A method, system, device and medium for distortion correction of in-vehicle fogging imaging

Abstract

The present application relates to a kind of vehicle-mounted atomization imaging distortion correction method, system, equipment and medium, wherein, method includes: response in-vehicle space imaging trigger signal, determine target imaging area in cabin;Target imaging area is supplied with atomization particle and forms constraint airflow, so that atomization particle forms dynamic imaging medium with predetermined thickness;The spatial state data of dynamic imaging medium and observer viewpoint information are collected, real-time imaging surface model is established and geometric distortion variable is determined;According to geometric distortion variable, the image to be projected is carried out reverse geometric pre-distortion, and compensation projection image that matches the current form of dynamic imaging medium is generated;Again, based on viewpoint information, the compensation projection image is carried out viewpoint compensation, and is projected to dynamic imaging medium;In the projection process, spatial state data and viewpoint information are cyclically updated, to adjust compensation projection image and viewpoint compensation.Therefore, the present application improves the geometric stability of projection image, display continuity and vehicle-mounted adaptability.

Description

Technical Field

[0001] This invention relates to the field of vehicle display technology, and in particular to a distortion correction method, system, device, and medium for vehicle-mounted fogging imaging. Background Technology

[0002] With the development of smart cockpit technology, in-vehicle display methods are gradually expanding from traditional screen displays to head-up displays, augmented reality displays, and spatial imaging. Existing in-vehicle head-up displays typically use the windshield as a reflective medium, with an image generation unit projecting a virtual image; in-vehicle transparent displays usually rely on side windows, partition glass, or transparent display films to carry the image. Since the above solutions all rely on fixed physical media such as windshields, windows, or screens, their imaging area and display position are usually limited by the vehicle structure, glass curvature, and installation location, making it difficult to form flexibly arranged spatial images in non-fixed areas inside the cockpit.

[0003] Fog projection or water mist projection utilizes a misting medium to project images into space, but this technology is mostly used in stage, exhibition, or outdoor settings. Because the misting medium differs from a rigid screen or fixed screen, its imaging area is often difficult to maintain a stable planar state; when the thickness, shape, or local concentration of the imaging medium changes, the projected image is prone to problems such as jitter, blurring, local stretching, or geometric distortion. Therefore, traditional projection correction methods designed for fixed screens, glass, or regular curved surfaces are difficult to adapt to the dynamically changing imaging medium in vehicle-mounted misting imaging.

[0004] Furthermore, within the vehicle cabin, the observer's position and posture may change depending on their seating position, altering their observation direction and distance relative to the imaging area. Since the display state of the projected image on the atomized medium is affected by the viewpoint, changes in the observer's viewpoint can lead to scaling changes, positional shifts, or parallax distortion. Consequently, existing projection methods struggle to maintain consistent image geometric stability in automotive atomized imaging scenarios. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a distortion correction method, system, device and medium for vehicle-mounted fog imaging, which solves the technical problem that the projected image is difficult to maintain geometric stability due to the spatial state of the dynamic imaging medium and the change of the observer's viewpoint in vehicle-mounted fog imaging scenarios.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0009] In a first aspect, embodiments of the present invention provide a distortion correction method for vehicle-mounted fogging imaging, comprising:

[0010] In response to the in-vehicle space imaging trigger signal, the target imaging area inside the cabin is determined;

[0011] Atomized particles are supplied to the target imaging area through the atomization generating component, and a constraining airflow is formed around the target imaging area through the airflow regulating component to constrain the atomized particles, so that the atomized particles form a dynamic imaging medium with a predetermined thickness in the target imaging area.

[0012] Collect spatial state data of the dynamic imaging medium and obtain the viewpoint information of the observer in the cockpit relative to the target imaging area;

[0013] A real-time imaging surface model of the dynamic imaging medium is established based on spatial state data, and the geometric distortion of the image to be projected onto the dynamic imaging medium is determined based on the real-time imaging surface model.

[0014] Based on the geometric distortion, reverse geometric pre-distortion processing is performed on the image to be projected to generate a compensated projection image;

[0015] The viewpoint compensation is performed on the compensated projection image based on the viewpoint information, and the viewpoint-compensated compensated projection image is then projected onto the dynamic imaging medium.

[0016] During the projection process, the spatial state data and viewpoint information are updated cyclically, and the projected image is adjusted and compensated based on the updated spatial state data and viewpoint information, as well as viewpoint compensation is performed.

[0017] Optionally, atomized particles are supplied to the target imaging region via an atomization generating component, and a constraining airflow is formed around the target imaging region via an airflow regulating component to constrain the atomized particles, thereby forming a dynamic imaging medium with a predetermined thickness within the target imaging region, including:

[0018] Based on the position and size of the target imaging area in the cockpit coordinate system, determine the supply direction and amount of atomized particles, as well as the target thickness of the dynamic imaging medium.

[0019] The ultrasonic transducer array in the atomization generation component is controlled to vibrate, so as to break the atomized liquid into micron-sized atomized particles and output the atomized particles to the target imaging area according to the supply direction and supply amount.

[0020] Based on the target thickness of the dynamic imaging medium, the airflow adjustment component is controlled to output constrained airflow around the target imaging area.

[0021] By adjusting at least one of the airflow flow rate, airflow velocity, and airflow pressure difference of the constrained airflow through the airflow adjustment component, an airflow boundary is formed around the target imaging area. The atomized particles output according to the supply direction and supply amount are constrained within the target imaging area through the airflow boundary, so that the atomized particles are distributed along the preset imaging surface to form a dynamic imaging medium with a predetermined thickness.

[0022] Optionally, spatial state data of the dynamic imaging medium is acquired, and viewpoint information of the observer in the cockpit relative to the target imaging area is obtained, including:

[0023] Before the dynamic imaging medium is formed, the space sensing component is used to perform a global scan of the cabin interior space to obtain three-dimensional spatial reference data of the cabin interior.

[0024] After the dynamic imaging medium is formed, the spatial sensing component is used to perform a local scan of the target imaging area to obtain local spatial data of the dynamic imaging medium at multiple sampling times.

[0025] By performing coordinate registration between local spatial data and three-dimensional spatial reference data, spatial state data of the dynamic imaging medium relative to the cockpit coordinate system is obtained. The spatial state data includes the three-dimensional morphological data and density distribution data of the dynamic imaging medium.

[0026] The cockpit monitoring component is used to acquire the eye position data of the observer in the cockpit. Based on the eye position data and the position of the target imaging area in the cockpit coordinate system, the viewpoint information of the observer relative to the target imaging area is determined. The viewpoint information includes the three-dimensional position of the observer's eyes in the cockpit coordinate system and the direction of the observer's line of sight relative to the target imaging area.

[0027] Optionally, a real-time imaging surface model of the dynamic imaging medium is established based on spatial state data, and the geometric distortion of the image to be projected onto the dynamic imaging medium is determined according to the real-time imaging surface model, including:

[0028] Extract multiple spatial sampling points of the dynamic imaging medium within the target imaging area from the three-dimensional morphological data in the spatial state data;

[0029] A real-time imaging surface model consisting of multiple local imaging units is generated by fitting a surface based on multiple spatial sampling points. Each local imaging unit includes the spatial coordinates, surface depth, and normal vector of the corresponding spatial sampling point.

[0030] Based on the spatial coordinates, surface depth, and normal vector of each local imaging unit, determine the projection point when the image to be projected is projected onto each local imaging unit;

[0031] The image to be projected is divided into multiple image grids, and the target grid position of each image grid is determined according to the preset ideal imaging surface;

[0032] Establish a projection mapping relationship between the target grid position of each image grid and the projection landing point of the corresponding local imaging unit, and determine at least one of the positional deviation, scale deviation, and perspective deviation of the projection landing point relative to the target grid position based on the projection mapping relationship, so as to serve as the geometric distortion variable of each image grid.

[0033] Optionally, the image to be projected is subjected to inverse geometric pre-distortion processing based on the geometric distortion to generate a compensated projection image, including:

[0034] Based on the geometric distortion of each image grid, determine the inverse compensation amount used to offset the geometric distortion, wherein the inverse compensation amount includes at least one of inverse displacement, inverse scaling, and inverse perspective transformation.

[0035] The vertex correction amount of each image grid is determined based on the reverse compensation amount, and the vertex position of each image grid is adjusted in reverse according to the vertex correction amount, so that the grid boundary of each image grid is deformed as the vertex position is adjusted, thereby generating a pre-distorted image that matches the current spatial morphology of the dynamic imaging medium.

[0036] Based on the density distribution data in the spatial state data, the brightness compensation amount corresponding to different image regions in the pre-distorted image is determined.

[0037] The brightness gain of the corresponding image region in the pre-distorted image is adjusted according to the brightness compensation amount to generate a compensated projection image.

[0038] Optionally, viewpoint compensation is performed on the compensated projection image based on viewpoint information, and the viewpoint-compensated compensated projection image is projected onto a dynamic imaging medium, including:

[0039] Determine the three-dimensional position of the observer's eye in the cockpit coordinate system and the direction of the observer's line of sight relative to the target imaging area from the viewpoint information;

[0040] The observer's observation ray relative to the dynamic imaging medium is determined based on the three-dimensional position and the direction of the line of sight. The target projection area of ​​the compensated projection image on the dynamic imaging medium is determined based on the intersection relationship between the observation ray and the real-time imaging surface model.

[0041] Based on the positional relationship between the target projection area and the corresponding local imaging unit in the real-time imaging surface model, the viewpoint compensation parameters of the compensated projection image are determined. The viewpoint compensation parameters include at least one of the following: image projection position, display scale, viewing angle transformation relationship, projection focus, projection phase, polarization angle, and projection screen window.

[0042] The compensated projection image is corrected according to the viewpoint compensation parameters to obtain the compensated projection image after viewpoint compensation.

[0043] The projection component projects the compensated projection image, which has undergone viewpoint compensation, onto the dynamic imaging medium according to the viewpoint compensation parameters.

[0044] Optionally, during projection, spatial state data and viewpoint information are cyclically updated, and the projected image is adjusted and compensated based on the updated spatial state data and viewpoint information, as well as viewpoint compensation is performed, including:

[0045] During the process of compensating the projected image onto the dynamic imaging medium, the current spatial state data of the dynamic imaging medium and the current viewpoint information of the observer relative to the target imaging area are collected according to a preset update cycle, and the current spatial state data and the current viewpoint information are used as the updated spatial state data and the updated viewpoint information, respectively.

[0046] The real-time imaging surface model, geometric distortions, and density distribution data are redefined based on the updated spatial state data, and the compensated projection image is adjusted according to the redefined geometric distortions and density distribution data.

[0047] The projection process of the adjusted compensated projection image is subjected to viewpoint compensation based on the updated viewpoint information, and the viewpoint-compensated compensated projection image is projected onto the dynamic imaging medium.

[0048] During the process of compensating for the projection image being projected onto the dynamic imaging medium, environmental status information inside the cabin is also collected according to a preset update cycle or in response to an environmental detection trigger signal. The environmental status information includes at least one of humidity information, air conditioning airflow status information, and driver visibility status information.

[0049] When the humidity information reaches the preset humidity threshold, a humidity adjustment operation is performed. The humidity adjustment operation includes at least one of reducing the supply of atomized particles from the atomization generation component and controlling the vehicle air conditioning system to perform dehumidification or return air operation.

[0050] When the air conditioning airflow state information indicates that the target imaging area is disturbed by the air conditioning airflow, at least one of the following is adjusted: airflow rate, airflow velocity, and airflow pressure difference of the constraint airflow output by the airflow adjustment component.

[0051] When the driver's field of vision status information indicates that the dynamic imaging medium obstructs the driver's field of vision, the target imaging area is changed, the supply of atomized particles to the atomization generation component is reduced, or projection is stopped.

[0052] Secondly, embodiments of the present invention provide a distortion correction system for vehicle-mounted fogging imaging, comprising:

[0053] The area determination module is used to determine the target imaging area inside the cabin in response to the in-vehicle space imaging trigger signal;

[0054] The medium forming module is used to supply atomized particles to the target imaging area through the atomization generating component, and to form a constraining airflow around the target imaging area through the airflow regulating component to constrain the atomized particles, so that the atomized particles form a dynamic imaging medium with a predetermined thickness in the target imaging area.

[0055] The information acquisition module is used to collect spatial state data of the dynamic imaging medium and obtain the viewpoint information of the observer in the cockpit relative to the target imaging area.

[0056] The model building module is used to build a real-time imaging surface model of the dynamic imaging medium based on spatial state data, and to determine the geometric distortion when the image to be projected is projected onto the dynamic imaging medium based on the real-time imaging surface model.

[0057] The image compensation module is used to perform inverse geometric pre-distortion processing on the image to be projected based on the geometric distortion, and generate a compensated projection image.

[0058] The projection compensation module is used to perform viewpoint compensation on the compensated projection image based on viewpoint information, and then project the viewpoint-compensated compensated projection image onto the dynamic imaging medium.

[0059] The cyclic correction module is used to cyclically update spatial state data and viewpoint information during projection, adjust and compensate the projected image based on the updated spatial state data and viewpoint information, and perform viewpoint compensation.

[0060] Thirdly, embodiments of the present invention provide a distortion correction device for vehicle-mounted fogging imaging, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor performs the distortion correction method for vehicle-mounted fogging imaging as described above.

[0061] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the distortion correction method for vehicle-mounted fogging imaging as described above.

[0062] (III) Beneficial Effects

[0063] The beneficial effects of this invention are:

[0064] First, the present invention determines the target imaging area within the cabin in response to an in-vehicle space imaging trigger signal, and forms a dynamic imaging medium with a predetermined thickness within the target imaging area through a fogging component and an airflow adjustment component. Therefore, compared to imaging methods that rely on windshields, windows, or fixed screens, the present invention can form an imaging medium for carrying projected images at a target spatial location within the cabin, thereby improving the flexibility of the in-vehicle space imaging area layout.

[0065] Secondly, this invention acquires spatial state data of the dynamic imaging medium and establishes a real-time imaging surface model of the dynamic imaging medium based on the spatial state data. Since the morphology of the dynamic imaging medium may change over time, the real-time imaging surface model can characterize the current spatial morphology of the dynamic imaging medium, reducing the reliance on assumptions of fixed planes or regular curved surfaces for subsequent projection correction, thereby improving the matching degree between the distortion correction basis and the actual imaging medium state.

[0066] Furthermore, this invention determines the geometric distortion of the image to be projected onto the dynamic imaging medium based on a real-time imaging surface model, and performs reverse geometric pre-distortion processing on the image to be projected based on the geometric distortion to generate a compensated projection image. Therefore, it is possible to pre-compensate for geometric deviations caused by changes in the spatial shape of the dynamic imaging medium before image projection, reducing the occurrence of stretching, offset, or perspective distortion in the projected image on the dynamic imaging medium.

[0067] Furthermore, this invention acquires the viewpoint information of an observer within the cockpit relative to the target imaging area, and performs viewpoint compensation on the projection process of the compensated projection image based on the viewpoint information. Therefore, when the observer's position, viewing direction, or viewing distance changes, the projection process can be adjusted according to the relative relationship between the observer and the target imaging area, reducing image scaling changes, positional shifts, or parallax distortion caused by changes in viewpoint.

[0068] Finally, this invention cyclically updates spatial state data and viewpoint information during projection, and adjusts and compensates the projected image and performs viewpoint compensation based on the updated spatial state data and viewpoint information. Therefore, it can continuously correct the image compensation and projection process even when the spatial state of the dynamic imaging medium and the observer's viewpoint are constantly changing, thereby improving the geometric stability and display continuity of the projected image during vehicle-mounted fogging imaging. Attached Figure Description

[0069] Figure 1 A schematic diagram of the overall process of the distortion correction method for vehicle-mounted fogging imaging provided in an embodiment of the present invention;

[0070] Figure 2 A schematic diagram of the specific process of step S2 in the distortion correction method for vehicle-mounted fogging imaging provided in an embodiment of the present invention;

[0071] Figure 3 A schematic diagram of the specific process of step S3 in the distortion correction method for vehicle-mounted fogging imaging provided in an embodiment of the present invention;

[0072] Figure 4 A schematic diagram illustrating the specific process of step S4 in the distortion correction method for vehicle-mounted fogging imaging provided in an embodiment of the present invention;

[0073] Figure 5 A schematic diagram of the specific process of step S5 in the distortion correction method for vehicle-mounted fogging imaging provided in an embodiment of the present invention;

[0074] Figure 6 A schematic diagram of the specific process of step S6 in the distortion correction method for vehicle-mounted fogging imaging provided in an embodiment of the present invention;

[0075] Figure 7 This is a schematic diagram illustrating the specific process of step S7 in the distortion correction method for vehicle-mounted fogging imaging provided in an embodiment of the present invention. Detailed Implementation

[0076] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0077] like Figure 1 As shown in the figure, an embodiment of the present invention proposes a distortion correction method for vehicle-mounted fogging imaging, comprising: determining a target imaging area in the cabin in response to an in-vehicle spatial imaging trigger signal; supplying fogging particles to the target imaging area through a fogging generating component, and forming a constraining airflow around the target imaging area through an airflow regulating component to constrain the fogging particles, so that the fogging particles form a dynamic imaging medium with a predetermined thickness in the target imaging area; acquiring spatial state data of the dynamic imaging medium and obtaining the viewpoint information of the observer in the cabin relative to the target imaging area; establishing a real-time imaging surface model of the dynamic imaging medium based on the spatial state data, and determining the geometric distortion of the image to be projected onto the dynamic imaging medium according to the real-time imaging surface model; and performing reverse geometric pre-distortion processing on the image to be projected according to the geometric distortion to generate a compensated projection image.

[0078] Viewpoint compensation is performed on the compensated projection image based on viewpoint information, and the compensated projection image is projected onto the dynamic imaging medium. During the projection process, spatial state data and viewpoint information are updated cyclically, and the compensated projection image is adjusted based on the updated spatial state data and viewpoint information, as well as viewpoint compensation is performed.

[0079] First, the present invention determines the target imaging area within the cabin in response to an in-vehicle space imaging trigger signal, and forms a dynamic imaging medium with a predetermined thickness within the target imaging area through a fogging component and an airflow adjustment component. Therefore, compared to imaging methods that rely on windshields, windows, or fixed screens, the present invention can form an imaging medium for carrying projected images at a target spatial location within the cabin, thereby improving the flexibility of the in-vehicle space imaging area layout.

[0080] Secondly, this invention acquires spatial state data of the dynamic imaging medium and establishes a real-time imaging surface model of the dynamic imaging medium based on the spatial state data. Since the morphology of the dynamic imaging medium may change over time, the real-time imaging surface model can characterize the current spatial morphology of the dynamic imaging medium, reducing the reliance on assumptions of fixed planes or regular curved surfaces for subsequent projection correction, thereby improving the matching degree between the distortion correction basis and the actual imaging medium state.

[0081] Furthermore, this invention determines the geometric distortion of the image to be projected onto the dynamic imaging medium based on a real-time imaging surface model, and performs reverse geometric pre-distortion processing on the image to be projected based on the geometric distortion to generate a compensated projection image. Therefore, it is possible to pre-compensate for geometric deviations caused by changes in the spatial shape of the dynamic imaging medium before image projection, reducing the occurrence of stretching, offset, or perspective distortion in the projected image on the dynamic imaging medium.

[0082] Furthermore, this invention acquires the viewpoint information of an observer within the cockpit relative to the target imaging area, and performs viewpoint compensation on the projection process of the compensated projection image based on the viewpoint information. Therefore, when the observer's position, viewing direction, or viewing distance changes, the projection process can be adjusted according to the relative relationship between the observer and the target imaging area, reducing image scaling changes, positional shifts, or parallax distortion caused by changes in viewpoint.

[0083] Finally, this invention cyclically updates spatial state data and viewpoint information during projection, and adjusts and compensates the projected image and performs viewpoint compensation based on the updated spatial state data and viewpoint information. Therefore, it can continuously correct the image compensation and projection process even when the spatial state of the dynamic imaging medium and the observer's viewpoint are constantly changing, thereby improving the geometric stability and display continuity of the projected image during vehicle-mounted fogging imaging.

[0084] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention can be understood more clearly and thoroughly, and that the scope of the present invention can be fully conveyed to those skilled in the art.

[0085] Specifically, embodiments of the present invention provide a distortion correction method for vehicle-mounted fogging imaging, characterized in that it includes:

[0086] S1. In response to the in-vehicle space imaging trigger signal, determine the target imaging area inside the cabin.

[0087] In this step, the in-vehicle space imaging trigger signals include occupant gesture trigger signals, voice interaction trigger signals, display trigger signals issued by the vehicle's infotainment system, enhanced display trigger signals issued by the navigation system, occupant positioning trigger signals generated by the seat or occupant detection module, or projection task trigger signals generated when cabin scenarios such as entertainment, conferencing, and AI assistant interaction are initiated. After the processor responds to the in-vehicle space imaging trigger signals, it wakes up the projection task and determines the target imaging area for forming the atomized imaging medium based on the occupant's position in the cabin, the interaction position, the type of content to be displayed, and the driving safety area. The target imaging area is located in the cabin space, such as in front of the passenger's fingertips, in the middle area of ​​the seat, or in the rear entertainment area, so that the image to be imaged is separated from the windshield, side windows, or physical screen and formed within the cabin space. When the occupant's position or interaction position changes, the target imaging area can also be updated to achieve spatial following of the image.

[0088] S2. Atomized particles are supplied to the target imaging area through the atomization generating component, and a constraining airflow is formed around the target imaging area through the airflow regulating component to constrain the atomized particles, so that the atomized particles form a dynamic imaging medium with a predetermined thickness in the target imaging area.

[0089] In this step, the atomization component employs a directional dry fog generation structure, and, in conjunction with pressure differential compensation, delivers the atomized particles to the target imaging area. Constrained airflow forms an airflow protective layer around the target imaging area, preventing the atomized particles from dispersing naturally. Instead, they maintain a relatively concentrated spatial distribution even under conditions of airflow from the vehicle's air conditioning, vehicle vibration, or air pressure disturbances caused by open windows. Thus, the atomized particles can form an imaging medium similar to a dynamic air curtain near the preset imaging surface, providing a suitable spatial position and thickness for projection.

[0090] Furthermore, such as Figure 2 As shown, step S2 includes:

[0091] S21. Based on the position and size of the target imaging area in the cockpit coordinate system, determine the supply direction and quantity of atomized particles, as well as the target thickness of the dynamic imaging medium.

[0092] The cockpit coordinate system is a three-dimensional reference coordinate system used to describe the spatial relationships within the cockpit. It is established based on the vehicle coordinate system, seat mounting reference points, dashboard center point, or projection component mounting reference points. The processor determines the supply direction of the atomized particles based on the spatial position of the target imaging area in the cockpit coordinate system, ensuring the particles enter the target imaging area in a direction towards it. It also determines the supply amount of atomized particles based on the size of the target imaging area, ensuring the particles reach the appropriate particle concentration for imaging within the target imaging area. Furthermore, it determines the target thickness of the dynamic imaging medium based on the size of the target imaging area, enabling the dynamic imaging medium to form an imaging layer that matches the target imaging area. For example, a target thickness of 5cm to 10cm is set to ensure the atomized particles are relatively flat within the target imaging area, balancing image carrying capacity and cockpit environment stability.

[0093] S22. Control the vibration of the ultrasonic transducer array in the atomization generating component to break the atomized liquid into micron-sized atomized particles, and output the atomized particles to the target imaging area according to the supply direction and supply amount.

[0094] The ultrasonic transducer array in the atomization generating component can break the atomized liquid into smaller and more uniformly distributed atomized particles through high-frequency vibration, for example, forming micron-sized droplets through 2.4MHz high-frequency vibration. In some embodiments, the processor generates a pulse width modulation (PWM) control signal based on a determined atomized particle supply, and controls at least one of the drive duty cycle, drive duration, or drive power of the ultrasonic transducer array through the PWM control signal to adjust the atomization intensity and atomized particle output of the atomization generating component, so that the atomized particles enter the target imaging area according to the supply. In some embodiments, the particle size of the atomized particles can be controlled below 5μm, or within the range of 3μm to 5μm, to form a dry fog medium that is not easily condensed in the cabin, thereby reducing the risk of glass fogging, interior moisture, or electronic component moisture, and providing a relatively uniform scattering medium for subsequent projection.

[0095] S23. Based on the target thickness of the dynamic imaging medium, control the airflow adjustment component to output constrained airflow around the target imaging area.

[0096] Therefore, based on the target thickness of the dynamic imaging medium, at least one of the following is determined: the coverage width of the constrained airflow, the initial flow rate, the initial flow velocity, or the initial pressure difference. This ensures that the constrained range matches the target thickness. A directional airflow is output around at least a portion of the boundary of the target imaging area via an airflow adjustment component. This directional airflow coordinates with the supply airflow of atomized particles to form a constrained airflow around the target imaging area. When the vehicle's air conditioning is operational, the processor can further adjust the constrained airflow based on the air conditioning outlet direction, airflow intensity, or local airflow conditions in the cabin. This coordinates the constrained airflow with the existing airflow within the cabin, reducing the impact of air conditioning turbulence on the dynamic imaging medium. For example, before or simultaneously with the supply of atomized particles, the processor can control the vehicle's air conditioning system to reduce / close the airflow towards the target imaging area and activate lateral guide airflow or an air protection curtain.

[0097] S24. By adjusting at least one of the airflow flow rate, airflow velocity, and airflow pressure difference of the constrained airflow through the airflow adjustment component, an airflow boundary is formed around the target imaging area, and the atomized particles output according to the supply direction and supply amount are constrained within the target imaging area through the airflow boundary, so that the atomized particles are distributed along the preset imaging surface to form a dynamic imaging medium with a predetermined thickness.

[0098] Furthermore, the airflow regulation component can adjust the airflow parameters of multiple angular nozzles via proportional solenoid valves, or adjust the airflow parameters of nozzles located on the side of the target imaging area, to form a constrained airflow boundary around the target imaging area, i.e., an air curtain for confining atomized particles. With the help of this constrained airflow boundary, atomized particles can flow close to the preset imaging surface under the guidance of airflow and be constrained within the target imaging area, thereby reducing the disturbance of the dynamic imaging medium caused by the turbulence formed by the cabin air conditioning vents, and limiting the diffusion of atomized particles into the driver's field of vision area, electronic component area, or non-imaging area. Thus, the atomized particles are stabilized into a fluid imaging medium with a predetermined thickness that can be dynamically maintained according to control commands, reducing dynamic imaging medium dispersion, jitter, and edge drift caused by air conditioning airflow, occupant movement, or vehicle vibration.

[0099] S3. Collect spatial state data of the dynamic imaging medium and obtain the viewpoint information of the observer in the cockpit relative to the target imaging area.

[0100] In this step, spatial state data is used to characterize the real-time morphology and particle distribution of the dynamic imaging medium within the cockpit space, while viewpoint information is used to characterize the observer's spatial position and line of sight when viewing the target imaging area. By acquiring these two types of data, geometric distortion calculations are performed on the non-flat, easily jittery, foggy imaging medium, and the projected image is adapted to the observer's actual viewing angle.

[0101] Furthermore, such as Figure 3 As shown, step S3 includes:

[0102] S31. Before the dynamic imaging medium is formed, the space perception component is used to perform a global scan of the cabin interior space to obtain three-dimensional spatial reference data of the cabin interior.

[0103] The spatial perception components include at least one of a time-of-flight sensor, lidar, and structured light sensor. The processor can perform a global scan of the cockpit interior space before the atomized particles enter the target imaging area, obtaining three-dimensional point cloud data including the seats, interior trim, air vents, projection units, and the spatial positions surrounding the target imaging area, and using this data as a spatial reference for subsequent local scanning and coordinate registration.

[0104] S32. After the dynamic imaging medium is formed, the spatial sensing component is used to perform a local scan of the target imaging area to obtain local spatial data of the dynamic imaging medium at multiple sampling times.

[0105] Accordingly, after the atomized particles form a dynamic imaging medium, the spatial sensing component can perform a local fine scan of the target imaging area to obtain the spatial contour, surface undulations, thickness changes, and atomized particle density changes of the area where the dynamic imaging medium is located at different sampling times. This local scan can be performed at a higher sampling frequency, such as 120Hz, to capture dynamic deformations caused by airflow from the vehicle's air conditioning system, vehicle vibration, or the movement of the atomized particles themselves.

[0106] S33. Perform coordinate registration between the local spatial data and the three-dimensional spatial reference data to obtain the spatial state data of the dynamic imaging medium relative to the cockpit coordinate system. The spatial state data includes the three-dimensional morphological data and density distribution data of the dynamic imaging medium.

[0107] Therefore, the processor can perform coordinate registration between the local spatial data obtained from partial scanning and the three-dimensional spatial reference data obtained from global scanning to obtain the spatial state data of the dynamic imaging medium relative to the cockpit coordinate system. Among them, the three-dimensional morphological data includes the spatial position, boundary contour, local thickness, and surface undulation state of the dynamic imaging medium, and the density distribution data includes the point cloud density or scattering intensity distribution of atomized particles in the target imaging area.

[0108] S34. Use the cockpit monitoring component to obtain the eye position data of the observer in the cockpit, and determine the viewpoint information of the observer relative to the target imaging area based on the eye position data and the position of the target imaging area in the cockpit coordinate system. The viewpoint information includes the three-dimensional position of the observer's eyes in the cockpit coordinate system and the direction of the observer's line of sight relative to the target imaging area.

[0109] Furthermore, the cockpit monitoring component includes cameras, infrared cameras, eye-tracking modules, or combinations thereof in the cockpit monitoring system, used to acquire or assist in determining the observer's three-dimensional eye coordinates, head posture, or gaze vector. Based on the relative positional relationship between the observer's eye position and the target imaging area, the observer's viewpoint position and gaze direction when viewing the dynamic imaging medium are determined. In the case of multiple observers, eye position data for different observers can be acquired separately, and the target observer's viewpoint information for subsequent viewpoint compensation can be determined based on a preset display object or the current interactive object.

[0110] S4. Establish a real-time imaging surface model of the dynamic imaging medium based on spatial state data, and determine the geometric distortion when the image to be projected is projected onto the dynamic imaging medium based on the real-time imaging surface model.

[0111] In this step, based on the three-dimensional morphological data and density distribution data of the dynamic imaging medium, a calculable real-time imaging surface model is established based on the actual non-uniform fluid morphology of the dynamic imaging medium. The real-time imaging surface model is a curved surface model composed of multiple local imaging units, used to characterize the spatial offset, surface depth variation, and local tilt state of the dynamic imaging medium relative to a preset ideal imaging surface. Since the dynamic imaging medium is affected by air conditioning airflow, vehicle vibration, gravity, and the random motion of atomized particles, resulting in local undulations and edge drift, the geometric distortion generated when the image to be projected is imaged on the actual dynamic imaging medium can be determined based on this real-time imaging surface model.

[0112] Furthermore, such as Figure 4 As shown, step S4 includes:

[0113] S41. Extract multiple spatial sampling points of the dynamic imaging medium within the target imaging area from the three-dimensional morphological data in the spatial state data.

[0114] Specifically, spatial sampling points belonging to the dynamic imaging medium can be selected from the point cloud data corresponding to the target imaging area, while background points corresponding to seats, interior trim, air vents, occupant bodies, or other fixed structures in the cabin can be removed. These spatial sampling points are distributed on the front surface, rear surface, or intermediate layer of the dynamic imaging medium, reflecting its spatial coordinates, thickness variations, and surface undulations at different local locations.

[0115] S42. Perform surface fitting based on multiple spatial sampling points to generate a real-time imaging surface model composed of multiple local imaging units, wherein each local imaging unit includes the spatial coordinates, surface depth, and normal vector of the corresponding spatial sampling point.

[0116] Specifically, surface fitting can be performed based on multiple spatial sampling points to generate a surface model with real-time morphology. This surface fitting can be achieved using non-uniform rational B-spline surface fitting, mesh surface fitting, or piecewise surface fitting, allowing the real-time imaging surface model to update with the dynamic fluctuations of the imaging medium. Each local imaging unit corresponds to a local region on the dynamic imaging medium; where spatial coordinates represent the position of the local region in the cockpit coordinate system, surface depth represents the front-to-back offset of the local region relative to the preset ideal imaging surface, and the normal vector represents the local tilt direction of the local region and is used to determine the incident relationship of the projected light rays relative to the local region.

[0117] S43. Based on the spatial coordinates, surface depth, and normal vector of each local imaging unit, determine the projection point when the image to be projected is projected onto each local imaging unit.

[0118] In this step, the spatial coordinates of each local imaging unit are used as the base landing point position when the image to be projected is projected onto the dynamic imaging medium. The base landing point position is then corrected for forward and backward offset based on the surface depth, and further corrected for local tilt direction based on the local tilt direction represented by the normal vector. This determines the projection landing point of the image points or grid vertices in the image to be projected onto each local imaging unit. For local areas with large changes in surface depth or rapid changes in normal vector, the corresponding landing point correction amount can be increased; for local areas with small changes in surface depth or relatively gentle changes in normal vector, the corresponding landing point correction amount can be decreased, so that the projection landing point can reflect the non-flat shape of the dynamic imaging medium.

[0119] S44. Divide the image to be projected into multiple image grids, and determine the target grid position of each image grid according to the preset ideal imaging surface.

[0120] The image to be projected can be divided into multiple image grids according to a regular grid or an adaptive grid. The preset ideal imaging surface is a reference imaging surface within the target imaging area used to characterize the desired display effect. It can be a virtual plane or a reference curved surface that matches the target imaging area. The position of the grid vertex or grid region corresponding to each image grid when it is presented without distortion on the preset ideal imaging surface is the target grid position of the corresponding image grid.

[0121] S45. Establish the projection mapping relationship between the target grid position of each image grid and the projection landing point of the corresponding local imaging unit, and determine at least one of the position deviation, scale deviation and perspective deviation of the projection landing point relative to the target grid position according to the projection mapping relationship, so as to serve as the geometric distortion variable of each image grid.

[0122] Furthermore, a projection mapping relationship between the target mesh position and the projection landing point can be established through homography matrix calculations, mesh deformation mapping, ray tracing, or inverse mapping. This projection mapping relationship characterizes the correspondence between the actual and ideal imaging positions of the projected image on the dynamic imaging medium. By comparing the differences between the target mesh position and the actual projection landing point, geometric distortions such as positional offset, local stretching or compression, and perspective tilt generated by each image mesh on the dynamic imaging medium can be obtained. These geometric distortions characterize the impact of the unevenness and dynamic deformation of the dynamic imaging medium on the projected image.

[0123] S5. Perform reverse geometric pre-distortion processing on the image to be projected based on the geometric distortion to generate a compensated projection image.

[0124] In this step, based on the geometric distortion, the image to be projected is corrected in the opposite direction to the actual distortion, so that the projected image presents the desired geometric shape on the dynamic imaging medium. Since the dynamic imaging medium has local undulations, thickness variations, and local tilts, it is difficult to eliminate local distortions by simply performing overall translation or uniform scaling. Therefore, inverse geometric pre-distortion processing based on the image grid can be used to make the compensated projected image match the current spatial shape of the dynamic imaging medium.

[0125] Furthermore, such as Figure 5 As shown, step S5 includes:

[0126] S51. Based on the geometric distortion of each image grid, determine the inverse compensation amount used to offset the geometric distortion, wherein the inverse compensation amount includes at least one of inverse displacement, inverse scaling, and inverse perspective transformation.

[0127] The geometric distortion includes the positional deviation, scale deviation, and perspective deviation of each image grid relative to the target grid. Correspondingly, the inverse compensation is used to apply an inverse adjustment to the corresponding image grid before projection, so that after the image grid is projected and falls into the dynamic imaging medium, it can offset the offset, stretching, compression, or perspective distortion caused by factors such as the surface undulation and local tilt of the dynamic imaging medium.

[0128] S52. Determine the vertex correction amount of each image grid according to the reverse compensation amount, and adjust the vertex position of each image grid in reverse according to the vertex correction amount, so that the grid boundary of each image grid is deformed as the vertex position is adjusted, and generate a pre-distorted image that matches the current spatial shape of the dynamic imaging medium.

[0129] By inversely adjusting the vertex positions of the image grid, the grid boundaries defined by adjacent vertices shift, stretch, or tilt as the vertex positions change. The pixel or texture content within the image grid is then interpolated or resampled according to the adjusted grid boundaries. This causes both the grid boundaries and internal textures to undergo a preset deformation corresponding to the inverse compensation amount, resulting in a pre-distorted image that appears to be distorted in reverse before projection but appears nearly distortion-free after being projected onto the dynamic imaging medium. This process corresponds to real-time grid deformation processing, which can generate matching image geometry frame-by-frame for dynamic surface changes in the dynamic imaging medium.

[0130] S53. Based on the density distribution data in the spatial state data, determine the brightness compensation amount corresponding to different image regions in the pre-distorted image.

[0131] Accordingly, the brightness compensation amount for each image region in the pre-distorted image can be determined based on the point cloud density or scattering intensity distribution of atomized particles within the target imaging region, as characterized by density distribution data. Specifically, the point cloud density or scattering intensity corresponding to each image region is normalized to obtain the medium intensity coefficient K, where K is a value between 0 and 1. A larger K indicates a higher point cloud density of atomized particles or a stronger scattering intensity in the dynamic imaging medium region corresponding to that image region. The brightness gain G is determined based on the medium intensity coefficient K and used as the brightness compensation amount. For example, the brightness gain G is determined according to the following relationship:

[0132] G = 1 + α × (1 - K);

[0133] Wherein, α is a preset compensation coefficient, and the value of α ranges from 0 to 1. Thus, when K=1, G=1, indicating that no brightness enhancement is applied to the corresponding image area; when K=0, G=1+α, indicating that maximum brightness enhancement is applied to the corresponding image area; when K is between 0 and 1, G decreases as K increases, in order to reduce the unevenness of brightness caused by differences in the density or scattering intensity distribution of fogged particle point clouds in different areas.

[0134] S54. Adjust the brightness gain of the corresponding image region in the pre-distorted image according to the brightness compensation amount to generate a compensated projection image.

[0135] Furthermore, based on the luminance gain G, which serves as the luminance compensation amount, the pixel luminance values ​​of the corresponding image regions in the pre-distorted image are adjusted. For any pixel in the pre-distorted image, its adjusted pixel luminance value I' can be determined according to the following relationship:

[0136] I' = clip(G×I);

[0137] Where I represents the pixel brightness value before adjustment, G represents the brightness gain of the image region where the pixel is located, and clip indicates that the adjusted pixel brightness value is limited to a preset brightness range. The compensated projection image generated after the above geometric pre-distortion and brightness compensation can simultaneously improve the geometric fidelity and brightness uniformity of the image after being projected onto a dynamic imaging medium.

[0138] S6. Perform viewpoint compensation on the compensated projection image based on the viewpoint information, and project the compensated projection image after viewpoint compensation onto the dynamic imaging medium.

[0139] In this step, the compensated projection image is corrected based on the observer's viewpoint information relative to the target imaging area, so that the image projected onto the dynamic imaging medium is adapted to the observer's current viewing direction. Because the fogged imaging medium has scattering display characteristics, observers at different positions may experience changes in image scale, spatial drift, or ghosting when viewing the same dynamic imaging medium. Viewpoint compensation enables the observer to obtain a more stable image position and display scale within the corresponding visual window.

[0140] Furthermore, such as Figure 6 As shown, step S6 includes:

[0141] S61. Determine the three-dimensional position of the observer's eye in the cockpit coordinate system and the direction of the observer's line of sight relative to the target imaging area from the viewpoint information.

[0142] S62. Determine the observer's observation ray relative to the dynamic imaging medium based on the three-dimensional position and line of sight. Then, determine the target projection area of ​​the compensated projection image on the dynamic imaging medium based on the intersection relationship between the observation ray and the real-time imaging surface model. This intersection relationship characterizes the relative spatial relationship between the observer's line of sight and the dynamic imaging medium, and is used to determine the target projection area corresponding to the observer's current viewpoint.

[0143] S63. Based on the positional relationship between the target projection area and the corresponding local imaging unit in the real-time imaging surface model, determine the viewpoint compensation parameters of the compensated projection image. The viewpoint compensation parameters include at least one of the following: image projection position, display scale, viewing angle transformation relationship, projection focus, projection phase, polarization angle, and projection screen window.

[0144] Therefore, based on the positional relationship between the target projection area and the corresponding local imaging unit in the real-time imaging surface model, the viewpoint compensation parameters of the compensated projection image are determined. These parameters are used to ensure that the compensated projection image maintains the expected spatial position, depth, and scale from a specific observer's perspective. Specifically, the image projection position determines the display position of the compensated projection image on the dynamic imaging medium; the display scale adjusts the display size of the compensated projection image from the observer's perspective; the viewpoint transformation relationship compensates for perspective deviations caused by changes in the observer's line of sight; the projection focus ensures the image is clearly presented near the target projection area; the projection phase and polarization angle adjust the display directionality or visual window of the projection beam in the dynamic imaging medium; and the projection image window defines the spatial range from which the observer obtains a stable image.

[0145] S64. Correct the compensated projection image according to the viewpoint compensation parameters to obtain the compensated projection image after viewpoint compensation.

[0146] Furthermore, based on at least one of the following: image projection position, display size, perspective shape, and screen window range, the display position, display ratio, viewing angle change relationship, and the projection screen window range of the compensated projection image are corrected to obtain the compensated projection image after viewpoint compensation. Furthermore, during the projection process, the optical projection parameters of the projection component can be adjusted according to at least one of the following: projection focus, projection phase, and polarization angle, so that the image projected onto the dynamic imaging medium matches the observer's current viewpoint, thereby reducing image misalignment, drift, scaling distortion, or ghosting caused by the observer's movement or change of posture.

[0147] S65. The compensated projection image after viewpoint compensation is projected onto the dynamic imaging medium by the projection component according to the viewpoint compensation parameters.

[0148] Therefore, the projection component projects the image onto the target projection area according to the viewpoint compensation parameters, so that the compensated projected image forms a display screen facing the observer's current viewpoint on the dynamic imaging medium. This projection component includes a DLP projection unit or other vehicle-mounted projection optical engine, and during the projection process, it adjusts the focus, phase, or polarization parameters to lock a relatively stable viewing window for the observer. The projection component can also perform automatic focusing based on the target projection area and the observer's viewpoint information to match the projection focal plane with the target projection area.

[0149] S7. During the projection process, the spatial state data and viewpoint information are updated cyclically, and the projected image is adjusted and compensated based on the updated spatial state data and viewpoint information, and viewpoint compensation is performed.

[0150] In this step, as the compensated projected image is projected onto the dynamic imaging medium, the morphological changes of the dynamic imaging medium and the changes in the observer's viewpoint are continuously acquired, and the image pre-distortion and viewpoint compensation are dynamically updated based on the changes. Since the dynamic imaging medium is affected by air conditioning airflow, vehicle vibration, occupant movement, and the diffusion of atomized particles, and the observer's head position or line of sight may also change, cyclic updates enable the projected image to continuously adapt to the current spatial morphology of the dynamic imaging medium and the observer's current viewing angle.

[0151] Furthermore, such as Figure 7 As shown, step S7 includes:

[0152] S71. During the process of projecting the compensated projection image onto the dynamic imaging medium, the current spatial state data of the dynamic imaging medium and the current viewpoint information of the observer relative to the target imaging area are collected according to a preset update cycle, and the current spatial state data and the current viewpoint information are used as the updated spatial state data and the updated viewpoint information, respectively.

[0153] The spatial perception component repeatedly scans the target imaging area according to the preset update cycle to obtain the current spatial state data of the dynamic imaging medium. The cockpit monitoring component synchronously or asynchronously acquires the observer's current viewpoint information. The preset update cycle can be set according to the shape change rate of the dynamic imaging medium, the projection frame rate, the vehicle vibration state, or the change rate of the observer's viewpoint. For example, it can update the current spatial state data of the dynamic imaging medium based on a 120Hz sampling frequency to capture the real-time fluctuations of the dynamic imaging medium.

[0154] S72. Based on the updated spatial state data, redetermine the real-time imaging surface model, geometric distortion variables, and density distribution data, and adjust the compensated projection image according to the redetermined geometric distortion variables and density distribution data.

[0155] In this step, the real-time imaging surface model of the dynamic imaging medium is refitted based on the updated spatial state data, and the geometric distortion of each image grid relative to the preset ideal imaging surface is recalculated. For regions where the surface of the dynamic imaging medium experiences local undulations, edge drift, or thickness changes, the inverse displacement, inverse scaling, or inverse perspective transformation of the corresponding image grid is redefined; for regions where point cloud density or scattering intensity changes, the brightness compensation of the corresponding image region is redefined. Thus, the compensated projected image can be updated frame-by-frame or periodically according to changes in shape and density distribution.

[0156] S73. Based on the updated viewpoint information, viewpoint compensation is performed on the projection process of the adjusted compensated projection image, and the compensated projection image after viewpoint compensation is projected onto the dynamic imaging medium.

[0157] Therefore, based on the updated three-dimensional position and line of sight of the observer's eyes, the observer's observation ray, target projection area, and viewpoint compensation parameters relative to the dynamic imaging medium are redefined. The adjusted compensated projection image is then corrected according to the redefined viewpoint compensation parameters, resulting in a viewpoint-compensated compensated projection image. When projecting the viewpoint-compensated compensated projection image onto the dynamic imaging medium, the optical projection parameters of the projection component can also be adjusted according to the redefined viewpoint compensation parameters. Through this process, the projected image remains relatively stable within the observer's corresponding visual window even when the observer's head moves, their posture changes, or their line of sight shifts, reducing image misalignment, scale distortion, or ghosting.

[0158] S74. During the process of projecting the compensated projection image onto the dynamic imaging medium, environmental status information inside the cabin is also collected according to a preset update cycle or in response to an environmental detection trigger signal. The environmental status information includes at least one of humidity information, air conditioning airflow status information, and driver visibility status information.

[0159] Furthermore, environmental status information is used to reflect the impact of dynamic imaging media on the cabin environment and driving safety. Humidity information can be obtained from the cabin humidity sensor to determine whether continuous fogging leads to an increase in cabin humidity; air conditioning airflow status information can be obtained from the vehicle's air conditioning system operating parameters, vent status, or local airflow sensors to determine whether the air conditioning vents disturb the dynamic imaging media; and driver visibility status information can be obtained from the cabin camera, forward visibility detection module, or driver line-of-sight detection results to determine whether the dynamic imaging media or projected image enters the driver's sensitive visual field.

[0160] S75. When the humidity information reaches a preset humidity threshold, a humidity adjustment operation is performed. The humidity adjustment operation includes at least one of reducing the supply of atomized particles from the atomization generating component and controlling the vehicle air conditioning system to perform dehumidification or recirculation operation. When the air conditioning airflow status information indicates that the target imaging area is disturbed by the air conditioning airflow, at least one of the airflow flow rate, airflow velocity, and airflow pressure difference of the constrained airflow output by the airflow adjustment component is adjusted. When the driving vision status information indicates that the dynamic imaging medium obstructs the driving vision, the target imaging area is changed, the supply of atomized particles from the atomization generating component is reduced, or projection is stopped.

[0161] When any of the above environmental adjustment conditions are met, the atomization generation component, airflow adjustment component, vehicle air conditioning system, or projection component can be linked for control. When the cabin humidity approaches or reaches a preset humidity threshold, the risk of fogging of windows, damp interior, or dampness of electronic components is reduced by decreasing the supply of atomized particles, reducing the particle size or spray intensity of atomized particles, activating vehicle air conditioning dehumidification, or recirculating excess moisture. When the air conditioning airflow disturbs the dynamic imaging medium, the spatial constraint of the constrained airflow boundary on the atomized particles is maintained by increasing or redistributing the flow rate, velocity, or pressure difference of the constrained airflow. When the dynamic imaging medium or projected image obstructs the driver's view, the target imaging area is reselected, the thickness of the dynamic imaging medium is reduced, or atomization and projection are stopped to ensure that the in-vehicle display process is coordinated with the driving safety status.

[0162] Furthermore, this invention provides a distortion correction system for vehicle-mounted fogging imaging, comprising: a region determination module, used to determine a target imaging region within the cabin in response to an in-vehicle spatial imaging trigger signal; a medium forming module, used to supply fogging particles to the target imaging region through a fogging generating component, and to form a constraining airflow around the target imaging region through an airflow regulating component to constrain the fogging particles, so that the fogging particles form a dynamic imaging medium with a predetermined thickness within the target imaging region; an information acquisition module, used to acquire spatial state data of the dynamic imaging medium and obtain viewpoint information of an observer within the cabin relative to the target imaging region; and a model building module, used to... The system establishes a real-time imaging surface model of the dynamic imaging medium based on spatial state data, and determines the geometric distortion of the image to be projected onto the dynamic imaging medium based on the real-time imaging surface model. The image compensation module performs inverse geometric pre-distortion processing on the image to be projected based on the geometric distortion to generate a compensated projection image. The projection compensation module performs viewpoint compensation on the compensated projection image based on viewpoint information and projects the viewpoint-compensated compensated projection image onto the dynamic imaging medium. The cyclic correction module updates the spatial state data and viewpoint information cyclically during the projection process, adjusts the compensated projection image based on the updated spatial state data and viewpoint information, and performs viewpoint compensation.

[0163] In some embodiments, the region determination module, media formation module, information acquisition module, model building module, image compensation module, projection compensation module, and cyclic correction module are executed by the same processor, or are deployed separately in the cockpit domain controller, projection controller, fogging imaging control module, or vehicle electronic control unit. The modules interact with each other via an in-vehicle communication bus or in-vehicle Ethernet to form a closed loop for dynamic imaging media generation, spatial perception, image compensation, viewpoint compensation, and environmental safety control.

[0164] In some embodiments, the media forming module connects the atomization generating component and the airflow regulating component. The atomization generating component includes an ultrasonic transducer array for breaking the atomized liquid into micron-sized atomized particles through high-frequency vibration; the airflow regulating component includes a proportional solenoid valve and multiple nozzles disposed around the target imaging area. The proportional solenoid valve is used to regulate the flow rate, velocity, or pressure difference of the airflow output from each nozzle. The multiple nozzles are arranged at the four corners or sides of the target imaging area to form an air curtain, a low-pressure guide zone, or a high-speed airflow wall around the target imaging area, causing the atomized particles to be distributed close to a preset imaging surface and forming a dynamic imaging medium.

[0165] In some embodiments, the airflow adjustment component is also linked to the vehicle's air conditioning system. When the vehicle's air conditioning system is in the air supply state, the medium forming module adjusts the flow rate, velocity, or pressure difference of the constrained airflow according to the air conditioning outlet direction, air outlet intensity, or local airflow state near the target imaging area, so that the constrained airflow is coordinated with the original airflow in the cabin, thereby reducing the disturbance of the air conditioning turbulence to the dynamic imaging medium.

[0166] In some embodiments, the information acquisition module connects the spatial perception component and the cockpit monitoring component. The spatial perception component includes at least one of a time-of-flight sensor, lidar, structured light sensor, ultrasonic ranging array, or speckle interferometry detection component, used to acquire three-dimensional spatial reference data of the cockpit interior and local spatial data of the dynamic imaging medium. The cockpit monitoring component includes at least one of a camera, infrared camera, or eye-tracking module, used to acquire the observer's three-dimensional eye coordinates, head posture, or gaze vector.

[0167] In some embodiments, the projection compensation module is connected to a projection component, which includes a DLP projection unit, a laser projection unit, or other vehicle-mounted projection optical engine. Based on the observer's viewpoint information, the projection compensation module corrects the image projection position, display ratio, viewing angle transformation relationship, or projection window of the compensated projection image. During projection, it adjusts the projection focus, projection phase, or polarization angle of the projection component to form a display image facing the observer's current viewpoint on a dynamic imaging medium.

[0168] In some embodiments, the cyclic correction module is also connected to a cabin humidity sensor, an air conditioning controller, and a driver visibility detection component. During projection, the cyclic correction module monitors cabin humidity, air conditioning airflow status, and driver visibility status, and performs humidity adjustment, constrained airflow adjustment, or safety avoidance control respectively when cabin humidity reaches a preset humidity threshold, the target imaging area is disturbed by air conditioning airflow, or the driver's view is obstructed by a dynamic imaging medium.

[0169] Furthermore, this embodiment of the invention also provides a distortion correction device for vehicle-mounted fogging imaging, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor performs the distortion correction method for vehicle-mounted fogging imaging as described above. The processor may be located in a cockpit domain controller, a vehicle electronic control unit, a projection controller, or a separate fogging imaging control module.

[0170] Furthermore, embodiments of the present invention also provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the distortion correction method for vehicle-mounted fogging imaging as described above.

[0171] In some alternative embodiments, the dynamic imaging medium, spatial state acquisition method, image compensation method, or medium constraint method of the present invention may also employ at least one of the following alternatives. The dynamic imaging medium is not limited to water mist formed by ultrasonic atomization, but may also be an ionized air medium, a physical micropowder medium, or a phase change material microcapsule medium. When using an ionized air medium, a local medium capable of scattering the projected beam is formed in the target imaging area by an ionization device; when using a physical micropowder medium, a recyclable microparticle medium meeting vehicle safety requirements is suspended in the target imaging area by an airflow field or electromagnetic field; when using a phase change material microcapsule medium, microcapsules form a temporary imaging medium in the target imaging area through a spraying and recovery assembly, and are recovered after imaging is completed.

[0172] Regarding spatial state acquisition methods, structured light vision, ultrasonic array ranging, or speckle interferometry can be used to replace or supplement time-of-flight sensors and / or lidar. When using structured light vision, a preset grid or stripe pattern is projected onto the target imaging area in the non-visible light band, and the spatial contour, thickness changes, or local undulations of the dynamic imaging medium are inferred from the deformation of the grid or stripe pattern on the dynamic imaging medium. When using ultrasonic array ranging, multiple ultrasonic sensors are arranged around the target imaging area, and the spatial position or central region change of the dynamic imaging medium is determined based on the sound wave return time difference. When using speckle interferometry, minute fluctuations in the dynamic imaging medium are detected through changes in laser speckle, thereby improving the detection accuracy of jitter-like distortions in the dynamic imaging medium.

[0173] Regarding image compensation methods, deep learning-based end-to-end correction models, edge computing methods, light field folding algorithms, or spatial phase modulation methods can be employed. Deep learning-based end-to-end correction models take the spatial state data of the dynamic imaging medium, distortion feedback images, or viewpoint information as input, and output compensated projected images, image grid correction amounts, or projection control parameters. Edge computing methods deploy geometric pre-distortion, brightness compensation, or viewpoint compensation within the edge computing chip local to the projection component to reduce data transmission latency between the cockpit's central computing unit and the projection component. Light field folding algorithms or spatial phase modulation methods adjust the spatial phase distribution of the projection beam according to the spatial position of the observer's viewpoint and the target imaging area, ensuring that the compensated projected image forms a clear image within a predetermined depth or predetermined visual window of the dynamic imaging medium.

[0174] Regarding media confinement methods, in addition to air curtain confinement, electrostatic field confinement or vacuum adsorption cyclic confinement can also be used. When using electrostatic field confinement, the atomized particles are given a predetermined charge, and electrodes or transparent grids are placed around the target imaging area. The electric field restricts the diffusion of atomized particles into non-imaging areas. When using vacuum adsorption cyclic confinement, atomized particles are supplied to one side of the target imaging area, while an air intake or recovery channel is set on the other side. This creates a circulating flow field between the ejection and recovery of the atomized particles, thus balancing imaging stability and cabin humidity control.

[0175] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0176] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions.

[0177] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the claims should be interpreted to include both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0178] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, then this invention should also include these modifications and variations.

Claims

1. A distortion correction method for vehicle-mounted fogging imaging, characterized in that, include: In response to the in-vehicle space imaging trigger signal, the target imaging area inside the cabin is determined; Atomized particles are supplied to the target imaging area through the atomization generating component, and a constraining airflow is formed around the target imaging area through the airflow regulating component to constrain the atomized particles, so that the atomized particles form a dynamic imaging medium with a predetermined thickness in the target imaging area. Collect spatial state data of the dynamic imaging medium and obtain the viewpoint information of the observer in the cockpit relative to the target imaging area; A real-time imaging surface model of the dynamic imaging medium is established based on spatial state data, and the geometric distortion of the image to be projected onto the dynamic imaging medium is determined based on the real-time imaging surface model. Based on the geometric distortion, reverse geometric pre-distortion processing is performed on the image to be projected to generate a compensated projection image; The viewpoint compensation is performed on the compensated projection image based on the viewpoint information, and the viewpoint-compensated compensated projection image is then projected onto the dynamic imaging medium. During the projection process, the spatial state data and viewpoint information are updated cyclically, and the projected image is adjusted and compensated based on the updated spatial state data and viewpoint information, as well as viewpoint compensation is performed.

2. The distortion correction method for vehicle-mounted fogging imaging according to claim 1, characterized in that, Atomizing particles are supplied to the target imaging area via an atomization generating component, and a constraining airflow is formed around the target imaging area via an airflow regulating component to confine the atomizing particles, thereby forming a dynamic imaging medium with a predetermined thickness within the target imaging area. This includes: Based on the position and size of the target imaging area in the cockpit coordinate system, determine the supply direction and amount of atomized particles, as well as the target thickness of the dynamic imaging medium. The ultrasonic transducer array in the atomization generation component is controlled to vibrate, so as to break the atomized liquid into micron-sized atomized particles and output the atomized particles to the target imaging area according to the supply direction and supply amount. Based on the target thickness of the dynamic imaging medium, the airflow adjustment component is controlled to output constrained airflow around the target imaging area. By adjusting at least one of the airflow flow rate, airflow velocity, and airflow pressure difference of the constrained airflow through the airflow adjustment component, an airflow boundary is formed around the target imaging area. The atomized particles output according to the supply direction and supply amount are constrained within the target imaging area through the airflow boundary, so that the atomized particles are distributed along the preset imaging surface to form a dynamic imaging medium with a predetermined thickness.

3. The distortion correction method for vehicle-mounted fogging imaging according to claim 2, characterized in that, Acquire spatial state data of the dynamic imaging medium and obtain viewpoint information of the observer in the cockpit relative to the target imaging area, including: Before the dynamic imaging medium is formed, the space sensing component is used to perform a global scan of the cabin interior space to obtain three-dimensional spatial reference data of the cabin interior. After the dynamic imaging medium is formed, the spatial sensing component is used to perform a local scan of the target imaging area to obtain local spatial data of the dynamic imaging medium at multiple sampling times. By performing coordinate registration between local spatial data and three-dimensional spatial reference data, spatial state data of the dynamic imaging medium relative to the cockpit coordinate system is obtained. The spatial state data includes the three-dimensional morphological data and density distribution data of the dynamic imaging medium. The cockpit monitoring component is used to acquire the eye position data of the observer in the cockpit. Based on the eye position data and the position of the target imaging area in the cockpit coordinate system, the viewpoint information of the observer relative to the target imaging area is determined. The viewpoint information includes the three-dimensional position of the observer's eyes in the cockpit coordinate system and the direction of the observer's line of sight relative to the target imaging area.

4. The distortion correction method for vehicle-mounted fogging imaging according to claim 3, characterized in that, A real-time imaging surface model of the dynamic imaging medium is established based on spatial state data, and the geometric distortions when the image to be projected is determined based on the real-time imaging surface model, including: Extract multiple spatial sampling points of the dynamic imaging medium within the target imaging area from the three-dimensional morphological data in the spatial state data; A real-time imaging surface model consisting of multiple local imaging units is generated by fitting a surface based on multiple spatial sampling points. Each local imaging unit includes the spatial coordinates, surface depth, and normal vector of the corresponding spatial sampling point. Based on the spatial coordinates, surface depth, and normal vector of each local imaging unit, determine the projection point when the image to be projected is projected onto each local imaging unit; The image to be projected is divided into multiple image grids, and the target grid position of each image grid is determined according to the preset ideal imaging surface; Establish a projection mapping relationship between the target grid position of each image grid and the projection landing point of the corresponding local imaging unit, and determine at least one of the positional deviation, scale deviation, and perspective deviation of the projection landing point relative to the target grid position based on the projection mapping relationship, so as to serve as the geometric distortion variable of each image grid.

5. The distortion correction method for vehicle-mounted fogging imaging according to claim 4, characterized in that, Based on the geometric distortion, inverse geometric pre-distortion processing is performed on the image to be projected to generate a compensated projection image, including: Based on the geometric distortion of each image grid, determine the inverse compensation amount used to offset the geometric distortion, wherein the inverse compensation amount includes at least one of inverse displacement, inverse scaling, and inverse perspective transformation. The vertex correction amount of each image grid is determined based on the reverse compensation amount, and the vertex position of each image grid is adjusted in reverse according to the vertex correction amount, so that the grid boundary of each image grid is deformed as the vertex position is adjusted, thereby generating a pre-distorted image that matches the current spatial morphology of the dynamic imaging medium. Based on the density distribution data in the spatial state data, the brightness compensation amount corresponding to different image regions in the pre-distorted image is determined. The brightness gain of the corresponding image region in the pre-distorted image is adjusted according to the brightness compensation amount to generate a compensated projection image.

6. The distortion correction method for vehicle-mounted fogging imaging according to claim 5, characterized in that, Viewpoint compensation is performed on the compensated projection image based on viewpoint information, and the viewpoint-compensated compensated projection image is projected onto a dynamic imaging medium, including: Determine the three-dimensional position of the observer's eye in the cockpit coordinate system and the direction of the observer's line of sight relative to the target imaging area from the viewpoint information; The observer's observation ray relative to the dynamic imaging medium is determined based on the three-dimensional position and the direction of the line of sight. The target projection area of ​​the compensated projection image on the dynamic imaging medium is determined based on the intersection relationship between the observation ray and the real-time imaging surface model. Based on the positional relationship between the target projection area and the corresponding local imaging unit in the real-time imaging surface model, the viewpoint compensation parameters of the compensated projection image are determined. The viewpoint compensation parameters include at least one of the following: image projection position, display scale, viewing angle transformation relationship, projection focus, projection phase, polarization angle, and projection screen window. The compensated projection image is corrected according to the viewpoint compensation parameters to obtain the compensated projection image after viewpoint compensation. The projection component projects the compensated projection image, which has undergone viewpoint compensation, onto the dynamic imaging medium according to the viewpoint compensation parameters.

7. The distortion correction method for vehicle-mounted fogging imaging according to claim 6, characterized in that, During projection, spatial state data and viewpoint information are updated cyclically, and the projected image is adjusted and compensated based on the updated spatial state data and viewpoint information. Viewpoint compensation is also performed, including: During the process of compensating the projected image onto the dynamic imaging medium, the current spatial state data of the dynamic imaging medium and the current viewpoint information of the observer relative to the target imaging area are collected according to a preset update cycle, and the current spatial state data and the current viewpoint information are used as the updated spatial state data and the updated viewpoint information, respectively. The real-time imaging surface model, geometric distortions, and density distribution data are redefined based on the updated spatial state data, and the compensated projection image is adjusted according to the redefined geometric distortions and density distribution data. The projection process of the adjusted compensated projection image is performed based on the updated viewpoint information, and the compensated projection image after viewpoint compensation is projected onto the dynamic imaging medium. During the process of compensating for the projection image being projected onto the dynamic imaging medium, environmental status information inside the cabin is also collected according to a preset update cycle or in response to an environmental detection trigger signal. The environmental status information includes at least one of humidity information, air conditioning airflow status information, and driver visibility status information. When the humidity information reaches the preset humidity threshold, a humidity adjustment operation is performed. The humidity adjustment operation includes at least one of reducing the supply of atomized particles from the atomization generation component and controlling the vehicle air conditioning system to perform dehumidification or return air operation. When the air conditioning airflow state information indicates that the target imaging area is disturbed by the air conditioning airflow, at least one of the following is adjusted: airflow rate, airflow velocity, and airflow pressure difference of the constraint airflow output by the airflow adjustment component; When the driver's field of vision status information indicates that the dynamic imaging medium obstructs the driver's field of vision, the target imaging area is changed, the supply of atomized particles to the atomization generation component is reduced, or projection is stopped.

8. A distortion correction system for vehicle-mounted fogging imaging, characterized in that, include: The area determination module is used to determine the target imaging area inside the cabin in response to the in-vehicle space imaging trigger signal; The medium forming module is used to supply atomized particles to the target imaging area through the atomization generating component, and to form a constraining airflow around the target imaging area through the airflow regulating component to constrain the atomized particles, so that the atomized particles form a dynamic imaging medium with a predetermined thickness in the target imaging area. The information acquisition module is used to collect spatial state data of the dynamic imaging medium and obtain the viewpoint information of the observer in the cockpit relative to the target imaging area. The model building module is used to build a real-time imaging surface model of the dynamic imaging medium based on spatial state data, and to determine the geometric distortion when the image to be projected is projected onto the dynamic imaging medium based on the real-time imaging surface model. The image compensation module is used to perform inverse geometric pre-distortion processing on the image to be projected based on the geometric distortion, and generate a compensated projection image. The projection compensation module is used to perform viewpoint compensation on the compensated projection image based on viewpoint information, and then project the viewpoint-compensated compensated projection image onto the dynamic imaging medium. The cyclic correction module is used to cyclically update spatial state data and viewpoint information during projection, adjust and compensate the projected image based on the updated spatial state data and viewpoint information, and perform viewpoint compensation.

9. A distortion correction device for vehicle-mounted fogging imaging, characterized in that, include: At least one processor; and memory that is communicatively connected to at least one processor; The memory stores instructions that can be executed by at least one processor. When the instructions are executed by at least one processor, the at least one processor performs the distortion correction method for vehicle-mounted fogging imaging as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions thereon, characterized in that, When the computer-executable instructions are executed by the processor, they implement the distortion correction method for vehicle-mounted fogging imaging as described in any one of claims 1 to 7.

Images