Image dynamic correction method, device and system and computer storage medium

Abstract

The invention provides an image dynamic correction method, device and system and a computer storage medium, the device comprises a memory and a processor, and the memory is configured to store the current position of the eyes of an observer in a three-dimensional space of a vehicle; and the processor is configured to perform real-time and regional distortion correction on an image to be displayed based on the current position and a pre-established mapping relation representing a relation between coordinates in the three-dimensional space and image distortion.

Description

Technical Field

[0001] This disclosure relates to the field of image processing technology, and in particular to an image dynamic correction method, apparatus, system, and computer storage medium. Background Technology

[0002] Head-up display (HUD) systems are projection technologies widely used in the automotive industry. They project important driving information, such as vehicle speed, navigation, and warnings, onto the windshield in front of the driver, allowing the driver to access information without taking their eyes off the road, thus greatly improving driving safety.

[0003] However, existing HUD technology still has significant limitations in image quality. The HUD system is a complex optical system, and its imaging effect is affected by a variety of factors, such as the shape of the windshield and the shape of the imaging system inside the HUD (such as the mirrors), which can easily cause image distortion, especially in the edge areas of the image.

[0004] Existing technologies typically employ static compensation methods, while the driver's actual eye position is in a three-dimensional space and is dynamically changing. Drivers of different heights and sitting postures will have different initial eye positions. Even for the same driver, head movement is inevitable during driving. Existing static compensation methods lack sufficient accuracy. Summary of the Invention

[0005] This disclosure provides an image dynamic correction method, apparatus, system, and computer storage medium; it can solve the technical problem of insufficient compensation accuracy in existing static compensation correction methods.

[0006] The technical solution disclosed herein is implemented as follows: In a first aspect, this disclosure provides an image dynamic correction apparatus, comprising: The memory is configured to store the current position of the observer's eye in the vehicle's three-dimensional space; The processor is configured to perform real-time, region-by-region distortion correction on the image to be displayed based on the current position and a pre-established mapping relationship that characterizes the relationship between coordinates and image distortion in the three-dimensional space.

[0007] Secondly, this disclosure provides an image dynamic correction system, including: Projection components; At least one vehicle-mounted camera; And the apparatus as described in the first aspect.

[0008] Thirdly, this disclosure provides an image dynamic correction method, including: Real-time acquisition of the observer's current eye position in the vehicle's three-dimensional space; Based on the current position and a pre-established mapping relationship that characterizes the relationship between coordinates and image distortion in the three-dimensional space, real-time, region-by-region distortion correction is performed on the image to be displayed.

[0009] Fourthly, this disclosure provides a computer-readable storage medium storing a computer program configured to implement the image dynamic correction method described in the third aspect when executed by a processor.

[0010] This disclosure provides an image dynamic correction method, apparatus, system, and computer storage medium. By configuring a processor, the current position of the observer stored in the memory is dynamically associated with a pre-established mapping relationship characterizing the relationship between coordinates in the three-dimensional space and image distortion. This apparatus can apply distortion correction in real time and in different regions. This allows correction to move beyond a few fixed two-dimensional design points and follow the observer's movements in three-dimensional space in real time, ensuring that the HUD image maintains stability and low distortion under dynamic viewing conditions, significantly improving imaging quality and viewing comfort. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the structure of an image dynamic correction system provided in this disclosure.

[0012] Figure 2 This is a schematic diagram of the structure of a head-up display device provided in this disclosure.

[0013] Figure 3 This is a schematic diagram of the structure of an image dynamic correction device provided in this disclosure.

[0014] Figure 4 This is a flowchart for establishing a mapping relationship provided in this disclosure.

[0015] Figure 5 This is a schematic diagram of a data acquisition plane for establishing a mapping relationship, as provided in this disclosure.

[0016] Figure 6 This is a schematic diagram of a correction sub-region provided in this disclosure.

[0017] Figure 7 A flowchart of the image dynamic correction method provided in this disclosure. Detailed Implementation

[0018] The technical solutions in this disclosure will now be clearly and completely described with reference to the accompanying drawings.

[0019] The fundamental flaw in existing static compensation correction methods lies in their static and two-dimensional nature. First, traditional sampling and calibration only collect data at the eye-point design plane, ignoring the observer's eye movement in the forward-backward direction (i.e., the depth direction) perpendicular to that plane. Second, the calibration data is static, meaning that one set of compensation data corresponds to only one fixed observation point.

[0020] In real-world driving scenarios, the driver's actual eye position is located in a three-dimensional space and is dynamically changing. Drivers of different heights and sitting postures will have different initial eye positions. Even for the same driver, their head will inevitably move during driving.

[0021] When the driver's actual eye position deviates from the preset calibration positions, the poor edge imaging becomes apparent due to the nonlinear effects of complex optical factors such as the windshield surface shape, and the original static compensation data cannot compensate for this. In other words, existing technology attempts to solve a dynamic, three-dimensional observation problem with a static, two-dimensional correction scheme. This fundamental mismatch leads to poor viewing comfort and imaging stability of the HUD in real driving environments.

[0022] Therefore, how to overcome the limitations of static, two-dimensional plane correction and achieve a clear and low-distortion imaging effect of HUD image when the observer's eyes move dynamically in the three-dimensional eye box space is a technical problem that urgently needs to be solved in this field.

[0023] Based on this, this disclosure first proposes an image dynamic correction device, which can be applied in, for example... Figure 1 In the illustrated image motion correction system, image motion correction system 10 is shown with an exemplary vehicle 11. Although a bus is illustrated, it should be understood that vehicle 11 can be any type of vehicle without departing from the scope of this disclosure. System 10 generally includes image motion correction device 12, vehicle sensor 13, head-up display device 14, and human-machine interface (HMI) 15.

[0024] The image dynamic correction device 12 is used to display content related to the current vehicle and its environment on the projected component 16 of the vehicle 11. The image dynamic correction device 12 includes at least one processor 121 and a memory 122, which may be a computer-readable storage device or medium. The processor 121 may be a custom or commercially available processor, a central processing unit (CPU), a graphics processing unit (GPU), an auxiliary processor among several processors associated with the image dynamic correction device 12, a semiconductor-based microprocessor (in the form of a microchip or chipset), a macroprocessor, a combination thereof, or generally a device for executing instructions. The computer-readable storage device or medium may include, for example, volatile and non-volatile memory in read-only memory (ROM), random access memory (RAM), and keep-alive memory (KAM). KAM is a persistent or non-volatile memory that can be used to store various operational variables when the processor 121 is powered off. Computer-readable storage devices or media may be implemented using multiple storage devices (such as PROM (Programmable Read-Only Memory), ePROM (Electrically Powered PROM), EEPROM (Electrically Erasable PROM), flash memory, or other electrical, magnetic, optical, or combined storage devices capable of storing data), some of which represent executable instructions used by the image motion correction device 12 to control various systems of the vehicle 11. The image motion correction device 12 may also consist of multiple controllers that are electrically in communication with each other.

[0025] The image motion correction device 12 communicates electrically with the vehicle sensors 13, the head-up display 14, and the HMI 15. Electrical communication can be established using, for example, a CAN bus, Wi-Fi network, cellular data network, etc. It should be understood that various other wired and wireless technologies and communication protocols used for communicating with the image motion correction device 12 are within the scope of this disclosure.

[0026] Vehicle sensor 13 is used to acquire information about the environment surrounding vehicle 11, i.e., to acquire external environmental parameters of vehicle 11. In an exemplary embodiment, vehicle sensor 13 includes an external camera 131, a vehicle communication system 132, and an electronic distance sensor 133. It should be understood that, without departing from the scope of this disclosure, vehicle sensor 13 may include additional sensors for determining characteristics of vehicle 11, such as vehicle speed, road curvature, and / or vehicle steering. As discussed above, vehicle sensor 13 is in electrical communication with image dynamic correction device 12.

[0027] An external camera 131 is used to capture images and / or videos of the environment surrounding the vehicle 11. In an exemplary embodiment, the external camera 131 is a photographic and / or video camera positioned to observe the environment in front of the vehicle 11. In one example, the external camera 131 is fixed inside the vehicle 11 (e.g., in the roof lining of the vehicle 11) and has a field of view through the projection member 16. In another example, the external camera 131 is fixed outside the vehicle 11, for example, on the roof of the vehicle 11, and has a view of the environment in front of the vehicle 11. It should be understood that cameras with various sensor types, including, for example, charge-coupled device (CCD) sensors, complementary metal-oxide-semiconductor (CMOS) sensors, and / or high dynamic range (HDR) sensors, are all within the scope of this disclosure. Furthermore, cameras with various lens types, including, for example, wide-angle lenses and / or narrow-angle lenses, are also within the scope of this disclosure.

[0028] The image motion correction device 12 uses the vehicle communication system 132 to communicate with other systems outside the vehicle 11. For example, the vehicle communication system 132 includes the ability to communicate with vehicles, infrastructure, remote call centers, and / or personal devices. The vehicle communication system 132 may include one or more antennas and / or communication transceivers for receiving and / or transmitting signals, such as Coordination Sensing Messages (CSM). The vehicle communication system 132 is configured to wirelessly transmit information between the vehicle 11 and another vehicle. Furthermore, the vehicle communication system 132 is configured to wirelessly transmit information between the vehicle 11 and infrastructure or other vehicles.

[0029] Electronic distance sensor 133 is used to determine the range (i.e., distance) between vehicle 11 and objects in the environment surrounding the vehicle. Electronic distance sensor 133 may utilize electromagnetic waves (e.g., radar), sound waves (e.g., ultrasound), and / or light (e.g., lidar) to determine the distance. Figure 1 In the exemplary embodiments shown, the electronic ranging sensor 133 is a lidar sensor. It should be understood that embodiments in which the electronic ranging sensor 133 includes radar sensors, ultrasonic sensors, lidar sensors, and / or other sensors configured to determine range (i.e., distance) fall within the scope of this disclosure.

[0030] refer to Figure 2This diagram illustrates a system diagram of a head-up display device 14 used by an exemplary occupant 21. Within the scope of this disclosure, in a non-limiting example, occupant 21 includes the driver, passengers, and / or any other person in the vehicle 11. The head-up display device 14 is used to display projected images (i.e., notification symbols providing visual information to occupant 21) on a projection component 16 of the vehicle 11. The head-up display device 14 includes a projection component 141 and an occupant status acquisition device 142, which may be an in-vehicle camera. As discussed above, the head-up display device 14 is in electrical communication with an image motion correction device 12.

[0031] Projection component 141 is used to project a projected image onto a projected component 16 of the vehicle 11, typically the windshield. It should be understood that various devices designed for projecting images, including, for example, optical collimators, laser projectors, digital light projectors (DLP), etc., are within the scope of this disclosure.

[0032] The occupant status acquisition device 142 is used to determine the position of the occupant 21 in the vehicle 11 and the driver's visual state parameters. For example, the occupant status acquisition device 142 can track the position of the occupant 21's head 211 or eyes 212, as well as the driver's facial expressions. The position and visual state parameters of the occupant 21 in the vehicle 11 obtained from the occupant status acquisition device 142 are used to locate a projected image on the windshield of the vehicle 11 and adjust the display color of the projected image. In an exemplary embodiment, the occupant status acquisition device 142 is one or more cameras disposed in the vehicle 11.

[0033] To operate the head-up display 14, the processor 121 in the image dynamic correction device 12 may include multiple software modules, including a system manager 144. During operation of the system 10, the system manager 144 receives at least a first input 24, a second input 23, and a third input 22. The first input 24 indicates the location of the vehicle 11 in space (i.e., the geographic location of the vehicle 11), the second input 23 indicates the location of the vehicle occupant 21 within the vehicle 11 and visual state parameters (e.g., the position of the occupant 21's eyes and / or head within the vehicle 11), and the third input 22 is data related to the expected lighting state of at least one indicator of a distant vehicle, which will be discussed in more detail below. The first input 24 may include data such as GNSS data (e.g., GPS data), vehicle speed, road curvature, and vehicle steering, and this data is collected from vehicle sensors 13. The second input 23 is received from occupant state acquisition device 142. The third input 22 is vehicle external environment parameters concerning the distant vehicle in the environment surrounding the vehicle 11. System manager 144 is configured to determine (e.g., calculate) the type, size, shape, and color of the projected image to be displayed using projection component 141 based on a first input 24 (i.e., vehicle position in the environment), a second input 23 (e.g., the position of the eyes 212 and / or head 211 of the occupant 21 in vehicle 11), and a third input 22. System manager 144 instructs image engine 143 to display the projected image using projection component 141. Image engine 143 is a software module or integrated circuit of projection component 141 or image dynamic correction device 12. Image engine 143 displays the projected image on the projected component 16 of vehicle 11 using projection component 141 based on the type, size, shape, and color of the projected image determined by system manager 144. When the head-up display device is an AR-HUD, the projected image is projected by projection component 141 onto the projected component 16 to display the projected image along the road surface 26.

[0034] In some exemplary embodiments of this disclosure, the image dynamic correction device 12 and the head-up display device 14 described above are combined to constitute the head-up display device of this disclosure.

[0035] In some examples of this disclosure, refer to Figure 3 The image dynamic correction device 12 may include a memory 122 and a processor 121. The memory 122 stores the current position of the observer's eye in the three-dimensional space of the vehicle. The processor 121 is configured to perform real-time, regional distortion correction on the image to be displayed based on the current position and a pre-established mapping relationship that characterizes the relationship between coordinates and image distortion in the three-dimensional space.

[0036] In some examples of this disclosure, memory 122 is configured to store the current position of an observer's (occupant 21's) eyes in the vehicle's three-dimensional space. This current position data is dynamically updated. For example, processor 121 calculates the current position, such as coordinates in the vehicle coordinate system, in real time based on input from occupant status acquisition device 142, and writes this coordinate value into a cache (such as RAM) of memory 122 for immediate retrieval in subsequent correction calculations.

[0037] Furthermore, memory 122 is also configured to store a key data structure: a pre-established mapping relationship characterizing the relationship between coordinates in the three-dimensional space and image distortion. This mapping relationship can be generated through measurement before the vehicle leaves the factory or during system calibration.

[0038] Mapping relationships can be stored in various forms, such as look-up tables (LUTs), function models, and neural network weights.

[0039] For a look-up table (LUT), memory 122 can store a large database or a multidimensional array. The index of the look-up table is a discretized three-dimensional spatial coordinate, such as (X, Y, Z) grid points in millimeters, while the contents of the look-up table are the distortion correction parameters corresponding to those coordinates.

[0040] For the function model, memory 122 can store a set of parameters of the function model (e.g., higher-order polynomials, B-spline functions, etc.). After obtaining the current position (X, Y, Z), processor 121 uses these coordinates as input to the function model and calculates the required distortion correction parameters in real time.

[0041] For the neural network weights, this mapping relationship can be implicitly represented by a trained neural network, such as a deformable Convolutional Neural Network (CNN). In this case, memory 122 stores the model structure and training weights of the deformable CNN.

[0042] Processor 121 is configured to execute instructions stored in memory 122. In this embodiment, the core configuration of processor 121 is to perform real-time, region-by-region distortion correction on the image to be displayed based on the current position and a pre-established mapping relationship that characterizes the relationship between coordinates in three-dimensional space and image distortion.

[0043] Specifically, the processor 121 retrieves the current position data from the memory 122 and uses the current position to query the mapping relationship stored in the memory 122. By querying the mapping relationship, the processor 121 reads the imaging data at that position, thus obtaining a set of distortion correction parameters corresponding to the current position.

[0044] The processor 121 acquires the raw frame data of the image to be displayed, generated by the system manager 144 or the image engine 143. It then processes this frame image using the distortion correction parameters it just acquired.

[0045] In some example implementations, the real-time nature of this disclosure is performed continuously and repeatedly over a very short period of time (e.g., 30, 60, or more times per second). When device 142 detects that the occupant 21's head 211 has moved, causing a change in its current position, processor 121 rapidly calculates new correction parameters in the next frame or several frames and applies them to the displayed image, thereby achieving real-time dynamic compensation.

[0046] In this disclosure, "regionalization" refers to the fact that the distortion correction parameters applied by the processor 121 are different for different regions of the image. Optical distortion (especially distortion caused by the windshield type 16) is typically spatially nonlinear, meaning it varies greatly from location to location. Therefore, the processor 121 does not apply a uniform transformation to the entire image, but instead divides the image into multiple regions and applies its own specific correction parameters to each region.

[0047] In some examples, the mapping is established based on distortion data collected on a reference plane of the preset eye box, and on at least one first spatial plane in front of the reference plane and at least one second spatial plane behind the reference plane.

[0048] Specifically, this setup process is typically completed in a laboratory using specialized equipment, such as optical platforms, multi-axis displacement platforms, and high-speed cameras, as described above. Figure 4 The establishment process may include steps S410 to S440.

[0049] In step S410, the platform is built.

[0050] Specifically, an imaging device (such as a high-speed camera) is mounted on a multi-axis displacement platform. The HUD system to be tested is set in front, including the projected component (windshield) 16 and the projection unit 141.

[0051] Then execute step S420 to define the plane.

[0052] In some examples disclosed herein, a reference plane may first be defined within the eyebox space. This is typically the theoretical eyepoint plane in the HUD.

[0053] Then, step S430 is executed for three-dimensional sampling.

[0054] The multi-axis displacement platform is controlled to move the imaging device point by point within a reference plane in millimeter (mm) increments and capture test images (e.g., a standard grid) displayed on the HUD, while saving distortion data.

[0055] Reference Figure 5 The control platform moves the imaging device to at least one first spatial plane 502 (e.g., Z+10mm, Z+20mm) in front of the reference plane 501. On this plane, millimeter-level in-plane scanning and data acquisition are performed again. The control platform moves the imaging device to at least one second spatial plane 503 (e.g., Z-10mm, Z-20mm) behind the reference plane. On this plane, millimeter-level in-plane scanning and data acquisition are performed again.

[0056] After sampling is completed, step S440 can be executed to establish a mapping relationship.

[0057] After step S430, a large number of coordinate points and their corresponding image distortion data are obtained. These data are processed and integrated to establish a functional relationship or lookup table between the stereoscopic spatial coordinates of each point during camera movement and the observed changes (distortion) in the HUD image, and then stored in memory 122.

[0058] In some example embodiments of this disclosure, the current position is obtained by analyzing the acquired image of the observer (occupant 21). This analysis process can be performed by processor 121 (or a dedicated coprocessor within it). Processor 121 runs one or more artificial intelligence (AI) algorithms to perform big data AI analysis and identification on the video stream from occupant status acquisition device 142.

[0059] This analysis process may include, but is not limited to, face detection, key point localization, pupil localization, tracking and filtering, and 3D solution.

[0060] Specifically, the facial region of occupant 21 is located in the image frame. Facial keypoint models are used to locate the contour edges of the eyes 212, nose, mouth, etc. The eye image is converted to grayscale, contrast is enhanced, and Gaussian filtering is used to remove noise; edge detection is used to refine the eye contour; the Hough Circle Transform algorithm can be used to accurately locate the center position coordinates of the pupil (i.e., the pixel coordinates on the two-dimensional image plane). Since the video is continuous, to improve stability and anti-interference capabilities, the processor 121 can use Kalman filtering to continuously track and predict the center position coordinates. This prevents drastic coordinate jumps caused by momentary occlusion or blinking. After obtaining the center position coordinates of the pupil on the two-dimensional image, the spatial position of the human eye can be calculated and determined based on anthropometry.

[0061] In some examples, if the occupant status acquisition device 142 includes a camera, the current position can be determined based on the image ratio between fixed parts in the environment where the observer (occupant 21) is located and the pupil of the observer's eye.

[0062] A fixed component refers to an object inside the vehicle 11 whose spatial position relative to the occupant status acquisition device 142 is fixed or within a preset range (e.g., seat adjustment range). For example, the fixed component is the driver's headrest or the edge of the seat back. The selection of the fixed component can also be customized based on user needs, which will not be elaborated in this disclosure.

[0063] The specific implementation process may include adjusting the seats to a fixed position uniformly when the vehicle leaves the factory. At this time, the processor 121 (or calibration device) measures and records the actual width of the headrest and the fixed distance between the headrest and the occupant status acquisition device 142. The processor 121 establishes a correspondence, that is, establishes a scale between the image pixel size and the actual spatial distance. In real-time operation, the processor 121 simultaneously detects the edges of the headrest and backrest and the pupils of the human eyes. The processor 121 calculates the pixel position difference of the pupil relative to the headrest in the current image frame. The spatial position of the eyes is calculated based on the ratio of the pupil position to the headrest. For example, if the driver's head (and its pupils) is detected to have become larger in the image, it means that the driver's head has moved forward relative to the headrest (whose size in the image is fixed). Through this change in scale, the processor 121 can deduce the amount of change in depth of the driver's head (pupils), thereby determining its current position.

[0064] In some examples, if the occupant status acquisition device 142 includes multiple cameras, the current position can be determined by binocular ranging based on images of the observer acquired from multiple different locations.

[0065] Specifically, for example, when the occupant status acquisition device 142 includes two cameras, namely a first camera and a second camera, the positions of the first camera and the second camera are precisely calibrated at the time the vehicle leaves the factory. The baseline distance between the first camera and the second camera, as well as their precise positions and orientations relative to the vehicle coordinate system, are determined. The processor 121 locates the first coordinates of the pupil in the image from the first camera and locates the second coordinates of the same pupil in the image from the second camera.

[0066] Because of the parallax between the first and second coordinates, the processor 121 can use the principle of triangulation to directly calculate the three-dimensional spatial coordinates of the pupil point relative to the dual-camera system, i.e., the current position.

[0067] In some exemplary embodiments of this disclosure, such as Figure 6 As shown, the display area 600 of the HUD is divided into multiple correction sub-areas 601.

[0068] The division of the correction sub-region 601 can be determined based on the imaging change trend reflected in the mapping relationship. During the calibration process, after analyzing the 3D distortion data, the processor can find that some regions show very little change in image distortion and the image is very stable when the eye point moves; while the distortion of some regions changes drastically with eye point movement. Therefore, the regional distortion correction performed by the processor 121 can specifically include dividing the display area 600 into multiple correction sub-regions 601 based on the aforementioned imaging change trend. When performing correction, the processor 121 applies different correction algorithms or parameter sets to each correction sub-region 601. For example, for image regions with large horizontal deviations, the UI adjusts the number of line width pixels; for image regions with large vertical deviations, the overall height of the UI is adjusted.

[0069] Furthermore, this embodiment can also adjust the layout of interface elements displayed in the multiple correction sub-regions 601 according to their distortion characteristics. Specifically, regions with first distortion characteristics (e.g., small distortion, gentle imaging change trend) can be used to display first type elements, such as key, large-sized images or icons. Regions with second distortion characteristics (e.g., small variation only in the left-right direction) can be used to display second type elements, such as narrow images or text lines. Regions with third distortion characteristics (e.g., small variation only in the up-down direction) can be used to display third type elements, such as long-sized images or lists. Pixel space is reserved between regions to prevent overlap or tearing during correction. This embodiment cleverly combines underlying optical distortion issues with upper-level UI / UX design (product experience), achieving optimal visual effects through a combination of design avoidance and technical correction.

[0070] Furthermore, this disclosure also provides an image dynamic correction method, referring to... Figure 7 The image dynamic correction method may include steps S710 to S720.

[0071] In step S710, the current position of the observer's eye in the three-dimensional space of the vehicle is acquired in real time.

[0072] In step S720, based on the current position and a pre-established mapping relationship that characterizes the relationship between coordinates and image distortion in the three-dimensional space, real-time, region-based distortion correction is performed on the image to be displayed.

[0073] It should be noted that the specific details of the image dynamic correction method can be found in the above description of the image dynamic correction device, and will not be repeated here.

[0074] This disclosure also provides a computer-readable storage medium storing at least one instruction that is executed by a processor to implement the image dynamic correction method as described in the above embodiments.

[0075] This disclosure also provides a computer program product including computer instructions stored in a computer-readable storage medium; a processor of a computing device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computing device to perform the image dynamic correction method described in the above embodiments.

[0076] Those skilled in the art will recognize that the functions described in this disclosure in one or more of the examples above can be implemented using hardware, software, firmware, or any combination thereof. When implemented in software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transfer of a computer program from one place to another. Storage media can be any available medium accessible to a general-purpose or special-purpose computer.

[0077] It should be noted that the technical solutions described in this disclosure can be combined arbitrarily as long as they do not conflict.

[0078] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An image dynamic correction device, characterized in that, The device comprises: a memory configured to store a current position of an observer's eyes in a three-dimensional space of a vehicle; a processor configured to perform real-time and regional distortion correction on an image to be displayed based on the current position and a pre-established mapping relationship representing a relationship between coordinates in the three-dimensional space and image distortion.

2. The image dynamic correction apparatus according to claim 1, characterized by The mapping relationship is established based on distortion data collected on a reference plane of a preset eyebox, at least one first spatial plane in front of the reference plane, and at least one second spatial plane behind the reference plane.

3. The image dynamic correction apparatus according to claim 1 or 2, characterized by, The regional distortion correction comprises: dividing a display region into a plurality of correction sub-regions based on imaging variation trends reflected in the mapping relationship; adjusting a layout of interface elements displayed in the correction sub-regions according to distortion characteristics of the correction sub-regions; wherein a region with a first distortion characteristic is used to display a first type of element.

4. The image dynamic correction apparatus according to claim 1, characterized by The current position is obtained by analyzing an image of the observer collected.

5. The image dynamic correction apparatus according to claim 4, characterized in that, The current position is determined based on an image proportion relationship between a fixed component in an environment where the observer is located and the observer's eye pupils.

6. The image dynamic correction apparatus according to claim 4, wherein The current position is determined by binocular distance measurement based on images of the observer collected at a plurality of different positions.

7. An image dynamic correction system, characterized in that, The device comprises: a projection component; at least one vehicle-mounted camera; and the device of any one of claims 1 to 6.

8. An image dynamic correction method, characterized in that, The method comprises: collecting a current position of an observer's eyes in a three-dimensional space of a vehicle in real time; performing real-time and regional distortion correction on an image to be displayed based on the current position and a pre-established mapping relationship representing a relationship between coordinates in the three-dimensional space and image distortion.

9. The image dynamic correction method of claim 8, wherein, The regional distortion correction comprises: dividing a display region into a plurality of correction sub-regions based on imaging variation trends reflected in the mapping relationship; adjusting a layout of interface elements displayed in the correction sub-regions according to distortion characteristics of the correction sub-regions; wherein a region with a first distortion characteristic is used to display a first type of element.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program configured to implement the image dynamic correction method of claim 8 or 9 when executed by a processor.

Images