Electronic outside rearview mirror follow-up field of view adjustment method and electronic device

Abstract

The application relates to the technical field of intelligent auxiliary driving, and provides an electronic outside rearview mirror follow-up visual field adjusting method and an electronic device, so as to track a target setting mark line, and the visual field tracking range is determined according to the mark line, the clock of the tracking target is ensured to be located in the center area of a display picture, and then the driving visual field is ensured; meanwhile, when the collection is used, the image feature point set of the collected image frame is synchronously acquired, the movement distribution change of the tracking target in the current image collection area is further determined, and then the mark line is adaptively adjusted, the visual field tracking area of the current collected image frame is moved to perform follow-up visual field adjustment, the stable tracking of the visual field tracking area is realized through point set registration, the display is smooth, the method is suitable for the matching combination of different tractors and trailers, the universality is good, and the safety of the vehicle is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent assisted driving technology, and in particular to an electronic rearview mirror with a dynamic field of view adjustment method and electronic device. Background Technology

[0002] Currently, semi-trailer trucks play a crucial role in my country's logistics and transportation industry. However, due to their large overall length and the articulated connection between the tractor and trailer, traditional physical rearview mirrors cannot detect traffic conditions near the rear of the trailer when the vehicle is turning, resulting in a significant blind spot. Most semi-trailer truck accidents are caused by this blind spot during turns. Therefore, it is necessary to have rearview mirrors that can be adjusted dynamically to maintain a clear view of the rear of the truck while it is turning.

[0003] The working principle of the electronic exterior rearview system is first to capture video using a digital camera. Since the camera does not move after installation, its field of view will cover a relatively large area. After video processing, a certain area is usually selected within the overall field of view for display, similar to how the driver manually adjusts the angle of the rearview mirror according to their own preferences and field of vision. Finally, the image area selected by the driver is displayed on the in-vehicle display screen.

[0004] Because semi-trailer trucks have large blind spots, which easily lead to traffic accidents, effectively reducing these blind spots has become a key technological direction for the industry. For example, the steering angle of the tractor's front wheels can be calculated by using the steering wheel rotation angle and steering ratio. The relationship between the angle of the tractor's front wheels and the angle between the tractor and the vehicle body can be collected at different vehicle speeds. Algorithms can then be used to calculate and crop the field of vision of the trailer's rear on the steering side, so that the field of vision changes with the steering wheel angle and vehicle speed.

[0005] However, in actual driving, the angle between the tractor's front wheels and the vehicle body varies greatly depending on the vehicle speed, making it difficult to capture all possible scenarios. When the acquired data is incomplete, the field of vision becomes abrupt, resulting in sudden jumps and an inability to smoothly adapt to changing conditions. Furthermore, since semi-trailer trucks are sold separately from their trailers, tractors can carry trailers of different lengths and sizes, and trailers can be towed by different tractors. The matching combination of tractors and trailers varies from person to person. After changing the tractor or trailer, the pre-collected data on the angle between the tractor's front wheels and the vehicle body at different speeds cannot be adapted to the current field of vision adjustment, resulting in poor versatility and making it unsuitable for actual production and transportation work. Summary of the Invention

[0006] This invention provides a method and electronic device for adjusting the field of view of an electronic exterior rearview mirror, which solves the technical problems of existing electronic exterior rearview mirror field of view adjustment technology, which relies on pre-acquisition and comprehensive data acquisition, and has the risk of jumps, poor smoothness and poor versatility.

[0007] To solve the above technical problems, the present invention provides a method for adjusting the field of view of an electronic exterior rearview mirror, comprising: Marker lines are set based on the target being tracked, and the field of view tracking range is defined according to the marker lines; Real-time acquisition of captured image frames, and acquisition of image feature point set of the field of view tracking area from the captured image frames according to the field of view tracking range; The movement and distribution changes of the tracking target in the current image acquisition area are determined based on the image feature point set; The marker line is adjusted according to the change in the movement distribution, thereby moving the field of view tracking area of ​​the currently acquired image frame to perform follow-up field of view adjustment.

[0008] This basic solution uses target tracking markers to define the field of view tracking range, ensuring the target clock is centered on the display screen, thus guaranteeing the driver's field of view. Simultaneously, during image acquisition, it synchronously acquires image feature point sets from the captured image frames to further determine the movement and distribution changes of the target in the current image acquisition area. This allows for adaptive adjustment of the markers, moving the field of view tracking area of ​​the current image frame to perform dynamic field of view adjustment. Stable tracking of the field of view tracking area is achieved through point set registration, resulting in smooth display. It is applicable to different combinations of tractor and trailer vehicles, exhibiting good versatility and effectively improving vehicle safety.

[0009] In a further implementation, determining the movement distribution changes of the tracking target in the current image acquisition area based on the image feature point set includes: Obtain the image feature point set of the current frame, the image feature point set including ORB feature points; Extract the abscissa data of each ORB feature point in the image feature point set, and calculate the feature distribution parameters of the image feature point set in the field of view tracking region based on the abscissa data, assuming that the ORB feature points follow a Gaussian distribution. By comparing the feature distribution parameters of adjacent acquired image frames, the change in the movement distribution of the tracked target in the current image acquisition area is calculated.

[0010] This scheme uses the image point set of the current frame as a basis, and the ORB feature points follow a Gaussian distribution. Based on the abscissa data of each ORB feature point in the image feature point set, it calculates the feature distribution parameters of the image feature point set in the field of view tracking area. It then compares and analyzes these parameters with the image positioning of the previous frame (the feature distribution parameters of the previous frame) to calculate the change in the movement distribution of the tracked target in the current image acquisition area. This serves as the basis for judging whether the field of view tracking area of ​​the current frame is continuous and smooth with the field of view tracking area of ​​the previous frame. It has strong anti-interference ability and can continuously provide a stable judgment basis to facilitate the execution of follow-up field of view adjustment, further improving driving safety.

[0011] In a further embodiment, the formula for calculating the characteristic distribution parameter is as follows:

[0012] In the formula, This represents the total number of feature points in the field-of-view tracking region. This represents the i-th feature point in the field-of-view tracking region. and Let represent the mean and variance of the set of image feature points within the field-of-view tracking region in the k-th frame, respectively.

[0013] In a further embodiment, the formula for calculating the change in the movement distribution is as follows:

[0014] In the formula, Represents the difference in expected change. This represents the difference in variance.

[0015] This scheme is based on the Gaussian distribution of feature points. It determines whether to perform view tracking based on the distribution of the feature point set. The target area is described only by the mean (center position) and variance (distribution shape), which naturally filters out outliers that deviate from the main distribution. It only needs to maintain the statistics of the distribution (mean and variance) instead of storing all feature points, which saves memory and computing resources. Moreover, the probability density function (PDF) of the Gaussian distribution can be calculated efficiently, which is suitable for the real-time feedback mechanism of driving. By comparing the KL divergence of the feature point set distribution in the previous and next frames, it can be determined whether the scene has changed significantly. It has low computational difficulty and high sensitivity.

[0016] In a further implementation, the marker line is adjusted according to the change in movement distribution, thereby moving the field-of-view tracking region of the currently acquired image frame to perform adaptive field-of-view adjustment, including: Obtain the expected change difference from the movement distribution change, move the marker line according to the expected change difference, and obtain the field of view tracking area after the movement; Obtain the image feature point set of the moved field-of-view tracking area and calculate the corresponding feature distribution parameters; The feature distribution parameters of the field of view tracking region after the movement are compared with the feature distribution parameters of the field of view tracking region in the previous frame to calculate the corresponding movement distribution change, and then the variance change difference is obtained from the movement distribution change as the target matching parameter. A matching reference threshold is preset, and the target matching parameters are compared with the matching reference threshold to determine whether the feature point set matching is successful. If so, the current frame display is replaced with the moved field of view tracking area to achieve dynamic field of view adjustment.

[0017] This scheme adjusts the marker line based on the expected change difference extracted from the motion distribution change when acquiring the motion distribution change of the current frame and the previous frame. It supports preset adjustments based on actual data (actual expected change difference). Then, based on the preset motion tracking area and the motion distribution change of the previous frame, it compares the variance change difference obtained from the motion distribution change with the matching reference threshold to quantify the dispersion characteristics of the feature point set extracted from the region of interest (view tracking area) after the motion. This determines whether the current motion adjustment optimizes the display. If so, the view tracking area is replaced to achieve dynamic view adjustment, thereby ensuring that the target is always closely followed while avoiding jumps in view display.

[0018] In a further implementation, a marker line is set based on the tracking target, and the field of view tracking range is defined according to the marker line, including: A1. Based on the tracking target on the vehicle body, set the initial position of the marker line; A2. Using the marked line as the center, plan the initial field of view area according to the initial width; A3. Perform ORB feature detection on the visual region and extract feature descriptors and the number of feature points; A4. Determine whether feature extraction was successful based on the number of feature points. If not, adjust the width of the current field of view incrementally and return to step A3. If yes, output the width of the corresponding field of view to define the field of view tracking range.

[0019] In a further implementation, the success of feature extraction is determined based on the number of feature points. If not, the initial width is adjusted incrementally. Specifically, it is determined whether the number of feature points is less than the minimum feature value of the field of view tracking area. If so, the feature extraction is determined to have failed, and the width value of the current field of view area is adjusted by a preset step size increment to obtain the adjusted target width.

[0020] This solution performs real-time detection of the number of feature points within the field of view tracking. If insufficient features are extracted, then... To incrementally increase the width of the current field of view, it relies on the number of feature points rather than absolute pixel values, is not sensitive to changes in lighting, and automatically performs incremental adjustments to obtain the target width. It is robust, computationally efficient, and can better adapt to different environments in driving scenarios.

[0021] In a further implementation, the method further includes: obtaining the adjusted target width, determining whether the target width is less than the maximum field of view, and if so, returning to step A3; otherwise, returning to step A1 to reset the initial position of the marker line.

[0022] This solution further sets a maximum field of view width as a limit for width adjustment, based on practical considerations, and supplements this with adjustments to the accurate positioning of the marker lines.

[0023] The present invention also provides an electronic device, including a processor, a memory, and an image acquisition module; The image acquisition module is used to acquire image frames in real time and transmit them to the processor; Memory, used to store computer programs; When the processor executes a computer program stored in the memory, it implements the electronic exterior rearview mirror follow-up field of view adjustment method as described above. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating the process of an electronic exterior rearview mirror follow-up field of view adjustment method provided in an embodiment of the present invention; Figure 2 This is a flowchart of the field-of-view tracking range definition provided in an embodiment of the present invention; Figure 3 This is a flowchart of the field-of-view tracking area adjustment provided in an embodiment of the present invention; Figure 4 This is a field-of-view area division view of the electronic exterior rearview mirror provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of the field of view provided in an embodiment of the present invention. Detailed Implementation

[0025] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. The embodiments are given for illustrative purposes only and should not be construed as limiting the present invention. The accompanying drawings are for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0026] Example 1 This invention provides a method for adjusting the field of view of an electronic exterior rearview mirror, such as... Figure 1 As shown, this embodiment includes the following steps: S1. Set marker lines based on the tracking target, and define the field of view tracking range according to the marker lines. See [link / reference] Figure 2 ,include: A1. Based on the tracking target on the vehicle body, set the initial position of the marker line; Among them, the marking lines are used to determine the area that needs to be tracked by the driver's field of vision when the vehicle is turning. The selection is based on the driver's habits, and is generally selected at the rear of the vehicle, or a clearly distinguishable reflective strip can also be selected.

[0027] Due to road regulations, the color of trailer cargo boxes is generally highly distinguishable from the environment, and reflective strips and other conspicuous markings are usually required at the rear of the vehicle. Therefore, the color at this location often exhibits certain texture features and contrast variations with the surrounding environment. In other words, the setting of the marking line tracking area follows the principle that a sufficient number of effective features can be easily extracted in its vicinity.

[0028] The specific steps for setting the lane markings are as follows: Park the trailer truck in a safe parking area, and the driver selects the lane markings on the display screen of the intelligent electronic rearview mirror system by adjusting knobs or buttons.

[0029] A2. Using the marked line as the center, plan the initial field of view area according to the initial width; For details, see Figure 5 To reduce computational load, the driver-set markings (denoted as ) are used as the basis for the entire image captured by the camera. A rectangular area centered at W, with an initial width of W and a height equal to the image height, is designated as the initial field of view for planning.

[0030] A3. Perform ORB feature detection on the visual region and extract feature descriptors and the number of feature points; A4. Determine whether feature extraction was successful based on the number of feature points. If not, adjust the width of the current field of view incrementally and return to step A3. If yes, output the width of the corresponding field of view to define the field of view tracking range.

[0031] In this embodiment, the success of feature extraction is determined based on the number of feature points; otherwise, the initial width is incrementally adjusted. Specifically: A41. Determine whether the number of feature points is less than the minimum feature value of the view tracking region. If so, feature extraction is deemed to have failed. The width of the current field of view is adjusted by a preset step increment to obtain the adjusted target width (W=W').

[0032] Specifically: The target width of the field of view tracking range can be determined by whether enough feature points have been extracted. When not enough features have been extracted, the range is increased by a step size increment. The width of the extraction field of view is continuously increased incrementally until the rear features of the vehicle are extracted.

[0033] A42. Obtain the adjusted target width and determine whether the target width is less than the maximum field of view width. If yes, return to step A3; otherwise, return to step A1 to reset the initial position of the marker line.

[0034] This embodiment performs real-time detection of the number of feature points within the field of view tracking. When insufficient features are extracted, it then... To incrementally increase the width of the current field of view, it relies on the number of feature points rather than absolute pixel values, is not sensitive to changes in lighting, and automatically performs incremental adjustments to obtain the target width. It is robust, computationally efficient, and can better adapt to different environments in driving scenarios.

[0035] This embodiment further sets a maximum field of view width as a limitation for width adjustment, based on actual conditions, and supplements this with adjustments to the accurate positioning of the marker lines.

[0036] S2. Acquire image frames in real time, and obtain the image feature point set of the field of view tracking area from the acquired image frames according to the field of view tracking range; ORB (Oriented FAST and Rotated BRIEF) is a fast and efficient image feature detection algorithm that combines FAST corner detection and the BRIEF descriptor, with improvements to provide scale invariance, rotation invariance, and some noise resistance. It is widely used in SLAM systems. In this embodiment, ORB feature points are extracted as the feature point set.

[0037] Furthermore, the ORB feature points extracted in this step are the feature points of the marker line and its surrounding area.

[0038] S3. Determine the movement distribution changes of the tracking target in the current image acquisition area based on the image feature point set, including: S31. Obtain the image feature point set of the current frame, wherein the image feature point set includes ORB feature points; S32. Extract the abscissa data of each ORB feature point in the image feature point set, and calculate the feature distribution parameters of the image feature point set in the field of view tracking area based on the abscissa data, assuming that the ORB feature points follow a Gaussian distribution. In this embodiment, the formula for calculating the feature distribution parameter is as follows:

[0039] In the formula, This represents the total number of feature points in the field-of-view tracking region. This represents the i-th feature point in the field-of-view tracking region. and Let represent the mean and variance of the set of image feature points within the field-of-view tracking region in the k-th frame, respectively.

[0040] S33. Compare the feature distribution parameters of adjacent acquired image frames to calculate the change in the movement distribution of the tracked target in the current image acquisition area.

[0041] Similarly, after extracting feature points from the (k+1)th frame of the image, the feature distribution parameters (i.e., the distribution of the horizontal coordinates) of the image feature point set are calculated using the same method. ,in

[0042] in, This represents the total number of feature points in the field-of-view tracking region. This represents the j-th feature point in the field-of-view tracking region. and These are the mean and variance of the set of image feature points within the field-of-view tracking region in the (k+1)th frame of the acquired image.

[0043] In this embodiment, the calculation formula for the change in the movement distribution is as follows:

[0044] In the formula, Represents the difference in expected change. This represents the difference in variance.

[0045] This embodiment is based on the Gaussian distribution of feature points. It determines whether to perform field-of-view tracking based on the distribution of the feature point set. The target area is described only by the mean (center position) and variance (distribution shape), which naturally filters out outliers that deviate from the main distribution. It only needs to maintain the statistics of the distribution (mean and variance) instead of storing all feature points, which can save memory and computing resources. Moreover, the probability density function (PDF) of the Gaussian distribution can be calculated efficiently, which is suitable for the real-time feedback mechanism of driving. By comparing the KL divergence of the feature point set distribution in the previous and next frames, it can be determined whether the scene has changed significantly. The calculation is simple and the sensitivity is high.

[0046] This embodiment uses the image point set of the current frame as a basis, and the ORB feature points follow a Gaussian distribution. Based on the abscissa data of each ORB feature point in the image feature point set, it calculates the feature distribution parameters of the image feature point set in the field of view tracking area. It then compares and analyzes the image positioning of the previous frame (the feature distribution parameters of the previous frame) to calculate the change in the movement distribution of the tracked target in the current image acquisition area. This serves as the basis for judging whether the field of view tracking area of ​​the current frame is continuous and smooth with the field of view tracking area of ​​the previous frame. It has strong anti-interference ability and can continuously provide a stable judgment basis to facilitate the execution of follow-up field of view adjustment and further improve driving safety.

[0047] S4. Adjusting the marker line according to the change in the movement distribution, thereby moving the field of view tracking area of ​​the currently acquired image frame to perform follow-up field of view adjustment, including: Obtain the expected change difference from the movement distribution change, move the marker line according to the expected change difference, and obtain the field of view tracking area after the movement; Obtain the image feature point set of the moved field-of-view tracking area and calculate the corresponding feature distribution parameters; The feature distribution parameters of the field of view tracking region after the movement are compared with the feature distribution parameters of the field of view tracking region in the previous frame to calculate the corresponding movement distribution change, and then the variance change difference is obtained from the movement distribution change as the target matching parameter. A matching reference threshold is preset, and the target matching parameters are compared with the matching reference threshold to determine whether the feature point set matching is successful. If so, the current frame display is replaced with the moved field of view tracking area to achieve dynamic field of view adjustment.

[0048] See Figure 3 The specific process for adjusting the field of view tracking area is as follows: The first step is to obtain the feature distribution parameters (i.e., the distribution of feature points) extracted from the field-view tracking region at time k and time k+1, and the expected change difference between the two frames. Set the maximum variance threshold. This serves as a pre-set upper limit for variance.

[0049] The second step is to calculate the expected change difference between the two frames. Move the center line of the vision tracking area at time k+1 to The location.

[0050] The third step is to recalculate the feature distribution parameters (i.e., the probability distribution of the image feature point set, denoted as) of the moved field-of-view tracking region in the acquired image frame at time k+1. The expected change difference is obtained by taking the difference between the feature distribution parameters of the view tracking region in the k-th frame and the feature distribution parameters of the k-th frame. Variance variation difference .

[0051] Step 4: Compare the differences in variance after shifting. With the maximum variance threshold Relationship: when If the dispersion of the image feature point set extracted from the moved field of view tracking area exceeds a certain threshold, then the matching of the feature point sets of the two frames of images fails, and no field of view adjustment is performed.

[0052] when If the dispersion of the image feature points extracted from the moved region of interest is within a certain range, then the feature point sets of the two frames are successfully matched. In this case, the field of view tracking region of the (k+1)th frame is replaced with the moved field of view tracking region. The display field of view is adjusted dynamically according to the following formula:

[0053]

[0054] That is, only the horizontal coordinate position of the view tracking area is adjusted, and the height of the view tracking area is not adjusted.

[0055] Thus, by combining the initial display area range (the field of view tracking area at time k), the field of view display range of the camera (the field of view tracking area at time k+1) is adjusted, thereby realizing the dynamic field of view adjustment of the semi-trailer truck. This enables the electronic rearview mirror to follow the view when turning, ensuring that the tracked target, such as the rear of the vehicle, remains in a fixed position on the electronic rearview mirror screen.

[0056] In this embodiment, when acquiring the change in motion distribution between the current frame and the previous frame, the marker line is adjusted based on the expected change difference extracted from the change in motion distribution. Preset adjustments are supported based on actual data (the actual expected change difference). Furthermore, based on the preset motion tracking area and the change in motion distribution between the previous frame and the previous frame, the dispersion characteristics of the feature point set extracted from the region of interest (motion tracking area) after the motion are numerically compared with the variance change difference obtained from the change in motion distribution and the matching reference threshold. This determines whether the current motion adjustment optimizes the display. If so, the motion tracking area is replaced to achieve dynamic motion adjustment, thereby ensuring that the target is always closely followed while avoiding jumps in the motion display.

[0057] Example 2 This invention also provides an electronic device, including a processor, a memory, and an image acquisition module; The image acquisition module is used to acquire image frames in real time and transmit them to the processor; Memory, used to store computer programs; When the processor executes a computer program stored in the memory, it implements the electronic exterior rearview mirror follow-up field of view adjustment method as described in any of the embodiments in 1 above. This invention uses a target tracking marker line to define the field of view tracking range, ensuring that the target clock is centered in the display screen, thus guaranteeing the driver's field of view. Simultaneously, during image acquisition, the feature point set of the acquired image frame is synchronously obtained to further determine the movement and distribution changes of the target in the current image acquisition area. The marker line is then adaptively adjusted, moving the field of view tracking area of ​​the current image frame to perform dynamic field of view adjustment. Stable tracking of the field of view tracking area is achieved through point set registration, resulting in smooth display. This invention is applicable to different combinations of tractor and trailer vehicles, exhibiting good versatility and effectively improving vehicle safety.

[0058] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A method for adjusting the field of view of an electronic exterior rearview mirror, characterized in that, include: Marker lines are set based on the target being tracked, and the field of view tracking range is defined according to the marker lines; Real-time acquisition of captured image frames, and acquisition of image feature point set of the field of view tracking area from the captured image frames according to the field of view tracking range; The movement and distribution changes of the tracking target in the current image acquisition area are determined based on the image feature point set; The marker line is adjusted according to the change in the movement distribution, thereby moving the field of view tracking area of ​​the currently acquired image frame to perform follow-up field of view adjustment; The step of setting marker lines based on the tracking target and defining the field of view tracking range according to the marker lines includes: A1. Based on the tracking target on the vehicle body, set the initial position of the marker line; A2. Using the marked line as the center, plan the initial field of view area according to the initial width; A3. Perform ORB feature detection on the visual region and extract feature descriptors and the number of feature points; A4. Determine whether feature extraction is successful based on the number of feature points. If not, adjust the width of the current field of view incrementally and return to step A3. If yes, output the width of the corresponding field of view to define the field of view tracking range. The steps for determining the movement distribution changes of the tracking target in the current image acquisition area based on the image feature point set include: Obtain the image feature point set of the current frame, the image feature point set including ORB feature points; Extract the abscissa data of each ORB feature point in the image feature point set, and calculate the feature distribution parameters of the image feature point set in the field of view tracking region based on the abscissa data, assuming that the ORB feature points follow a Gaussian distribution. By comparing the feature distribution parameters of adjacent acquired image frames, the change in the movement distribution of the tracked target in the current image acquisition area is calculated; The formula for calculating the characteristic distribution parameter is as follows: In the formula, This represents the total number of feature points in the field-of-view tracking region. This represents the i-th feature point in the field-of-view tracking region. and Let represent the mean and variance of the set of image feature points within the field-of-view tracking region in the k-th frame, respectively. The formula for calculating the change in the movement distribution is as follows: In the formula, Represents the difference in expected change. This represents the difference in variance.

2. The electronic exterior rearview mirror's field of view adjustment method as described in claim 1, characterized in that, Adjusting the marker line according to the changes in the movement distribution, thereby moving the field-of-view tracking area of ​​the currently acquired image frame to perform follow-up field-of-view adjustment, including: Obtain the expected change difference from the movement distribution change, move the marker line according to the expected change difference, and obtain the field of view tracking area after the movement; Obtain the image feature point set of the moved field-of-view tracking area and calculate the corresponding feature distribution parameters; The feature distribution parameters of the field of view tracking region after the movement are compared with the feature distribution parameters of the field of view tracking region in the previous frame to calculate the corresponding movement distribution change, and then the variance change difference is obtained from the movement distribution change as the target matching parameter. A matching reference threshold is preset, and the target matching parameters are compared with the matching reference threshold to determine whether the feature point set matching is successful. If so, the current frame display is replaced with the moved field of view tracking area to achieve dynamic field of view adjustment.

3. The electronic exterior rearview mirror's field of view adjustment method as described in claim 1, characterized in that, The success of feature extraction is determined based on the number of feature points. If not, the initial width is adjusted incrementally. Specifically, the number of feature points is determined to be less than the minimum feature value of the field of view tracking area. If so, the feature extraction is determined to have failed. The width value of the current field of view area is adjusted by a preset step size increment to obtain the adjusted target width.

4. The electronic exterior rearview mirror's field of view adjustment method as described in claim 3, characterized in that, Also includes: Obtain the adjusted target width, and determine whether the target width is less than the maximum field of view. If so, return to step A3; otherwise, return to step A1 to reset the initial position of the marker line.

5. An electronic device, characterized in that: The electronic device includes a processor, a memory, and an image acquisition module; The image acquisition module is used to acquire image frames in real time and transmit them to the processor; Memory, used to store computer programs; When a processor executes a computer program stored in a memory, it implements the electronic exterior rearview mirror follow-up field of view adjustment method as described in any one of claims 1-4.

Images