Calibration method of intelligent self-closed-loop car lamp

Abstract

The invention discloses a calibration method for an intelligent self-closed-loop automobile lamp, and belongs to the technical field of automobile lighting. The method comprises the following steps: establishing a two-dimensional coordinate system of a double-lamp projection image, and carrying out parameter definition; calculating a light starting point coordinate, a brightness edge point coordinate and a midpoint coordinate between the light starting point and the brightness edge point; brightness distribution uniformity adjustment is carried out on the light starting point, the brightness edge point and a middle point between the light starting point and the brightness edge point; fusing the double-lamp projection images by adjusting the coordinates of the light starting point, calculating the number of overlapped pixel points and the brightness distribution proportion, and adjusting the double-lamp projection parameters; monitoring the brightness uniformity index of the double-lamp projection fusion image, and readjusting the double-lamp projection parameters; and acquiring an optimized double-lamp projection fusion image, and verifying a parameter adjustment effect of double-lamp projection. According to the calibration method of the intelligent self-closed-loop car lamp, precise calibration and fusion of the projection images of the double lamps are achieved, the lighting effect is improved, and the driving safety is improved.

Description

Technical Field

[0001] The present invention relates to a calibration method for an intelligent self-closed-loop vehicle lamp, belonging to the technical field of automotive lighting. Background Art

[0002] Currently, in autonomous driving and night driving, long-distance, high-brightness, and uniform lighting are crucial for improving the recognition accuracy of the vehicle lamp projection image. In traditional methods, the lighting brightness is increased by increasing the number of LED chips or raising the current and voltage, but these methods often bring problems such as increased cost, increased power consumption, and uneven lighting.

[0003] In recent years, people have begun to explore dual-lamp or multi-lamp fusion technologies. By reasonably arranging multiple vehicle lamp modules and using advanced algorithms to achieve precise calibration and fusion of the projection image, a high-brightness, long-distance, and uniform lighting effect can be achieved while maintaining low cost and low power consumption. However, the existing dual-lamp fusion technologies still have problems such as insufficient calibration accuracy and unsatisfactory fusion effect, and it is difficult to meet the requirements of autonomous driving and night driving for high-precision lighting. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a calibration method for an intelligent self-closed-loop vehicle lamp, which realizes precise calibration and fusion of the dual-lamp projection image, is beneficial to significantly improving the lighting brightness, ensuring the uniform distribution of the lighting light, thereby enhancing the lighting effect, improving the image recognition accuracy, and improving the driving safety. Compared with the traditional method, it has higher calibration accuracy, better fusion effect, stronger self-adaptability, and lower cost and power consumption.

[0005] To solve the above technical problem, the technical solution of the present invention is as follows:

[0006] The present invention provides a calibration method for an intelligent self-closed-loop vehicle lamp, characterized in that it includes the following steps:

[0007] Step S1: Taking the lower left corner of the dual-lamp projection image as the origin, establish a two-dimensional coordinate system, and define the parameters of the dual-lamp projection image. The parameters of the dual-lamp projection image include the initial coordinates of the bright points, the coordinates of the bright points, the starting coordinates of the light rays, the coordinates of the brightness edge points, and the midpoint coordinates between the starting point of the light ray and the brightness edge point of the dual-lamp projection image;

[0008] Step S2: Calculate the starting coordinates of the light ray and the coordinates of the brightness edge point according to the initial coordinates of the bright points of the dual-lamp projection image, and calculate the midpoint coordinates between the starting point of the light ray and the brightness edge point according to the starting coordinates of the light ray and the coordinates of the brightness edge point;

[0009] Step S3: Adjust the brightness distribution uniformity of the light starting point, the brightness edge point, and the midpoint between the light starting point and the brightness edge point;

[0010] Step S4: Fusion of the dual-lamp projection images by adjusting the light starting point coordinates. During the fusion process, use a high-definition camera to collect the images during the dual-lamp projection fusion. Calculate the number of overlapping pixel points and the brightness distribution ratio based on the images during the dual-lamp projection fusion. Adjust the dual-lamp projection parameters based on the number of overlapping pixel points and the brightness distribution ratio, and finally complete the fusion of the dual-lamp projection images;

[0011] Step S5: Real-time monitor the brightness uniformity index of the dual-lamp projection fusion image, readjust the dual-lamp projection parameters according to the brightness uniformity index, and optimize the dual-lamp projection fusion image;

[0012] Step S6: Use a high-definition camera to collect the optimized dual-lamp projection fusion image, verify the effect of the parameter adjustment of the dual-lamp projection. If the preset image quality is achieved, the dual-lamp projection image fusion task ends. If the preset image quality is not achieved, jump to Step S5.

[0013] Furthermore, in the said Step S2, the calculation formula of the light starting point coordinates is as follows:

[0014] When F0 x -F0 y =0, O1 x1 =F0 x , O1 y1 =F0 y ;

[0015] When F0 x >F0 y , O1 x1 =F0 x , O1 y1 =0;

[0016] When F0 x <F0 y , O1 x1 =0, O1 y1 =F0 y ;

[0017] Among them, (F0 x , F0 y ) is the initial coordinate of the bright point in the dual-lamp projection image;

[0018] (O1 x1 , O1 y1 ) is the light starting point coordinate.

[0019] Furthermore, in the said Step S2, the calculation formula of the brightness edge point coordinates is as follows:

[0020] When F0 x ≤F0 y O2 x2 = F0 x - F0 y ×tanβ, O2 y2 = F0 y ;

[0021] When F0 x > F0 y O2 x2 = F0 x , O2 y2 = F0 x ×tan(π / 2 - β)- F0 y ;

[0022] Among them, (F0 x , F0 y ) is the initial coordinate of the bright spot in the double - lamp projection image;

[0023] (O2 x2 , O2 y2 ) is the coordinate of the brightness edge point;

[0024] β is the angle of the intersection of the projection light rays of the double - lamp projection image.

[0025] Furthermore, in the step S3, the brightness distribution uniformity of the light starting point, the brightness edge point, and the mid - point between the light starting point and the brightness edge point is adjusted, specifically including the following steps:

[0026] Step S31: Obtain the current brightness values of the light starting point, the brightness edge point, and the mid - point between the light starting point and the brightness edge point through image processing technology;

[0027] Step S32: Calculate the brightness difference between the current brightness values of the light starting point, the brightness edge point, and the mid - point between the light starting point and the brightness edge point and the target brightness value;

[0028] Step S33: Adjust the projection brightness or pixel brightness response according to the brightness difference.

[0029] Furthermore, in the step S4, the calculation formula for the number of overlapping pixel points is as follows:

[0030]

[0031] Among them, N 重合 is the number of overlapping pixel points;

[0032] W is the total width of the double - lamp projection image;

[0033] H is the total height of the double - lamp projection image;

[0034] (x, y) are the pixel coordinates in the double - lamp projection image;

[0035] I1(x, y) and I2(x, y) are the pixel values of the projection images of two single lamps at the position (x, y) respectively;

[0036] δ(I1(x, y), I2(x, y)) is the Kronecker function.

[0037] Furthermore, in the step S4, the calculation formula of the brightness distribution ratio is as follows:

[0038]

[0039] where H(k) is the proportion of the pixels with brightness value k in the double - lamp projection image to the total number of pixels;

[0040] W is the total width of the double - lamp projection image;

[0041] H is the total height of the double - lamp projection image;

[0042] (x, y) are the pixel coordinates in the double - lamp projection image;

[0043] I(x, y) is the pixel brightness value of the double - lamp projection image at the position (x, y);

[0044] k is the brightness value;

[0045] L{I(x, y) = k} is the indicator function.

[0046] Furthermore, in the step S5, the brightness uniformity index of the double - lamp projection fusion image is monitored in real time, which specifically includes the following steps:

[0047] Step S51: Collect the double - lamp projection fusion image through the cameras built in the two lamps, and send the collected double - lamp projection fusion image to the image processing unit;

[0048] Step S52: The image processing unit pre - processes the double - lamp projection fusion image;

[0049] Step S53: Extract features from the pre - processed double - lamp projection fusion image;

[0050] Step S54: Calculate the brightness uniformity index of the double - lamp projection fusion image after feature extraction. The brightness uniformity index includes the standard deviation and the coefficient of variation.

[0051] Furthermore, in the step S54, the calculation formula of the standard deviation is as follows:

[0052]

[0053] Among them, σ is the standard deviation of the double-light projection image;

[0054] N is the total number of pixels in the double-light projection image;

[0055] I i is the brightness value of the i-th pixel;

[0056] μ is the average brightness of the pixels in the double-light projection image.

[0057] Furthermore, the double-light projection parameters include the light starting point coordinates, projection angle, and projection brightness. The calculation formula for the projection brightness is as follows:

[0058]

[0059] Among them, I corrected is the corrected pixel brightness;

[0060] I measured is the actually measured pixel brightness;

[0061] L avg is the average brightness value;

[0062] ΔL is the brightness difference between the current pixel point brightness and the average brightness.

[0063] Furthermore, the calculation formula for the projection angle is as follows:

[0064]

[0065] Among them, θ optimal is the optimal projection angle;

[0066] θ initial is the initial projection angle;

[0067] Δx is the offset of the double-light projection image in the horizontal direction;

[0068] Δy is the offset of the double-light projection image in the vertical direction;

[0069] β is the angle of the intersection of the projection light rays of the double-light projection image and the projection line.

[0070] Adopting the above technical solution, the present invention has the following beneficial effects:

[0071] By establishing a two-dimensional coordinate system for the double-lamp projection image, a benchmark is provided for subsequent parameter calculation and image fusion. By adjusting the brightness distribution uniformity of the light starting point, the brightness edge point, and the midpoint between the light starting point and the brightness edge point, it is beneficial to ensure the uniform brightness distribution of the light starting point, the brightness edge point, and the midpoint between the light starting point and the brightness edge point, and avoid the situation of local over-dark or over-bright. By adjusting the light starting point coordinates, and adjusting the double-lamp projection parameters based on the calculated number of overlapping pixel points and the brightness distribution ratio, the fusion of the double-lamp projection image is realized, and the quality of the double-lamp projection image is improved. By monitoring the brightness uniformity index of the double-lamp projection fusion image, and readjusting the double-lamp projection parameters according to the brightness uniformity index and verifying the parameter adjustment effect, the further optimization of the double-lamp projection fusion image is realized, which is beneficial to improving the calibration accuracy, achieving a better fusion effect, enhancing the recognition accuracy of the projection image, improving the performance and stability of the lighting system, and improving the driving safety of autonomous driving and night driving. Compared with the traditional method, the present invention has higher calibration accuracy, better fusion effect and stronger self-adaptability, and can realize the improvement of the lighting effect only through algorithm optimization and parameter adjustment, without increasing additional hardware costs, and also has lower power consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a flowchart of the calibration method for the intelligent self-closed-loop vehicle lamp of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0073] In order to make the content of the present invention be more clearly understood, the present invention will be further described in detail below according to specific embodiments in conjunction with the drawings.

[0074] As Figure 1 shown, this embodiment provides a calibration method for an intelligent self-closed-loop vehicle lamp, which includes the following steps:

[0075] Step S1: Taking the lower left corner of the double-lamp projection image as the origin 0, establish a two-dimensional coordinate system, wherein the O x axis is horizontally to the right, and the O y axis is vertically downward, providing a benchmark for subsequent parameter calculation and image fusion;

[0076] Then, define the parameters of the double-lamp projection image, including the bright point F of the double-lamp projection image, the initial bright point F0 of the double-lamp projection image, the light starting point O1, the brightness edge point O2, the midpoint O3 between the light starting point and the brightness edge point. The bright point coordinates of the double-lamp projection image are (F x , F y ), the initial bright point coordinates of the double-lamp projection image are (F0 x , F0 y ), and the light starting point coordinates are (O1 x1, O1 y1 ), the brightness edge point coordinates are (O2 x2 , O2 y2 ), and the midpoint coordinates between the light ray starting point and the brightness edge point are (O3 x3 , O3 y3 ).

[0077] Step S2: According to the initial coordinates (F0 x , F0 y ) of the bright point in the double - lamp projection image, calculate the light ray starting point coordinates (O1 x1 , O1 y1 ) and the brightness edge point coordinates (O2 x2 , O2 y2 ), and according to the light ray starting point coordinates and the brightness edge point coordinates, calculate the midpoint coordinates (O3 x3 , O3 y3 );

[0078] Specifically, the calculation formula for the light ray starting point coordinates is as follows:

[0079] When F0 x - F0 y = 0, O1 x1 = F0 x , O1 y1 = F0 y ;

[0080] When F0 x > F0 y , O1 x1 = F0 x , O1 y1 = 0;

[0081] When F0 x < F0 y , O1 x1 = 0, O1 y1 = F0 y ;

[0082] Among them, (F0 x , F0 y ) are the initial coordinates of the bright point in the double - lamp projection image;

[0083] (O1 x1 , O1 y1 ) are the light ray starting point coordinates;

[0084] Specifically, the calculation formula for the brightness edge point coordinates is as follows:

[0085] When F0 x ≤ F0 y , O2 x2= F0 x -F0 y × tanβ, O2 y2 = F0 y ;

[0086] When F0 x > F0 y , O2 x2 = F0 x , O2 y2 = F0 x × tan(π / 2 - β) - F0 y ;

[0087] Among them, (F0 x , F0 y ) is the initial coordinate of the bright spot in the double - lamp projection image;

[0088] (O2 x2 , O2 y2 ) is the coordinate of the brightness edge point;

[0089] β is the angle of the intersection of the projection light rays of the double - lamp projection image. The projection light rays of the double - lamp projection image are the light rays emitted from the light source and finally form a light spot on the road surface or other object surfaces. The projection rays are the light rays emitted from the light source and projected in a specific direction after being adjusted by an optical system (such as a lens, a mirror);

[0090] Specifically, the calculation formula for the mid - point coordinate between the light - ray starting point and the brightness edge point is as follows:

[0091]

[0092] Among them, (O1 x1 , O1 y1 ) is the light - ray starting - point coordinate;

[0093] (O2 x2 , O2 y2 ) is the coordinate of the brightness edge point;

[0094] (O3 x3 , O3 y3 ) is the mid - point coordinate between the light - ray starting point and the brightness edge point.

[0095] Step S3: Adjust the brightness distribution uniformity of the light - ray starting point O1, the brightness edge point O2, and the mid - point O3 between the light - ray starting point and the brightness edge point to ensure that the brightness of points O1, O2, and O3 is evenly distributed and avoid obvious local over - darkness or over - brightness;

[0096] Specifically, it includes the following steps:

[0097] Step S31: Obtain the current brightness values B1, B2, and B3 of the light starting point, the brightness edge point, and the midpoint between the light starting point and the brightness edge point respectively through image processing technology, where the image processing technology can adopt histogram equalization technology;

[0098] Step S32: Calculate the brightness differences between the current brightness values B1, B2, B3 of the light starting point, the brightness edge point, and the midpoint between the light starting point and the brightness edge point and the target brightness value B tar get That is, ΔB1 = N target - B1, ΔB2 = B target - B2, ΔB3 = B target - B3;

[0099] Step S33: Adjust the projection brightness or pixel brightness response according to the brightness differences, so that B1, B2, and B3 are close to B target .

[0100] Step S4: Fusion of the dual-lamp projection image by adjusting the light starting point coordinates. During the fusion process, use a high-definition camera to collect the images during the dual-lamp projection fusion process, calculate the number of overlapping pixel points and the brightness distribution ratio based on the images during the dual-lamp projection fusion process, accurately adjust the dual-lamp projection parameters based on the number of overlapping pixel points and the brightness distribution ratio, and finally complete the fusion of the dual-lamp projection image, so that the dual-lamp projection image is fused into a more complete and uniform image on the ground;

[0101] Specifically, the calculation formula for the number of overlapping pixel points is as follows:

[0102]

[0103] where N 重合 is the number of overlapping pixel points;

[0104] W is the total width of the dual-lamp projection image;

[0105] H is the total height of the dual-lamp projection image;

[0106] (x, y) is the pixel point coordinate in the dual-lamp projection image;

[0107] I1(x, y) and I2(x, y) are the pixel values of the projection images of the two single lamps at the position (x, y) respectively;

[0108] δ(I1(x, y), I2(x, Y)) is the Kronecker function. When I1(x, y) = I2(x, y), δ(I1(x, y), I2(x, y)) = 1; when I1(x, y) ≠ I2(x, y), δ(I1(x, y), I2(x, y)) = 0. It is used to determine whether the pixel values of the projection images of two single lights at the same position are the same. If they are the same, a count is made.

[0109] This formula traverses each pixel point of the double - light projection image through double summation, uses the Kronecker function δ to determine whether the pixel values of the projection images of two single lights at the corresponding positions are the same. If they are the same, a count is made. By introducing the Kronecker function δ, the formula is made more concise and easier to understand. At the same time, this formula can be applied to any two images of the same size without additional pre - processing steps.

[0110] Specifically, the calculation formula for the brightness distribution ratio is as follows:

[0111]

[0112] Among them, H(k) is the proportion of pixel points with brightness value k in the double - light projection image to the total pixel points;

[0113] W is the total width of the double - light projection image;

[0114] H is the total height of the double - light projection image;

[0115] (x, y) are the pixel point coordinates in the double - light projection image;

[0116] I(x, y) is the pixel brightness value of the double - light projection image at position (x, y);

[0117] k is the brightness value. For an 8 - bit grayscale image, the usual value range is from 0 to 255;

[0118] L{I(x, y) = k} is the indicator function. When I(x, y) = k, L{I(x, y) = k} = 1; when I(x, y) ≠ k, L{I(x, y) = k} = 0. It is used to determine whether the brightness value of each pixel point in the double - light projection image is equal to k. If it is equal, a count is made.

[0119] This formula traverses each pixel point of the double - light projection image through double summation, uses the indicator function to determine whether the brightness value of each pixel point is equal to k, divides the number of pixel points with brightness value k by the total number of pixel points to obtain the proportion of pixel points with brightness value k to the total pixel points. By introducing the indicator function, the calculation of the brightness distribution is transformed into a probability problem, which more intuitively reflects the distribution of brightness values in the projection image. In addition, this formula can be easily extended to color images, and only need to calculate the brightness distribution of each color channel separately.

[0120] Step S5: To further improve the calibration accuracy and fusion effect, an intelligent self-closed-loop feedback mechanism is introduced to monitor the brightness uniformity index of the dual-lamp projection fusion image in real time. According to the brightness uniformity index, the dual-lamp projection parameters are readjusted to optimize the dual-lamp projection fusion image. For example, the calculated brightness uniformity index is compared with a preset threshold to determine whether the quality of the dual-lamp projection fusion image meets the standard. If the brightness uniformity index exceeds the preset threshold range, the feedback mechanism is triggered to automatically readjust the dual-lamp projection parameters, which can ensure that the dual-lamp projection fusion image always remains in the best state and improves the stability and reliability of the lighting system;

[0121] Specifically, the steps for monitoring the brightness uniformity index of the dual-lamp projection fusion image in real time are as follows:

[0122] Step S51: Collect the dual-lamp projection fusion image through the cameras built into the two lamps and send the collected dual-lamp projection fusion image to the image processing unit;

[0123] Step S52: The image processing unit preprocesses the dual-lamp projection fusion image, including denoising and enhancing the contrast, to improve the accuracy of subsequent image analysis;

[0124] Step S53: Extract features from the preprocessed dual-lamp projection fusion image;

[0125] Step S54: Calculate the brightness uniformity index of the dual-lamp projection fusion image after feature extraction to quantify the brightness uniformity of the dual-lamp projection fusion image. The brightness uniformity index includes the standard deviation and the coefficient of variation. The standard deviation is used to measure the degree of dispersion of the brightness distribution of the projection image. The larger the standard deviation, the more uneven the brightness distribution. The coefficient of variation is the ratio of the standard deviation to the average brightness and is used to standardize the degree of dispersion of the brightness distribution so that it is not affected by the average brightness level;

[0126] Among them, the calculation formula for the standard deviation is as follows:

[0127]

[0128] Among them, σ is the standard deviation of the dual-lamp projection image;

[0129] N is the total number of pixels in the dual-lamp projection image;

[0130] I i is the brightness value of the i-th pixel;

[0131] μ is the average brightness of the pixels in the dual-lamp projection image;

[0132] The calculation formula for the coefficient of variation is as follows:

[0133] CV = σ / μ;

[0134] Wherein, CV is the coefficient of variation;

[0135] σ is the standard deviation of the double - lamp projection image;

[0136] μ is the average brightness of the pixels in the double - lamp projection image;

[0137] Specifically, the double - lamp projection parameters include the light starting point coordinates, projection angle, and projection brightness. For example, if the brightness in the central area of the projection image is too high and the brightness in the peripheral area is too low, the brightness uniformity can be improved by adjusting the projection angle or projection brightness of the double lamps;

[0138] Among them, the calculation formula for the projection brightness is as follows:

[0139]

[0140] Wherein, I corrected is the corrected pixel brightness;

[0141] I measured is the actually measured pixel brightness;

[0142] L avg is the average brightness value;

[0143] ΔL is the brightness difference between the current pixel point brightness and the average brightness;

[0144] Among them, the calculation formula for the projection angle is as follows:

[0145]

[0146] Wherein, θ optimal is the optimal projection angle;

[0147] θ initial is the initial projection angle;

[0148] Δx is the offset of the double - lamp projection image in the horizontal direction;

[0149] Δy is the offset of the double - lamp projection image in the vertical direction;

[0150] β is the angle of the intersection of the projection light rays of the double - lamp projection image. The projection light rays of the double - lamp projection image are the light rays emitted from the light source and finally form light spots on the road surface or other object surfaces. The projection rays are the light rays emitted from the light source and projected in a specific direction after being adjusted by the optical system.

[0151] Step S6: Use a high-definition camera to collect the optimized dual-lamp projection fusion image, verify the parameter adjustment effect of the dual-lamp projection. If the preset image quality is achieved, the dual-lamp projection image fusion task ends. If the preset image quality is not achieved, jump to step S5;

[0152] During the entire projection image fusion process, the system continuously monitors the quality change of the projection image and dynamically adjusts the optimization strategy according to the monitoring results. It can appropriately adjust and optimize the algorithm parameters according to conditions such as the specific vehicle model and lighting requirements. At the same time, it continuously collects and analyzes actual data to iteratively update the algorithm to further improve its performance and stability.

[0153] The working principle of the present invention is as follows:

[0154] Taking the lower left corner of the dual-lamp projection image as the origin, establish a two-dimensional coordinate system and define the parameters of the dual-lamp projection image; calculate the light starting point coordinates, brightness edge point coordinates, and the midpoint coordinates between the light starting point and the brightness edge point; adjust the brightness distribution uniformity of the light starting point, brightness edge point, and the midpoint between the light starting point and the brightness edge point; fuse the dual-lamp projection image by adjusting the light starting point coordinates. During the fusion process, use a high-definition camera to collect the images during the dual-lamp projection fusion process, calculate the number of overlapping pixel points and the brightness distribution ratio, and adjust the dual-lamp projection parameters. Finally, complete the fusion of the dual-lamp projection image; continuously monitor the brightness uniformity index of the dual-lamp projection fusion image, readjust the dual-lamp projection parameters according to the brightness uniformity index, and optimize the dual-lamp projection fusion image; use a high-definition camera to collect the optimized dual-lamp projection fusion image, verify the parameter adjustment effect of the dual-lamp projection. If the preset image quality is achieved, the dual-lamp projection image fusion task ends. If the preset image quality is not achieved, continue to iteratively adjust the projection parameters

[0155] By establishing a two-dimensional coordinate system for the double-lamp projection image, a benchmark is provided for subsequent parameter calculation and image fusion. By adjusting the luminance distribution uniformity of the light starting point, the luminance edge point, and the midpoint between the light starting point and the luminance edge point, it is beneficial to ensure the uniform luminance distribution of the light starting point, the luminance edge point, and the midpoint between the light starting point and the luminance edge point, and avoid the situation of local over-dark or over-bright. By adjusting the light starting point coordinates and adjusting the double-lamp projection parameters based on the calculated number of overlapping pixel points and the luminance distribution ratio, the fusion of the double-lamp projection image is achieved, and the quality of the double-lamp projection image is improved. By monitoring the luminance uniformity index of the double-lamp projection fusion image, readjusting the double-lamp projection parameters according to the luminance uniformity index, and verifying the parameter adjustment effect, the further optimization of the double-lamp projection fusion image is realized, which is beneficial to improving the calibration accuracy, achieving a better fusion effect, enhancing the recognition accuracy of the projection image, improving the performance and stability of the lighting system, and improving the driving safety of autonomous driving and night driving. Compared with the traditional method, the present invention has higher calibration accuracy, better fusion effect and stronger self-adaptability. The improvement of the lighting effect can be achieved only through algorithm optimization and parameter adjustment, without increasing additional hardware costs and with lower power consumption.

[0156] The specific embodiments described above further elaborate on the technical problems solved, the technical solutions, and the beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A calibration method for an intelligent self-closed-loop vehicle lamp, characterized in that, It includes the following steps: Step S1: Taking the lower left corner of the double-lamp projection image as the origin, establishing a two-dimensional coordinate system, and defining the parameters of the double-lamp projection image. The parameters of the double-lamp projection image include the initial coordinates of the bright points, the coordinates of the bright points, the starting coordinates of the light rays, the coordinates of the brightness edge points, and the midpoint coordinates between the starting point of the light ray and the brightness edge point of the double-lamp projection image; Step S2: Calculating the starting coordinates of the light ray and the coordinates of the brightness edge point according to the initial coordinates of the bright points of the double-lamp projection image, and calculating the midpoint coordinates between the starting point of the light ray and the brightness edge point according to the starting coordinates of the light ray and the coordinates of the brightness edge point; Step S3: Adjusting the uniformity of the brightness distribution of the starting point of the light ray, the brightness edge point, and the midpoint between the starting point of the light ray and the brightness edge point; Step S4: Fusing the double-lamp projection image by adjusting the starting coordinates of the light ray. During the fusion process, a high-definition camera is used to collect the images during the double-lamp projection fusion process. The number of overlapping pixel points and the brightness distribution ratio are calculated according to the images during the double-lamp projection fusion process. The double-lamp projection parameters are adjusted based on the number of overlapping pixel points and the brightness distribution ratio, and finally the fusion of the double-lamp projection image is completed; Step S5: Real-time monitoring the brightness uniformity index of the double-lamp projection fusion image, readjusting the double-lamp projection parameters according to the brightness uniformity index, and optimizing the double-lamp projection fusion image; Step S6: Using a high-definition camera to collect the optimized double-lamp projection fusion image, verifying the effect of the parameter adjustment of the double-lamp projection. If the preset image quality is achieved, the double-lamp projection image fusion task ends. If the preset image quality is not achieved, it jumps to Step S5.

2. The calibration method of the intelligent self-closed-loop vehicle lamp according to claim 1, wherein In the above Step S2, the calculation formula for the starting coordinates of the light ray is as follows: When F0 x -F0 y = 0, O1 x1 = Fo x , O1 y1 = F0 y ; When F0 x > F0 y then, O1 x1 = F0 x , O1 y1 = 0; When F0 x <F0 y then, O1 x1 = 0, O1 y1 = F0 y ; Among them, (F0 x , F0 y ) is the initial coordinate of the bright point in the double-light projection image; (O1 x1 , O1 y1 ) is the starting coordinate of the light ray.

3. The calibration method of the intelligent self-closed-loop vehicle lamp according to claim 2, wherein, In the above Step S2, the calculation formula for the coordinates of the brightness edge point is as follows: When F0 x ≤F0 y At this time, O2 x2 =F0 x -F0 y ×tanβ, O2 y2 =F0 y ; When F0 x >F0 y , O2 x2 =F0 x , O2 y2 =F0 x ×tan(π / 2-β)-F0 y ; Among them, (F0 x , F0 y ) is the initial coordinate of the bright point in the double-light projection image; (O2 x2 ,O2 y2 ) are the coordinates of the luminance edge points; β is the angle of the intersection of the projection light ray and the projection line of the double-lamp projection image.

4. The calibration method of the intelligent self-closed-loop vehicle lamp according to claim 3, characterized in that, In the above Step S3, adjusting the uniformity of the brightness distribution of the starting point of the light ray, the brightness edge point, and the midpoint between the starting point of the light ray and the brightness edge point specifically includes the following steps: Step S31: Obtaining the current brightness values of the starting point of the light ray, the brightness edge point, and the midpoint between the starting point of the light ray and the brightness edge point through image processing technology; Step S32: Calculating the brightness difference between the current brightness values of the starting point of the light ray, the brightness edge point, and the midpoint between the starting point of the light ray and the brightness edge point and the target brightness value; Step S33: Adjusting the projection brightness or the pixel brightness response according to the brightness difference.

5. The calibration method of the intelligent self-closed-loop vehicle lamp according to claim 4, characterized in that, In the above Step S4, the calculation formula for the number of overlapping pixel points is as follows: Among them, N 重合 is the number of overlapping pixel points; W is the total width of the double-lamp projection image; H is the total height of the double-lamp projection image; (x, y) are the pixel coordinates in the double-lamp projection image; I1(x, y) and I2(x, y) are the pixel values of the projection images of two single lamps at the position (x, y) respectively; δ(I1(x, y), I2(x, y)) is the Kronecker function.

6. The calibration method of the intelligent self-closed-loop vehicle lamp according to claim 5, wherein In the above Step S4, the calculation formula for the brightness distribution ratio is as follows: Among them, H(k) is the proportion of the pixel points with the brightness value of k in the double-lamp projection image to the total pixel points; W is the total width of the double-lamp projection image; H is the total height of the double-lamp projection image; (x, y) are the pixel coordinates in the double-lamp projection image; I(x, y) is the pixel brightness value of the dual-lamp projection image at position (x, y); k is the brightness value; L{I(x, y) = k} is the indicator function.

7. The calibration method of the intelligent self-closed-loop vehicle lamp according to claim 6, wherein, In step S5, the brightness uniformity index of the dual-lamp projection fusion image is monitored in real time, which specifically includes the following steps: Step S51: Collect the dual-lamp projection fusion image through the cameras built in the dual lamps, and send the collected dual-lamp projection fusion image to the image processing unit; Step S52: The image processing unit preprocesses the dual-lamp projection fusion image; Step S53: Extract features from the preprocessed dual-lamp projection fusion image; Step S54: Calculate the brightness uniformity index of the dual-lamp projection fusion image after feature extraction, and the brightness uniformity index includes the standard deviation and the coefficient of variation.

8. The calibration method of the intelligent self-closed-loop vehicle lamp according to claim 7, characterized in that, In step S54, the calculation formula of the standard deviation is as follows: where σ is the standard deviation of the dual-lamp projection image; N is the total number of pixels in the dual-lamp projection image; I i is the luminance value of the i-th pixel; μ is the average brightness of the pixels in the dual-lamp projection image.

9. The calibration method of the intelligent self-closed-loop vehicle lamp according to claim 8, characterized in that, The dual-lamp projection parameters include the light starting point coordinates, the projection angle, and the projection brightness. The calculation formula of the projection brightness is as follows: where I corrected is the corrected pixel luminance; I measured is the actually measured pixel luminance; L avg is the average luminance value; ΔL is the brightness difference between the current pixel point brightness and the average brightness.

10. The calibration method of the intelligent self-closed-loop vehicle lamp according to claim 9, characterized in that, The calculation formula of the projection angle is as follows: Among them, θ optimal is the optimal projection angle; θ initial is the initial projection angle; Δx is the offset of the dual-lamp projection image in the horizontal direction; Δy is the offset of the dual-lamp projection image in the vertical direction; β is the angle of the intersection of the projection light of the dual-lamp projection image and the projection ray.

Images