In-vehicle facial recognition method, storage medium and vehicle

Abstract

The present application relates to the technical field of facial recognition. Particularly provided are an in-vehicle facial recognition method, a storage medium and a vehicle. The method is applied to an in-vehicle facial recognition system. The system comprises a first image collection apparatus and a second image collection apparatus, wherein the first image collection apparatus is used for collecting a facial image in a driver's area, and the second image collection apparatus is used for collecting a facial image in an in-vehicle area. The method comprises: when a first preset instruction is received, controlling a first image collection apparatus to collect a facial image in a driver's area as a first image, and controlling a second image collection apparatus to collect a facial image in the driver's area as a second image; and on the basis of the first image and the second image, recognizing a face in the driver's area, so as to obtain a recognition result indicating whether the recognition of the face in the driver's area is successful. In the technical solution provided in the present application, the face in the driver's area can be recognized more accurately and efficiently, so that a user has better intelligent driving experience.

Description

In-vehicle face recognition method, storage medium, and vehicle

[0001] This application claims priority to Chinese patent application No. 202311764988.4 filed on December 20, 2023, with the invention name “In-vehicle face recognition method, storage medium and vehicle”. The entire contents of the above Chinese patent application are incorporated into this application by reference. Technical Field

[0002] The present application relates to the field of face recognition technology, and specifically provides an in-vehicle face recognition method, a storage medium, and a vehicle. Background Art

[0003] In autonomous driving technology, multiple image acquisition devices will be installed in the car to identify user identity, monitor user status, etc., and using facial recognition technology to determine the identity and number of users in the car is also a key technology in the current smart cockpit field.

[0004] The existing technology uses different image acquisition devices to capture images of different areas to perform facial recognition on users in different areas of the vehicle. For example, a dedicated image acquisition device captures facial images of people in the driver's seat area for facial recognition in the driver's seat area, while another dedicated image acquisition device captures facial images of people in non-driver's seat areas (such as the passenger seat and rear seats) for facial recognition in non-driver's seat areas. However, in actual applications, the above-mentioned methods often result in inaccurate facial recognition, which leads to unsuccessful recognition. Even if recognition is successful, it takes a long time, affecting the user experience. Summary of the Invention

[0005] In order to solve the above technical problems, this application proposes an in-vehicle face recognition method, storage medium and vehicle, which can more accurately and efficiently recognize faces in the main driving area, thereby giving users a better smart driving experience.

[0006] To achieve the above objectives, the technical solution of this application is implemented as follows:

[0007] In a first aspect, the present application provides an in-vehicle face recognition method, which is applied to an in-vehicle face recognition system; the system includes: a first image acquisition device and a second image acquisition device; the first image acquisition device is used to acquire a face image of a person in the main driving area; the second image acquisition device is used to acquire a face image of an in-vehicle area; wherein the in-vehicle area includes the main driving area; the method includes:

[0008] When a first preset instruction is received, the first image acquisition device is controlled to acquire the facial image of the person in the main driving area as a first image, and the second image acquisition device is controlled to acquire the facial image of the person in the main driving area as a second image;

[0009] Based on the first image and the second image, the face of the person in the main driving area is recognized to obtain a recognition result of whether the face recognition of the person in the main driving area is successful.

[0010] In some embodiments, the recognizing a face in the main driver's seat area based on the first image and the second image to obtain a recognition result of whether the face recognition in the main driver's seat area is successful includes:

[0011] Acquire pre-stored facial feature information corresponding to the first image acquisition device as first reference feature information;

[0012] Recognize the face of the person in the main driving area based on the first image and the first reference feature information, and obtain a first recognition result of whether the recognition is successful;

[0013] Acquiring pre-stored facial feature information corresponding to the second image acquisition device as second reference feature information;

[0014] Recognizing a face in the driver's seat area based on the second image and the second reference feature information to obtain a second recognition result of whether the recognition is successful;

[0015] Based on the first recognition result and the second recognition result, it is determined whether the recognition of the face of the person in the main driving area is successful.

[0016] In some embodiments, the step of recognizing a face in the driver's seat area based on the first image and the first reference feature information to obtain a first recognition result indicating whether the recognition is successful includes:

[0017] performing feature extraction on the first image to obtain first feature information;

[0018] calculating a similarity between the first feature information and the first reference feature information;

[0019] The first recognition result is obtained based on the similarity between the first feature information and the first reference feature information.

[0020] In some embodiments, the step of recognizing a face in the driver's seat area based on the second image and the second reference feature information to obtain a second recognition result indicating whether the recognition is successful includes:

[0021] performing feature extraction on the second image to obtain second feature information;

[0022] calculating a similarity between the second feature information and the second reference feature information;

[0023] The second recognition result is obtained based on the similarity between the second feature information and the second reference feature information.

[0024] In some embodiments, determining whether the face recognition of the primary driver's seat area is successful based on the first recognition result and the second recognition result includes:

[0025] When one of the first recognition result and the second recognition result is a successful recognition, determining that the recognition of the face of the person in the main driving area is successful;

[0026] When both the first recognition result and the second recognition result are recognition failures, it is determined that recognition of the face in the main driving area has failed.

[0027] In some embodiments, the method further comprises:

[0028] When the first recognition result is a recognition success and the second recognition result is a recognition failure, updating the second reference feature information based on the facial information of the main driving area to obtain updated second reference feature information;

[0029] The step of recognizing the face of the person in the main driving area based on the second image and the second reference feature information to obtain a second recognition result of whether the recognition is successful includes:

[0030] Based on the second image and the updated second reference feature information, the face of the person in the main driving area is recognized to obtain the second recognition result.

[0031] In some embodiments, when the first recognition result is a successful recognition and the second recognition result is a failed recognition, updating the second reference feature information based on the facial information of the main driver's seat area to obtain the updated second reference feature information includes:

[0032] When the first recognition result is a recognition success and the second recognition result is a recognition failure, controlling the second image acquisition device to acquire the facial image of the main driving area as a third image;

[0033] performing feature extraction on the third image to obtain third feature information;

[0034] The second reference characteristic information is updated using the third characteristic information to obtain the updated second reference characteristic information.

[0035] In some embodiments, the method further comprises:

[0036] When the second recognition result is a recognition success and the first recognition result is a recognition failure, updating the first reference feature information based on the facial information of the main driving area to obtain updated first reference feature information;

[0037] The step of recognizing a face in the main driving area based on the first image and the first reference feature information to obtain a first recognition result of whether the recognition is successful includes:

[0038] Based on the first image and the updated first reference feature information, the face of the person in the main driving area is recognized to obtain the first recognition result.

[0039] In some embodiments, when the second recognition result is a successful recognition and the first recognition result is a failed recognition, updating the first reference feature information based on the facial information of the main driver's seat area to obtain the updated first reference feature information includes:

[0040] When the second recognition result is a recognition success and the first recognition result is a recognition failure, controlling the first image acquisition device to acquire the facial image of the person in the main driving area as a fourth image;

[0041] performing feature extraction on the fourth image to obtain fourth feature information;

[0042] The first reference feature information is updated using the fourth feature information to obtain the updated first reference feature information.

[0043] In some embodiments, the first reference feature information and the second reference feature information are obtained based on pre-recorded faces of people in the main driving area.

[0044] In some embodiments, the first reference feature information and the second reference feature information are obtained in the following manner:

[0045] When the second preset instruction is received, a first preset interaction strategy is obtained; wherein the first preset interaction strategy is used to guide the face of the person in the main driving area to perform a first predetermined action;

[0046] controlling the first image acquisition device to acquire a facial image of the person in the main driving area performing the first predetermined action as a fifth image, and controlling the second image acquisition device to acquire a facial image of the person in the main driving area performing the first predetermined action as a sixth image;

[0047] performing feature extraction on the fifth image to obtain the first reference feature information;

[0048] Perform feature extraction on the sixth image to obtain the second reference feature information.

[0049] In some embodiments, the second preset instruction is issued by a user; or, the second preset instruction is triggered by a first predetermined condition.

[0050] In some embodiments, triggering the second preset instruction includes:

[0051] determining whether both the first reference characteristic information and the second reference characteristic information do not exist;

[0052] When neither the first reference feature information nor the second reference feature information exists, detecting the face positions and the number of faces in the vehicle interior area;

[0053] When it is determined that there is a face only in the main driving area, the second preset instruction is triggered.

[0054] In some embodiments, the vehicle interior area also includes a passenger area; and triggering the second preset instruction further includes:

[0055] When it is determined that there is a face in both the driver's seat area and the passenger seat area, a preset entry position selection page is displayed;

[0056] receiving an entry location selected by a user based on the entry location selection page;

[0057] When the entry location selected by the user is the main driving area, the second preset instruction is triggered.

[0058] In some embodiments, the interior area of ​​the vehicle also includes a passenger seat area; and the second reference feature information is obtained based on pre-recording of a face in the passenger seat area.

[0059] In some embodiments, the second reference feature information is obtained in the following manner:

[0060] When the third preset instruction is received, a second preset interaction strategy is obtained; wherein the second preset interaction strategy is used to guide the face of the person in the passenger seat area to perform a second predetermined action;

[0061] controlling the second image acquisition device to acquire a facial image of the passenger seat area including the second predetermined action as a seventh image;

[0062] Perform feature extraction on the seventh image to obtain the second reference feature information.

[0063] In some embodiments, the third preset instruction is issued by a user; or, the third preset instruction is triggered by a second predetermined condition.

[0064] In some embodiments, triggering the third preset instruction includes:

[0065] determining whether both the first reference characteristic information and the second reference characteristic information do not exist;

[0066] When neither the first reference feature information nor the second reference feature information exists, detecting the face positions and the number of faces in the vehicle interior area;

[0067] When it is determined that there is a face only in the passenger seat area, the third preset instruction is triggered.

[0068] In some embodiments, triggering the third preset instruction further includes:

[0069] When it is determined that there is a face in both the driver's seat area and the passenger seat area, a preset entry position selection page is displayed;

[0070] receiving an entry location selected by a user based on the entry location selection page;

[0071] When the entry location selected by the user is the passenger seat area, the third preset instruction is triggered.

[0072] In some embodiments, the vehicle interior area further includes a passenger area; and the method further includes:

[0073] When receiving the fourth preset instruction, controlling the second image acquisition device to acquire the facial image of the passenger seat area as the eighth image;

[0074] Based on the eighth image, the face of the person in the passenger seat area is recognized to obtain a recognition result of whether the face recognition in the passenger seat area is successful.

[0075] In some embodiments, the vehicle interior area further includes a rear area; and the method further includes:

[0076] When receiving the fifth preset instruction, controlling the second image acquisition device to acquire the facial image of the person in the rear row area as the ninth image;

[0077] Based on the ninth image, faces of people in the rear area are recognized to obtain a recognition result of whether face recognition of the rear area is successful.

[0078] In a second aspect, the present application provides a computer-readable storage medium storing a plurality of program codes, wherein the program codes are suitable for being loaded and run by a processor to execute the in-vehicle face recognition method described in any one of the technical solutions in the first aspect.

[0079] In a third aspect, the present application provides a vehicle comprising a storage device and a processor, wherein the storage device stores program code that can be run on the processor, and when the program code is executed by the processor, the in-vehicle face recognition method described in any one of the technical solutions in the first aspect is implemented.

[0080] The in-vehicle face recognition method, storage medium, and vehicle provided in embodiments of the present application, upon receiving a first preset instruction, control a first image acquisition device to capture a face image of a person in the main driver's seat as a first image, and control a second image acquisition device to capture a face image of a person in the main driver's seat as a second image. Based on the first and second images, the face of the person in the main driver's seat is recognized to obtain a recognition result indicating whether the face recognition in the main driver's seat is successful. This allows face recognition in the main driver's seat to be performed based on images captured by the two image acquisition devices. Because the two image acquisition devices can capture face images from more angles and thus obtain more facial feature information, face recognition can be performed more accurately and efficiently, thereby providing users with a better smart car experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] The disclosure of this application will become more easily understood with reference to the accompanying drawings. Those skilled in the art will readily appreciate that these drawings are merely for the purpose of illustrating this application and are not intended to limit the scope of protection of this application. Furthermore, similar numbers in the drawings represent similar components, where:

[0082] FIG1 is a flow chart showing the main steps of the in-vehicle face recognition method provided by an embodiment of the present application;

[0083] FIG2 is a schematic diagram of the operation of a face recording solution based on a single camera in an embodiment of the present application;

[0084] FIG3 is a flow chart of a face recording solution based on a single camera in an embodiment of the present application;

[0085] FIG4 is a flowchart of a face recording and updating technical solution in an embodiment of the present application;

[0086] FIG5 is a diagram showing the overall interaction logic for in-vehicle face entry and recognition in an embodiment of the present application;

[0087] FIG6 is a logic flow chart of the in-vehicle face recognition method according to an embodiment of the present application;

[0088] FIG7 is a flowchart of a face entry interaction process based on a DMS system and an OMS system according to an embodiment of the present application;

[0089] FIG8 is a flow chart of actively entering the main driving mode in an embodiment of the present application;

[0090] FIG9 is a flow chart of position detection input in an embodiment of the present application. DETAILED DESCRIPTION

[0091] Some embodiments of the present application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and are not intended to limit the scope of protection of the present application.

[0092] In the description of this application, "module" and "processor" may include hardware, software, or a combination of both. A module may include hardware circuitry, various suitable sensors, communication ports, and memory. It may also include software components, such as program code, or a combination of software and hardware. A processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. A processor has data and / or signal processing capabilities. A processor may be implemented in software, hardware, or a combination of both. Non-transitory computer-readable storage media include any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and / or B" refers to all possible combinations of A and B, such as only A, only B, or both A and B. The terms "at least one of A or B" or "at least one of A and B" have similar meanings to "A and / or B" and may include only A, only B, or both A and B. The singular forms "one" and "the" may also include the plural forms.

[0093] In order to solve the technical problems of inaccurate face recognition and long recognition time caused by the use of a single image acquisition device to capture images in the prior art, the present application provides an in-vehicle face recognition method, which is applied to an in-vehicle face recognition system. The system includes: a first image acquisition device and a second image acquisition device, the first image acquisition device is used to capture face images of the main driving area, and the second image acquisition device is used to capture face images of the in-vehicle area. The in-vehicle area includes the main driving area. As shown in Figure 1, the in-vehicle face recognition method in the embodiment of the present application includes the operations described in the following steps S101-S102:

[0094] Step S101, when a first preset instruction is received, controlling the first image acquisition device to acquire a facial image of a person in the main driving area as a first image, and controlling the second image acquisition device to acquire a facial image of a person in the main driving area as a second image;

[0095] In this embodiment, the first preset instruction is a command for performing facial recognition in the driver's seat area. This first preset instruction can be initiated by the user or automatically triggered by the system when certain predetermined conditions are met. That is, the facial recognition process can be triggered by both active and passive triggering. Active triggering can include active interaction methods such as voice triggering and screen touch triggering; passive triggering can include passive triggering methods such as changes in seat sensors or door signal sensors, and changes in the number of people based on vision.

[0096] In practice, users can trigger facial recognition in the driver's seat by sending a voice message to the system, or by clicking or pressing a dedicated function button. Alternatively, the in-car system can automatically monitor changes in the vehicle's interior and, when it detects a user taking the driver's seat, automatically initiate facial recognition to authenticate the driver.

[0097] In this embodiment, two image acquisition devices are used to simultaneously capture facial images of the main driving area, which can obtain more facial feature information from different angles. In order to more conveniently obtain facial images, the above-mentioned image acquisition devices in this embodiment are all cameras.

[0098] In order to more accurately and efficiently capture facial images of people in the main driving area, the first image acquisition device described in this embodiment is a DMS (Driver Monitor System) camera, and the second image acquisition device is an OMS (Occupant monitor system) camera. Accordingly, the DMS camera can be installed in locations such as the A-pillar or steering column of the car, with an HFOV (Horizontal Field of View) of less than 60° and a visual range covering the main driving area, so that it can capture the driver's facial image from the front and obtain a clearer facial image. The OMS camera can be installed in the rearview mirror area of ​​the car, with an HFOV greater than 120°, so that its visual field covers the entire interior area, so that it can capture facial images of people in all areas of the car.

[0099] In this embodiment, the DMS camera can be a 2D camera such as an IR (Infrared) camera or an RGB-IR camera, or a 3D camera based on TOF (Time of Flight) or structured light, or other types of cameras. In actual applications, it can be selected according to actual conditions, and this embodiment does not impose specific restrictions on this; the OMS camera can be an RGB camera or an RGB-IR camera, or other types of cameras. In actual applications, it can be selected according to actual conditions, and this embodiment does not impose specific restrictions on this.

[0100] Step S102: Based on the first image and the second image, recognize the face of the person in the main driving area to obtain a recognition result of whether the face recognition of the person in the main driving area is successful.

[0101] With the increasing number of in-car cameras, facial recognition has become a key factor influencing the intelligent operation of smart cockpits. This embodiment simultaneously recognizes faces in the driver's seat based on the first image captured by the DMS camera and the second image captured by the OMS camera, enabling more efficient and accurate facial recognition results.

[0102] In order to further accurately and efficiently obtain the recognition result of whether the face recognition in the main driver's area is successful, the present embodiment describes recognizing the face in the main driver's area based on the first image and the second image to obtain the recognition result of whether the face recognition in the main driver's area is successful, including: obtaining pre-stored facial feature information corresponding to the first image acquisition device as first reference feature information; recognizing the face in the main driver's area based on the first image and the first reference feature information, and obtaining a first recognition result of whether the recognition is successful; obtaining pre-stored facial feature information corresponding to the second image acquisition device as second reference feature information; recognizing the face in the main driver's area based on the second image and the second reference feature information, and obtaining a second recognition result of whether the recognition is successful; and determining whether the recognition of the face in the main driver's area is successful based on the first recognition result and the second recognition result.

[0103] Specifically, taking the first image acquisition device as a DMS camera and the second image acquisition device as an OMS camera as an example, the DMS camera belongs to the corresponding DMS face recognition system (hereinafter referred to as the DMS system), and the OMS camera belongs to the corresponding OMS face recognition system (hereinafter referred to as the OMS system). The DMS system stores facial feature information pre-recorded through the DMS camera. Similarly, the OMS system stores facial feature information pre-recorded through the OMS camera. When performing facial recognition of the main driver, the DMS system can obtain a recognition result through the facial image collected by the DMS camera and the facial feature information pre-stored by the system; the OMS system can obtain another recognition result through the facial image collected by the OMS camera and the facial feature information pre-stored by the system. Based on the above two recognition results, it can be determined whether the recognition of the face in the main driver area is successful. In this way, the situation where inaccurate recognition results may be caused by using only one camera for face recognition is effectively avoided.

[0104] Furthermore, in order to more accurately obtain the above-mentioned first recognition result, the embodiment described in which the face of the person in the main driving area is recognized based on the first image and the first reference feature information to obtain the first recognition result of whether the recognition is successful includes: extracting features from the first image to obtain first feature information; calculating the similarity between the first feature information and the first reference feature information; and obtaining the first recognition result based on the similarity between the first feature information and the first reference feature information.

[0105] Similarly, in order to obtain the above-mentioned second recognition result more accurately, the embodiment described in this invention recognizes the face of the person in the main driving area based on the second image and the second reference feature information to obtain the second recognition result of whether the recognition is successful, including: extracting features from the second image to obtain second feature information; calculating the similarity between the second feature information and the second reference feature information; and obtaining the second recognition result based on the similarity between the second feature information and the second reference feature information.

[0106] That is, in this embodiment, the DMS system and the OMS system each perform face recognition on the face image acquired by its corresponding camera. During the recognition process, the recognition result is determined by calculating feature similarity.

[0107] Specifically, taking the DMS system as an example, the system pre-stores facial feature information of at least one user. When performing facial recognition, the system first extracts features from the collected facial image to obtain the feature information of the facial image. Then, the similarity between the feature information and each stored facial feature information is calculated to obtain at least one similarity value. The system can pre-set a similarity threshold, for example, 80%. If any of the above similarity values ​​is greater than 80%, it indicates that the current user is a legitimate user, that is, the facial recognition of the current user is successful; if no similarity value is greater than 80%, it indicates that the current user is an illegal user or a user who has not entered the face, that is, the facial recognition of the current user has failed. For the above recognition results of success or failure, the system can display them to the user in the form of voice prompts or text prompts.

[0108] In this embodiment, the specific identification process of the OMS system is the same as the above process of the DMS system, and will not be repeated here.

[0109] In order to more accurately determine whether the recognition of the face of the person in the main driver's seat area is successful, the present embodiment describes determining whether the recognition of the face of the person in the main driver's seat area is successful based on the first recognition result and the second recognition result, including: when one of the first recognition result and the second recognition result is a successful recognition, determining that the recognition of the face of the person in the main driver's seat area is successful; when both the first recognition result and the second recognition result are recognition failures, determining that the recognition of the face of the person in the main driver's seat area is a failure.

[0110] That is, in this embodiment, facial recognition in the driver's seat is considered successful when either the OMS or the DMS successfully recognizes the face. Only when both systems fail is facial recognition in the driver's seat considered a failure. This effectively avoids recognition failures caused by a single camera not capturing a sufficiently clear facial image. Furthermore, because this embodiment simultaneously recognizes faces in the driver's seat through both the OMS and the DMS, the initial facial registration process is significantly simplified compared to existing technologies. For example, in one application scenario, if a user initially enters their facial information through the OMS but not through the DMS, their face can still be successfully recognized when they are in the driver's seat, even if the DMS does not have their facial information. In contrast, in existing technologies, if a user does not enter their facial information through the DMS, their face will not be successfully recognized. This shows that compared to existing technologies, this embodiment effectively reduces facial registration steps and simplifies the facial registration process.

[0111] Furthermore, to ensure that the facial feature information stored in the OMS system can be continuously updated based on the user's actual facial state, so that subsequent facial recognition of the user can be performed more accurately and efficiently, the method described in this embodiment also includes: when the first recognition result is a successful recognition and the second recognition result is a failed recognition, updating the second reference feature information based on the facial information of the person in the main driver's seat area to obtain updated second reference feature information. Under this premise, the method described in this embodiment of recognizing the face of the person in the main driver's seat area based on the second image and the second reference feature information to obtain a second recognition result indicating whether the recognition is successful includes: recognizing the face of the person in the main driver's seat area based on the second image and the second reference feature information to obtain the second recognition result.

[0112] Specifically, when the first recognition result is a success and the second recognition result is a failure, that is, when the DMS system succeeds but the OMS system fails, this indicates that the facial feature information stored in the OMS system cannot accurately reflect the user's current facial features and needs to be updated so that the facial feature information stored in the OMS system can most accurately reflect the user's current facial features. After the OMS system updates the facial feature information, the next time the OMS system performs facial recognition on the user in the main driving area, the OMS system can calculate the facial feature similarity based on the latest facial feature information, thereby more accurately identifying the user's identity.

[0113] In order to update the above-mentioned second reference feature information without the user noticing, so as to further improve the user experience, the embodiment described in this invention updates the second reference feature information based on the facial information of the main driver's seat area to obtain the updated second reference feature information when the first recognition result is a successful recognition and the second recognition result is a failed recognition, including: when the first recognition result is a successful recognition and the second recognition result is a failed recognition, controlling the second image acquisition device to acquire the facial image of the main driver's seat area as a third image; performing feature extraction on the third image to obtain third feature information; and using the third feature information to update the second reference feature information to obtain the updated second reference feature information.

[0114] Specifically, as described above, when the DMS successfully identifies a user but the OMS fails, this indicates that the facial feature information stored in the OMS does not accurately reflect the user's current facial features. In this case, the OMS camera automatically captures the user's current facial image and, based on this current facial image, obtains facial feature information corresponding to the current facial image. This facial feature information is then used to replace the second reference feature information. This new facial feature information replaces the old facial feature information in the OMS, thus obtaining updated second reference feature information.

[0115] During the above-mentioned update process, the OMS camera automatically collects the current facial image of the user, and the user does not need to make the corresponding predetermined action of face entry. That is, for the user, the above-mentioned facial feature update process is completely imperceptible, and the user does not need additional operations. The OMS system can continuously update its stored facial features, greatly improving the user experience.

[0116] In the above process, the reason why the user does not need to perform the corresponding predetermined action of face entry is that at this time the DMS system can successfully identify the user's identity, which means that the DMS system can successfully obtain a facial image of good quality, and the OMS system can also obtain a facial image of good quality accordingly. Based on this facial image, the OMS system can already obtain sufficient facial features. Therefore, the OMS system no longer needs to guide the user to perform the corresponding predetermined action.

[0117] Furthermore, to ensure that the facial feature information stored in the DMS system is continuously updated based on the user's actual facial state, thereby enabling more accurate and efficient subsequent facial recognition of the user, the method described in this embodiment further includes: when the second recognition result is a successful recognition and the first recognition result is a failed recognition, updating the first reference feature information based on the facial information of the person in the main passenger seat area to obtain updated first reference feature information. Under this premise, the method described in this embodiment of recognizing the face of the person in the main passenger seat area based on the first image and the first reference feature information to obtain a first recognition result indicating whether the recognition is successful includes: recognizing the face of the person in the main passenger seat area based on the first image and the first reference feature information to obtain the first recognition result.

[0118] Specifically, when the second recognition result is a success and the first recognition result is a failure, that is, when the OMS system succeeds but the DMS system fails, this indicates that the facial feature information stored in the DMS system does not accurately reflect the user's current facial features and needs to be updated so that the facial feature information stored in the DMS system can most accurately reflect the user's current facial features. After the DMS system updates the facial feature information, the next time the DMS system performs facial recognition on the user in the main driving area, the DMS system can calculate the facial feature similarity based on the latest facial feature information, thereby more accurately identifying the user's identity.

[0119] In order to update the above-mentioned first reference feature information without the user noticing, so as to further improve the user experience, the embodiment described in this invention updates the first reference feature information based on the facial information of the main driver's seat area to obtain the updated first reference feature information when the second recognition result is a successful recognition and the first recognition result is a failed recognition, including: when the second recognition result is a successful recognition and the first recognition result is a failed recognition, controlling the first image acquisition device to acquire the facial image of the main driver's seat area as a fourth image; performing feature extraction on the fourth image to obtain fourth feature information; and using the fourth feature information to update the first reference feature information to obtain the updated first reference feature information.

[0120] Specifically, as described above, when the OMS successfully identifies a user but the DMS fails, this indicates that the facial feature information stored in the DMS does not accurately reflect the user's current facial features. In this case, the DMS camera automatically captures the user's current facial image and, based on this current facial image, obtains facial feature information corresponding to the current facial image. This facial feature information is then used to replace the first reference feature information. This means that the new facial feature information replaces the old facial feature information in the DMS, thus obtaining updated first reference feature information.

[0121] During the above-mentioned update process, the DMS camera automatically collects the current facial image of the user, and the user does not need to make the corresponding predetermined action of face entry. That is, for the user, the above-mentioned facial feature update process is completely imperceptible, and the user does not need additional operations. The DMS system can continuously update its stored facial features, greatly improving the user experience.

[0122] In the above process, the reason why the user does not need to perform the corresponding predetermined action of face entry is that at this time the OMS system can successfully identify the user's identity, which means that the OMS system can successfully obtain a facial image of good quality, and the DMS system can also obtain a facial image of good quality accordingly. Based on this facial image, the DMS system can already obtain sufficient facial features. Therefore, the DMS system no longer needs to guide the user to perform the corresponding predetermined action.

[0123] In order to more accurately identify the faces of people in the main driving area, the first reference feature information and the second reference feature information described in this embodiment are obtained based on pre-entry of the faces of people in the main driving area.

[0124] In this embodiment, the system recognizes faces in the driver's seat by comparing the current facial image captured in the driver's seat with a previously recorded facial image in the driver's seat. Therefore, when the in-vehicle facial recognition system is activated, the user's face or facial features must be recorded for subsequent identification.

[0125] In order to enter user facial information more accurately and effectively, the first reference feature information and the second reference feature information described in this embodiment are obtained in the following manner: when a second preset instruction is received, a first preset interaction strategy is obtained; wherein, the first preset interaction strategy is used to guide the face of the person in the main driving area to perform a first predetermined action; the first image acquisition device is controlled to acquire a facial image of the person in the main driving area including the first predetermined action as the fifth image, and the second image acquisition device is controlled to acquire a facial image of the person in the main driving area including the first predetermined action as the sixth image; feature extraction is performed on the fifth image to obtain the first reference feature information; feature extraction is performed on the sixth image to obtain the second reference feature information.

[0126] Specifically, in the initial state of the DMS system and the OMS system, that is, when no facial feature information is stored in the DMS system and the OMS system, the user's face needs to be entered to obtain the above-mentioned first reference feature information and second reference feature information. When the user enters the face in the main driving area, the system will automatically obtain the interaction strategy corresponding to the main driving area to guide the user's face to make predetermined movements, such as looking straight ahead, looking at the car instrument screen, looking at the car central control screen, etc., so that the DMS camera and the OMS camera can both obtain facial images from different angles. Afterwards, the DMS system performs feature extraction on the facial image obtained by the DMS camera, that is, it can obtain the above-mentioned first reference feature information; the OMS system performs feature extraction on the facial image obtained by the OMS camera, that is, it can obtain the above-mentioned second reference feature information. In actual applications, before the feature extraction of the acquired facial image, operations such as facial image detection, feature point detection, posture detection, and image quality judgment are also included to obtain more effective facial feature information.

[0127] In order to facilitate user operation and further improve user experience, the second preset instruction described in this embodiment is issued by the user; or, the second preset instruction described in this embodiment is triggered by the first predetermined condition.

[0128] Specifically, the second preset command is a command to record a face in the main driver's area. This command can be initiated by the user or automatically triggered by the system when it detects that a predetermined condition is met. As shown in Figure 8, the user can trigger the main driver's registration process by clicking the specially set "Main Driver Registration" button, or the user can trigger the main driver's registration process by issuing the "Main Driver Registration" command through voice.

[0129] In order to facilitate user operations and further improve user experience, this embodiment triggers the second preset instruction, including: determining whether both the first reference feature information and the second reference feature information do not exist; when both the first reference feature information and the second reference feature information do not exist, detecting the face position and number of faces in the in-vehicle area; when it is determined that faces exist only in the main driving area, triggering the second preset instruction.

[0130] That is, in this embodiment, the system will detect whether the first reference feature information and the second reference feature information currently exist. When neither of the above two information exists, it indicates that the user face has not yet been entered. At this time, as shown in Figure 9, the system automatically detects the face position and number of faces in the vehicle interior area. When it is determined that there is a face only in the main driver's seat area, it automatically enters the main driver's seat entry process. The above face position refers to the user position in the vehicle, and the above face number refers to the number of users in the vehicle. In actual applications, the user position and the number of users can be detected by seat sensors or by visual signal detection. This embodiment does not impose any specific restrictions on the specific detection method.

[0131] In this embodiment, the interior area also includes the passenger seat area. To further facilitate user operation and enhance the user experience, in this embodiment, triggering the second preset instruction further includes: displaying a preset entry location selection page when it is determined that a face is present in both the driver's seat area and the passenger seat area; receiving an entry location selected by the user based on the entry location selection page; and triggering the second preset instruction when the user-selected entry location is the driver's seat area.

[0132] That is, in this embodiment, when the system detects that there are people in both the main driver's seat and the co-driver's seat, a selection page pops up for the user to select the face entry location. When the user chooses to enter the face in the main driver's seat, the main driver's seat entry process is entered.

[0133] As mentioned above, the vehicle interior area described in this embodiment also includes the passenger area. In order to more accurately recognize faces in the passenger area or the main driver area, the second reference feature information described in this embodiment is obtained based on pre-recorded faces in the passenger area.

[0134] That is, in this embodiment, when the system recognizes a face in the passenger seat, it obtains the recognition result by comparing the facial image currently captured in the passenger seat with the facial image previously recorded in the passenger seat. When the system recognizes a face in the driver's seat, it can obtain the recognition result by comparing the facial image currently captured in the driver's seat with the facial image previously recorded in the passenger seat. Therefore, when the in-vehicle facial recognition system is initially activated, the user's face or facial features must be recorded in order to subsequently identify the user.

[0135] In order to enter user facial information more accurately and effectively, the second reference feature information described in this embodiment is obtained in the following manner: when a third preset instruction is received, a second preset interaction strategy is obtained; wherein, the second preset interaction strategy is used to guide the face of the person in the passenger seat area to perform a second predetermined action; the second image acquisition device is controlled to acquire a facial image of the person in the passenger seat area including the second predetermined action as a seventh image; and feature extraction is performed on the seventh image to obtain the second reference feature information.

[0136] Specifically, in the initial state of the DMS system and the OMS system, that is, when no facial feature information is stored in the DMS system and the OMS system, it is necessary to enter the user's face to at least obtain the above-mentioned second reference feature information. When the user enters the face in the co-pilot area, the system will automatically obtain the interaction strategy corresponding to the co-pilot area to guide the user's face to make predetermined movements, such as looking straight ahead, looking at the car's central control screen, etc., so that the OMS camera can obtain facial images at different angles. Afterwards, the OMS system extracts features from the facial image obtained by the OMS camera, and the above-mentioned second reference feature information can be obtained. In actual applications, before the feature extraction of the acquired facial image, operations such as facial image detection, feature point detection, posture detection, and image quality judgment are also included to obtain more effective facial feature information.

[0137] In order to facilitate user operation and further improve user experience, the third preset instruction described in this embodiment is issued by the user; or, the third preset instruction described in this embodiment is triggered by the second predetermined condition.

[0138] Specifically, the third preset command is a command to record a face in the passenger seat area. This command can be initiated by the user or automatically triggered by the system when it detects that a predetermined condition is met. The user can trigger the passenger seat registration process by clicking a specially set "Passenger seat registration" button or by issuing a voice command to trigger the passenger seat registration process.

[0139] In order to facilitate user operations and further improve user experience, this embodiment triggers the third preset instruction, including: determining whether the first reference feature information and the second reference feature information do not exist; when the first reference feature information and the second reference feature information do not exist, detecting the face position and number of faces in the in-vehicle area; when it is determined that there are faces only in the co-pilot area, triggering the third preset instruction.

[0140] That is, in this embodiment, the system will detect whether the first reference feature information and the second reference feature information currently exist. When neither of the above exists, it indicates that the user's face has not been entered. At this time, as shown in Figure 9, the system automatically detects the face position and number of faces in the area inside the car. When it is determined that there is a face only in the co-pilot area, it automatically enters the co-pilot entry process.

[0141] In order to further facilitate user operations and further improve user experience, this embodiment triggers the third preset instruction and also includes: when it is determined that there are faces in both the main driver area and the passenger seat area, displaying a preset entry position selection page; receiving the entry position selected by the user based on the entry position selection page; when the entry position selected by the user is the passenger seat area, triggering the third preset instruction.

[0142] That is, in this embodiment, when the system detects that there are people in both the main driver's seat and the co-pilot seat, a selection page pops up for the user to select the face entry location. When the user chooses to enter the face in the co-pilot seat, the co-pilot entry process is entered.

[0143] As mentioned above, the vehicle interior area described in this embodiment also includes the passenger seat area. To effectively recognize faces in the passenger seat area, the method described in this embodiment further includes: upon receiving a fourth preset instruction, controlling the second image acquisition device to capture an image of the face of the person in the passenger seat area as an eighth image; and recognizing the face of the person in the passenger seat area based on the eighth image to obtain a recognition result of whether the face recognition in the passenger seat area is successful.

[0144] In this embodiment, the vehicle interior area also includes a rear area. To effectively recognize faces in the rear area, the method of this embodiment further includes: upon receiving a fifth preset instruction, controlling the second image acquisition device to capture an image of faces in the rear area as a ninth image; and recognizing faces in the rear area based on the ninth image to determine whether face recognition in the rear area is successful.

[0145] That is, in this embodiment, the OMS system can also be used to recognize faces in the passenger and rear passenger areas. Specifically, for the passenger passenger area, the current facial image of the person in the passenger passenger area currently captured by the OMS camera is compared with the facial image previously recorded in the passenger passenger area to obtain a facial recognition result for the passenger passenger area; for the rear passenger area, the current facial image of the person in the rear passenger area currently captured by the OMS camera is compared with the facial image previously recorded in the passenger passenger area to obtain a facial recognition result for the rear passenger area.

[0146] The overall process of the above technical solution is shown in Figure 4, which includes the following steps:

[0147] S1. Obtain facial images captured by the DMS camera and the OMS camera and provide corresponding interactive guidance for facial entry; wherein, the interactive guidance for facial entry can guide the user to enter the face through voice prompts, facial model prompts, real-time facial image stream prompts, etc.

[0148] S2. During the boot process, facial images captured by the DMS and OMS cameras are processed and the facial recording results are output. This facial recording process includes feature detection and extraction of facial images by the DMS and OMS systems, as well as sequential logic processing between the DMS and OMS systems. The specific sequential logic processing is shown in Figure 7.

[0149] S3. During the face recognition process, the face features entered in the DMS and OMS systems are updated based on the recognition results of the DMS and OMS systems. The triggering condition for the face update process is the face recognition results of the DMS and OMS systems.

[0150] Figure 5 is the overall interactive logic diagram for in-vehicle face entry and recognition in this embodiment. As shown in Figure 5, when entering a face in the main driver's seat, the DMS system and the OMS system can simultaneously obtain facial features; when entering a face in the passenger seat, only the OMS system can obtain facial features. Afterwards, when performing face recognition, when recognizing a face in the main driver's seat, the DMS system and the OMS system are used to jointly recognize the face; when recognizing a face in the passenger seat, only the OMS system is used to recognize the face. The selection of the entry location can be triggered manually by the user or based on the position detection results of the OMS vision.

[0151] Based on the face entry interaction between the DMS system and the OMS system, on the premise that the face entry is successful, at least one facial feature of the face entry can be obtained, and then based on the process shown in Figure 6, the feature update method is used to ensure that the DMS system and the OMS system have their own facial features.

[0152] As shown in Figure 6, first, the user initiates facial recognition verification, at which time the occupancy detection of people in the car will be triggered. If there is someone in the main driver's seat, facial recognition of both the DMS system and the OMS system will be triggered simultaneously. If both fail to recognize successfully, a failure result will be returned. If both are successfully recognized, a success result will be returned, and no facial update processing will be performed. If the DMS system successfully recognizes but the OMS system fails to recognize successfully, a facial update of the OMS system will be triggered, and the facial features of the current OMS system will be updated. If the OMS system successfully recognizes but the DMS system fails to recognize successfully, a facial update of the DMS system will be triggered, and the facial feature library of the current DMS system will be updated. Except for the main driver's seat, facial recognition of other positions will be recognized by the OMS system, and the corresponding facial recognition results will be fed back.

[0153] In this embodiment, after obtaining a face in any of the DMS and OMS systems, when the user changes location, any system will recognize the user. At this time, updating the system without facial features can well ensure that the user has DMS+OMS facial features in subsequent actual facial recognition experiences.

[0154] Figure 7 shows the face entry interaction processing method based on the DMS system and the OMS system. As shown in Figure 7, the face entry position information is first obtained through the entry position detection module, and it is further determined whether the request is initiated by the main driver. For the co-driver's request, the face entry will be performed only based on the OMS camera, and the user will be allowed to perform relevant entry interactions through the central control screen. If the entry is successful, the user will be prompted in time that the entry is successful, and if the entry fails, the user will also be prompted. For the main driver's request, this embodiment will trigger the face entry of the DMS system and the OMS system at the same time, allowing the user to perform DMS face entry on the instrument screen and OMS face entry on the central control screen at the same time. As long as the DMS entry is successful or the OMS entry is successful, the entry can be determined to be successful. Of course, if both fail, it will be considered that the face entry failed.

[0155] Among them, for the input interaction of the main driver, the input triggering timing of DMS and OMS can be selected as simultaneous triggering or sequential triggering.

[0156] It should be noted that the reason for the difference in the operation process of the driver and co-driver input in Figure 7 is that the installation positions of the DMS camera and the OMS camera are inconsistent. The driver's DMS + OMS can capture a clear face, while the co-driver's OMS only captures a clear face.

[0157] The above-mentioned entry location can be acquired through active or passive methods. Passive methods include detection based on seat sensors or visual signals, while active methods can be acquired through user clicks, voice interaction, etc. The flowchart for entering the main driver's seat using the active method is shown in Figure 8. The entry location detection flowchart is shown in Figure 9.

[0158] Based on the above steps S101-S102, the present application solves the technical problems in the prior art of inaccurate face recognition and long recognition time caused by using a single image acquisition device to capture images.

[0159] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effect of the present application, different steps do not have to be performed in such an order. They can be performed simultaneously (in parallel) or in other orders. These changes are within the scope of protection of the present application.

[0160] The technical solution provided by an embodiment of the present application, upon receiving a first preset instruction, controls the first image acquisition device to capture a facial image of a person in the main driver's seat as a first image, and controls the second image acquisition device to capture a facial image of the person in the main driver's seat as a second image. Based on the first and second images, the face of the person in the main driver's seat is recognized to obtain a recognition result indicating whether the facial recognition in the main driver's seat is successful. This allows facial recognition in the main driver's seat to be performed based on images captured by both image acquisition devices. Because the two image acquisition devices can capture facial images from more angles and thus obtain more facial feature information, facial recognition can be performed more accurately and efficiently, thereby providing users with a better smart car experience.

[0161] Furthermore, in practical applications, a simple and convenient facial entry and recognition process is an important foundation for users to use facial recognition-related functions. However, because existing in-car facial recognition uses independent cameras to capture and recognize facial images in different areas, facial entry requires multiple face entry for different recognition systems, resulting in frequent interactions between users and the system. In other words, the facial entry and recognition process in existing technologies is basically entry for a single recognition system or independent face entry based on multiple recognition systems, and recognition is also for faces in a single location. The facial entry solution based on a single recognition system is shown in Figure 2. During the entry process, the user obtains a facial image of the main driver's area based on the DMS camera. The user performs relevant entry action matching based on the prompts during the entry process. The underlying algorithm service collects the DMS image and performs relevant face detection, feature point detection, posture detection, facial feature extraction, and other algorithms to build a facial feature database, which serves as the basis for subsequent facial recognition.

[0162] The solution based on independent face entry with multiple cameras is shown in Figure 3. When using the face recognition function, users need to perform face-related entry based on the DMS system and OMS system respectively. Faces need to be entered twice and only face recognition in two specific positions (driver's seat / passenger seat) is supported.

[0163] Based on Figures 2 and 3 and the above, the existing technology still has the following drawbacks: a. Face entry based on the DMS and OMS systems requires two interactions, and the user must move each time (from the driver's seat to the front passenger seat or the back seat), which is cumbersome and can easily cause user frustration; b. Face entry based on the DMS system only covers the driver's seat and cannot cover face recognition for the entire vehicle. In summary, the existing technology solutions have the problems of multiple face entry interactions and a small face recognition area in the entire vehicle during in-vehicle face interaction, which affects the user's face interaction experience.

[0164] Specific scenario examples are as follows:

[0165] a. The user completes facial entry and stores facial features using the OMS system in the passenger or rear seat. When the user returns to the driver's seat, the DMS system cannot be used for identity authentication. If the user wants to use the DMS system for facial recognition, the user needs to sit back in the driver's seat and re-enter the face. b. The user completes facial entry and stores facial features using the DMS system in the driver's seat. When the user switches to the passenger or rear seat, the OMS system cannot complete facial verification.

[0166] Based on the above steps S101-S102, this application can also solve the problems of the existing technology in that the number of face entry interactions is large and the area where faces can be recognized in the entire vehicle is small.

[0167] Specifically, when a user enters their face in the driver's seat, the DMS and OMS systems can obtain facial features at the same time without having to enter the face twice. During the entry process, the user only needs to interact with the system once. When a user enters their face in the passenger seat, although only the OMS system obtains the facial features, when facial recognition is subsequently performed in the driver's seat, the facial features in the DMS system can be updated based on the recognition results of the OMS system. Regardless of which system the facial features are updated in, the update is performed without the user's knowledge. That is, during the entire process, the user only needs to enter the face once in the driver's seat or the passenger seat, and no additional operations are required afterwards, which greatly improves the user experience.

[0168] Based on the above analysis, this application actually solves the following problems: the design of a solution for related face updates using the DMS system and the OMS system reduces the number of face entries and the switching of entry positions, increases the face recognition area in the car, and improves the user's smart cockpit interaction experience.

[0169] This application also has the following technical advantages:

[0170] The present application provides a function of realizing full-cabin face recognition in a vehicle based on an input interaction method between the DMS system and the OMS system, which can effectively realize the function of one-time face input of users in the vehicle and face recognition of the entire vehicle. Among them, the face update method based on the recognition results of the DMS system and the OMS system in Figure 6 can solve the problem of "the co-driver's input cannot be recognized by the main driver" and the problem of "the main driver's input cannot be recognized by the entire vehicle"; the face input interaction based on the DMS system and the OMS system in Figure 7 can reduce the number of user face inputs and shorten the user input experience link in the vehicle.

[0171] Furthermore, the present application also provides a computer-readable storage medium. In a computer-readable storage medium embodiment according to the present application, the computer-readable storage medium can be configured to store a program for executing the in-vehicle face recognition method of the above-mentioned method embodiment, and the program can be loaded and run by the processor to implement the above-mentioned in-vehicle face recognition method. For ease of explanation, only the parts related to the embodiment of the present application are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present application. The computer-readable storage medium can be a storage device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present application is a non-transitory computer-readable storage medium.

[0172] Furthermore, the present application also provides a vehicle, which includes a storage device and a processor, wherein the storage device stores program code that can be run on the processor, and when the program code is executed by the processor, the in-vehicle face recognition method as described in any of the above-mentioned in-vehicle face recognition method technical solutions is implemented.

[0173] In this embodiment, the storage device in the vehicle may be configured to store a program for executing the in-vehicle face recognition method of the above-described method embodiment, and the processor may be configured to execute the program in the storage device, including but not limited to a program for executing the in-vehicle face recognition method of the above-described method embodiment. For ease of illustration, only the portions relevant to the embodiments of this application are shown. For specific technical details not disclosed, please refer to the method section of the embodiments of this application.

[0174] In the embodiment of the present application, the vehicle may be a control device formed by various electronic devices. In some possible implementations, the device may include multiple storage devices and multiple processors. The program for executing the in-vehicle face recognition method of the above-mentioned method embodiment can be divided into multiple subroutines, and each subroutine can be loaded and run by the processor to execute different steps of the in-vehicle face recognition method of the above-mentioned method embodiment. Specifically, each subroutine can be stored in different memories respectively, and each processor can be configured to execute the program in one or more memories to jointly implement the in-vehicle face recognition method of the above-mentioned method embodiment, that is, each processor executes different steps of the in-vehicle face recognition method of the above-mentioned method embodiment respectively to jointly implement the in-vehicle face recognition method of the above-mentioned method embodiment.

[0175] The multiple processors may be deployed on the same device. For example, the vehicle may be a high-performance device comprised of multiple processors, and the multiple processors may be processors configured on the high-performance device. Furthermore, the multiple processors may be deployed on different devices. For example, the vehicle may be a server cluster, and the multiple processors may be processors on different servers within the server cluster.

[0176] It will be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment of the present application can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of each of the above method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium can include: any entity or device, medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium that can carry the computer program code. It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable storage media do not include electric carrier signals and telecommunication signals.

[0177] Thus far, the technical solutions of the present application have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present application.

Claims

1. A method for in-vehicle face recognition, characterized in that: Applicable to an in-vehicle face recognition system; the system comprises: a first image acquisition device and a second image acquisition device; the first image acquisition device is used to acquire a face image of a main driving area; the second image acquisition device is used to acquire a face image of an in-vehicle area; wherein the in-vehicle area includes the main driving area; the method comprises: When a first preset instruction is received, the first image acquisition device is controlled to acquire the face image of the person in the main driving area as the first image, and the second image acquisition device is controlled to acquire the face image of the person in the main driving area as the second image; Based on the first image and the second image, the face of the person in the main driving area is recognized to obtain a recognition result of whether the face recognition of the person in the main driving area is successful.

2. The in-vehicle face recognition method according to claim 1, characterized in that: The step of recognizing a face in the main driving area based on the first image and the second image to obtain a recognition result of whether the face recognition in the main driving area is successful includes: Acquire pre-stored facial feature information corresponding to the first image acquisition device as first reference feature information; Based on the first image and the first reference feature information, recognize the face of the person in the main driving area to obtain a first recognition result of whether the recognition is successful; Acquire pre-stored facial feature information corresponding to the second image acquisition device as second reference feature information; Based on the second image and the second reference feature information, recognize the face of the person in the main driving area to obtain a second recognition result of whether the recognition is successful; Based on the first recognition result and the second recognition result, it is determined whether recognition of the face of the person in the main driving area is successful.

3. The in-vehicle face recognition method according to claim 2, characterized in that: The step of recognizing a face in the main driving area based on the first image and the first reference feature information to obtain a first recognition result of whether the recognition is successful includes: Performing feature extraction on the first image to obtain first feature information; Calculating the similarity between the first feature information and the first reference feature information; The first recognition result is obtained based on the similarity between the first feature information and the first reference feature information.

4. The in-vehicle face recognition method according to claim 2, characterized in that: The step of recognizing the face of the person in the main driving area based on the second image and the second reference feature information to obtain a second recognition result of whether the recognition is successful includes: performing feature extraction on the second image to obtain second feature information; Calculating the similarity between the second feature information and the second reference feature information; The second recognition result is obtained based on the similarity between the second feature information and the second reference feature information.

5. The in-vehicle face recognition method according to claim 2, characterized in that: The determining whether the recognition of the face of the person in the main driving area is successful based on the first recognition result and the second recognition result includes: When one of the first recognition result and the second recognition result is a successful recognition, determining that the recognition of the face of the person in the main driving area is successful; When both the first recognition result and the second recognition result are recognition failures, it is determined that recognition of the face in the main driving area has failed.

6. The in-vehicle face recognition method according to claim 5, characterized in that: The method further comprises: When the first recognition result is recognition success and the second recognition result is recognition failure, updating the second reference feature information based on the face information of the main driving area to obtain updated second reference feature information; The step of recognizing the face of the person in the main driving area based on the second image and the second reference feature information to obtain a second recognition result of whether the recognition is successful includes: Based on the second image and the updated second reference feature information, the face of the person in the main driving area is recognized to obtain the second recognition result.

7. The in-vehicle face recognition method according to claim 6, characterized in that: When the first recognition result is a recognition success and the second recognition result is a recognition failure, updating the second reference feature information based on the face information of the main driving area to obtain the updated second reference feature information includes: When the first recognition result is a recognition success and the second recognition result is a recognition failure, controlling the second image acquisition device to acquire the face image of the main driving area as a third image; Performing feature extraction on the third image to obtain third feature information; The second reference characteristic information is updated using the third characteristic information to obtain the updated second reference characteristic information.

8. The in-vehicle face recognition method according to claim 5, characterized in that: The method further comprises: When the second recognition result is recognition success and the first recognition result is recognition failure, updating the first reference feature information based on the face information of the main driving area to obtain updated first reference feature information; The step of recognizing a face in the main driving area based on the first image and the first reference feature information to obtain a first recognition result of whether the recognition is successful includes: Based on the first image and the updated first reference feature information, the face of the person in the main driving area is recognized to obtain the first recognition result.

9. The in-vehicle face recognition method according to claim 8, characterized in that: When the second recognition result is a recognition success and the first recognition result is a recognition failure, updating the first reference feature information based on the face information of the main driving area to obtain the updated first reference feature information includes: When the second recognition result is recognition success and the first recognition result is recognition failure, controlling the first image acquisition device to acquire the face image of the main driving area as the fourth image; performing feature extraction on the fourth image to obtain fourth feature information; The first reference characteristic information is updated by using the fourth characteristic information to obtain the updated first reference characteristic information.

10. The in-vehicle face recognition method according to claim 2, characterized in that: The first reference feature information and the second reference feature information are obtained based on pre-entering a face of a person in the main driving area.

11. The in-vehicle face recognition method according to claim 10, characterized in that: The first reference feature information and the second reference feature information are obtained in the following manner: When the second preset instruction is received, a first preset interaction strategy is obtained; wherein the first preset interaction strategy is used to guide the face of the person in the main driving area to perform a first predetermined action; Controlling the first image acquisition device to acquire a facial image of the person in the main driving area including the first predetermined action as a fifth image, and controlling the second image acquisition device to acquire a facial image of the person in the main driving area including the first predetermined action as a sixth image; performing feature extraction on the fifth image to obtain the first reference feature information; Perform feature extraction on the sixth image to obtain the second reference feature information.

12. The in-vehicle face recognition method according to claim 11, characterized in that: The second preset instruction is issued by a user; or, the second preset instruction is triggered by a first predetermined condition.

13. The in-vehicle face recognition method according to claim 12, characterized in that: Triggering the second preset instruction includes: Determining whether both the first reference feature information and the second reference feature information do not exist; When neither the first reference feature information nor the second reference feature information exists, detecting the face positions and the number of faces in the vehicle interior area; When it is determined that there is a face only in the main driving area, the second preset instruction is triggered.

14. The in-vehicle face recognition method according to claim 13, characterized in that: The vehicle interior area also includes a passenger area; Triggering the second preset instruction further includes: When it is determined that there are faces in both the main driver area and the co-driver area, displaying a preset input position selection page; receiving an entry location selected by a user based on the entry location selection page; When the entry location selected by the user is the main driving area, the second preset instruction is triggered.

15. The in-vehicle face recognition method according to claim 2, characterized in that: The in-vehicle area also includes a passenger seat area; the second reference feature information is obtained based on pre-entering a face of a person in the passenger seat area.

16. The in-vehicle face recognition method according to claim 15, characterized in that: The second reference characteristic information is obtained in the following manner: When the third preset instruction is received, a second preset interaction strategy is obtained; wherein the second preset interaction strategy is used to guide the face in the passenger seat area to perform a second predetermined action; Controlling the second image acquisition device to acquire a facial image of the passenger area including the second predetermined action as a seventh image; Perform feature extraction on the seventh image to obtain the second reference feature information.

17. The in-vehicle face recognition method according to claim 16, characterized in that: The third preset instruction is issued by a user; or, the third preset instruction is triggered by a second predetermined condition.

18. The in-vehicle face recognition method according to claim 17, characterized in that: Triggering the third preset instruction includes: Determining whether both the first reference feature information and the second reference feature information do not exist; When neither the first reference feature information nor the second reference feature information exists, detecting the face positions and the number of faces in the vehicle interior area; When it is determined that there is a face only in the passenger seat area, the third preset instruction is triggered.

19. The in-vehicle face recognition method according to claim 18, characterized in that: Triggering the third preset instruction further includes: When it is determined that there are faces in both the main driver area and the co-driver area, displaying a preset input position selection page; receiving an entry location selected by a user based on the entry location selection page; When the entry position selected by the user is the passenger seat area, the third preset instruction is triggered.

20. The in-vehicle face recognition method according to claim 1, characterized in that: The vehicle interior area also includes a passenger area; and the method further includes: When receiving the fourth preset instruction, controlling the second image acquisition device to acquire the face image of the passenger area as the eighth image; Based on the eighth image, the face of the person in the passenger seat area is recognized to obtain a recognition result of whether the face recognition of the person in the passenger seat area is successful.

21. The in-vehicle face recognition method according to claim 1, characterized in that: The vehicle interior area also includes a rear area; and the method further includes: When the fifth preset instruction is received, controlling the second image acquisition device to acquire the face image of the rear area as the ninth image; Based on the ninth image, faces of people in the rear area are recognized to obtain a recognition result of whether face recognition of the rear area is successful.

22. A computer-readable storage medium storing a plurality of program codes, characterized in that: The program code is suitable for being loaded and run by a processor to execute the in-vehicle face recognition method according to any one of claims 1 to 21.

23. A vehicle, characterized in that: The vehicle includes a storage device and a processor, the storage device stores a program code that can be run on the processor, and when the program code is executed by the processor, the in-vehicle face recognition method as described in any one of claims 1 to 21 is implemented.

Images