recognition library; wherein the first distance and the second distance refer to the distance between the shooting point of the smart wearable device acquiring the palm image and the palm of the wearer of the smart wearable device.

[0050] See also Figure 3 , which is a schematic diagram showing the calibration of the target recognition method of the present application in one embodiment. Figure 3As shown, a simple example of optical image data conversion is presented. The gesture stored in the device feature recognition library is a palm image obtained by the device user at a distance of 50 cm, and the palm image is vertical. However, the gesture obtained during the recognition process is a palm image at a distance of 70 cm. At the same time, the palm image has a certain inclination angle. Therefore, through geometric principles such as the cosine theorem, the palm image at a distance of 70 cm is corrected to a 50 cm image.

[0051] In another embodiment, step S12 may also be to first calibrate the characteristic parameters of the identification marker according to the depth image, and then extract the identification marker from the optical image.

[0052] S13: Extracting target recognition parameters of the first recognition marker based on the corrected first marker image.

[0053] S14, comparing the target identification parameter with pre-stored target authentication parameters to generate a target identification result; wherein the target identification result includes an identity identification result.

[0054] See also Figure 4 , which shows a flow chart of comparison and recognition of the target recognition method of the present application in one embodiment. Figure 4 As shown, S14 specifically includes the following steps:

[0055] S141 , in response to the target identification parameter being consistent with the pre-stored target authentication parameter, the target identification result is target identification success.

[0056] S142, in response to the target recognition parameter not being consistent with the pre-stored target authentication parameter, the target recognition result is a target recognition failure; wherein the target recognition result includes an identity recognition result and / or a scene recognition result.

[0057] Specifically, for example, if the error in the lengths of all five fingers is set to be within 2mm, then the target recognition of the five fingers is successful. If the error in the lengths of one or more fingers exceeds 2mm, the target recognition of the five fingers fails. The threshold of 2mm is generally set based on the resolution of the TOF hardware. The better the hardware, the smaller the threshold.

[0058] In one embodiment, the target identification parameters include biometric parameters of the target to be identified; and the step of comparing the target identification parameters with pre-stored target authentication parameters to generate a target identification result includes:

[0059] In response to the biometric parameter being consistent with the pre-stored biometric parameter, the target recognition result is a target recognition success; in response to the biometric parameter being inconsistent with the pre-stored biometric parameter, the target recognition result is a target recognition failure.

[0060] Specifically, the target person's biometric parameters can be physical parameters of the user's body, such as the length of five fingers, the length of the index finger knuckles, the width of the fingers, the arm, the location of a specific mole or scar on the arm, the length of the arm and forearm, the characteristics of the palm prints, and birthmarks of a specific size; the biometric parameters of a specific object are also the physical characteristic parameters, such as the size and shape of the object.

[0061] In another embodiment, the target recognition parameters further include posture feature parameters of the target to be recognized; and the step of comparing the target recognition parameters with pre-stored target authentication parameters to generate a target recognition result includes:

[0062] In response to the biometric parameters being consistent with the pre-stored biometric parameters and the posture characteristic parameters being consistent with the pre-stored posture characteristic parameters, the target recognition result is a target recognition success; in response to the biometric parameters not being consistent with the pre-stored biometric parameters or the posture characteristic parameters not being consistent with the pre-stored posture characteristic parameters, the target recognition result is a target recognition failure.

[0063] Specifically, the target person's posture characteristic parameters can be the specific posture of the user's body during the recognition process, such as the angle between the palm and the forearm, the OK gesture made by the left hand, arms crossed across the chest, etc.; the posture characteristic parameters of a specific object can be its placement position, the distance from the background wall, etc.

[0064] In one embodiment, the method further includes: extracting a second marker image containing a second identification marker from the first optical image, and correcting the characteristic parameters of the second identification marker in the second marker image according to the first depth image, wherein the second identification marker includes a first item associated with the target to be identified or a second item associated with the scene; based on the corrected second marker image, extracting target identification parameters of the second identification marker; based on the target identification parameters of the first identification marker and the target parameters of the second identification marker, comparing them with pre-stored target authentication parameters to generate a target identification result.

[0065] In one embodiment, the target recognition parameters of the second recognition marker are physical feature parameters and posture feature parameters.

[0066] In another embodiment, the step of comparing the target identification parameter with pre-stored target authentication parameters to generate a target identification result includes:

[0067] In response to at least two of the target identification parameters being consistent with pre-stored target authentication parameters, the target identification result is that the target identification is successful.

[0068] Specifically, to further improve recognition accuracy, for identity or scene recognition, two or more biometric parameters can be used for identification, or two or more posture characteristic parameters can be used for identification, or one or more biometric parameters can be combined with one or more posture characteristic parameters for identification. In this way, by increasing the complexity of recognition, recognition accuracy can be improved. For example, the hand characteristics of the user of the recognition device can be combined with the pen used by the user in the study.

[0069] Regarding scene recognition, for example: 1. A specific painting is hung or a cabinet is placed in a conference room scene, and the electronic device used by the user executes the target recognition method, for example, after scanning, it recognizes that the current scene is a conference room, and automatically connects to the meeting in that scene. 2. In the exhibition scene, the electronic device used by the user executes the target recognition method, for example, after scanning a specific product, it recognizes the relevant product information corresponding to the exhibition scene, and further completes the exchange of personal information and manufacturer information of the exhibition products through the presented interface. 3. In a scene outside a user's car, the electronic device used by the user executes the target recognition method, for example, after scanning a specific pendant inside the car from outside the car, it recognizes that it is the user's vehicle, thereby remotely activating or shutting down the vehicle or other specific functions. 4. In a scene inside a user's car, the electronic device used by the user executes the target recognition method, for example, after scanning a specific pendant inside the car from outside the car, the call function of the AR glasses is turned off, and the call function is switched to the in-vehicle device.

[0070] Furthermore, the external human-computer interaction interface configured or presented by the software program can be flexibly configured for different identifiers and characteristic parameters to achieve user and usage scenario limitations. For example, a dialog box can be presented through the human-computer interaction interface. In actual applications, for biometric parameters, multiple biometric parameters form a biometric parameter list, and multiple posture characteristic parameters form a posture characteristic parameter list. Biometric parameters and posture characteristic parameters can be automatically configured through active guidance, such as presenting a biometric parameter list for user selection, or they can be customized, such as allowing users to create new gestures.

[0071] In one embodiment, after step S11 and before step S12, the target recognition method further includes:

[0072] Preprocessing the first optical image and the first depth image includes at least image fusion and denoising of the first optical image and the first depth image. Image fusion refers to associating distance data from the depth image with the optical image to increase the dimensionality of the optical image data.

[0073] See also Figure 5 , which shows a target recognition flow chart of the target recognition method of the present application in one embodiment. Figure 5 As shown, the entire process of target recognition includes: S51, inputting optical camera and TOF camera data, and performing preprocessing such as image fusion. S52, extracting image data of the identification marker (optical image and TOF image data). S53, correcting the optical image of the identification marker based on the TOF camera data. S54, extracting the biometric parameters and posture characteristic parameters of the characteristic marker from the corrected identification marker image. S55, comparing the biometric parameters and posture characteristic parameters with the identity authentication information stored in the device. If the biometric parameters and posture characteristic parameters are consistent with the identity authentication information stored in the device, it is determined that the identity recognition is successful; if the biometric parameters and posture characteristic parameters are inconsistent with the identity authentication information stored in the device, it is determined that the identity recognition has failed.

[0074] The protection scope of the target identification method described in this application is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing, or replacing steps in the prior art based on the principles of this application are included in the protection scope of this application.

[0075] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the target recognition method is implemented.

[0076] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with a computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments. The aforementioned computer-readable storage medium includes any computer storage medium capable of storing program code, such as ROM, RAM, a magnetic disk, or an optical disk.

[0077] See also Figure 6 , which is a schematic diagram showing the structural connection of an electronic device in one embodiment of the present application. Figure 6As shown, this embodiment provides an electronic device 6, specifically comprising: a processor 61 and a memory 62; the memory 62 is used to store a computer program, and the processor 61 is used to execute the computer program stored in the memory 62, so that the electronic device 6 performs each step of the target recognition method. The target recognition method includes: respectively acquiring an optical image and a depth image containing a test target, wherein the depth image refers to an image containing distance information of the test target; extracting an identification marker from the optical image, and correcting the characteristic parameters of the identification marker based on the depth image; extracting target recognition parameters of the identification marker based on the corrected identification marker; and comparing the target recognition parameters with pre-stored target authentication parameters to generate a target recognition result.

[0078] The above-mentioned processor 61 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be 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.

[0079] The memory 62 may include a random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0080] In actual applications, the electronic device may be a computer including all or part of the components such as a memory, a storage controller, one or more processing units (CPUs), a peripheral interface, an RF circuit, an audio circuit, a speaker, a microphone, an input / output (I / O) subsystem, a display, other output or control devices, and an external port; the computer includes but is not limited to personal computers such as desktop computers, laptops, tablet computers, smart phones, smart TVs, and personal digital assistants (PDAs). The electronic device may also be a vehicle terminal. In other embodiments, the electronic device may also be a server, which may be arranged on one or more physical servers based on various factors such as function and load, or may be a cloud server composed of a distributed or centralized server cluster, which is not limited in this embodiment.

[0081] The smart wearable device described in this application includes:

[0082] An optical camera, which collects optical images including a test target;

[0083] Depth camera, which collects depth images containing the test target;

[0084] One or more processors, the one or more processors communicating with the optical camera and the depth camera; and a memory, the memory configured to store instructions that, when executed by the one or more processors, cause the one or more processors to perform steps, the steps comprising:

[0085] Acquire a first optical image and a first depth image containing the target to be identified, respectively, wherein the first depth image refers to an image containing distance information of the target to be identified;

[0086] extracting a first marker image containing a first identification marker from the first optical image, and correcting characteristic parameters of the first identification marker in the first marker image according to the first depth image, wherein the first identification marker includes a first part of the target to be identified;

[0087] extracting target recognition parameters of the first recognition marker based on the corrected first marker image;

[0088] The target recognition parameters are compared with the pre-stored target authentication parameters to generate the target recognition result. Depth image Optical image Optical image Depth image

[0089] Preferably, this application can be applied to AR / VR smart glasses, smart watches and other smart wearable devices for identity recognition and scene recognition. AR products can be used with AR devices in smart homes, smart transportation, and assisted driving, which is highly playable. For example, in smart homes, AR is used to identify objects at home. When you get home, you can use AR to recognize specific gestures to start furniture and other equipment; in smart transportation, AR is used to browse traffic information; in assisted driving, the driver wears AR glasses for navigation, etc. Therefore, compared with fingerprint recognition, through optical cameras and TOF cameras, there is no need to touch the device every time when recognition is not required.

[0090] In summary, the target recognition method, storage medium and device described in this application are based on the TOF camera combined with the optical camera to collect static images, or multiple frames of dynamic images, perform data processing and recognition, and realize the identification of the wearer or user of the device. It can be applied to wearable devices such as AR and VR, and has application prospects in many fields such as smart home, smart transportation, and assisted driving. The TOF camera in this application can provide distance data, which can serve as a powerful supplement to the image data of traditional optical cameras. The TOF camera is combined with a high-resolution optical camera. It can not only measure the size of the target object through the TOF camera, but also obtain more details of the target object through the optical camera. This application combines the two to provide a flexible identity recognition and scene recognition method for wearable devices. This application can realize contactless recognition; improve the accuracy and robustness of recognition by combining biometrics with gestures and other posture features; and can flexibly set each recognition feature, which is highly playable. This application effectively overcomes the various shortcomings of the existing technology and has high industrial utilization value.

[0091] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical concepts disclosed in this application shall be covered by the claims of this application.

Claims

1. A target recognition method, characterized in that: The method comprises: Acquire a first optical image and a first depth image containing the target to be identified, respectively, wherein the first depth image refers to an image containing distance information of the target to be identified; Extracting a first marker image containing a first identification marker from the first optical image, and correcting characteristic parameters of the first identification marker in the first marker image according to the first depth image, wherein the first identification marker includes a first part of the target to be identified; comprising: determining, from the first depth image, that the characteristic parameter of the first part in the first marker image is a first distance; obtaining a second depth image of the same first part of the target to be identified in a feature recognition library, and correcting the first distance in the first marker image to a second distance corresponding to the second depth image; wherein the first distance refers to the distance information of the first part in the first depth image, and the second distance refers to the distance information of the first part in the second depth image; extracting target recognition parameters of the first recognition marker based on the corrected first marker image; The target identification parameters are compared with pre-stored target authentication parameters to generate a target identification result.

2. The target recognition method according to claim 1, characterized in that: The step of extracting a first marker image containing a first identification marker from the first optical image comprises: The first optical image is input into a pre-trained marker image model to obtain an image containing a first portion of the target to be identified as the first marker image.

3. The target recognition method according to claim 1, characterized in that: The step of comparing the target identification parameter with the pre-stored target authentication parameter to generate a target identification result includes: In response to the target identification parameter being consistent with the pre-stored target authentication parameter, the target identification result is a target identification success; In response to the target identification parameter not matching the pre-stored target authentication parameter, the target identification result is a target identification failure; The target recognition result includes an identity recognition result.

4. The target recognition method according to claim 3, characterized in that: The target identification parameters include biometric parameters of the target to be identified; the step of comparing the target identification parameters with pre-stored target authentication parameters to generate a target identification result includes: In response to the biometric parameter being consistent with the pre-stored biometric parameter, the target recognition result is a successful target recognition; In response to the biometric parameter not matching the pre-stored biometric parameter, the target recognition result is target recognition failure.

5. The target recognition method according to claim 4, characterized in that: The target recognition parameters also include posture feature parameters of the target to be recognized; the step of comparing the target recognition parameters with pre-stored target authentication parameters to generate a target recognition result includes: In response to the biometric characteristic parameter being consistent with the pre-stored biometric characteristic parameter and the posture characteristic parameter being consistent with the pre-stored posture characteristic parameter, the target recognition result is successful target recognition; In response to the biometric characteristic parameter not matching the pre-stored biometric characteristic parameter or the posture characteristic parameter not matching the pre-stored posture characteristic parameter, the target recognition result is target recognition failure.

6. The target recognition method according to claim 1, characterized in that: The method further comprises: extracting a second marker image containing a second identification marker from the first optical image, and correcting characteristic parameters of the second identification marker in the second marker image based on the first depth image, wherein the second identification marker includes a first object associated with the target to be identified or a second object associated with the scene; extracting target recognition parameters of the second recognition marker based on the corrected second marker image; The target recognition parameter of the first recognition marker and the target parameter of the second recognition marker are compared with pre-stored target authentication parameters to generate a target recognition result.

7. The target recognition method according to claim 6, characterized in that: The target recognition parameters of the second recognition marker are physical feature parameters and posture feature parameters.

8. The target recognition method according to claim 1, characterized in that: Before the step of extracting the first identification marker from the first optical image, the method further includes: The first optical image and the first depth image are preprocessed, where the preprocessing includes at least image fusion and denoising of the first optical image and the first depth image.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the target recognition method according to any one of claims 1 to 8 is implemented.

10. An electronic device, characterized in that: include: processor and memory; The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the electronic device performs the target recognition method according to any one of claims 1 to 8.

11. A smart wearable device, characterized in that: The smart wearable device includes: An optical camera, which collects optical images including a test target; Depth camera, which collects depth images containing the test target; One or more processors, the one or more processors communicating with the optical camera and the depth camera; and a memory, the memory configured to store instructions that, when executed by the one or more processors, cause the one or more processors to perform steps, the steps comprising: Acquire a first optical image and a first depth image containing the target to be identified, respectively, wherein the first depth image refers to an image containing distance information of the target to be identified; Extracting a first marker image containing a first identification marker from the first optical image, and correcting characteristic parameters of the first identification marker in the first marker image according to the first depth image, wherein the first identification marker includes a first part of the target to be identified; comprising: determining, from the first depth image, that the characteristic parameter of the first part in the first marker image is a first distance; obtaining a second depth image of the same first part of the target to be identified in a feature recognition library, and correcting the first distance in the first marker image to a second distance corresponding to the second depth image; wherein the first distance refers to the distance information of the first part in the first depth image, and the second distance refers to the distance information of the first part in the second depth image; extracting target recognition parameters of the first recognition marker based on the corrected first marker image; The target identification parameters are compared with pre-stored target authentication parameters to generate a target identification result.

Images