A method of adjusting a vehicle mirror, a vehicle and a system

Abstract

The present disclosure relates to a method, a vehicle and a system for adjusting a rearview mirror of a vehicle. The method is applied to a vehicle, and the method comprises: receiving at least one frame of image sent by smart glasses which establishes a connection with the vehicle, the at least one frame of image being obtained by the smart glasses shooting a current field of view of a driver under the condition that the smart glasses detect that the driver performs a head turning action; determining a current field of view coverage of a target rearview mirror at a current angle according to the at least one frame of image, the target rearview mirror being a target rearview mirror which the driver is interested in; determining a target angle of the target rearview mirror according to the current field of view coverage, and driving the target rearview mirror to adjust to the target angle.

Description

Technical Field

[0001] This disclosure relates to the field of vehicle auxiliary system technology, and more specifically, to a method for adjusting a vehicle rearview mirror, a vehicle, and a vehicle rearview mirror adjustment system. Background Technology

[0002] Vehicle exterior rearview mirrors, as important auxiliary devices for drivers to observe road conditions behind them, play a crucial role in driving safety. With the development of intelligent driving technology, drivers have placed higher demands on the ease of adjustment and the comprehensiveness of the field of vision covered by rearview mirrors.

[0003] In existing technologies, vehicle exterior rearview mirrors mostly rely on manual adjustment by the driver or preset memory positions for angle setting. This adjustment method has the following limitations: In complex driving environments, different drivers have significant differences in height, sitting posture, and observation habits, and fixed preset positions are difficult to meet individual needs; it is difficult for drivers to frequently manually adjust the rearview mirror angle while driving, resulting in large blind spots in scenarios such as lane changes and reversing; traditional adjustment methods cannot dynamically optimize the rearview mirror angle according to the driver's real-time line of sight, affecting driving safety. Summary of the Invention

[0004] One object of this disclosure is to provide a technical solution for adjusting vehicle rearview mirrors.

[0005] According to one aspect of the present invention, a method for adjusting a vehicle rearview mirror is provided, the method being applied to a vehicle, the method comprising: Receive at least one frame of image sent by smart glasses that have established a connection with the vehicle, wherein the at least one frame of image is captured by the smart glasses when the driver performs a head turning action and the driver's current field of vision is detected; The current field of view coverage of the target rearview mirror of interest to the driver is determined at the current angle based on the at least one frame of image. The target angle of the target rearview mirror is determined based on the current field of view coverage, and the target rearview mirror is driven to adjust to the target angle.

[0006] Optionally, the method further includes: receiving infrared light signals emitted by the smart glasses when the driver performs a head turning action by using an infrared receiving array set at the two rearview mirrors of the vehicle; analyzing the received infrared light signals corresponding to each rearview mirror to obtain the corresponding infrared light signal intensity; and determining the target rearview mirror among the two rearview mirrors based on the infrared light signal intensity corresponding to each rearview mirror.

[0007] Optionally, the method further includes: parsing the received infrared light signal corresponding to each rearview mirror to obtain the corresponding recognition confidence level; wherein, the recognition confidence level represents the confidence level that the infrared light signal received by the corresponding infrared receiving array is the infrared light signal emitted by the smart glasses; obtaining a confidence level threshold; and, if the recognition confidence level is greater than or equal to the confidence level threshold, performing the step of determining the current field of view coverage of the target rearview mirror of interest to the driver at the current angle based on the at least one frame of image.

[0008] Optionally, obtaining the confidence threshold includes: detecting the current brightness of the environment in which the vehicle is located, and determining the confidence threshold based on the current brightness.

[0009] Optionally, the method further includes: parsing the infrared light signal received by the target rearview mirror to obtain the current ideal observation direction of the target rearview mirror; generating a theoretically optimal field of view area based on the current ideal observation direction; and determining the current field of view coverage based on the theoretically optimal field of view area.

[0010] Optionally, determining the current field of view coverage of the target rearview mirror of interest to the driver at the current angle based on the at least one frame of image includes: performing a fusion process on the at least one frame of image to obtain a fused image; identifying target image regions containing target objects in the fused image and determining the proportion of each target image region in the fused image; and determining the current field of view coverage based on the proportion of each target image region in the fused image.

[0011] Optionally, the target angle includes an angular component in a first direction and an angular component in a second direction; determining the target angle of the target rearview mirror based on the current field of view coverage includes: identifying the centroid of a key target in the fused image and the center of the target rearview mirror in the fused image; calculating a first angular deviation between the centroid of the key target and the center of the target rearview mirror in the first direction and a second angular deviation in the second direction based on the centroid of the key target and the center of the target rearview mirror; wherein the first direction is perpendicular to the second direction; obtaining the angular component of the target angle in the first direction based on the first angular deviation and the angular component of the current angle in the first direction; obtaining the angular component of the target angle in the second direction based on the second angular deviation and the angular component of the current angle in the second direction.

[0012] Optionally, determining the target angle of the target rearview mirror based on the current field of view coverage includes: identifying the proportion of the blind spot range in the fused image; determining a current field of view score based on the current field of view coverage, the proportion of the blind spot range, and the visibility of the following vehicle; and determining the target angle of the target rearview mirror based on the current field of view coverage if the current field of view score is less than a set score threshold.

[0013] According to a second aspect of this disclosure, a vehicle is provided, including a memory and a processor, the memory for storing a computer program; the processor for executing the computer program to implement the following method: Receive at least one frame of image sent by smart glasses that have established a connection with the vehicle, wherein the at least one frame of image is captured by the smart glasses when the driver performs a head turning action and the driver's current field of vision is detected; The current field of view coverage of the target rearview mirror of interest to the driver is determined at the current angle based on the at least one frame of image. The target angle of the target rearview mirror is determined based on the current field of view coverage, and the target rearview mirror is driven to adjust to the target angle.

[0014] According to a third aspect of this disclosure, a vehicle rearview mirror adjustment system is provided, including a vehicle and smart glasses connected to the vehicle, the smart glasses being used to capture the driver's current field of vision when a head turning action is detected, to obtain at least one frame of image, and to send the at least one frame of image to the vehicle. The vehicle is used for: Receive at least one frame of image sent by smart glasses that have established a connection with the vehicle, wherein the at least one frame of image is captured by the smart glasses when the driver performs a head turning action and the driver's current field of vision is detected; The current field of view coverage of the target rearview mirror of interest to the driver is determined at the current angle based on the at least one frame of image. The target angle of the target rearview mirror is determined based on the current field of view coverage, and the target rearview mirror is driven to adjust to the target angle.

[0015] According to a fourth aspect of this disclosure, a computer-readable storage medium is provided, on which a computer program is stored, the computer program implementing the method according to a first aspect of this disclosure when executed by a processor.

[0016] The smart glasses in this embodiment only trigger image acquisition and transmission after detecting the driver's actual head movement towards the rearview mirror, avoiding the continuous uploading of meaningless images that occupy the vehicle's communication bandwidth and reducing wireless transmission power consumption and data processing pressure. Based on the driver's real-time field of vision image, the current field of vision coverage corresponding to the current angle of the target rearview mirror is quantitatively calculated. Then, based on the quantified current field of vision coverage, the appropriate target angle of the rearview mirror is accurately solved and the mirror is automatically driven to complete the adjustment. This forms a complete automated link of head movement triggering - field of vision quantification assessment - rearview mirror adaptive adjustment. The entire process does not require the driver to manually adjust the rearview mirror, making it convenient to operate. It can also specifically fill in the side and rear blind spots under the current view, continuously optimize the rear field of vision coverage, and improve driving safety and driving experience in driving scenarios.

[0017] Other features and advantages of the embodiments of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the disclosure.

[0019] Figure 1 This is a schematic diagram of the structure of a vehicle rearview mirror adjustment system to which the method of the present disclosure can be applied.

[0020] Figure 2 This is a flowchart illustrating a method for adjusting a vehicle rearview mirror according to some embodiments.

[0021] Figure 3 This is a block diagram of a vehicle according to some embodiments.

[0022] Figure 4 This is a block diagram of a vehicle rearview mirror adjustment system according to one embodiment. Detailed Implementation

[0023] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.

[0024] The following description of at least one exemplary embodiment is merely illustrative and is not intended to limit the scope of this disclosure or its application or use.

[0025] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0026] In all the examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0027] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0028] This disclosure relates to a method for adjusting a vehicle rearview mirror. Figure 1 This is a schematic diagram of a vehicle rearview mirror adjustment system that can utilize the vehicle rearview mirror adjustment method according to embodiments of the present disclosure. Figure 1 As shown, the vehicle rearview mirror adjustment system includes a vehicle 1000 and smart glasses 2000. A wireless communication connection is established between the vehicle 1000 and the smart glasses 2000, enabling data interaction through this wireless communication connection.

[0029] The smart glasses 2000 include an active light-emitting module 2100, an image acquisition module 2200, an attitude perception module 2300, and a wireless communication module 2400. The attitude perception module has a built-in inertial sensing unit that collects real-time data on the driver's head pitch, roll, and yaw attitudes, identifying head movements towards the vehicle's left and right side mirrors. When the driver's head is detected turning towards the mirror area, the attitude perception module 2300 outputs a trigger signal to the active light-emitting module 2100. The active light-emitting module 2100 uses an infrared LED array to output an infrared light signal that is multi-frequency modulated, Manchester encoded, and includes CRC verification, serving as a positioning marker for the driver's gaze towards the mirrors. The image acquisition module 2200 uses a wide-angle camera to continuously collect real-time images of the driver's field of vision. The wireless communication module 2400 uses a low-latency BLE / UWB wireless transmission scheme to upload the glasses' attitude data, captured images, and luminous encoding information to the vehicle 1000 in real time.

[0030] The vehicle 1000 includes an infrared light signal recognition module 1100, a central control module 1200, a rearview mirror drive module 1300, and an environmental perception module 1400. The infrared light signal recognition module 1100 has left and right infrared receiving arrays located around the rearview mirrors on both sides of the vehicle, specifically receiving infrared light signals emitted by the smart glasses 2000 and analyzing light intensity, incident angle, and device encoding information. The environmental perception module 1400 collects real-time data on driving environment illuminance and rain / fog image contrast and transmits it to the central control module 1200. The rearview mirror drive module 1300 uses a dual-axis motor actuator, enabling independent adjustment of the rearview mirrors in the horizontal and vertical directions. The central control module 1200, as the core scheduling unit of the system, establishes electrical signal connections with the infrared light signal recognition module 1100, the rearview mirror drive module 1300, and the environmental perception module 1400, and incorporates algorithm units for coordinate calibration, image fusion, target detection, field of view evaluation, PID closed-loop regulation, and gradient descent optimization.

[0031] The signal interaction link between vehicle 1000 and smart glasses 2000 is as follows: smart glasses 2000 establishes a two-way data interaction channel with the central control module 1200 of vehicle 1000 through its own wireless communication module; the active light emission module 2100 of smart glasses 2000 and the infrared light signal recognition module 1100 of vehicle 1000 form an independent optical positioning link, which uses infrared light to quickly determine the driver's focus on the side of the rearview mirror; the image acquisition module 2200 of smart glasses 2000 uploads the acquired field of vision image to the central control module 1200 through the wireless channel to complete the image data interaction; the central control module 1200 outputs adjustment commands based on the infrared light signal positioning result and the image field of vision analysis result, and directly controls the rearview mirror drive module 1300 to perform adaptive fine adjustment of the rearview mirror angle, thereby forming a complete closed-loop collaborative adjustment link of "driver-smart glasses-vehicle controller-exterior rearview mirror".

[0032] In related technologies, vehicle exterior rearview mirrors often rely on manual adjustment by the driver or preset memory positions for angle setting. In complex driving environments, different drivers have significant differences in height, posture, and observation habits, making it difficult for fixed preset positions to meet individual needs. Drivers also find it difficult to frequently adjust the rearview mirror angle manually while driving, resulting in large blind spots in scenarios such as lane changes and reversing. One possible implementation is a gaze-tracking scheme based on an onboard camera. This scheme uses an onboard camera to track the driver's gaze to achieve automatic rearview mirror adjustment. However, in complex lighting conditions, the recognition accuracy is low, and it cannot capture the actual field of vision observed by the driver, resulting in insufficient gaze detection accuracy, long response time, and inaccurate field of vision coverage optimization.

[0033] While smart glasses possess image acquisition and display capabilities, their application in in-vehicle driver assistance systems remains in the exploratory stage. In related technologies, there is a lack of effective linkage mechanisms between smart glasses and the vehicle's external vision system, failing to fully realize their potential in driver assistance.

[0034] To this end, this embodiment of the present disclosure uses smart glasses to capture at least one frame of the driver's current field of vision when the driver performs a head turning action, and sends the image to the vehicle. The vehicle determines the current field of vision coverage of the target rearview mirror at the current angle based on the image, and determines the target angle and drives the adjustment based on the current field of vision coverage. This can form a complete automated link of head action triggering - field of vision quantification assessment - rearview mirror adaptive adjustment. The entire process does not require the driver to manually adjust the rearview mirror, making it convenient to operate. It can also specifically fill in the side and rear blind spots under the current view, continuously optimize the rear field of vision coverage, and improve driving safety and driving experience in driving scenarios.

[0035] The following combination Figure 1 The system described herein illustrates various embodiments.

[0036] <First Embodiment> Figure 2 This is a flowchart illustrating a method for adjusting a vehicle rearview mirror according to some embodiments. This method for adjusting the vehicle rearview mirror can be implemented by the vehicle itself; specifically, it can be implemented by, for example... Figure 1 The vehicle shown is 1000. (As shown) Figure 2 As shown, the vehicle rearview mirror adjustment method of this embodiment may include the following steps S210 to S230.

[0037] Step S210: Receive at least one image frame sent by the smart glasses that have established a connection with the vehicle. The at least one image frame is captured by the smart glasses when they detect that the driver is performing a head turning action, showing the driver's current field of vision.

[0038] In this embodiment, the vehicle and smart glasses can use low-latency wireless communication protocols such as Bluetooth, UWB, or BLE to interact with each other, ensuring real-time data transmission.

[0039] The smart glasses' posture perception module collects head posture data from inertial sensors in real time, continuously calculating the driver's head movement trends. When the posture perception module detects that the driver's head is turning towards the left / right exterior rearview mirror area, it immediately sends a shooting trigger command to the image acquisition module. Upon receiving the trigger command, the image acquisition module immediately captures the driver's current field of vision, generating one or more frames of images. The smart glasses then transmit the images synchronously to the vehicle's central control module via a wireless communication link, allowing the central control module to perform subsequent analysis such as target detection and field of vision coverage calculation.

[0040] If the vehicle is in motion (the vehicle's speed is greater than or equal to the set speed threshold), the smart glasses can continuously capture multiple frames of images and perform weighted multi-frame fusion based on head posture data before sending them to the vehicle. This eliminates image distortion caused by motion blur, blinking, and sudden changes in illumination. When the vehicle is stationary (the vehicle's speed is less than the speed threshold), the glasses can send a single frame image or multiple frames of images to the vehicle.

[0041] In this embodiment, the image captured by the head turning action only contains the real-time field of view when the driver is looking at the rearview mirror.

[0042] In some embodiments, before executing the method of this embodiment, the adjustment system of the vehicle rearview mirror, which consists of the vehicle and the smart glasses, may be initialized. Specifically, the initialization of the vehicle rearview mirror adjustment system may be completed when the vehicle is powered on.

[0043] During initialization, the smart glasses establish a two-way wireless communication connection with the vehicle via a UWB or BLE Bluetooth module. After pairing and authentication, the device spatial calibration process begins. The vehicle reads the current seat position parameters, steering wheel position parameters, and real-time position parameters of the left and right exterior rearview mirrors in real time via the CAN bus. Simultaneously, the smart glasses collect head posture data in real time through their built-in inertial measurement unit, specifically including pitch angle, roll angle, and yaw angle. Based on the head posture data, the vehicle constructs a local coordinate system for the glasses to determine the spatial posture reference of the smart glasses.

[0044] Simultaneously, a standard vehicle coordinate system is established, defining the X-axis as the vehicle's forward direction, the Y-axis as the horizontal direction to the left of the vehicle, and the Z-axis as the vertical upward direction. Then, a pre-defined spatial coordinate transformation algorithm is used to solve for the coordinate transformation matrix between the glasses coordinate system and the vehicle coordinate system. This matrix achieves spatial unification of the two coordinate systems, enabling precise mapping between the smart glasses' line-of-sight coordinates, rearview mirror pixel coordinates, and vehicle spatial coordinates, eliminating spatial calculation errors caused by device installation deviations and wearing position deviations.

[0045] After completing the device spatial calibration, the personalized driver seating posture benchmark calibration process is initiated. During the calibration process, the driver is guided to maintain a natural driving posture, with their head upright and looking directly ahead of the vehicle, maintaining a stable posture for 3-5 seconds. During this period, the smart glasses continuously collect 100 or more frames of high-frequency head posture and eye gaze data, and the head spatial position and eye gaze direction corresponding to each frame are statistically analyzed.

[0046] The average head position of the driver is calculated by averaging the head spatial position corresponding to each frame of data; the average eye gaze direction of the driver is calculated by averaging the eye gaze direction corresponding to each frame of data. Based on the above benchmark parameters, the driver's core spatial parameters are further calculated and solidified, including the driver's eye center height, the straight-line distance between the eye position and the steering wheel, and the relative distance between the eye position and the left and right rearview mirrors. Finally, a personalized human spatial observation model for the driver is constructed, providing a personalized benchmark for subsequent determination of the gaze point, calculation of target offset, and assessment of visual field coverage, adapting to drivers with different heights, sitting postures, and driving habits.

[0047] In some embodiments, the vehicle may have a rearview mirror angle pre-stored with at least one driver, and if it is recognized that the driver is one of the drivers, the rearview mirror angle may be adjusted to the rearview mirror angle bound to that driver.

[0048] Step S220: Determine the current field of view coverage of the target rearview mirror of interest to the driver at the current angle based on at least one frame of image.

[0049] This step aims to quantitatively evaluate the actual rearview mirror's visibility at a fixed angle by using real-world images of the driver's field of vision captured by smart glasses. It objectively quantifies the completeness of the rearview mirror's coverage of the road, other road users, and safe areas behind the vehicle, resulting in a quantitative evaluation index—the current field of vision coverage—that can be used for adaptive angle adjustment of the rearview mirror. This provides core data for subsequent target angle calculation and closed-loop optimization adjustment of the rearview mirror.

[0050] Current field of view coverage represents the proportion of the driver's desired viewing area that the target rearview mirror can cover at the current angle. By evaluating current field of view coverage, the visual quality of the target rearview mirror at the current angle can be quantified, providing a basis for subsequent angle optimization.

[0051] In some embodiments, the central control module can process the input image based on at least one received image frame using edge detection and semantic segmentation algorithms to extract the closed contour of the rearview mirror surface in the image, obtaining the pixel bounding box corresponding to the mirror area; using the horizontal center line of the image as the dividing reference, the horizontal coordinate of the center of the mirror bounding box is determined: if the horizontal coordinate of the mirror center is less than the pixel coordinate of the horizontal center line of the image, the left rearview mirror can be determined as the target rearview mirror of the driver's attention; if the horizontal coordinate of the mirror center is greater than the pixel coordinate of the horizontal center line of the image, the right rearview mirror can be determined as the target rearview mirror of the driver's attention.

[0052] Current field of view coverage is a core evaluation indicator that quantitatively characterizes the completeness of the rear view at the current angle of the target rearview mirror. Current field of view coverage comprehensively reflects the overall coverage effect of the rearview mirror on static road conditions and dynamic traffic participants. It serves as a crucial bridge connecting image perception and rearview mirror angle adjustment, transforming the abstract concept of driver field of view quality into a quantifiable, calculable, and iterative numerical indicator. This provides a core quantitative basis for subsequent objective function construction, field of view compliance judgment, rearview mirror target angle calculation, and closed-loop adaptive adjustment, forming the core foundation for achieving intelligent, personalized, and precise rearview mirror adjustment.

[0053] In some embodiments, at least one frame of image may be input into a pre-trained machine learning model to obtain the current field of view coverage of the target rearview mirror that the driver is interested in at the current angle.

[0054] Step S230: Determine the target angle of the target rearview mirror based on the current field of view coverage, and drive the target rearview mirror to adjust to the target angle.

[0055] This step uses real-time field of view coverage as the core evaluation criterion to solve for the target angle of the target rearview mirror, and finally achieves adaptive and precise correction of the rearview mirror attitude, closed-loop optimization of the driver's rear observation field of view, and completes the entire process from quantitative assessment of field of view to adaptive adjustment of hardware.

[0056] The target angle refers to the rearview mirror angle that optimizes the field of vision coverage or meets preset requirements. By dynamically determining the target angle based on the current field of vision coverage, the vehicle can adaptively adjust the rearview mirrors, improving the driver's field of vision and reducing blind spot risks. The target angle can include independent angular components in two vertical dimensions: a first direction and a second direction.

[0057] The target rearview mirror is adjusted to the target angle. This can be achieved by hardware-executed actions based on the target angle. The vehicle controller sends the dual-axis target angle command to the rearview mirror motor drive module, which controls the motors corresponding to the first direction and the motors corresponding to the second direction to operate synchronously, adjusting the target rearview mirror to the target angle. After adjustment, the smart glasses are linked to re-collect the field of view image, forming a closed-loop iterative optimization mechanism.

[0058] As can be seen from steps S210 to S230 above, in this embodiment, the smart glasses only trigger image acquisition and transmission after detecting the driver's actual head movement towards the rearview mirror. This avoids the continuous uploading of meaningless images, which would occupy the vehicle's communication bandwidth and reduce wireless transmission power consumption and data processing pressure. Based on the driver's real-time field of view image, the current field of view coverage corresponding to the current angle of the target rearview mirror is quantitatively calculated. Then, based on the quantified current field of view coverage, the appropriate target angle of the rearview mirror is accurately calculated and the mirror is automatically driven to complete the adjustment. This forms a complete automated link of head movement triggering - field of view quantification evaluation - rearview mirror adaptive adjustment. The entire process does not require the driver to manually adjust the rearview mirror, making it convenient to operate. It can also specifically fill in the blind spots on the side and rear under the current view, continuously optimize the rear field of view coverage, and improve driving safety and driving experience in driving scenarios.

[0059] In some embodiments, the method may further include: receiving infrared light signals emitted by smart glasses when the driver performs a head turning action by an infrared receiving array disposed at two rearview mirrors of the vehicle; analyzing the received infrared light signals corresponding to each rearview mirror to obtain the corresponding infrared light signal intensity; and determining the target rearview mirror among the two rearview mirrors based on the infrared light signal intensity corresponding to each rearview mirror.

[0060] In this embodiment, an infrared light signal-assisted positioning process can be configured to quickly distinguish between the left and right rearview mirrors.

[0061] Specifically, when the smart glasses' posture perception module detects that the driver's head is turning towards the left / right side mirror area, it activates the active light-emitting module to emit infrared light signals. Infrared receiving arrays are installed at both side mirrors of the vehicle, and the infrared receiving arrays on the left and right sides simultaneously receive infrared light signals.

[0062] The vehicle's infrared light signal recognition module analyzes the infrared light signals collected by the infrared receiving arrays corresponding to the left and right rearview mirrors, and outputs the intensity of the infrared light signal received by the left infrared receiving array. Intensity of infrared light signal received by the infrared receiving array on the right side .

[0063] Calculate the difference in light intensity between the two paths Combined with a preset direction determination threshold Target rearview mirror determination: If The system determines that the driver is focusing on the left rearview mirror; if The system determines that the driver is focusing on the right rearview mirror; if If no valid target is identified, the rearview mirror angle adjustment process will not be initiated.

[0064] In these examples, the intensity of infrared light signals collected by dual-sided infrared receiver arrays is used to quickly distinguish between left and right target rearview mirrors. This method has low signal analysis computation and fast response speed, enabling the driver to lock onto the target rearview mirror in milliseconds, reducing image processing computational overhead. On the other hand, it is combined with a redundant verification mechanism for image recognition to distinguish between the sides of the rearview mirror. The two mutually verify each other, effectively overcoming the shortcomings of single recognition methods that are prone to misjudgment under adverse conditions such as strong light, rain, fog, and infrared signal obstruction. After accurately locking onto the target rearview mirror, the field of view coverage calculation and angle closed-loop adjustment are performed only on the corresponding mirror area, avoiding invalid calculations and adjustments for non-target rearview mirrors. This reduces the wear and tear of frequent actuator movements, narrows the field of view analysis calculation range, and improves the real-time performance, accuracy, and smoothness of the rearview mirror's dynamic adjustment. It also comprehensively reduces blind spots and significantly improves driving safety and user experience.

[0065] In some embodiments, the method further includes: parsing the received infrared light signals corresponding to each rearview mirror to obtain the corresponding recognition confidence level; wherein, the recognition confidence level represents the confidence level that the infrared light signal received by the corresponding infrared receiving array is the infrared light signal emitted by the smart glasses; obtaining a confidence level threshold; and, if the recognition confidence level is greater than or equal to the confidence level threshold, performing the step of determining the current field of view coverage of the target rearview mirror of interest to the driver at the current angle based on at least one frame of image.

[0066] The vehicle's infrared light signal recognition module analyzes the infrared light signals collected by the infrared receiving arrays of the left and right rearview mirrors, respectively, and calculates the recognition confidence level for each of the two infrared light signals. This recognition confidence level characterizes the credibility of the signal received by the corresponding infrared receiving array originating from the infrared light signal emitted by the vehicle's paired smart glasses. A fixed confidence level threshold can be pre-stored in the vehicle. This threshold is read in real time, and the recognition confidence levels of the two signals are compared with it. Only when the recognition confidence level corresponding to either rearview mirror is greater than or equal to the confidence level threshold is the infrared light signal triggered by the head turn determined to be a valid signal, and the subsequent steps of calculating the current field of view coverage of the target rearview mirror at the current angle based on the received image are executed. If the recognition confidence levels of both infrared receiving arrays are lower than the confidence level threshold, the infrared light signal is determined to be an interference signal such as sunlight or ambient infrared clutter, and the image acquisition and analysis and rearview mirror angle adjustment process is skipped directly, without initiating the subsequent field of view calculation and adjustment logic.

[0067] In some examples, the confidence level can be obtained using the following formula:

[0068] in, Indicates the confidence level of identification; Indicates the intensity of infrared light signal; This represents the normalized light intensity score, reflecting the strength of the signal; The infrared light signal direction matching score is used to quantify the degree of matching between the infrared light incident direction and the theoretical expected direction of the driver looking at the rearview mirror. The higher the score, the higher the credibility of the infrared light signal coming from the corresponding rearview mirror. This indicates the encoding check flag (0 / 1), which is 1 only when the frequency, Manchester encoding, and CRC checks all pass. , , The weights are obtained through offline calibration in advance and stored in the vehicle's memory. , , The sum is 1.

[0069] In some examples, the standard angle range of the left rearview mirror can be pre-stored in the vehicle. Standard angle range of the right rearview mirror .

[0070] Angle of incidence of infrared light signal Standard center angle of the target rearview mirror The deviation can be expressed as , The smaller the value, the closer the direction of the infrared light signal incident is to the ideal angle for the driver to look at the rearview mirror.

[0071] Set the maximum allowable angle deviation threshold The direction matching score of infrared light signals can be calculated using a linear normalization formula. :

[0072] At that time, the received infrared light signal is directly facing the infrared receiving array, and the infrared light signal direction matching score is... , indicating a perfect match in direction. When the received infrared light signal deviates from the rearview mirror area, it indicates that the direction is invalid.

[0073] In this embodiment, the deviation between the incident angle of the infrared light signal and the standard center angle of the target rearview mirror is... The unit of measurement is angle, and the maximum permissible angle deviation threshold is... The dimension of the infrared light signal is also angle, and the obtained direction matching score is... , is a dimensionless reliability index in the interval [0,1], representing the degree of agreement between the direction of infrared light and the theoretical direction of the driver's gaze at the rearview mirror.

[0074] By adding a pre-judgment mechanism based on recognition confidence, invalid triggers caused by natural light and infrared stray signals from surrounding devices can be filtered out in advance. This avoids unnecessary high-computational calculations such as current field of view coverage calculation and target angle calculation of the target rearview mirror when the vehicle is not actually watching the rearview mirror, thus saving computing resources of the vehicle controller. At the same time, the image processing process is only entered when the infrared light signal is reliable, eliminating erroneous adjustment actions caused by misidentification of the rearview mirror due to stray light from the source, reducing unnecessary start-stop of the rearview mirror motor, reducing mechanical wear, and further improving the overall operational stability and adjustment accuracy of the system.

[0075] In some embodiments, obtaining a confidence threshold includes: detecting the current brightness of the environment in which the vehicle is located, and determining a confidence threshold based on the current brightness.

[0076] For example, when the ambient brightness is high, a higher confidence threshold can be set to filter out strong light interference; when the ambient brightness is low, a lower confidence threshold can be set to ensure recognition sensitivity. By adaptively adjusting the confidence threshold according to the ambient brightness, the system can maintain stable recognition performance under different lighting conditions, improving its all-weather adaptability.

[0077] In this embodiment, the environmental illuminance of the vehicle's current environment can be detected in real time by the vehicle's environmental perception module, thereby representing the current brightness of the environment.

[0078] In this embodiment, a first mapping data reflecting the mapping relationship between brightness and confidence threshold can be preset; based on the current brightness and the first mapping data, the confidence threshold corresponding to the current brightness is obtained.

[0079] The first mapping data can be the first mapping function, or the first lookup table, etc., and is not limited here.

[0080] For the first mapping function, the dependent variable is the confidence threshold and the independent variable is brightness. Thus, by substituting the current brightness into the first mapping function, the confidence threshold corresponding to the current brightness can be obtained.

[0081] For the first lookup table, the confidence threshold corresponding to the current brightness can be found in the first lookup table. If the current brightness cannot be found directly in the first lookup table, two values ​​adjacent to the current brightness can be found, and the confidence threshold corresponding to the current brightness can be obtained by interpolation based on these two values ​​and the confidence thresholds corresponding to these two values.

[0082] For example, in a bright, sunny daytime scene with an ambient illuminance greater than 20,000 Lux, where there is strong infrared stray light interference, a higher confidence threshold is matched to filter out sunlight noise; in a low-light nighttime scene with an ambient illuminance less than 500 Lux, where there is less infrared noise, a lower confidence threshold is matched to improve signal capture sensitivity; and in rain and fog conditions, the confidence threshold is finely adjusted in conjunction with image contrast to achieve adaptive dynamic adjustment of the threshold according to ambient brightness.

[0083] This embodiment relies on the dynamic adaptive adjustment of the confidence threshold based on ambient brightness. In strong light environments, raising the threshold can effectively filter out a large amount of infrared interference signals brought by sunlight, avoiding false system triggering. In low-light nighttime scenes, lowering the threshold can accurately capture weak infrared light signals and prevent the failure to recognize real gaze actions. In special working conditions such as rain and fog, the threshold is simultaneously fine-tuned to balance recognition sensitivity and anti-interference ability, solving the defect that fixed thresholds cannot adapt to all lighting environments. It balances the system's recognition accuracy and response sensitivity, and greatly improves the reliability of the rearview mirror collaborative adjustment function under different weather conditions and time periods.

[0084] In some embodiments, the smart glasses can also detect the current brightness of the driver's environment and adjust the luminous power of the active light-emitting module according to the current brightness.

[0085] In this embodiment, second mapping data reflecting the mapping relationship between brightness and power can be preset; based on the current brightness and the second mapping data, the power corresponding to the current brightness is obtained as the luminous power of the active light-emitting module.

[0086] The second mapping data can be a second mapping function, a second lookup table, etc., and is not limited here.

[0087] For the second mapping function, the dependent variable is power and the independent variable is brightness. Thus, by substituting the current brightness into the second mapping function, the power corresponding to the brightness can be obtained, which can be used as the luminous power of the active light-emitting module.

[0088] For the second lookup table, the power corresponding to the current brightness can be found in the second lookup table and used as the luminous power of the active light-emitting module. If the current brightness cannot be directly found in the second lookup table, two values ​​adjacent to the current brightness can be found, and based on these two values ​​and the power corresponding to these two values, an interpolation method can be used to obtain the power corresponding to the current brightness and used as the luminous power of the active light-emitting module.

[0089] For example, during strong daylight with ambient brightness above 20,000 Lux, the active light-emitting module is adjusted to 100% of its rated luminous power to ensure that the infrared light signal has sufficient recognition under strong light clutter; when the ambient brightness is in the range of 5,000 Lux to 20,000 Lux, the luminous power is adjusted to 70%; in nighttime scenes with ambient brightness below 500 Lux, the luminous power is reduced to 30% to avoid excessive infrared light intensity causing overexposure and signal saturation of the image on the vehicle-end receiving module; if a low-contrast environment of rain or fog is detected, the luminous power is increased by an additional 10% to 20% on the corresponding basic luminous power to compensate for the attenuation loss of infrared light due to fog.

[0090] The smart glasses dynamically adjust the luminous power based on the ambient brightness. In strong light environments, the transmission power is increased to enhance the infrared signal-to-noise ratio, which, together with the confidence threshold simultaneously raised on the vehicle side, suppresses interference from sunlight stray light. At night, the luminous power is reduced, saving the smart glasses' battery power and preventing the vehicle's infrared light signal recognition module from saturating and distorting due to strong infrared light. In rain and fog scenarios, the luminous intensity is slightly increased to compensate for light attenuation, forming a two-way closed-loop control with the vehicle's adaptive confidence threshold. This achieves coordinated adaptation of infrared signal transmission and recognition under different lighting and weather conditions, effectively balancing signal recognition sensitivity and system anti-interference capability, reducing the probability of false triggering and missed recognition, and improving the overall adaptability of the coordinated adjustment system across all scenarios.

[0091] In some embodiments, determining the current field of view coverage of the target rearview mirror of interest to the driver at the current angle based on at least one frame of image includes: performing fusion processing on at least one frame of image to obtain a fused image; identifying target image regions containing target objects in the fused image and determining the proportion of each target image region in the fused image; wherein the target objects include road boundaries, vehicle outlines, and blind spot ranges; and determining the current field of view coverage based on the proportion of each target image region in the fused image.

[0092] In this embodiment, at least one frame of image can be registered by combining the head posture data output by the posture perception module of the smart glasses. A time-weighted fusion strategy is adopted, in which images closer to the current moment are assigned higher weights, and the images are superimposed to generate a fused image that eliminates motion blur, blinking, and instantaneous illumination interference.

[0093] Target detection and region segmentation are performed on the fused image to identify various target objects such as roads, vehicles, guardrails, lane lines, pedestrians, and cyclists. The target image regions corresponding to each type of target object are divided. The proportion of visible pixels of each type of target image region in the effective mirror area of ​​the fused image is calculated to obtain the visible proportion of the region corresponding to each type of target object.

[0094] Calculate the sum of the visible area percentages corresponding to various target objects to obtain the current field of view coverage rate under the current adjustment angle of the target rearview mirror.

[0095] In some examples, a blind spot penalty can also be added to the current field of view coverage.

[0096] Specifically, it can be done by calculating the percentage of visible pixels in the blind zone within the effective mirror area of ​​the fused image, and then subtracting the product of the percentage of visible pixels in the blind zone within the effective mirror area of ​​the fused image and its corresponding weight after calculating the sum of the percentages of visible pixels in the area corresponding to various target objects, to obtain the current field of view coverage of the target rearview mirror at the current adjustment angle.

[0097] In these examples, multi-frame weighted fusion processing effectively suppresses image distortion caused by factors such as vehicle motion blur, human blinking, sudden changes in illumination, and partial occlusion, improving image integrity and target recognition accuracy. It identifies multiple target objects in the fused image and divides corresponding target image regions. Different weights are assigned based on the safety priority of each target type to calculate the current field of view coverage, moving beyond a simple statistical calculation of image area. This allows for precise differentiation of the visibility of high-risk traffic participants such as motor vehicles, non-motor vehicles, and pedestrians, as well as basic road condition elements such as roads, guardrails, and lane lines. The quantified indicators better reflect real-world driving safety needs. Based on this, closed-loop adjustment of the rearview mirror angle can specifically fill blind spots for various targets to the side and rear of the vehicle, significantly improving the precision, rationality, and driving safety assurance capabilities of the automatic rearview mirror adjustment logic.

[0098] In some embodiments, the method further includes: parsing the infrared light signal received by the target rearview mirror to obtain the current ideal viewing direction of the target rearview mirror; generating a theoretically optimal field of view area based on the current ideal viewing direction; and determining the current field of view coverage based on the theoretically optimal field of view area.

[0099] The current ideal observation direction is the driver's reference line of sight after responding to observation, representing the driver's actual line of sight when looking at the target in the rearview mirror. The current ideal observation direction is dynamically updated with the driver's head rotation and gaze movements, accurately reflecting which direction the driver wants to look in the rearview mirror at the moment, and serves as the reference axis for personalized field of vision calculation.

[0100] Specifically, the infrared light signal angle can be mapped onto the three-dimensional space of the vehicle body, and then corrected according to the head posture data output by the smart glasses to correct the wearing position deviation and obtain the current ideal observation direction.

[0101] The theoretical optimal field of view is a standard optimal observation area generated by taking the current ideal observation direction as the central axis, combined with the vehicle type, rearview mirror physical size, lane width, vehicle blind spot range, and driving conditions. It represents the range of rear safe field of view that the rearview mirror should theoretically completely cover under the driver's current gaze posture.

[0102] Specifically, taking the current ideal observation direction as the center, the rearview mirror field of view parameters stored in the vehicle's calibration are retrieved, and a fixed safe offset range is extended upwards, downwards, leftwards, and rightwards to form a rectangular or polygonal standard field of view area, which is then uniformly converted to the image pixel coordinate system, i.e., the theoretically optimal field of view area.

[0103] The theoretical optimal field of view area includes the rear lane, adjacent lane-changing areas, vehicle blind spots, and other standard visible ranges that require key observation. The central control module transforms the actual visible area corresponding to the target rearview mirror in the image uploaded by the smart glasses to the same coordinate system as the theoretical optimal field of view area, calculates the ratio of the overlapping area of ​​the two areas to the total area of ​​the theoretical optimal field of view area, and thus obtains the current field of view coverage of the target rearview mirror at the current angle.

[0104] This embodiment obtains the ideal observation direction corresponding to the driver's actual gaze through infrared light signal analysis, thereby generating a theoretically optimal field of view area that fits the driver's observation needs. Unlike fixed standard field of view determination methods, it can match the personalized observation needs brought about by different drivers' height, sitting posture, and observation habits. Based on the personalized theoretical field of view area, the current field of view coverage is calculated, and the quantitative indicators are more in line with the driver's actual driving observation needs. It provides a precise and personalized evaluation basis for the adaptive adjustment of the rearview mirror angle, effectively reducing the problem of blind spots still existing after the rearview mirror is adjusted, and improving the rear field of view coverage and driving safety.

[0105] In some embodiments, determining the target angle of the target rearview mirror based on the current field of view coverage includes: identifying the proportion of blind spot range and the visibility of the following vehicle in the fused image; determining the current field of view score based on the current field of view coverage, the proportion of blind spot range, and the visibility of the following vehicle; and determining the target angle of the target rearview mirror based on the current field of view coverage if the current field of view score is less than a set score threshold.

[0106] The blind spot ratio is the ratio of the pixel area of ​​the blind spot (where there are no roads, vehicles, pedestrians, lane lines, or other effective observable targets) within the rearview mirror surface to the total area of ​​all effectively visible pixels on the mirror surface. As a penalty for visual field defects, the blind spot ratio quantifies the size of the dangerous area that cannot be observed from the side and rear of the rearview mirror. This compensates for the shortcomings of a single field-of-view coverage rate, which cannot reflect the hazards of blind spots, and prevents the system from pursuing only the total field-of-view area while retaining large areas of blind spots.

[0107] In this embodiment, the fused image may be segmented and blind spot detected to identify the real blind spot area within the target rearview mirror's field of view, and the proportion of the blind spot pixel area to the effective visible area of ​​the mirror may be calculated to obtain the current blind spot range percentage.

[0108] Rear vehicle visibility is a normalized score characterizing the degree to which vehicles (including motor vehicles and non-motor vehicles) can be identified in the rearview mirror. It is calculated by weighting and normalizing the results based on target detection, the number of rear targets, target image clarity, and whether the target is occluded. Rear vehicle visibility is used to quantify the observation effect on dynamic traffic participants, prioritizing the visibility of rear vehicles in lane changing and overtaking scenarios, and avoiding the safety hazard of having adequate static road visibility but completely losing sight of vehicles behind.

[0109] In this embodiment, a target detection algorithm can be used to identify vehicles behind, nearby non-motorized vehicles and pedestrians, determine the visibility status of the following vehicle and surrounding high-risk targets, and quantify the visibility score of the following vehicle.

[0110] The current field of vision score can be determined using the following formula:

[0111] in, Indicates the current field of view score. Indicates the current field of view coverage. Indicates the percentage of the blind spot area. Indicates the visibility of the vehicle behind. , , These represent the corresponding weights, which can be obtained through offline calibration and stored in the vehicle's memory.

[0112] In this embodiment, the current field of view coverage Dimensionless, value range [0,1], positive indicator; the higher the percentage of various road, vehicle, pedestrian, and other targets visible in the rearview mirror, the larger the value; percentage of blind spot area. Dimensionless, value range [0,1], negative index, representing the percentage of pixels in the ineffective blind zone within the mirror surface; rear vehicle visibility. Dimensionless, with a value range of [0,1], it is a positive indicator that represents the degree to which vehicles behind can be identified; the current field of view score is a dimensionless comprehensive evaluation index, and the higher the value, the better the overall effect of the rearview mirror's field of view.

[0113] The current field of view score uses the current field of view coverage as a positive gain term, the blind spot range as the first penalty term, and the state where the following vehicle is not visible as the second penalty term, so that the evaluation results simultaneously meet the requirements of road field of view integrity, blind spot suppression effect and dynamic driving target observation.

[0114] The current field of view score is further compared with a pre-defined optimal field of view score. If the current field of view score is greater than or equal to the threshold, the current rearview mirror angle is deemed to be in good condition, and no angle adjustment is required. If the current field of view score is less than the threshold, it indicates that the current mirror angle has problems such as insufficient field of view coverage, large blind spots, or limited observation of rear vehicles. In this case, the angle optimization process is initiated: the target angle of the target rearview mirror is determined based on the current field of view coverage, and the target rearview mirror is driven to adjust to the target angle.

[0115] In these examples, the current field of view score is determined by integrating multiple indicators such as current field of view coverage, blind spot ratio, and visibility of following vehicles. This comprehensively evaluates the actual field of view quality of the rearview mirror and avoids the defects of local optima caused by optimizing a single indicator. The adjustment is triggered on demand by comparing the current field of view score with the score threshold. Angle optimization is only initiated when the field of view does not meet the driving observation conditions, which effectively reduces the ineffective actions of the rearview mirror and motor wear.

[0116] In some embodiments, the method further includes: recording the current angle of the target rearview mirror when the current field of view score is greater than or equal to a score threshold, and binding the current angle of the target rearview mirror to the driver.

[0117] When the current field of view score is determined to be greater than or equal to the score threshold, it means that the rear field of view corresponding to the current target rearview mirror angle has reached the optimal observation state under the current sitting posture and observation habits of the driver. The controller reads the current horizontal angle component and vertical angle component of the rearview mirror in real time, stores the set of optimal angle parameters in the vehicle's local storage unit or the smart glasses binding storage area, and binds the set of rearview mirror angle data with the identity information of the currently paired and logged-in driver.

[0118] When the driver gets back into the car and the smart glasses are paired with the vehicle, the pre-bound and stored rearview mirror angle can be retrieved directly, and the rearview mirror can be automatically driven to restore to the historical optimal posture without repeating the entire adjustment process of image acquisition, target detection, and current field of view scoring and iterative optimization.

[0119] In these examples, the optimal rearview mirror posture is determined by comparing the current field of view score with the score threshold. Angle data is only stored after the comprehensive field of view index has been stably met, ensuring that the bound angle parameters have a high-quality observation effect with a complete field of view, low blind spot, and clear visibility of dynamic targets. The optimal angle is bound to the driver's identity for storage, realizing personalized rearview mirror posture memory reuse for the driver. When the driver uses the car again, they can restore their exclusive optimal field of view with one click, saving the waiting time of automatic iterative adjustment every time they get in the car. This reduces the computing power consumption of repeated image processing by the vehicle controller and the wear and tear of frequent operation of the rearview mirror motor, taking into account ease of use, personalized adaptation, and system operation energy consumption optimization.

[0120] In some embodiments, third mapping data reflecting the mapping relationship between field of view coverage and angle can be preset; based on the field of view coverage and the third mapping data, the angle corresponding to the current field of view coverage is obtained as the target angle.

[0121] The third mapping data can be a third mapping function, a third lookup table, etc., and is not limited here.

[0122] For the third mapping function, the dependent variable is the angle and the independent variable is the field of view coverage. By substituting the current field of view coverage into the third mapping function, the angle corresponding to the current field of view coverage can be obtained as the target angle.

[0123] For the third lookup table, the angle corresponding to the current field of view coverage can be found in the third lookup table and used as the target angle. If the current field of view coverage cannot be found directly in the third lookup table, two values ​​adjacent to the current field of view coverage can be found, and the angle corresponding to the current field of view coverage can be obtained by interpolation based on these two values ​​and the angles corresponding to these two values, and used as the target angle.

[0124] In some embodiments, the target angle includes an angular component in a first direction and an angular component in a second direction; determining the target angle of the target rearview mirror based on the current field of view coverage includes: identifying the centroid of a key target in the fused image and the center of the target rearview mirror in the fused image; calculating a first angular deviation between the centroid of the key target and the center of the target rearview mirror in a first direction and a second angular deviation in a second direction based on the centroid of the key target and the center of the target rearview mirror; wherein the first direction is perpendicular to the second direction; obtaining the angular component of the target angle in a first direction based on the first angular deviation and the angular component of the current angle in a first direction; and obtaining the angular component of the target angle in a second direction based on the second angular deviation and the angular component of the current angle in a second direction.

[0125] In this embodiment, the first direction can be the target angle of the rearview mirror, which includes a first direction angle component and a second direction angle component that are perpendicular to each other. The first direction is the horizontal adjustment direction of the rearview mirror, and the second direction is the vertical adjustment direction of the rearview mirror.

[0126] In some examples, the key target may be the object within the frame that is closest to the vehicle and has the highest risk level.

[0127] Center of mass of key objectives Center of the target rearview mirror The first angular deviation in the first direction can be expressed as The deviation of the centroid of the key target from the center of the target's rearview mirror in the second direction at a second angle can be expressed as: .in, and These represent the mirror imaging ratios for the first and second directions, respectively, and the unit can be degrees per pixel.

[0128] Based on this, a PID controller can be introduced to calculate the target angle.

[0129] To suppress reciprocating jitter and eliminate steady-state deviation, PID closed-loop control is used to update the dual-axis target angle of the rearview mirror. The update formula for the first direction is:

[0130] The formula for updating the second direction is:

[0131] in, The current angle of the target rearview mirror is the angular component in the first direction. The current angle of the target rearview mirror is the angular component in the second direction. Let be the angular component of the target angle in the first direction. The angular component of the target angle in the second direction. This is the proportionality coefficient. The integral coefficient is... The differential coefficient; the integral term , Cumulative historical bias eliminates long-term centering offset, differential term , It suppresses large single adjustments, buffers sudden angle changes, and prevents the mirror from oscillating back and forth.

[0132] In this embodiment, , , , The dimensions of all values ​​are angles. , , All are dimensionless calibration constants, and all components are in degrees. The sum is then output. , The unit remains degrees.

[0133] In these examples, the offset between the key target and the center of the mirror is first accurately quantified using pixel coordinates. Then, the pixel deviation is converted into a physical angle that can be directly used for motor control by relying on the calibrated imaging coefficient. The quantization process is intuitive and the conversion accuracy is stable. The PID algorithm is used to correct the horizontal and vertical angles respectively. The proportional term quickly responds to the centering requirement, the integral term eliminates the long-term steady-state deviation of the field of view, and the derivative term suppresses the large swing of the motor, solving the problem of repeated back-and-forth oscillation of the rearview mirror from the algorithm level. After the adjustment is completed, the image is re-acquired and iteratively verified to form a closed-loop control, continuously correcting the field of view deviation. It can stably keep high-risk key targets such as vehicles, pedestrians, and cyclists in the optimal observation area of ​​the center of the rearview mirror. The two-dimensional decoupled independent control does not interfere with each other, taking into account the adjustment response speed and operation smoothness, continuously reducing the blind spot to the side and rear, and significantly improving driving safety in lane changing and reversing scenarios.

[0134] <Second Embodiment> This embodiment provides a vehicle, Figure 3 A schematic diagram of the hardware structure of the electronic device is shown.

[0135] like Figure 3 As shown, the vehicle 1000 includes a processor 310 and a memory 320. The memory can be used to store a computer program, and the processor can be used to retrieve the computer program from the memory to execute the method of the first embodiment of this disclosure. The processor can be one or more processors, which can execute instructions individually or jointly. Similarly, the memory can be one or more memories, which can store the aforementioned computer program individually or jointly.

[0136] <Third Embodiment> This embodiment provides a vehicle rearview mirror adjustment system. Figure 4 A schematic diagram of the hardware structure of the vehicle's rearview mirror adjustment system is shown.

[0137] like Figure 4 As shown, the vehicle rearview mirror adjustment system 4000 includes a vehicle 1000 according to the second embodiment and a smart glasses 2000 connected to the vehicle 1000. The smart glasses 2000 is used to capture the driver's current field of vision when the driver performs a head turning action, obtain at least one frame of image, and send at least one frame of image to the vehicle.

[0138] In some embodiments, the smart glasses 2000 are also used to activate the active light-emitting module to emit infrared light signals when the driver performs a head turning action.

[0139] In some embodiments, the smart glasses 2000 are also used to detect the current brightness of the environment in which the driver is located, and adjust the luminous power of the active light-emitting module according to the current brightness.

[0140] This disclosure also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the vehicle rearview mirror adjustment method in any embodiment of this disclosure. Optionally, the computer-readable storage medium may be a non-transitory storage medium, but is not limited thereto; it may also be a temporary storage medium.

[0141] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0142] This disclosure may be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having a computer-readable program loaded thereon for causing a processor to implement any of the methods in the foregoing embodiments of this disclosure.

[0143] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media may include, for example, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), compact disc-read-only memory (CD-ROM), digital versatile disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any combination thereof. The computer-readable storage medium used herein is not to be interpreted as a transient signal itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0144] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include one or more of copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to computer-readable storage media in the respective computing / processing device.

[0145] The computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source or object programs written in any combination of one or more programming languages, including object-oriented programming languages ​​(such as Smalltalk, C++, etc.) and conventional procedural programming languages ​​(such as the "C" language or similar programming languages). The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network (e.g., a local area network or a wide area network), or it may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays, or programmable logic arrays, can execute computer-readable program instructions to implement various aspects of the embodiments of this disclosure by utilizing state information from the computer-readable program instructions.

[0146] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0147] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0148] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0149] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. It should be noted that implementation in hardware, implementation in software, and implementation using a combination of software and hardware are all equivalent.

[0150] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein. The scope of this disclosure is defined by the appended claims.

Claims

1. A method for adjusting a vehicle rearview mirror, characterized in that, The method is applied to a vehicle, and the method includes: Receive at least one frame of image sent by smart glasses that have established a connection with the vehicle, wherein the at least one frame of image is captured by the smart glasses when the driver performs a head turning action and the driver's current field of vision is detected; The current field of view coverage of the target rearview mirror of interest to the driver is determined at the current angle based on the at least one frame of image. The target angle of the target rearview mirror is determined based on the current field of view coverage, and the target rearview mirror is driven to adjust to the target angle.

2. The method according to claim 1, characterized in that, The method further includes: The infrared receiving arrays located at the two rearview mirrors of the vehicle receive the infrared light signals emitted by the smart glasses when the driver performs a head turning action. The infrared light signal received by each rearview mirror is analyzed to obtain the corresponding infrared light signal intensity; The target rearview mirror is determined based on the infrared light signal intensity corresponding to each rearview mirror.

3. The method according to claim 2, characterized in that, The method further includes: The infrared light signal received by each rearview mirror is analyzed to obtain the corresponding recognition confidence level; wherein, the recognition confidence level represents the confidence level that the infrared light signal received by the corresponding infrared receiving array is the infrared light signal emitted by the smart glasses. Obtain the confidence threshold; If the identification confidence level is greater than or equal to the confidence threshold, the step of determining the current field of view coverage of the target rearview mirror of interest to the driver at the current angle based on the at least one frame of image is performed.

4. The method according to claim 3, characterized in that, The process of obtaining the confidence threshold includes: The current brightness of the environment in which the vehicle is located is detected, and a confidence threshold is determined based on the current brightness.

5. The method according to claim 2, characterized in that, The method further includes: The current ideal viewing direction of the target rearview mirror is obtained by analyzing the infrared light signal received by the target rearview mirror. The theoretically optimal field of view is generated based on the current ideal observation direction, and the current field of view coverage is determined based on the theoretically optimal field of view.

6. The method according to claim 1, characterized in that, Determining the current field-of-view coverage of the target rearview mirror of interest to the driver at the current angle based on the at least one frame of image includes: The at least one frame of image is fused to obtain a fused image; Identify target image regions containing target objects in the fused image, and determine the proportion of each target image region in the fused image; The current field-of-view coverage is determined based on the proportion of each target image region in the fused image.

7. The method according to claim 6, characterized in that, The target angle includes an angle component in a first direction and an angle component in a second direction; Determining the target angle of the target rearview mirror based on the current field of view coverage includes: Identify the centroid of key targets in the fused image and the center of the rearview mirror of the target in the fused image; Based on the centroid of the key target and the center of the target rearview mirror, calculate the first angular deviation between the centroid of the key target and the center of the target rearview mirror in the first direction and the second angular deviation in the second direction; wherein, the first direction is perpendicular to the second direction; The angle component of the target angle in the first direction is obtained based on the first angle deviation and the angle component of the current angle in the first direction; the angle component of the target angle in the second direction is obtained based on the second angle deviation and the angle component of the current angle in the second direction.

8. The method according to claim 6, characterized in that, Determining the target angle of the target rearview mirror based on the current field of view coverage includes: Identify the proportion of blind spots in the fused image and the visibility of the following vehicle, which represents the degree to which the following vehicle is identifiable; The current field of view score is determined based on the current field of view coverage, the proportion of the blind spot area, and the visibility of the vehicle behind. If the current field of view score is less than a set score threshold, the target angle of the target rearview mirror is determined based on the current field of view coverage.

9. A vehicle, characterized in that, It includes a memory and a processor, the memory being used to store a computer program; the processor being used to execute the computer program to implement the following method: Receive at least one frame of image sent by smart glasses that have established a connection with the vehicle, wherein the at least one frame of image is captured by the smart glasses when the driver performs a head turning action and the driver's current field of vision is detected; The current field of view coverage of the target rearview mirror of interest to the driver is determined at the current angle based on the at least one frame of image. The target angle of the target rearview mirror is determined based on the current field of view coverage, and the target rearview mirror is driven to adjust to the target angle.

10. A vehicle rearview mirror adjustment system, characterized in that, The system includes a vehicle and smart glasses that establish a connection with the vehicle. The smart glasses are used to: capture the driver's current field of vision when the driver performs a head turning action, obtain at least one frame of image, and send the at least one frame of image to the vehicle. The vehicle is used for: Receive at least one frame of image sent by smart glasses that have established a connection with the vehicle, wherein the at least one frame of image is captured by the smart glasses when the driver performs a head turning action and the driver's current field of vision is detected; The current field of view coverage of the target rearview mirror of interest to the driver is determined at the current angle based on the at least one frame of image. The target angle of the target rearview mirror is determined based on the current field of view coverage, and the target rearview mirror is driven to adjust to the target angle.

Images