Image distortion correction method and device, vehicle, medium and program product

Abstract

The invention provides an image distortion correction method and device, a vehicle, a medium and a program product, and relates to the technical field of intelligent cabin image processing. The method comprises the following steps: determining an image to be projected on a vehicle windshield by a projection device; distortion correction is conducted on the image through a distortion correction model, target correction parameters of the distortion correction model are obtained according to the actual view angle position corresponding to the projection device and reference correction parameters corresponding to multiple reference view angle positions respectively, and the multiple reference view angle positions are located at different positions in the three-dimensional space of the vehicle; and projecting the image after distortion correction to a windshield of the vehicle for display. According to the method, the reference correction parameters corresponding to the multiple reference view angle positions are determined in advance in the three-dimensional space of the vehicle, image distortion correction of different view angles can be carried out according to actual requirements, the definition and accuracy of the image at any view angle in the space in the vehicle are ensured, and the individual requirements of users are met.

Description

Technical Field

[0001] This disclosure relates to the field of intelligent cockpit image processing technology, and in particular to an image distortion correction method, device, vehicle, medium and program product. Background Technology

[0002] In related technologies, key driving information of a vehicle is projected onto the windshield to facilitate the driver's observation of vehicle and navigation information during driving. However, due to factors such as uneven curvature and thickness of the windshield, the projected image is distorted, so distortion correction is required. Summary of the Invention

[0003] This disclosure provides an image distortion correction method, apparatus, vehicle, medium, and program product that can ensure the clarity and accuracy of images from any viewing angle inside a vehicle to meet the personalized needs of users.

[0004] According to a first aspect of the present disclosure, an image distortion correction method is provided, comprising: Determine the image to be projected onto the vehicle's windshield using the projection device; The image is distorted by a distortion correction model, wherein the target correction parameters of the distortion correction model are obtained based on the actual viewing position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing positions, which are located at different positions in the three-dimensional space of the vehicle. The distortion-corrected image is projected onto the vehicle's windshield for display.

[0005] In this way, by pre-determining the reference correction parameters corresponding to multiple reference viewpoint positions in the three-dimensional space of the vehicle, image distortion correction can be performed from different viewpoints according to actual needs, ensuring the clarity and accuracy of the image from any viewpoint inside the vehicle space, and meeting the personalized needs of users.

[0006] In some possible implementations, the target correction parameters are obtained through a three-dimensional interpolation algorithm based on the actual viewing angle position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing angle positions.

[0007] Thus, through the three-dimensional interpolation algorithm, the target correction parameters corresponding to the actual viewpoint position can be dynamically calculated based on the reference correction parameters corresponding to multiple reference viewpoint positions, ensuring that the image can be effectively corrected for distortion under different viewpoints, and better adapting to complex projection environments and changing viewpoint positions.

[0008] In some possible implementations, the target correction parameters are obtained in the following manner: For each reference viewpoint position, determine the Euclidean distance between the reference viewpoint position and the actual viewpoint position; Based on the Euclidean distance, determine the weight parameters corresponding to the reference viewpoint position; The target correction parameter is obtained based on the weight parameter and reference correction parameter corresponding to each of the plurality of reference viewpoint positions.

[0009] In this way, by using Euclidean distance and weight parameters, the weights of different reference correction parameters can be determined based on the spatial relationship between the actual viewpoint position and the reference viewpoint position, thereby calculating the target correction parameters more accurately.

[0010] In some possible implementations, the reference correction parameters corresponding to the reference viewpoint position are obtained in the following manner: When the projection device projects the calibration image to be corrected onto the windshield, for each of the plurality of reference viewpoints, the actual position corresponding to the feature point in the calibration image is determined. Based on the actual and ideal positions of the feature points, the reference correction parameters corresponding to the reference viewpoint position are determined.

[0011] In this way, by pre-calibrating the difference between the actual and ideal positions of the feature points in the image, the distortion of the image under different reference viewpoints can be determined, ensuring the reliability and effectiveness of the target correction parameters and providing an accurate basis for actual image distortion correction.

[0012] In some possible implementations, in the vehicle's corresponding three-dimensional coordinate system, the multiple reference viewpoint positions include: The origin of the three-dimensional coordinate system; The three axes of the three-dimensional coordinate system correspond to calibration points, and each axis includes at least one calibration point.

[0013] In this way, by selecting calibration points as reference viewpoints on the origin and axes of the three-dimensional coordinate system, the interior space of the vehicle can be fully covered. These locations encompass the possible installation positions and usage scenarios of projection devices inside the vehicle, ensuring the applicability and accuracy of the distortion correction model.

[0014] In some possible implementations, the origin of the three-dimensional coordinate system is the geometric center of the vehicle, and each axis includes at least two calibration points, wherein the at least two calibration points corresponding to an axis are distributed on the positive half-axis and the negative half-axis corresponding to the axis.

[0015] In this way, by selecting calibration points on the positive and negative half-axis of each axis, the space inside the vehicle can be fully covered, the distortion patterns in different directions inside the vehicle can be captured more accurately, and richer data can be provided, thereby improving the accuracy of the correction model.

[0016] In some possible implementations, the method further includes: In response to the angle adjustment operation corresponding to the projection device, a new actual angle position is obtained; Based on the reference correction parameters corresponding to the multiple reference viewpoint positions and the new actual viewpoint position, new target correction parameters are determined.

[0017] In this way, by responding to the perspective adjustment operation and calculating new target correction parameters in real time, the system can dynamically adapt to changes in the actual perspective position of the projection device, so that drivers and passengers can see clear and accurate images from different perspectives inside the vehicle, thus improving the user experience.

[0018] In some possible implementations, the viewpoint adjustment operation includes at least one of the following: The system obtains the user's movement operation on the vehicle's infotainment display screen regarding the actual viewing position of the projection device. The system detects a change in the eye position of a designated user corresponding to the projection device, where the designated user is a driver or passenger in a designated position within the vehicle. It is determined that the designated user corresponding to the projection device has been switched.

[0019] Thus, by providing multiple viewing angle adjustment options, it can adapt to various user scenarios. The viewing angle and correction parameters of the projection device can be dynamically adjusted according to the eye position and needs of different users, ensuring that each user obtains the best display effect and enhancing the system's personalization. In some possible implementations, the projection device includes multiple units, and the display area of ​​the windshield includes multiple display sub-areas, with each projection device corresponding to one display sub-area.

[0020] In this way, more key driving information can be displayed, and different users can be assigned to different projection devices. This not only meets the core needs of the driver, but also fully considers the experience of passengers in the vehicle, satisfying the personalized needs of users. In some possible implementations, the display area of ​​the windshield extends from one side A-pillar of the vehicle to the other side A-pillar.

[0021] In this way, the windshield provides a wider display area, offering a large display area for key driving information and reducing the need for the driver to shift their gaze.

[0022] In some possible implementations, the method further includes, prior to distortion correction of the image using a distortion correction model: Determine the actual viewing angle positions corresponding to the plurality of projection devices; For each projection device, the target correction parameters corresponding to the projection device are determined based on the actual viewing angle position and the reference correction parameters corresponding to the multiple reference viewing angle positions.

[0023] In this way, by automatically detecting the actual viewing angle position of each projection device and determining the target correction parameters, the image distortion of each display sub-area can be accurately corrected, and it can be used to set different specified users for different projection devices, further meeting the personalized needs of users.

[0024] According to a second aspect of the present disclosure, an image distortion correction apparatus is provided, comprising: The determination module is configured to determine the image to be projected onto the vehicle windshield by the projection device. The correction module is configured to perform distortion correction on the image using a distortion correction model, wherein the target correction parameters of the distortion correction model are obtained based on the actual viewing angle position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing angle positions, which are located at different positions in the three-dimensional space of the vehicle. The projection module is configured to project a distortion-corrected image onto the vehicle's windshield for display.

[0025] In some possible implementations, the target correction parameters are obtained through a three-dimensional interpolation algorithm based on the actual viewing angle position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing angle positions.

[0026] In some possible implementations, the target correction parameters are obtained in the following manner: For each reference viewpoint position, determine the Euclidean distance between the reference viewpoint position and the actual viewpoint position; Based on the Euclidean distance, determine the weight parameters corresponding to the reference viewpoint position; The target correction parameter is obtained based on the weight parameter and reference correction parameter corresponding to each of the plurality of reference viewpoint positions.

[0027] In some possible implementations, the reference correction parameters corresponding to the reference viewpoint position are obtained in the following manner: When the projection device projects the calibration image to be corrected onto the windshield, for each of the plurality of reference viewpoints, the actual position corresponding to the feature point in the calibration image is determined. Based on the actual and ideal positions of the feature points, the reference correction parameters corresponding to the reference viewpoint position are determined.

[0028] According to a third aspect of the present disclosure, a vehicle is provided, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to perform the image distortion correction method described in the first aspect of the present disclosure.

[0029] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the image distortion correction method described in the first aspect of the present disclosure.

[0030] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the image distortion correction method described in the first aspect of the present disclosure.

[0031] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects: This disclosure involves determining the image to be projected onto a vehicle's windshield using a projection device; correcting the image's distortion using a distortion correction model, wherein the target correction parameters of the distortion correction model are obtained based on the actual viewing angle position of the projection device and reference correction parameters corresponding to multiple reference viewing angle positions located at different positions in the vehicle's three-dimensional space; and projecting the distortion-corrected image onto the vehicle's windshield for display. Thus, by pre-determining the reference correction parameters corresponding to multiple reference viewing angle positions in the vehicle's three-dimensional space, image distortion correction can be performed at different viewing angles according to actual needs, ensuring the clarity and accuracy of the image from any viewing angle within the vehicle space and meeting the user's personalized needs.

[0032] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0033] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0034] Figure 1 This is a flowchart illustrating an image distortion correction method according to an exemplary embodiment.

[0035] Figure 2 This is a schematic diagram illustrating a scene of capturing a distorted image according to an exemplary embodiment.

[0036] Figure 3 This is a flowchart illustrating an image distortion correction method according to an exemplary embodiment.

[0037] Figure 4 This is a block diagram illustrating an image distortion correction device according to an exemplary embodiment.

[0038] Figure 5 This is a block diagram illustrating a vehicle according to an exemplary embodiment.

[0039] Figure 6 This is a block diagram illustrating a chip according to an exemplary embodiment. Detailed Implementation

[0040] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0041] In related technologies, when correcting distortion in an image projected onto a windshield, the correction parameters in the distortion correction algorithm can be calibrated using the driver's calibrated viewing angle. Specifically, a camera can be used to acquire a specific calibration image at the calibrated viewing angle. Reverse correction is then performed by analyzing the feature point positions of the calibration image and the corresponding feature point positions of the distorted image projected onto the windshield, ensuring that the image seen by the driver is clear and undistorted.

[0042] However, the calibration viewpoint primarily focuses on the driver, which has several limitations. For example, when passengers view the projected content from the front passenger seat or rear seats, the image will still be distorted and misaligned due to the different viewing angles, and may even become unrecognizable, thus limiting the usability of the projected content for other occupants. Furthermore, a single viewpoint calibration cannot adapt to changes in viewing position caused by vehicle seat adjustments or drivers of different heights. For instance, when the driver adjusts the seat position (forward / backward, height / lowering), or when drivers of different heights and sitting postures use the vehicle, the correction parameters corresponding to the original calibration viewpoint may not guarantee the accuracy of the image under the new viewpoint, leaving the driver still at risk of misinterpreting information due to image distortion.

[0043] Reference Figure 1 , Figure 1 This is a flowchart illustrating an image distortion correction method according to an exemplary embodiment, such as... Figure 1 As shown, the image distortion correction method can be applied to vehicles and includes the following steps.

[0044] In step S101, the image to be projected onto the vehicle windshield by the projection device is determined.

[0045] In step S102, the image is distorted using a distortion correction model. The target correction parameters of the distortion correction model are obtained based on the actual viewing angle position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing angle positions, which are located at different positions in the three-dimensional space of the vehicle.

[0046] In step S103, the distortion-corrected image is projected onto the vehicle's windshield for display.

[0047] For example, a projection device is a vehicle-mounted device used to generate images and project them onto the windshield. The projection device can project data such as key driving information as images onto a region or the entire bottom region of the windshield in front of the driver's line of sight. The key driving information may include at least one of the following: vehicle speed, engine speed, gear position, fuel level, fuel consumption, coolant temperature, navigation information, indicator lights, etc.

[0048] For example, the image is the original picture content that the projection device is preparing to project onto the windshield. The projection device can generate a light source array corresponding to the image and project it onto the windshield. The content projected onto the windshield can be numbers, arrows, icons, graphics, text, etc.

[0049] For example, distortion refers to the shape distortion of an image during projection caused by factors such as the optical system and projection angle, such as barrel distortion or pincushion distortion. In particular, the image projected by the projection device is distorted due to factors such as the curvature and thickness of the windshield, as well as the different viewing angles of the driver and passengers, and the different heights, driving habits, and sitting postures of different people.

[0050] For example, a distortion correction model is a mathematical model or algorithmic framework used to correct image distortion, describing the distortion characteristics of a windshield at a specific viewing angle. The distortion correction model can eliminate distortion by calculating and adjusting the pixel positions of an image.

[0051] For example, the mathematical model of the distortion correction model may include radial distortion correction functions and tangential distortion correction functions, or it may employ checkerboard corner detection, etc.; there are no limitations here.

[0052] For example, the three-dimensional space of a vehicle refers to the interior space of the vehicle, which can be described by a three-dimensional coordinate system with a point inside the vehicle as the origin. This three-dimensional coordinate system can be used to accurately describe the position of the projection device, the position of the reference point, and the eye position of the driver and passengers. The origin of the three-dimensional coordinate system can be the whole-vehicle three-dimensional coordinate system specified in standard documents, such as those in ISO 8855 / DIN 70000; or it can be a three-dimensional coordinate system established with any point inside the vehicle as the origin, which is not limited here.

[0053] For example, distortion correction models can be multinomial models, mesh mapping models, etc. In these models, pixel coordinate mapping relationships can be calculated by establishing a distortion correction mathematical model. For instance, methods such as nearest neighbor interpolation, bilinear interpolation, and 3D interpolation algorithms can be used to resample pixel values, ensuring that the image projected onto the windshield conforms to human visual perception. Distortion correction models can also be based on deep learning models that predict correction parameters and are trained using annotation information from specific regions.

[0054] For example, distortion correction models have correction parameters. For instance, if the distortion correction model is a mathematical model, the coefficients of the variables in a polynomial model are the correction parameters; if the distortion correction model is a deep learning model, then the weight parameters, bias parameters, or hyperparameters of the neural network are the correction parameters.

[0055] For example, different viewing angles correspond to different glass surface angles, thus resulting in different distortion phenomena. The actual viewing angle position refers to the actual position of the target occupant's eyes in the vehicle's three-dimensional space, which can be estimated by a vision sensor or determined by user preset. Reference viewing angle positions are a series of pre-selected calibrated positions within the vehicle's three-dimensional space, distributed across any possible viewing angle of the driver and passengers. These multiple reference viewing angle positions can be discrete positions or pre-set calibrated positions according to certain rules, such as points on the coordinate axes of a three-dimensional coordinate system.

[0056] In some embodiments, in the vehicle's corresponding three-dimensional coordinate system, the multiple reference viewpoint positions include: the origin of the three-dimensional coordinate system; calibration points corresponding to the three axes of the three-dimensional coordinate system, with each axis including at least one calibration point.

[0057] In some embodiments, the origin of the three-dimensional coordinate system is the geometric center of the vehicle, and each axis includes at least two calibration points, wherein the at least two calibration points corresponding to an axis are distributed on the positive half-axis and the negative half-axis corresponding to the axis.

[0058] Understandably, the viewpoint here can be a single point, or it can be the eye box area in related technologies, i.e., the area where the user's eyes can move at the current position, for example... Figure 2 The movable area of ​​the user's eye position in the YOZ plane.

[0059] For example, the reference correction parameters are for each pre-selected reference viewing angle position, and are the correction parameters corresponding to the windshield at that reference viewing angle position. The reference correction parameters can be reference data that has been precisely calculated in advance and stored in the system.

[0060] For example, during the calibration phase, reference correction parameters corresponding to the calibrated reference viewing angle positions can be predetermined to construct a three-dimensional spatial model. During actual use, the image content to be projected onto the vehicle's windshield by the projection device, such as navigation maps, vehicle speed information, and warning signs, can be determined. The actual viewing angle position of the driver's eyes in the vehicle's three-dimensional space can be estimated or determined using cameras, seat sensors, or other methods. Based on a certain interpolation or fitting algorithm, the target correction parameters corresponding to the actual viewing angle position can then be determined using multiple pre-calibrated reference correction parameters corresponding to the reference viewing angle positions.

[0061] For example, a distortion correction model is configured based on the calculated target distortion parameters. The original image is then digitally processed using the distortion correction model to generate a pre-distorted image, i.e., the distortion-corrected image. This distortion-corrected image is then projected onto the vehicle's windshield using a projection device. Ultimately, the image seen by the driver through the windshield is the clear, undistorted, and correctly positioned original information, avoiding misreading or visual fatigue caused by distortion. This provides the driver with a more comfortable and natural visual experience, enhancing user satisfaction with the vehicle's display system.

[0062] For example, by establishing a three-dimensional spatial model of distortion correction parameters, distortion correction can be achieved based on any viewpoint, meeting personalized needs. Moreover, it has strong synergy with systems such as smart cockpits, ADAS (Advanced Driving Assistance Systems), and navigation, enabling seamless information exchange and sharing, making it a highly promising development direction in the field of intelligent vehicles.

[0063] This disclosure involves determining an image to be projected onto a vehicle's windshield using a projection device; correcting the image's distortion using a distortion correction model, wherein the target correction parameters of the distortion correction model are obtained based on the actual viewing angle position corresponding to the projection device and reference correction parameters corresponding to multiple reference viewing angle positions located at different positions in the vehicle's three-dimensional space; and projecting the distortion-corrected image onto the vehicle's windshield for display. Thus, by pre-determining the reference correction parameters corresponding to multiple reference viewing angle positions in the vehicle's three-dimensional space, image distortion correction can be performed at different viewing angles according to actual needs, ensuring the clarity and accuracy of the image from any viewing angle within the vehicle's interior space.

[0064] In some possible implementations, the target correction parameters are obtained through a three-dimensional interpolation algorithm based on the actual viewing angle position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing angle positions.

[0065] For example, a 3D interpolation algorithm is a mathematical method for continuously reconstructing discrete data points in 3D space. It calculates the value at any new location in space using known values. Here, the target distortion parameters corresponding to the actual viewpoint position can be determined using the reference distortion parameters corresponding to the reference viewpoint position. 3D interpolation algorithms include trilinear interpolation, nearest neighbor interpolation, radial basis function interpolation, etc., and are not limited here.

[0066] For example, the three-dimensional interpolation algorithm can use the inverse distance weighted interpolation algorithm, which determines the weight coefficient corresponding to each reference view position by measuring the distance between multiple reference view positions and the actual view position, and then performs a weighted average to obtain the target distortion parameters corresponding to the actual view position.

[0067] For example, multiple reference viewpoint positions are preset in the three-dimensional space of the vehicle, and the image distortion at these positions is determined to obtain the corresponding reference correction parameters. Based on the actual viewpoint position corresponding to the projection device, the actual viewpoint position is spatially correlated with the multiple reference viewpoint positions. Using a three-dimensional interpolation algorithm, the target correction parameters corresponding to the actual viewpoint position can be determined based on the reference correction parameters and weights corresponding to the reference viewpoint positions.

[0068] Thus, through the three-dimensional interpolation algorithm, the target correction parameters corresponding to the actual viewpoint position can be dynamically calculated based on the reference correction parameters corresponding to multiple reference viewpoint positions, ensuring that the image can be effectively corrected for distortion under different viewpoints, and better adapting to complex projection environments and changing viewpoint positions.

[0069] In some possible implementations, the target correction parameters are obtained in the following manner: For each reference viewpoint position, determine the Euclidean distance between the reference viewpoint position and the actual viewpoint position; Based on the Euclidean distance, determine the weight parameters corresponding to the reference viewpoint position; The target correction parameter is obtained based on the weight parameter and reference correction parameter corresponding to each of the plurality of reference viewpoint positions.

[0070] For example, Euclidean distance is used to measure the spatial distance between a reference viewpoint and the actual viewpoint. The weighting parameter reflects the contribution of the reference correction parameter at the reference viewpoint to the target correction parameter at the actual viewpoint. Specifically, the closer the reference viewpoint is to the actual viewpoint, the larger the weighting parameter corresponding to the reference viewpoint; conversely, the farther the reference viewpoint is from the actual viewpoint, the smaller the weighting parameter corresponding to the reference viewpoint.

[0071] For example, for each reference viewpoint position, the Euclidean distance between it and the actual viewpoint position is determined. A weight parameter is assigned to each reference viewpoint position based on the Euclidean distance. Typically, the reciprocal of the Euclidean distance or another decay function is used to determine the weights, with closer reference viewpoint positions receiving greater weights. Here, the weight parameter corresponding to the reference viewpoint position can be determined by taking the reciprocal of the square of the Euclidean distance between the reference viewpoint position and the actual viewpoint position. Then, using the weight parameter of each reference viewpoint position and the corresponding reference correction parameter, a weighted average method is used to determine the target correction parameter corresponding to the actual viewpoint position.

[0072] In this way, by using Euclidean distance and weight parameters, the weights of different reference correction parameters can be determined based on the spatial relationship between the actual viewpoint position and the reference viewpoint position, thereby calculating the target correction parameters more accurately.

[0073] In some possible implementations, the reference correction parameters corresponding to the reference viewpoint position are obtained in the following manner: When the projection device projects the calibration image to be corrected onto the windshield, for each of the plurality of reference viewpoints, the actual position corresponding to the feature point in the calibration image is determined. Based on the actual and ideal positions of the feature points, the reference correction parameters corresponding to the reference viewpoint position are determined.

[0074] For example, a calibration image is a specific image used for distortion correction, typically containing easily identifiable feature points such as corner points and dots. Calibration images can be, for example, a standardized checkerboard pattern, a dot array, etc., and contain multiple identifiable feature points with well-defined locations and shapes. Examples include corner points in a checkerboard image or the centers of circles in a dot array image. The number of feature points can be set according to the actual situation, such as 20, 50, 100, etc., and is not limited here.

[0075] For example, the ideal position is the location of a feature point under ideal, distortion-free conditions. The ideal position is typically preset when designing a calibration image. The actual position is the location of the feature point in the actual display after being projected onto the windshield. Due to distortion, the actual position may differ from the ideal position.

[0076] For example, the calibration image is projected onto the windshield using a projection device. The calibration image contains feature points at multiple known ideal positions. For each calibration image obtained from a reference viewpoint, image recognition techniques, such as corner detection and template matching, are used to determine the actual position of the feature points in the calibration image on the windshield.

[0077] For example, the reference correction parameters corresponding to the reference viewpoint position can be determined by comparing the difference between the actual and ideal positions of feature points. The calculated reference correction parameters are stored for use in the actual correction process for calculating target correction parameters or other related operations.

[0078] For example, given the coordinate information corresponding to the actual and ideal positions of feature points in a calibrated image, if the distortion correction model is a polynomial mathematical model, the coefficients of the variables in the polynomial model, i.e., the target correction parameters, can be determined by substituting the coordinate information corresponding to the actual and ideal positions into the polynomial mathematical model. If the distortion correction model is a deep learning model, the positional distortion relationship between the actual and ideal positions can be learned by substituting the coordinate information corresponding to the actual and ideal positions into the deep learning model, thereby obtaining the weight parameters, bias parameters, or hyperparameters of the neural network, i.e., the target correction parameters.

[0079] In this way, by pre-calibrating the difference between the actual and ideal positions of the feature points in the image, the distortion of the image under different reference viewpoints can be determined, ensuring the reliability and effectiveness of the target correction parameters and providing an accurate basis for actual image distortion correction.

[0080] In some possible implementations, in the vehicle's corresponding three-dimensional coordinate system, the multiple reference viewpoint positions include: The origin of the three-dimensional coordinate system; The three axes of the three-dimensional coordinate system correspond to calibration points, and each axis includes at least one calibration point.

[0081] For example, the vehicle's three-dimensional coordinate system is a three-dimensional spatial coordinate system established with the vehicle as a reference. It can be used to describe the position and spatial relationship of various components inside the vehicle. For instance, a fixed point on the vehicle can be selected as the origin, such as the center of the vehicle or the center of a specific component, and three mutually perpendicular axes can be defined to represent the front-back, left-right, and up-down directions of the vehicle, respectively.

[0082] For example, a calibration point is a specific point selected on an axis of a three-dimensional coordinate system, used as a reference viewpoint. The calibration point can be selected at any location on the coordinate axis, but typically a representative point is chosen based on actual needs, such as the midpoint or endpoint of the axis; there are no restrictions here.

[0083] For example, a three-dimensional coordinate system is predefined with the geometric center of the vehicle as the origin, and its origin and three axes are determined. For example, the direction pointing forward of the vehicle is the X-axis, the direction pointing to the left of the driver's seat is the Y-axis, and the vertical direction pointing to the roof of the vehicle is the Z-axis.

[0084] For example, taking the pre-defined eye position of the passenger in the front passenger seat as the origin, a three-dimensional coordinate system can be predefined, and its origin and three axes can be determined. For example, the direction pointing forward of the vehicle is the X-axis, the direction pointing to the left of the driver's seat is the Y-axis, and the vertical direction pointing to the roof of the vehicle is the Z-axis.

[0085] For example, multiple reference viewpoints can be selected in a three-dimensional coordinate system, which may include the origin and calibration points on each axis, with at least one calibration point selected as a reference viewpoint on each axis. The calibration point can be a fixed position on the axis, such as a point at a certain distance from the origin.

[0086] For example, in addition to the origin, a calibration point can be set on each axis as a reference viewpoint. For instance, a calibration point can be set on the positive half-axis of the X-axis, Y-axis, and Z-axis, and the distance between each calibration point and the origin can be equal or unequal.

[0087] It is understandable that if a calibration point is set on each axis other than the origin, and the space between the origin and the three calibration points does not include all viewpoints inside the vehicle, the target correction parameters corresponding to all viewpoints inside the vehicle can still be calculated. Setting the space between the origin and the three calibration points to include all viewpoints inside the vehicle can further ensure the accuracy of the target correction parameters.

[0088] For example, at each reference viewpoint, the distortion of the image is measured using a calibration image, and the corresponding reference correction parameters are calculated. Using these correction parameters from the reference viewpoints, a complete distortion correction model can be constructed to describe the distortion patterns at different viewpoints inside the vehicle.

[0089] In this way, by selecting calibration points as reference viewpoints on the origin and axes of the three-dimensional coordinate system, the interior space of the vehicle can be fully covered. These locations encompass the possible installation positions and usage scenarios of projection devices inside the vehicle, ensuring the applicability and accuracy of the distortion correction model.

[0090] In some possible implementations, the origin of the three-dimensional coordinate system is the geometric center of the vehicle, and each axis includes at least two calibration points, wherein the at least two calibration points corresponding to an axis are distributed on the positive half-axis and the negative half-axis corresponding to the axis.

[0091] For example, the geometric center of a vehicle refers to its center of mass or center of symmetry in three-dimensional space. The geometric center is an important reference point in vehicle design, used to define the vehicle's overall coordinate system. The location of the geometric center can be calculated and determined based on the vehicle's dimensions, shape, and mass distribution.

[0092] For example, in addition to the origin, two calibration points can be set on each axis as reference viewpoint positions. For instance, one calibration point can be set on the positive half-axis of the X-axis, Y-axis, and Z-axis, and one calibration point can be set on the negative half-axis of the X-axis, Y-axis, and Z-axis. The distance between each calibration point and the origin can be equal or unequal.

[0093] It is understandable that if each axis, excluding the origin, has at least one calibration point on both the positive and negative half-axis, the target correction parameters can still be calculated for all viewpoints within the vehicle, even when the origin is any location in the three-dimensional space inside the vehicle. By setting the origin to the geometric center of the vehicle, all viewpoints within the vehicle can be contained as much as possible within the space connected by all reference viewpoints, further ensuring the accuracy of the target correction parameters.

[0094] In this way, by selecting calibration points on the positive and negative half-axis of each axis, the space inside the vehicle can be fully covered, the distortion patterns in different directions inside the vehicle can be captured more accurately, and richer data can be provided, thereby improving the accuracy of the correction model.

[0095] In some possible implementations, the method further includes: In response to the angle adjustment operation corresponding to the projection device, a new actual angle position is obtained; Based on the reference correction parameters corresponding to the multiple reference viewpoint positions and the new actual viewpoint position, new target correction parameters are determined.

[0096] For example, the viewing angle adjustment operation refers to the operation that triggers a change in the viewing angle position of the projection device, including user-initiated adjustment, automatic system adjustment, or other methods. For instance, the user can freely select the actual viewing angle position of the projection device on the vehicle's display screen, or the system can automatically determine the actual viewing angle position of the projection device based on the driver's eye position.

[0097] For example, after the viewing angle adjustment operation, the new actual viewing angle position of the projection device can be determined. The new actual viewing angle position may be different from the previous actual viewing angle position, and new target correction parameters corresponding to the new actual viewing angle position can be recalculated to ensure the accuracy of image display.

[0098] For example, when the system detects that the user or the system has adjusted the viewing angle of the projection device, the vehicle system can obtain the new actual viewing angle position by means of sensors or user input.

[0099] For example, based on the new actual viewpoint position and the reference correction parameters corresponding to multiple existing reference viewpoint positions, new target correction parameters can be determined using a 3D interpolation algorithm or other calculation methods. For instance, the Euclidean distance between the new actual viewpoint position and each reference viewpoint position can be determined. Based on the Euclidean distance, a weight parameter corresponding to each reference viewpoint position relative to the new actual viewpoint position is determined. Using a weighted average method, combining the reference correction parameters and the weight parameters, the new target correction parameters are calculated.

[0100] For example, new target correction parameters can be input into the distortion correction model to perform distortion correction processing on the image. The distortion-corrected image is then projected onto the vehicle's windshield using a projection device to ensure a clear and accurate display.

[0101] In this way, by responding to the perspective adjustment operation and calculating new target correction parameters in real time, the system can dynamically adapt to changes in the actual perspective position of the projection device, so that drivers and passengers can see clear and accurate images from different perspectives inside the vehicle, thus improving the user experience.

[0102] In some possible implementations, the viewpoint adjustment operation includes at least one of the following: The system obtains the user's movement operation on the vehicle's infotainment display screen regarding the actual viewing position of the projection device. The system detects a change in the eye position of a designated user corresponding to the projection device, where the designated user is a driver or passenger in a designated position within the vehicle. It is determined that the designated user corresponding to the projection device has been switched.

[0103] For example, the in-vehicle infotainment display is the central control display screen or other interactive interface inside the vehicle, where users can interact with the in-vehicle system through touch, buttons, or other means. Motion operation refers to the user's adjustment of the actual viewing angle of the projection device on the in-vehicle infotainment display screen through touch, dragging, or other methods.

[0104] For example, users can change the viewing position of the projection device by dragging a virtual slider or icon on the screen. After detecting the user's movement, the new actual viewing position is obtained, and the new target correction parameters are calculated based on the new actual viewing position and the existing reference correction parameters.

[0105] For example, eye position refers to the location of a specified user's eyes inside the vehicle, where the specified user may be, for example, the driver, front passenger, or rear passenger. Changes in the specified user's eye position may affect the optimal viewing angle of the projection device; for example, a change of driver, adjustment of the driver's seat, or a change in the passenger's seating position corresponding to the specified user. The viewing angle of the projection device can be adjusted according to the specified user's eye position to ensure optimal display results.

[0106] For example, a camera or other sensors can identify the eye position of a specified user. When a change in eye position is detected, the actual viewing angle of the projection device is automatically adjusted to ensure that the image is displayed in the specified user's optimal field of vision. Based on the new actual viewing angle and reference correction parameters, new target correction parameters are calculated.

[0107] For example, switching the designated user for the projection device refers to a change in the object corresponding to the projection device. For instance, the designated user of the projection device may be changed from the driver to a front passenger, a passenger in any rear seat, or from a passenger back to the driver. After the user switch, target correction parameters can be determined based on the actual viewing position of the designated user to accommodate the new user's observation needs.

[0108] In this way, the ability to provide multiple viewing angle adjustment options can adapt to various user scenarios. It can dynamically adjust the viewing angle and correction parameters of the projection device according to the eye position and needs of different users, ensuring that each user can obtain the best display effect and enhancing the personalization of the system.

[0109] In some possible implementations, the projection device includes multiple devices, and the display area of ​​the windshield includes multiple display sub-areas, with each projection device corresponding to one display sub-area.

[0110] For example, a vehicle may be equipped with multiple projection devices, each responsible for projecting onto a specific area of ​​the windshield. The display sub-area can be different locations on the windshield, such as the upper left corner, the lower right corner, etc., or it can be divided according to function, such as a navigation information area, a vehicle speed information area, etc.

[0111] For example, the display area of ​​the windshield can be set according to the actual situation, such as the entire windshield, the lower half of the windshield, the area in front of the driver in the windshield, etc. The shape is not limited, and the size is set according to the actual situation, which is not restricted here. In addition, the division position of the display sub-areas is also not limited.

[0112] In one example, such as Figure 2As shown, the display area of ​​the windshield can be the black area 200 corresponding to the bottom of the windshield (not shown), which can reduce optical reflection errors, improve display clarity and stability, and provide advantages such as high resolution, high brightness, and high contrast, enabling clear presentation of information even in strong light. The display area can be evenly divided into multiple display sub-areas 201, such as three display sub-areas 201, or multiple display sub-areas 201 can be set at intervals within the display area. Each display sub-area 201 corresponds to a projection device; there is no limitation here.

[0113] In this way, more key driving information can be displayed, and different users can be assigned to different projection devices. This not only meets the core needs of the driver, but also fully considers the experience of passengers in the vehicle and meets the personalized needs of users.

[0114] In some possible implementations, the display area of ​​the windshield extends from one side A-pillar of the vehicle to the other side A-pillar.

[0115] In this way, the windshield provides a wider display area, offering a large display area for key driving information and reducing the need for the driver to shift their gaze.

[0116] In some possible implementations, the method further includes, prior to distortion correction of the image using a distortion correction model: Determine the actual viewing angle positions corresponding to the plurality of projection devices; For each projection device, the target correction parameters corresponding to the projection device are determined based on the actual viewing angle position and the reference correction parameters corresponding to the multiple reference viewing angle positions.

[0117] For example, the actual viewing angle position for each projection device can be determined before distortion correction. This can be achieved, for instance, through user specification or automatic system detection.

[0118] For example, users can pre-set a designated user for each projection device according to their actual needs, and through the aforementioned perspective adjustment operations, such as the user moving the actual perspective position of the projection device on the vehicle's display screen, the changes in the eye position of the designated user corresponding to the projection device can be detected, and the actual perspective positions of the multiple projection devices can be determined.

[0119] For example, for each projection device, the target correction parameters for that projection device are calculated based on the actual viewing angle position corresponding to the projection device and the reference correction parameters corresponding to multiple existing reference viewing angle positions. The calculated target correction parameters are then input into the distortion correction model to correct the image corresponding to each projection device. The corrected image is then projected onto the corresponding display sub-area of ​​the windshield to ensure a clear and accurate display.

[0120] In this way, by automatically detecting the actual viewing angle position of each projection device and determining the target correction parameters, the image distortion of each display sub-area can be accurately corrected, and it can be used to set different specified users for different projection devices, further meeting the personalized needs of users.

[0121] Reference Figure 2-3 This disclosure provides a calibration scenario for an image distortion correction method for intelligent vehicles. For example, the display area of ​​a car windshield can be divided into three independent display sub-areas, and the display image corresponding to each projection device can be projected onto the corresponding display sub-area. Due to factors such as the curvature of the windshield and installation errors, the image projected onto the display sub-area is distorted. Therefore, distortion correction calibration can be performed in advance to ensure the clarity and accuracy of the image under different viewing angles.

[0122] For example, this disclosure calculates distortion correction parameters at different viewpoints by calibrating at multiple reference viewpoints within the vehicle's three-dimensional space, establishes a three-dimensional spatial model, and achieves distortion correction at any viewpoint in three-dimensional space through a three-dimensional interpolation algorithm. Furthermore, vehicle owners can customize distortion correction styles through the system application to meet the usage needs of different passengers.

[0123] like Figure 3 As shown, the calibration of the correction parameters of the projection device includes: S301, after automobile production is completed, the control projection devices independently display standard calibration images in their respective display sub-regions 201. The calibration images include the ideal positions corresponding to feature points. The calibration images contain uniformly distributed feature points covering the entire screen, serving as a reference for subsequent distortion correction.

[0124] S302, high-definition cameras are placed at seven reference viewpoints within the vehicle's three-dimensional space, and each camera captures distorted images of three display sub-regions 201.

[0125] Among them, such as Figure 2 As shown, the 7 reference viewpoints are: The center position of the vehicle is used as the origin of the three-dimensional coordinate system; This refers to the left side of the vehicle, near the driver's eye level, located on the positive half of the Y-axis. The location is on the right side of the vehicle, near the passenger's eye level, and situated on the negative half of the Y-axis. This is the position above the vehicle, located on the positive half of the Z-axis; This is the position below the vehicle, located on the negative half of the Z-axis; This is the position at the front of the vehicle, located on the positive half of the X-axis. This refers to the rear of the vehicle, near the eye level of the rear passengers, located on the negative half of the X-axis.

[0126] This includes ensuring that the shooting angle and position of the high-definition camera are fixed to guarantee the accuracy of the captured distorted images.

[0127] S303 extracts the set of actual position coordinates of feature points from the captured distorted image using an image processing algorithm.

[0128] S304 compares the actual position coordinates of the extracted feature points with the ideal position coordinates to calculate the distortion correction parameters.

[0129] The distortion correction parameters for each of the three display sub-regions at each reference viewpoint can be calculated by comparing the actual coordinates with the ideal coordinates of feature points in the distorted image.

[0130] The formula for calculating the distortion correction parameters is as follows:

[0131] in, Indicates the first At the first reference viewpoint position The distortion correction parameters corresponding to each display sub-region Indicates in The actual coordinates of the feature points in the distorted image captured at the reference viewpoint. This represents the ideal coordinates corresponding to the feature point. This represents a distortion correction model.

[0132] Among them, the distortion correction model can use a polynomial fitting method to model a pre-set polynomial model to ensure the accuracy of the calculation results.

[0133] Specifically, for each display sub-region, S302-304 is repeatedly executed at 7 reference viewing positions to obtain the distortion correction parameters for each reference viewing position for each display sub-region, i.e., the reference correction parameters.

[0134] S305 establishes a three-dimensional spatial model of distortion correction parameters based on the reference correction parameters corresponding to 7 reference viewpoint positions.

[0135] The three-dimensional spatial model can be used to calculate the coordinates of any point within a specified range using a three-dimensional interpolation algorithm. This serves as the target correction parameter corresponding to the actual viewpoint position. The specified range can be the interior area of ​​the vehicle.

[0136] The calculation involves a weighted average based on the distance between the actual viewpoint and the reference viewpoint, with closer viewpoints receiving a higher weight. The reference viewpoint position is known. The corresponding reference correction parameters are Then the reference correction parameter at point Q It can be represented as:

[0137] in, It is a point to the calibration point The Euclidean distance. It is a weighted index (usually taken as...) ).

[0138] With the S305, users can perform personalized image distortion correction according to their actual needs.

[0139] For example, the application of the vehicle system can provide a variety of preset calibration view positions, and the car owner can choose to perform image distortion correction from view positions such as the driver's seat, passenger seat, rear seat or the center of the vehicle.

[0140] For example, the in-vehicle infotainment system can allow drivers to specify a more precise viewing angle, such as a position slightly to the left and forward of the center of the vehicle. Drivers can select any position as the actual viewing angle through a simple drag-and-drop operation.

[0141] For example, the in-vehicle infotainment system can support assigning different actual viewing positions to the three display sub-areas 201, which can simultaneously meet the viewing needs of different passengers. For instance, the left display sub-area 201 is based on the driver's viewing position, the right display sub-area 201 is based on the passenger's viewing position, and the middle display sub-area 201 is based on the viewing position of the rear passengers.

[0142] This disclosure calibrates the image at seven calibrated viewing angles within the vehicle's three-dimensional space, covering all possible observation angles inside the vehicle. This ensures the clarity and accuracy of the image from different perspectives, thereby improving driving safety and comfort.

[0143] This disclosure establishes a three-dimensional spatial model of distortion correction parameters, allowing car owners to customize distortion correction styles through the system application to meet the viewing needs of different passengers. It enables distortion correction based on any viewing angle, satisfying personalized requirements and improving user experience.

[0144] The in-vehicle infotainment system application disclosed in this publication can also support real-time adjustment of distortion correction parameters to ensure optimal display performance under different driving scenarios. Through high-precision distortion correction, it enhances the visual experience for drivers and passengers, reduces visual fatigue, and improves driving safety.

[0145] Reference Figure 4 , Figure 4 This is a block diagram illustrating an image distortion correction device 400 according to an exemplary embodiment. (Refer to...) Figure 4 The image distortion correction device 400 includes a determining module 401, a correction module 402, and a projection module 403.

[0146] The determining module 401 is configured to determine the image to be projected onto the vehicle windshield by the projection device; The correction module 402 is configured to perform distortion correction on the image through a distortion correction model, wherein the target correction parameters of the distortion correction model are obtained based on the actual viewing position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing positions, which are located at different positions in the three-dimensional space of the vehicle. The projection module 403 is configured to project a distortion-corrected image onto the windshield of the vehicle for display.

[0147] In some possible implementations, the target correction parameters are obtained through a three-dimensional interpolation algorithm based on the actual viewing angle position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing angle positions.

[0148] In some possible implementations, the target correction parameters are obtained in the following manner: For each reference viewpoint position, determine the Euclidean distance between the reference viewpoint position and the actual viewpoint position; Based on the Euclidean distance, determine the weight parameters corresponding to the reference viewpoint position; The target correction parameter is obtained based on the weight parameter and reference correction parameter corresponding to each of the plurality of reference viewpoint positions.

[0149] In some possible implementations, the reference correction parameters corresponding to the reference viewpoint position are obtained in the following manner: When the projection device projects the calibration image to be corrected onto the windshield, for each of the plurality of reference viewpoints, the actual position corresponding to the feature point in the calibration image is determined. Based on the actual and ideal positions of the feature points, the reference correction parameters corresponding to the reference viewpoint position are determined.

[0150] In some possible implementations, in the vehicle's corresponding three-dimensional coordinate system, the multiple reference viewpoint positions include: The origin of the three-dimensional coordinate system; The three axes of the three-dimensional coordinate system correspond to calibration points, and each axis includes at least one calibration point.

[0151] In some possible implementations, the origin of the three-dimensional coordinate system is the geometric center of the vehicle, and each axis includes at least two calibration points, wherein the at least two calibration points corresponding to an axis are distributed on the positive half-axis and the negative half-axis corresponding to the axis.

[0152] In some possible implementations, the image distortion correction device 400 is further configured to: In response to the angle adjustment operation corresponding to the projection device, a new actual angle position is obtained; Based on the reference correction parameters corresponding to the multiple reference viewpoint positions and the new actual viewpoint position, new target correction parameters are determined.

[0153] In some possible implementations, the viewpoint adjustment operation includes at least one of the following: The system obtains the user's movement operation on the vehicle's infotainment display screen regarding the actual viewing position of the projection device. The system detects a change in the eye position of a designated user corresponding to the projection device, where the designated user is a driver or passenger in a designated position within the vehicle. It is determined that the designated user corresponding to the projection device has been switched.

[0154] In some possible implementations, the projection device includes multiple devices, and the display area of ​​the windshield includes multiple display sub-areas, with each projection device corresponding to one display sub-area.

[0155] In some possible implementations, the display area of ​​the windshield extends from one side A-pillar of the vehicle to the other side A-pillar.

[0156] In some possible implementations, before performing distortion correction on the image using a distortion correction model, the image distortion correction device 400 is further configured to: Determine the actual viewing angle positions corresponding to the plurality of projection devices; For each projection device, the target correction parameters corresponding to the projection device are determined based on the actual viewing angle position and the reference correction parameters corresponding to the multiple reference viewing angle positions.

[0157] Regarding the image distortion correction device 400 in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the image distortion correction method, and will not be elaborated here.

[0158] Based on the same inventive concept, this disclosure also provides a vehicle, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to execute the image distortion correction method described in this disclosure.

[0159] Based on the same inventive concept, this disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the image distortion correction method described in this disclosure.

[0160] Based on the same inventive concept, this disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the image distortion correction method described in this disclosure.

[0161] Reference Figure 5 , Figure 5 This is a block diagram illustrating a vehicle 500 according to an exemplary embodiment. For example, vehicle 500 can be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicle. Vehicle 500 can be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle.

[0162] like Figure 5 As shown, vehicle 500 may include various subsystems, such as infotainment system 510, perception system 520, decision control system 530, drive system 540, and computing platform 550. Vehicle 500 may also include more or fewer subsystems, and each subsystem may include multiple components. Furthermore, each subsystem and each component of vehicle 500 can be interconnected via wired or wireless means.

[0163] In some embodiments, the infotainment system 510 may include a communication system, an entertainment system, and a navigation system, etc.

[0164] The perception system 520 may include several sensors for sensing information about the environment surrounding the vehicle 500. For example, the perception system 520 may include a global positioning system (which may be GPS, BeiDou, or other positioning systems), an inertial measurement unit (IMU), lidar, millimeter-wave radar, ultrasonic radar, and a camera device.

[0165] The decision control system 530 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0166] The drive system 540 may include components that provide powered motion to the vehicle 500. In one embodiment, the drive system 540 may include an engine, an energy source, a transmission system, and wheels. The engine may be one or a combination of internal combustion engines, electric motors, and compressed air engines. The engine is capable of converting energy provided by the energy source into mechanical energy.

[0167] Some or all of the functions of vehicle 500 are controlled by computing platform 550. Computing platform 550 may include at least one processor 551 and memory 552, and processor 551 may execute instructions 553 stored in memory 552.

[0168] The processor 551 can be any conventional processor, such as a commercially available CPU. The processor may also include graphics processing units (GPUs), field-programmable gate arrays (FPGAs), systems on chips (SoCs), application-specific integrated circuits (ASICs), or combinations thereof.

[0169] The memory 552 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0170] In addition to instruction 553, memory 552 can also store data, such as road maps, route information, vehicle position, direction, speed, and other data. The data stored in memory 552 can be used by computing platform 550.

[0171] In this embodiment of the disclosure, the processor 551 may execute instructions 553 to complete all or part of the steps of the above-described image distortion correction method.

[0172] In another exemplary embodiment, a computer program product is also provided, the computer program product comprising a computer program executable by a programmable device, the computer program having a code portion for performing the above-described image distortion correction method when executed by the programmable device.

[0173] Some embodiments of this disclosure also provide a chip system, such as Figure 6As shown, the chip system includes at least one processor 601 and at least one interface circuit 602. The processor 601 and the interface circuit 602 are interconnected via lines. For example, the interface circuit 602 can be used to receive signals from other devices (e.g., the memory of an electronic device). As another example, the interface circuit 602 can be used to send signals to other devices (e.g., the processor 601). Exemplarily, the interface circuit 602 can read instructions stored in memory and send those instructions to the processor 601. When the instructions are executed by the processor 601, the image distortion correction device can perform the steps in the above embodiments. Of course, the chip system may also include other discrete components, and some embodiments of this disclosure do not specifically limit this.

[0174] In some embodiments of this disclosure, the interface circuit 602 can acquire data, program instructions, and / or information from the internal storage area of ​​the chip system; it can also acquire data, program instructions, and / or information from outside the chip system.

[0175] Optionally, the chip system may also include a memory for storing necessary computer programs and data.

[0176] Those skilled in the art will also understand that the various illustrative logical blocks and steps listed in the embodiments of this application can be implemented by electronic hardware, computer software, or a combination of both. Whether such functionality is implemented through hardware or software depends on the specific application and the overall system design requirements. Those skilled in the art can implement the described functionality using various methods for each specific application, but such implementation should not be construed as exceeding the scope of protection of the embodiments of this application.

[0177] Furthermore, the term “exemplary” is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as “exemplary” is not necessarily to be construed as advantageous compared to other aspects or designs. Rather, the use of the term “exemplary” is intended to present the concept in a concrete manner. As used herein, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise specified or clear from the context, “X applies A or B” is intended to mean any of the natural inclusive arrangements. That is, “X applies A or B” satisfies any of the foregoing instances if X applies A; X applies B; or both X applies A and B. Additionally, unless otherwise specified or clear from the context to refer to the singular form, the articles “a” and “an” as used in this application and the appended claims are generally understood to mean “one or more.”

[0178] Similarly, although this disclosure has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art upon reading and understanding this specification and the accompanying drawings. This disclosure includes all such modifications and variations and is limited only by the scope of the claims. In particular, with respect to the various functions performed by the components described above (e.g., elements, resources, etc.), unless otherwise indicated, the terminology used to describe such components is intended to correspond to any component (functionally equivalent) that performs the specific function of the described component, even if structurally not equivalent to the disclosed structure. Furthermore, although specific features of this disclosure may have been disclosed with respect to only one of several implementations, such features may be combined with one or more other features of other implementations, as may be desired and advantageous to any given or particular application. Moreover, with regard to the terms “comprising,” “owning,” “having,” “having,” or variations thereof as used in the detailed description or claims, such terms are intended to be inclusive in a manner similar to the term “including.”

[0179] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the appended claims.

[0180] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. An image distortion correction method, characterized in that, include: Determine the image to be projected onto the vehicle's windshield using the projection device; The image is distorted by a distortion correction model, wherein the target correction parameters of the distortion correction model are obtained based on the actual viewing position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing positions, which are located at different positions in the three-dimensional space of the vehicle. The distortion-corrected image is projected onto the vehicle's windshield for display.

2. The method according to claim 1, characterized in that, The target correction parameters are obtained through a three-dimensional interpolation algorithm based on the actual viewing angle position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing angle positions.

3. The method according to claim 1, characterized in that, The target correction parameters are obtained in the following manner: For each reference viewpoint position, determine the Euclidean distance between the reference viewpoint position and the actual viewpoint position; Based on the Euclidean distance, determine the weight parameters corresponding to the reference viewpoint position; The target correction parameter is obtained based on the weight parameter and reference correction parameter corresponding to each of the plurality of reference viewpoint positions.

4. The method according to claim 1, characterized in that, The reference correction parameters corresponding to the reference viewpoint position are obtained in the following manner: When the projection device projects the calibration image to be corrected onto the windshield, for each of the plurality of reference viewpoints, the actual position corresponding to the feature point in the calibration image is determined. Based on the actual and ideal positions of the feature points, the reference correction parameters corresponding to the reference viewpoint position are determined.

5. The method according to claim 1, characterized in that, In the vehicle's corresponding three-dimensional coordinate system, the multiple reference viewpoint positions include: The origin of the three-dimensional coordinate system; The three axes of the three-dimensional coordinate system correspond to calibration points, and each axis includes at least one calibration point.

6. The method according to claim 5, characterized in that, The origin of the three-dimensional coordinate system is the geometric center of the vehicle. Each axis includes at least two calibration points, wherein the at least two calibration points corresponding to an axis are distributed on the positive half-axis and the negative half-axis corresponding to the axis.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: In response to the angle adjustment operation corresponding to the projection device, a new actual angle position is obtained; Based on the reference correction parameters corresponding to the multiple reference viewpoint positions and the new actual viewpoint position, new target correction parameters are determined.

8. The method according to claim 7, characterized in that, The view adjustment operation includes at least one of the following: The system obtains the user's movement operation on the vehicle's infotainment display screen regarding the actual viewing position of the projection device. The system detects a change in the eye position of a designated user corresponding to the projection device, where the designated user is a driver or passenger in a designated position within the vehicle. It is determined that the designated user corresponding to the projection device has been switched.

9. The method according to any one of claims 1-6, characterized in that, The projection device includes multiple devices, and the display area of ​​the windshield includes multiple display sub-areas, with each projection device corresponding to one display sub-area.

10. The method according to claim 9, characterized in that, The display area of ​​the windshield extends from one A-pillar of the vehicle to the other A-pillar.

11. The method according to claim 9, characterized in that, Before performing distortion correction on the image using a distortion correction model, the method further includes: Determine the actual viewing angle positions corresponding to the plurality of projection devices; For each projection device, the target correction parameters corresponding to the projection device are determined based on the actual viewing angle position and the reference correction parameters corresponding to the multiple reference viewing angle positions.

12. An image distortion correction device, characterized in that, include: The determination module is configured to determine the image to be projected onto the vehicle's windshield by the projection device; The correction module is configured to perform distortion correction on the image using a distortion correction model, wherein the target correction parameters of the distortion correction model are obtained based on the actual viewing angle position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing angle positions, which are located at different positions in the three-dimensional space of the vehicle. The projection module is configured to project a distortion-corrected image onto the vehicle's windshield for display.

13. The apparatus according to claim 12, characterized in that, The target correction parameters are obtained through a three-dimensional interpolation algorithm based on the actual viewing angle position corresponding to the projection device and the reference correction parameters corresponding to multiple reference viewing angle positions.

14. The apparatus according to claim 12, characterized in that, The target correction parameters are obtained in the following manner: For each reference viewpoint position, determine the Euclidean distance between the reference viewpoint position and the actual viewpoint position; Based on the Euclidean distance, determine the weight parameters corresponding to the reference viewpoint position; The target correction parameter is obtained based on the weight parameter and reference correction parameter corresponding to each of the plurality of reference viewpoint positions.

15. The apparatus according to claim 12, characterized in that, The reference correction parameters corresponding to the reference viewpoint position are obtained in the following manner: When the projection device projects the calibration image to be corrected onto the windshield, for each of the plurality of reference viewpoints, the actual position corresponding to the feature point in the calibration image is determined. Based on the actual and ideal positions of the feature points, the reference correction parameters corresponding to the reference viewpoint position are determined.

16. A vehicle, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to perform the image distortion correction method according to any one of claims 1-11.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the image distortion correction method according to any one of claims 1-11.

18. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the image distortion correction method according to any one of claims 1-11.

Images