A lighting control method for intelligent pixel headlight

Abstract

The present invention provides a lighting control method for an intelligent pixel headlight, which integrates the light images of multiple pixel headlights projected in front of a vehicle on multiple planes. When the projection surface is a complex surface, changes in the relative positions of the pixel headlights and the complex surface may lead to projection occlusion problems or drastic changes in image resolution. It is necessary to obtain three-dimensional surface information of the multi-plane surface and establish a mapping relationship between the two-dimensional projection image and the three-dimensional complex surface, so as to achieve a multi-channel projection image with no geometric distortion and natural brightness transition, so that the picture projected by the pixel headlight meets the expected requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of smart pixel headlights, and more specifically, to a lighting control method for smart pixel headlights. Background Art

[0002] HD pixel headlights divide the projected light field into different shapes, and control the lights in related areas through the control of the lamp beads in the pixel headlights. The lighting area of ​​the pixel headlights is divided into several areas and mapped with the lamp beads in the matrix pixel headlights. Each lamp bead is individually controllable, and the divided lighting areas are also individually controllable, so the lighting of the related areas is more precise and detailed.

[0003] However, when pixel headlights project light images onto complex surfaces, the irregular geometry of the complex surface will cause the image displayed on the projection surface to be distorted, seriously affecting the visual observation effect of the observer. The pixel headlights need to cache the image for preprocessing, and the preprocessed image is then projected onto the complex surface by the pixel headlights so that the observer can view the display content of the projection image normally.

[0004] In view of this, the present invention provides a lighting control method for intelligent pixel headlights, which integrates the light images of multiple pixel headlights projected in front of a vehicle on multiple planes. When the projection surface is a complex surface, changes in the relative positions of the pixel headlights and the complex surface may lead to projection occlusion problems or drastic changes in image resolution. It is necessary to obtain three-dimensional surface information of the multi-plane surface and establish a mapping relationship between the two-dimensional projection image and the three-dimensional complex surface, so as to achieve a multi-channel projection image with the picture projected by the pixel headlights meeting the expected requirements, without geometric distortion and natural brightness transition. Summary of the invention

[0005] In order to solve the problem that the existing pixel headlights project light images onto complex surfaces, which cause distortion and deformation and affect user observation, the present invention proposes a light control method for intelligent pixel headlights. For the light fusion of multiple pixel headlights in front of a vehicle on multiple planes, when the projection surface is a complex surface, the change in the relative position of the pixel headlights and the complex surface may cause projection occlusion problems or drastic changes in image resolution. It is necessary to obtain three-dimensional surface information of the multi-plane surface and establish a mapping relationship between the two-dimensional projection image and the three-dimensional complex surface, so as to achieve a multi-channel projection image with no geometric distortion and natural brightness transition, so that the picture projected by the pixel headlights meets the expected requirements.

[0006] A lighting control method for a smart pixel headlight, characterized in that it comprises the following steps:

[0007] S1: Perform original projection of the car's pixel headlights and establish original projection coordinates of the pixel headlights; place a camera in front of the car and establish camera coordinates.

[0008] S2: The car pixel headlights are reprojected, the camera captures the reprojected coordinates, and the calibrated pixel headlight projection coordinates are obtained by comparing and adjusting the reprojected coordinates with the original projection coordinates.

[0009] S3: Based on the calibrated pixel headlight projection coordinates, the complex projection surface is reconstructed in three dimensions to obtain the target correction plane, the ideal viewpoint is determined based on the target correction plane, a viewpoint coordinate system is established at the ideal viewpoint, the conversion relationship between the world coordinate system and the viewpoint coordinate system is calculated, and the world coordinates on the complex projection surface are converted to the coordinates of the viewpoint coordinate system using the conversion relationship.

[0010] S4: The car pixel headlight projects the target projection space based on the viewpoint coordinate system. When the projection falls on the irregular surface of the target projection space, the original image pixel points of the feature points on the irregular projection surface are collected.

[0011] S5: Perform additional weight allocation based on the pixel points of the original image in S4 to obtain the brightness value of the pixel headlight at that point.

[0012] S6: Based on the brightness value obtained in S5, the original image obtained by the pixel headlight projection obtained in S4 is adjusted to obtain a multi-channel projection image without geometric distortion and with natural brightness transition.

[0013] In some embodiments, before S1, the camera is calibrated before being placed in front of the vehicle. The camera calibration is used to determine the coordinates and posture of the camera in the scene. The position of the camera on the vehicle is Ci(X,Y), and the angle between the camera and the target projection area is obtained. At each candidate position, an image is captured by the camera to ensure that the camera settings are appropriate and to use the same parameters and configuration for shooting. Based on the degree of distortion of the feedback image, the position with the best image quality is selected as the optimal position.

[0014] In some embodiments, S1 also includes the following steps: dividing the overall light image to be projected into N (N is the number of pixel headlights) sub-original images, and outputting each sub-original image to the corresponding pixel headlight to be projected on the target projection space.

[0015] Furthermore, the lighting of some lamp beads projects partial feature images in the target projection space (such as white spots in a black background), and the camera acquires local feature images of the partial feature image projection and converts them into local feature grayscale images, and the local feature grayscale images are converted into local feature binary images.

[0016] Furthermore, the lamp beads are fully lit to project a full-feature image in the target projection space, the camera acquires the full-feature image of the full-feature image projection and converts it into a full-feature grayscale image, and the full-feature grayscale image is converted into a full-feature binary image.

[0017] Furthermore, a marking algorithm is used to regard all connected components in the full feature binary image and the local feature binary image as projected features, and the coordinates of each projected feature are the coordinates of the feature in the camera image.

[0018] Furthermore, the coordinates of each projected feature correspond to the coordinates in the camera image through a binary coding scheme.

[0019] In some embodiments, S3 also includes the following steps: performing three-dimensional reconstruction of the complex projection surface based on the calibration pixel headlight projection coordinates to obtain a target correction plane, where the projected target area is the X, Y plane, and the distance from the plane to the camera is the Z axis.

[0020] In some embodiments, S4 also includes the following steps: the original image pixel point first uses a camera to obtain the brightness value of the actual projection target area, and then uses the camera to obtain the brightness value of the image to be projected.

[0021] Furthermore, the camera obtains the actual brightness value of the projection target area through Lc(Xc,Yc), which is the photometric response of a point in the target area captured by the camera, Lmin is the minimum distance from the camera to the center point of the projection target area, and Ac(Xc,Yc) represents the photometric attenuation matrix of the target area obtained by the camera. The calculation formula of the photometric attenuation matrix is: Ac(Xc,Yc)=Lmin / Lc(Xc,Yc).

[0022] Furthermore, the brightness value of the image to be projected obtained by the camera is captured by the camera to obtain the luminosity value of the image we need to project. The specific method is to capture the RGB value of each pixel of the projection, calculate the brightness value of each point through the RGB value, and convert RGB to grayscale value. The calculation formula is:

[0023] Lci(Xci,Yci)=0.299R(Xci,Yci)+0.587G(Xci,Yci)+0.144B(Xci,Yci); Lci(Xci,Yci) represents the RGB value of the light intensity value of the headlight of pixel number i at the position point of the feature image (Xci,Yci,) based on the camera coordinate system.

[0024] Furthermore, the luminance value Lpi(Xpi, Ypi) corresponding to each pixel point in the coordinate system where each pixel headlight Pi is located is obtained, and the calculation formula is: Lpi(Xpi, Ypi,)=Lci(Ac(Xc, Yc)).

[0025] In some embodiments, S5 also includes the following steps: assigning an additional weight coefficient to the pixel point of each sub-original image, the additional weight coefficient has a value range of [0, 1], and the additional weight coefficient is a weighted opacity alpha value. The alpha value is divided into 256 gray levels, and an alpha value of 0 is a fully transparent state; an alpha value of 1 is an opaque state; an alpha value of 0 to 1 is a semi-transparent state. For pixels in the non-overlapping area image, the brightness adjustment algorithm assigns them an alpha value of 1, that is, an opaque state; for pixels in the overlapping area image, the brightness adjustment algorithm makes their alpha value at a reasonable value between 0 and 1, that is, in a reasonable semi-transparent state.

[0026] In some embodiments, S6 also includes the following steps, and the formula of the brightness adjustment algorithm fusion function is:

[0027]

[0028] The x pixel is the relative position of the overlapping area, P is the gradient coefficient, p is the gradient coefficient of brightness, and α adjusts the brightness coefficient.

[0029] Beneficial effects of the present invention: The present invention proposes a lighting control method for intelligent pixel headlights, which integrates the light images of multiple pixel headlights projected in front of a vehicle on multiple planes. When the projection surface is a complex surface, changes in the relative positions of the pixel headlights and the complex surface may lead to projection occlusion problems or drastic changes in image resolution. It is necessary to obtain three-dimensional surface information of the multi-plane surface, establish a mapping relationship between the two-dimensional projection image and the three-dimensional complex surface, and achieve a multi-channel projection image with the picture projected by the pixel headlights meeting the expected requirements, without geometric distortion and natural brightness transition. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 The figure is a schematic diagram of the overall flow of the lighting control method of the smart pixel headlight of the present invention.

[0031] Figure 2 It is a schematic diagram of the three-dimensional reconstruction in S3 of the lighting control method of the smart pixel headlight of the present invention.

[0032] Figure 3 This is a schematic diagram of S6 splicing and fusion of the lighting control method of the smart pixel headlight in Example 1.

[0033] The following specific implementation manner will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0034] The following examples are described to assist the understanding of the present application, and the examples are not and should not be interpreted in any way as limiting the scope of protection of the present application.

[0035] In the following description, those skilled in the art will recognize that throughout this discussion, components may be described as separate functional units (which may include sub-units), but those skilled in the art will recognize that various components or portions thereof may be divided into separate components or may be integrated together (including within a single system or component).

[0036] At the same time, the connections between components or systems are not intended to be limited to direct connections, rather, the data between these components may be modified, reformatted, or otherwise changed by intermediate components. In addition, additional or fewer connections may be used. It should also be noted that the terms "coupled," "connected," or "input" should be understood to include direct connections, indirect connections through one or more intermediate devices, and wireless connections.

[0037] Embodiment 1:

[0038] like Figure 1 FIG. 1 is a schematic diagram of the overall process of the lighting control method of the smart pixel headlight of the present invention, and the lighting control method of the smart pixel headlight is characterized in that it includes the following steps:

[0039] S1: Perform original projection of the car's pixel headlights and establish original projection coordinates of the pixel headlights; place a camera in front of the car and establish camera coordinates.

[0040] S2: The car pixel headlights are reprojected, the camera captures the reprojected coordinates, and the calibrated pixel headlight projection coordinates are obtained by comparing and adjusting the reprojected coordinates with the original projection coordinates.

[0041] S3: Based on the calibrated pixel headlight projection coordinates, the complex projection surface is reconstructed in three dimensions to obtain the target correction plane, the ideal viewpoint is determined based on the target correction plane, a viewpoint coordinate system is established at the ideal viewpoint, the conversion relationship between the world coordinate system and the viewpoint coordinate system is calculated, and the world coordinates on the complex projection surface are converted to the coordinates of the viewpoint coordinate system using the conversion relationship.

[0042] S4: The car pixel headlight projects the target projection space based on the viewpoint coordinate system. When the projection falls on the irregular surface of the target projection space, the original image pixel points of the feature points on the irregular projection surface are collected.

[0043] S5: Perform additional weight allocation based on the pixel points of the original image in S4 to obtain the brightness value of the pixel headlight at that point.

[0044] S6: Based on the brightness value obtained in S5, the original image obtained by the pixel headlight projection obtained in S4 is adjusted to obtain a multi-channel projection image without geometric distortion and with natural brightness transition.

[0045] Before S1, the camera is calibrated before being placed in front of the car. The camera calibration is used to determine the coordinates and posture of the camera in the scene. The position of the camera on the car is Ci(X,Y), and the angle between the camera and the target projection area is obtained. At each candidate position, the camera is used to capture images to ensure that the camera settings are appropriate and use the same parameters and configuration for shooting. According to the different degrees of distortion of the feedback image, the position with the best image quality is selected as the best position.

[0046] S1 also includes the following steps: the whole picture of the light to be projected is divided into N (N is the number of pixel headlights) sub-original images, each sub-original image is output to the corresponding pixel headlight and projected on the target projection space, the lighting of some lamp beads is projected on the target projection space for partial feature images (such as white spots in a black background), the camera acquires local feature images for partial feature image projection and converts them into local feature grayscale images, the local feature grayscale images are converted into local feature binary images, the lamp beads are fully lit to project full feature images in the target projection space, the camera acquires full feature images for full feature image projection and converts them into full feature grayscale images, the full feature grayscale images are converted into full feature binary images, a marking algorithm is used to regard all connected components in the full feature binary image and the local feature binary image as projection features, the coordinates of each projection feature are the coordinates of the feature in the camera image, and the coordinates of each projection feature and the coordinates in the camera image are binary coded. Corresponding to the scheme, each lamp bead can project a feature image, and an id is assigned to the lamp bead. For example, there are 100 lamp beads, and the values ​​are assigned to the decimal numbers 1-100. The decimal numbers 1-100 are re-encoded in 8-bit binary format. For example, the binary format of the decimal number 8 is 00001000; 00001001=9; a feature image is projected for each bit. If the corresponding bit of the binary code of its ID is 1, the feature is displayed in the feature image to be projected (the function is to convert the feature image points projected by the lamp beads of decimal numbers 1-100 into binary format numbers); after projecting the feature image of the specified bit and acquiring the camera image accordingly, detect whether the feature is displayed in the light projection area. If the feature is displayed, place "1" in the corresponding bit of the ID associated with the feature, otherwise, place "0"; repeat the process to restore the ID of all features in the camera image; ensure that the feature in the projection corresponds to the feature with the same ID in the camera image.

[0047] In some embodiments, Figure 2As shown, S3 also includes the following steps: performing three-dimensional reconstruction on the complex projection surface based on the calibration pixel headlight projection coordinates to obtain a target correction plane, the projected target area is the X, Y plane, and the distance from the plane to the camera is the Z axis.

[0048] In some embodiments, S4 also includes the following steps: the original image pixel point first uses the camera to obtain the actual brightness value of the projection target area, the camera obtains the actual brightness value of the projection target area through Lc (Xc, Yc) is the photometric response of a point in the target area captured by the camera, Lmin is the minimum distance from the camera to the center point of the projection target area, Ac (Xc, Yc) represents the photometric attenuation matrix of the target area obtained by the camera, and the calculation formula of the photometric attenuation matrix is: Ac (Xc, Yc) = Lmin / Lc (Xc, Yc), and then the brightness value of the image to be projected is obtained by the camera, and the brightness value of the image to be projected obtained by the camera is captured by the camera to obtain the photometric value of the image we need to project. The method is to capture the RGB value of each pixel of the projection, calculate the brightness value of each point through the RGB value, and convert RGB to grayscale value. The calculation formula is: Lci(Xci,Yci)=0.299R(Xci,Yci)+0.587G(Xci,Yci)+0.144B(Xci,Yci); the Lci(Xci,Yci) represents the RGB value of the light intensity value from the i-pixel headlight at the position point of the feature image (Xci,Yci,) based on the camera coordinate system; the luminosity value Lpi(Xpi,Ypi) corresponding to each pixel point in the coordinate system where each pixel headlight Pi is located is obtained respectively, and the calculation formula is: Lpi(Xpi,Ypi,)=Lci(Ac(Xc,Yc)).

[0049] S5 also includes the following steps: assigning an additional weight coefficient to the pixel point of each sub-original image, the additional weight coefficient has a value range of [0, 1], the additional weight coefficient is a weighted opacity alpha value, the alpha value is divided into 256 gray levels, an alpha value of 0 is a fully transparent state; an alpha value of 1 is an opaque state; an alpha value of 0 to 1 is a semi-transparent state, for pixels in the non-overlapping area image, the brightness adjustment algorithm assigns them an alpha value of 1, that is, an opaque state; for pixels in the overlapping area image, the brightness adjustment algorithm makes their alpha value at a reasonable value between 0 and 1, that is, in a reasonable semi-transparent state.

[0050] S6 also includes the following steps: Figure 3As shown in the figure, multiple HD headlight projection images are geometrically spliced. In the splicing method, the overall projection image will have a "brightness mutation" phenomenon in a certain area at the splicing point, which directly affects the visual effect of the overall projection image. Therefore, it is necessary to adjust the brightness of the multi-channel projection image splicing point. The overlapping part of the projection image is subjected to brightness fusion processing, so that the brightness transition of the processed image near the overlapping area at the splicing point is natural, achieving the effect of "seamless splicing". First, the brightness adjustment algorithm fusion function is selected:

[0051] The x pixel point is the relative position of the overlapping area, P is the gradient coefficient, p is the gradient coefficient of the brightness, and α adjusts the brightness coefficient; F1 represents the original image of the HD headlight 1. In the non-overlapping area, the brightness is adjusted to itself. In the overlapping area, the brightness is adjusted to the product of F1 itself and the fusion function R(x). F2 represents the original image of the HD headlight 2. In the non-overlapping area, the brightness is adjusted to itself. In the overlapping area, the brightness is adjusted to the product of F itself and the fusion function [1-R(x)].

[0052] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A lighting control method for a smart pixel headlight, characterized in that : Includes the following steps: S1: Perform original projection of the car's pixel headlights and establish original projection coordinates of the pixel headlights; place a camera in front of the car and establish camera coordinates; S2: The car pixel headlight is re-projected, the camera captures the re-projected coordinates, and the calibration pixel headlight projection coordinates are obtained by comparing and adjusting the re-projected coordinates with the original projection coordinates; S3: Perform 3D reconstruction of the complex projection surface based on the calibration pixel headlight projection coordinates to obtain a target correction plane, determine the ideal viewpoint based on the target correction plane, establish a viewpoint coordinate system at the ideal viewpoint, calculate the conversion relationship between the world coordinate system and the viewpoint coordinate system, and use the conversion relationship to convert the world coordinates on the complex projection surface to the coordinates of the viewpoint coordinate system; S4: The car pixel headlight projects the target projection space based on the viewpoint coordinate system. When the projection falls on the irregular surface of the target projection space, the original image pixel points of the feature points on the irregular projection surface are collected; S5: performing additional weight allocation based on the pixel points of the original image in S4 to obtain the brightness value of the pixel headlight at that point; S6: Based on the brightness value obtained in S5, the original image obtained by the pixel headlight projection obtained in S4 is adjusted to obtain a multi-channel projection image without geometric distortion and with natural brightness transition.

2. The lighting control method of the smart pixel headlight according to claim 1, characterized in that: Before S1, the camera is calibrated before being placed in front of the car. The camera calibration is used to determine the coordinates and posture of the camera in the scene. The position of the camera on the car is Ci(X,Y), and the angle between the camera and the target projection area is obtained. At each candidate position, the camera is used to capture images to ensure that the camera settings are appropriate and use the same parameters and configuration for shooting. According to the different degrees of distortion of the feedback image, the position with the best image quality is selected as the best position.

3. The lighting control method of the smart pixel headlight according to claim 1, characterized in that: In S2, the overall light image projected in the secondary projection is divided into N sub-original images, and each sub-original image is output to the corresponding pixel headlight to be projected on the target projection space, where N is the number of pixel headlights.

4. The lighting control method of the smart pixel headlight according to claim 3, characterized in that: The lighting of some lamp beads projects partial feature images in the target projection space, and the camera acquires local feature images of the partial feature image projection and converts them into local feature grayscale images, which are converted into local feature binary images; when all lamp beads are on, full feature images are projected in the target projection space, and the camera acquires full feature images of the full feature image projection and converts them into full feature grayscale images, which are converted into full feature binary images.

5. The lighting control method of the smart pixel headlight according to claim 3, characterized in that: A marking algorithm is used to regard all connected components in the full feature binary image and the local feature binary image as projected features. The coordinates of each projected feature are the coordinates of the feature in the camera image. The coordinates of each projected feature correspond to the coordinates in the camera image through a binary coding scheme.

6. The lighting control method of the smart pixel headlight according to claim 1, characterized in that: S3 also includes the following steps: performing three-dimensional reconstruction on the complex projection surface based on the calibration pixel headlight projection coordinates to obtain a target correction plane, where the projected target area is the X, Y plane, and the distance from the plane to the camera is the Z axis.

7. The lighting control method of the smart pixel headlight according to claim 1, characterized in that: S4 also includes the following steps: the original image pixel point first uses the camera to obtain the brightness value of the actual projection target area, and then uses the camera to obtain the brightness value of the image to be projected.

8. The lighting control method of the smart pixel headlight according to claim 1, characterized in that: S5 also includes the following steps: assigning an additional weight coefficient to the pixel points of each sub-original image, the additional weight coefficient has a value range of [0, 1], and the additional weight coefficient is a weighted opacity alpha value. The alpha value is divided into 256 gray levels, and an alpha value of 0 is a fully transparent state; an alpha value of 1 is an opaque state; an alpha value of 0 to 1 is a semi-transparent state. For pixels in the non-overlapping area of ​​the image, the brightness adjustment algorithm assigns them an alpha value of 1, that is, an opaque state.

Images