A pixel-based vehicle lamp module lens calibration method based on light intensity or illumination distribution data fitting

Abstract

The application provides a pixel type vehicle lamp module lens calibration method based on light intensity or illumination distribution data fitting, comprising the following steps: lighting a preset calibration dot matrix pattern on a pixel type light source surface; each calibration dot in the calibration dot matrix has a known pixel coordinate and size; obtaining light intensity angle distribution data or illumination distribution data; analyzing the light intensity angle distribution or illumination distribution data, extracting a horizontal profile peak value, obtaining an initial horizontal field of view angle and a focal length initial value; converting the light intensity angle distribution data into a receiving surface illumination distribution according to a set distance; establishing a geometric model containing a focal length and a radial distortion coefficient, taking an angle domain residual error as a target function, and inversely calculating lens parameters by iteratively fitting a residual error between a measured light spot center and a theoretical projection point. The application realizes automatic and high-precision calibration of pixel type vehicle lamp module lens parameters, simplifies a calibration process, and reduces equipment dependence.

Description

Technical Field

[0001] This invention relates to the fields of automotive lighting, pixel-type automotive lamp modules, optical system calibration, and light distribution design, and particularly to a lens calibration method for pixel-type automotive lamp modules based on fitting light intensity or illuminance distribution data. Background Technology

[0002] With the development of features such as adaptive high beams, road projection, and interactive signal displays, more and more automotive headlights or signal lights are adopting headlight modules that combine pixel-based light-emitting devices with projection lenses. The light-emitting surface of such modules can be independently controlled by pixels or small areas, and the rear end forms a far-field light intensity or illuminance distribution through a projection lens. To achieve precise light distribution, designers need to know the lens's effective focal length, horizontal / vertical field of view, principal point position, radial distortion, and the mapping relationship between pixel coordinates and spatial angles or receiving surface coordinates.

[0003] The traditional imaging lens industry typically uses imaging measurement stations, MTF testers, calibration plate camera systems, or dedicated distortion testing equipment to measure the aforementioned parameters. However, automotive lighting manufacturers' standard testing resources mainly consist of distributed photometers, near-field / far-field photometers, light distribution screens, and IES / light intensity data processing software. Most automotive lighting manufacturers do not possess complete imaging lens testing equipment, nor is it convenient to introduce additional complex imaging testing processes during mass production or design iteration stages.

[0004] The existing methods have at least the following problems:

[0005] First, when light distribution design is based solely on the nominal field of view or structural design value of the lens, actual lens assembly errors, effective focal length deviations, and distortions can cause the theoretical pixel position to be inconsistent with the measured far-field spot position.

[0006] Second, dedicated imaging inspection equipment is expensive, has complex fixtures, and involves a heavy testing process, making it unsuitable for rapid design iterations within the existing testing systems of automotive lighting manufacturers.

[0007] Third, although distributed photometers can output IES or light intensity angle data, their traditional use is mainly to verify light distribution results, lacking a complete process of backfitting lens geometry models from photometric data.

[0008] Fourth, pixel-based modules have various light emission modes such as dot matrix, stripes, and area patterns; without repeatable calibration patterns and fitting algorithms, it is difficult to convert photometric scanning results into lens parameters that can be used for design and compensation. Summary of the Invention

[0009] How can we make full use of existing photometric measurement equipment in the automotive lighting industry and achieve automatic and high-precision calibration of lens parameters for pixel-type automotive lighting modules through a unified data fitting framework, while simplifying the calibration process and reducing equipment dependence?

[0010] A pixel-based automotive headlight module lens calibration method based on fitting light intensity or illuminance distribution data includes the following steps:

[0011] A preset calibration dot matrix pattern is illuminated on the pixel-type light source surface. The calibration dot matrix is ​​configured according to the size of the light source surface, the dot matrix density, and the order of the model to be fitted. Each calibration point in the calibration dot matrix has known pixel coordinates and size.

[0012] Acquire light intensity angular distribution data or illuminance distribution data. The light intensity angular distribution data is obtained by scanning the dot matrix pattern with a distributed photometer, and the illuminance distribution data is directly acquired by a projection screen or calibration plane in conjunction with a camera.

[0013] Analyze the light intensity angle distribution or illuminance distribution data, extract the horizontal profile peak value, and obtain the initial horizontal field of view and initial focal length values;

[0014] The light intensity angle distribution data is converted into the illuminance distribution on the receiving surface according to a set distance, or the illuminance distribution data is directly used to detect the center of the measured light spot at each calibration point on the receiving surface.

[0015] A geometric model including focal length and radial distortion coefficient is established. Based on the angular domain residual as the objective function, the lens parameters are obtained by iteratively fitting the residual between the measured spot center and the theoretical projection point.

[0016] Furthermore, in the step of converting the light intensity angular distribution data into the illuminance distribution of the receiving surface, the following conversion formula is used:

[0017] X = D·tan(θx), Y = D·tan(θy),

[0018] Where D is the set distance, θx and θy are the angles in the horizontal and vertical directions, respectively, and X and Y are the coordinates of the corresponding points on the receiving surface.

[0019] Furthermore, the step of detecting the center of the measured light spot includes:

[0020] Calculate the theoretical projection points of each calibration point on the receiving surface;

[0021] A search window is generated centered on the theoretical projection point;

[0022] In each search window, local bright areas are filtered out, and the center of the measured spot is calculated using the peak core weighted center algorithm.

[0023] Furthermore, in the step of establishing the geometric model, the pixel coordinates (u, v) are first converted into millimeter coordinates (x, y) of the light source surface:

[0024] x = (u - u0)·p,

[0025] y = (v0 - v)·p,

[0026] Where u and v are the pixel coordinates of the light source surface, (u0, v0) are the pixel coordinates of the optical center, and p is the pixel spacing.

[0027] Furthermore, the geometric model maps millimeter coordinates to the emission angle using the following formula:

[0028] r² = x² + y²

[0029] s(r) = 1 + Σ(k_i · r^(2i)), i=1,…,n, where n≥1, k_i is the radial distortion coefficient, and there is at least one radial distortion coefficient.

[0030] θx = atan((x / f)·s(r)) + Δθx,

[0031] θy = atan((y / f)·s(r)) + Δθy,

[0032] Where f is the focal length, and Δθx and Δθy are the angular offsets.

[0033] Furthermore, the inverse parameters of the fitting include: model focal length f, at least one radial distortion coefficient, horizontal angle offset Δθx, and vertical angle offset Δθy.

[0034] Furthermore, the method includes an iterative convergence judgment step: if the parameter change or residual does not reach the preset convergence threshold, the initial fitting value is updated and the model is refitted until convergence is achieved and the calibration result is output.

[0035] Furthermore, the method includes the step of applying the calibration results to ground projection compensation: inputting a ground target map, and calculating and outputting a light emission pattern that can be imported into the light source surface based on the focal length and distortion parameters of the calibration results.

[0036] Furthermore, the ground projection compensation step also includes performing perspective transformation, inverse square distance compensation, and incident angle cosine compensation on the luminous pattern based on the installation height and the vehicle's downward angle.

[0037] Furthermore, the method includes the step of applying the calibration results to the horizontal plane mode light distribution: inputting the target illuminance map on the vertical receiving plane at a fixed distance, performing back projection calculation based on the calibrated field of view and focal length mapping relationship, and outputting the horizontal mode light source surface pattern.

[0038] Furthermore, the method includes the step of applying the calibration results to simulation verification: inputting the target pattern of the light source surface, calculating the theoretical projection point position and coverage area based on the calibrated focal length and distortion parameters, and outputting simulation data for evaluating the light distribution effect.

[0039] The beneficial effects of this invention are as follows:

[0040] 1. This invention supports two data sources simultaneously: light intensity angle distribution data (IES files) acquired by a distributed photometer and illuminance distribution data collected by a projection screen in conjunction with a camera. Automotive lighting manufacturers do not need to purchase additional dedicated imaging lens calibration equipment; they can directly utilize existing distributed photometers on the production line or in the laboratory to complete lens geometry calibration, significantly reducing equipment investment and the complexity of the calibration process.

[0041] 2. The calibration process of this invention is performed directly on the assembled vehicle headlight module. The calibration result includes the combined effects of lens optical parameters, light source mounting position deviation, assembly tolerance, and system errors. Compared to the method of calibrating the lens separately and then relocating the light source, the geometric mapping relationship obtained by this invention is closer to the actual working state, and the accuracy of light distribution correction is higher.

[0042] 3. This invention can obtain seven complete parameters through a single fitting: model focal length f, at least one radial distortion coefficient, horizontal angle offset Δθx, and vertical angle offset Δθy, along with the field of view (FOV) and principal point coordinates. These parameters can be directly used for ground projection compensation, regulatory light distribution inverse calculation, light source surface pattern distortion correction, and simulation verification without repeated calibration or data conversion.

[0043] 4. This invention employs a two-level constraint strategy: First, the initial horizontal field of view and initial focal length are stably estimated using the peak value of the horizontal profile of the light intensity or illuminance distribution, providing a reliable starting point for subsequent nonlinear optimization; then, the measured spot centers of multiple calibration points are detected on the illuminance distribution of the receiving surface, providing multi-point spatial constraints. Fitting with an odd-order radial distortion model yields distortion parameters with clear physical meaning, which can be directly used for lens aberration analysis and light distribution correction.

[0044] 5. This invention defines the fitting objective function in the angular domain, that is, directly minimizing the deviation of the emission angle between the measured light spot center and the theoretical projection point. Automotive headlight beam distribution design typically uses the far-field (25 meters and above) angular distribution as the final evaluation criterion. Angular domain evaluation avoids the difficulty in interpreting residuals caused by changes in projection distance, allowing the accuracy of the calibration results to be more directly reflected in the actual beam distribution effect, which aligns with the far-field application scenarios of automotive headlights.

[0045] 6. The complete parameters obtained from the calibration of this invention can be directly applied to various pixel-based vehicle lighting function development scenarios: supporting ground projection compensation mode, horizontal receiving plane projection mode, and light source surface pattern pre-correction mode, providing basic geometric mapping for functions such as adaptive high beam, ground warning projection, and light distribution regulation optimization. Based on the calibration results, engineers can quickly complete the reverse mapping from the target light distribution to the light source surface pattern and evaluate the design effect with the help of simulation verification functions, thereby significantly shortening the design iteration cycle of pixel-based vehicle lighting. Attached Figure Description

[0046] Figure 1 This is a flowchart of the present invention.

[0047] Figure 2 A calibration fit preview image generated for the first embodiment (HD200 module IES fit).

[0048] Figure 3 A preview image of the ground pattern back projection and distortion correction generated for the first embodiment.

[0049] Figure 4 A calibration fit preview image generated for the second embodiment (HD160 module IES fit).

[0050] Figure 5 A preview image of the ground pattern back projection and distortion correction generated for the second embodiment. Detailed Implementation

[0051] To further understand the features, technical means, and specific objectives and functions achieved by the present invention, the present invention will be described in further detail below with reference to specific embodiments and accompanying drawings.

[0052] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0053] Reference Figure 1 As shown.

[0054] I. Method and Procedure

[0055] The automotive lamp optical calibration method based on light source surface dot matrix and IES (or illuminance distribution) provided by this invention includes the following steps:

[0056] Step 1: Input measurement information

[0057] The input includes a calibration point matrix coordinate table, an IES light intensity file or an illuminance surface distribution map, and an optional target projection map. The number of calibration points and their values ​​can be set by module. The point matrix coordinate table records the column coordinates, row coordinates, pixel range, and calibration point number of each calibration block on the light source surface; the IES file records the horizontal angle, vertical angle, and corresponding light intensity; the illuminance surface distribution map can be directly acquired from the screen or camera. The optional target map is used for subsequent back-projection to generate a perspective-only light source pattern or a distortion-corrected light source pattern.

[0058] Step 2: Horizontal profile peak detection and initial field of view fitting

[0059] 2.1 Read the IES file and parse the horizontal angle array, vertical angle array, and light intensity matrix.

[0060] 2.2 Construct a representative light intensity profile in the horizontal angular direction, and detect multiple horizontal peaks based on the number of calibration point arrays.

[0061] 2.3 Establish a correspondence between the coordinates of the light source plane and the detected horizontal peak angle, and first fit the center equivalent horizontal field of view, left and right edge angles and initial focal length.

[0062] The initial values ​​obtained in this step are used for subsequent receiving surface projection, search window setting, and radial model iteration, reducing the risk of global optimization getting trapped in erroneous local extrema.

[0063] Step 3: Conversion of IES to receiving surface illuminance

[0064] The angular light intensity data of the IES is converted into the illuminance distribution on the receiving surface at a set distance D. For each angular point, the correspondence between the angular domain and the coordinates of the receiving surface can be established according to X=D·tan(θx) and Y=D·tan(θy). In practice, a standard distance of 25m can be used, or other fixed distances can be set according to the test system. The converted planar data is used to search for the center of each light spot in a geometrically intuitive coordinate system.

[0065] It should be noted that this invention does not limit the illuminance of the receiving surface to be obtained from an IES file. If the illuminance surface or luminance surface distribution is obtained directly by a camera using a projection screen, diffuser, or calibration plane at the test site, as long as the illuminance surface has been geometrically calibrated, the process can proceed directly to the spot center detection and lens model fitting steps.

[0066] Step 4: Theoretical point generation, search window setting, and spot center detection

[0067] 4.1 Based on the current focal length, principal point position, and initial distortion value, project the light source surface dot matrix onto the receiving surface to obtain the theoretical position of each calibration point.

[0068] 4.2 The search window half-width is automatically calculated based on the calibration block size, horizontal field of view, distance, and spacing between adjacent points to avoid false detections caused by manual window settings.

[0069] 4.3 In each search window, filter out local bright areas that exceed the threshold, and use the weighted center of the peak core region as the center of the measured spot.

[0070] 4.4 It is preferable to use the peak core ratio or peak core centroid algorithm to reduce the influence of background light, trailing light and stray light on the center of the spot.

[0071] Step 5: Odd-order radial model fitting and iterative optimization

[0072] The measured center of the light spot is jointly fitted with the coordinates of the light source surface lattice. The fitting parameters include the model focal length f, at least one radial distortion coefficient, and horizontal angle offset Δθx and vertical angle offset Δθy. In practice, an odd-order radial projection model is used, which maps the millimeter coordinates of the light source surface to the horizontal and vertical angles after applying a radial scaling term.

[0073] The root mean square of the residual in the angular domain is preferred for fitting the target, because changes in the far-field distance of the headlights linearly amplify the planar coordinate residuals, while the angular residuals are more suitable for representing lens geometric errors.

[0074] After fitting, the theoretical points and search window are recalculated using the new parameters, and the spot center is detected again and fitted again. If the focal length change and the root mean square change of the residual are below the preset thresholds, convergence is considered achieved; otherwise, iteration continues until the maximum number of iterations is reached.

[0075] II. Geometric Model and Objective Function

[0076] Using the pixel coordinates (u, v) of the light source surface as input, and assuming the optical center (u0, v0), pixel pitch p, and focal length f, the pixel coordinates can be converted to millimeter coordinates:

[0077] x = (u - u0)·p,

[0078] y = (v0 - v)·p,

[0079] Let r² = x² + y². For the radial distortion model, the following can be used:

[0080] s(r) = 1 + Σ(k_i · r^(2i)), i=1,…,n, where n≥1, k_i is the radial distortion coefficient, and the radial distortion coefficient is at least one. In a specific embodiment of the present invention, n=4, that is...

[0081] s(r) = 1 + k1·r² + k2·r 4+k3·r 6 +k4·r 8 ;

[0082] The angle of departure can then be expressed as:

[0083] θx = atan((x / f)·s(r)) + Δθx,

[0084] θy = atan((y / f)·s(r)) + Δθy;

[0085] When projected onto a perpendicular receiving plane at a distance D, the theoretical position is:

[0086] X=D·tan(θx), Y=D·tan(θy);

[0087] The objective function can be the root mean square of the angular residuals at all valid calibration points, or the coordinate residuals of the receiving surface at a fixed distance can be used as an auxiliary evaluation. For far-field measurements of vehicle lights, angular error is preferred as the primary evaluation index.

[0088] III. Output Results and Subsequent Applications

[0089] Output files may include a results table, calibration preview image, summary file, perspective light source pattern only, distortion-corrected light source pattern, and ground mode projection preview image. The results table can record a results summary, horizontal peak value, light source surface lattice, theoretical points, test points, residual table, iteration table, and distortion parameters.

[0090] Reference Figure 2-3 As shown, Example 1: HD200 module IES fitting

[0091] This embodiment uses HD200 input data for verification. The input files include a 15×7 dot matrix coordinate table, an IES light intensity file, and a ground target map. The dot matrix in this embodiment contains 15 columns × 7 rows, totaling 105 calibration points. It should be noted that 105 points is only the dot matrix size used in this embodiment; this method does not limit the number or distribution of calibration points. The algorithm is executed according to the aforementioned steps 1 to 5, including: peak detection on the IES horizontal profile, fitting the initial horizontal field of view, detecting the center of each calibration spot on the receiving surface, and iteratively fitting the k1-k4 radial distortion model.

[0092] The calibration results of this embodiment are shown in the table below:

[0093]

[0094] Reference Figure 4-5 As shown, Example 2: HD160 module IES fitting

[0095] This embodiment uses HD160 input data for verification. The input files include a 15×7 dot matrix coordinate table, an IES light intensity file, and a ground target map. This embodiment also uses 105 calibration points. The algorithm outputs 105 effective light spots, indicating that all calibration points in this embodiment were successfully detected and participated in the fitting.

[0096] The calibration results of this embodiment are shown in the table below:

[0097]

[0098] Optional implementation methods

[0099] In other alternative embodiments of the present invention, those skilled in the art can make the following adjustments according to actual needs.

[0100] Optional Implementation Method 1: Calibration Pattern Design

[0101] A dot matrix of light source surface with several columns × several rows and several pixels per point is used as the calibration pattern. The dot matrix covers most of the effective width of the light source surface and takes multiple row positions in the vertical direction in order to constrain distortion in both the horizontal and vertical directions simultaneously.

[0102] Optional Implementation Method 2: Order Selection of the Distortion Model

[0103] When the residuals of a low-order distortion model exceed a preset threshold, a higher-order k-term is automatically attempted; if the higher-order term does not significantly improve the root mean square or stability of the residuals, the lower-order model is maintained to avoid overfitting.

[0104] Optional Implementation Method 3: Ground Pattern Projection

[0105] For the ground mode module, input the installation height and downward angle, back-project the ground target image onto the light source surface according to perspective geometry, and overlay radial distortion compensation and illuminance compensation.

[0106] Optional Implementation Mode 4: Image Correction Mode

[0107] For low-resolution light source area signal maps, continuous grayscale correction, nearest neighbor binary correction, or region correction mode that preserves linewidth can be provided.

[0108] Optional Implementation Method 5: Parameter Adaptation Generalization

[0109] For modules with different resolutions, pixel pitches, or field of view, only the light source width and height, pixel pitch, initial field of view, and distance parameters can be adjusted, while retaining the IES fitting process.

[0110] Example 3: Calibration Application at a Standard Distance of 25 Meters

[0111] This embodiment demonstrates a typical application at a standard distance of 25 meters. In this embodiment, IES data is converted to receiver surface illuminance according to D=25 m. After detecting the center of the light spot, it is fitted according to the aforementioned model formula. After the fitting is completed, the root mean square of the angular residual is checked; if it is less than 0.01°, the calibration is considered qualified. This embodiment verifies the stability and repeatability of the method in a real standard test bench environment.

[0112] Example 4: Calibration based on direct illuminance surface distribution

[0113] This embodiment demonstrates an application using an alternative input source. Instead of using IES files, the illuminance surface distribution map is directly captured by a camera on a projection screen. This illuminance surface has been pre-calibrated geometrically. The algorithm skips step 3 (IES conversion) and directly executes steps 4 (spot center detection) and 5 (model iterative fitting). This embodiment proves that the invention is not limited to IES as the sole input source and can be directly adapted to on-site acquisition modes.

[0114] The embodiments described above merely illustrate the optimal implementation of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A pixel-based automotive headlight module lens calibration method based on fitting light intensity or illuminance distribution data, characterized in that, Includes the following steps: A preset calibration dot matrix pattern is illuminated on the pixel-type light source surface. The calibration dot matrix is ​​configured according to the size of the light source surface, the dot matrix density, and the order of the model to be fitted. Each calibration point in the calibration dot matrix has known pixel coordinates and size. Acquire light intensity angular distribution data or illuminance distribution data. The light intensity angular distribution data is obtained by scanning the dot matrix pattern with a distributed photometer, and the illuminance distribution data is directly acquired by a projection screen or calibration plane in conjunction with a camera. Analyze the light intensity angle distribution or illuminance distribution data, extract the horizontal profile peak value, and obtain the initial horizontal field of view and initial focal length values; The light intensity angle distribution data is converted into the illuminance distribution on the receiving surface according to a set distance, or the illuminance distribution data is directly used to detect the center of the measured light spot at each calibration point on the receiving surface. A geometric model including focal length and radial distortion coefficient is established. Based on the angular domain residual as the objective function, the lens parameters are obtained by iteratively fitting the residual between the measured spot center and the theoretical projection point.

2. The pixel-based automotive headlight module lens calibration method based on light intensity or illuminance distribution data fitting according to claim 1, characterized in that, In the step of converting the light intensity angle distribution data into the illuminance distribution of the receiving surface, the following conversion formula is used: X = D·tan(θx), Y = D·tan(θy), Where D is the set distance, θx and θy are the angles in the horizontal and vertical directions, respectively, and X and Y are the coordinates of the corresponding points on the receiving surface.

3. The pixel-based automotive headlight module lens calibration method based on light intensity or illuminance distribution data fitting according to claim 1, characterized in that, The steps for detecting the center of the measured light spot include: Calculate the theoretical projection points of each calibration point on the receiving surface; A search window is generated centered on the theoretical projection point; In each search window, local bright areas are filtered out, and the center of the measured spot is calculated using the peak core weighted center algorithm.

4. The pixel-based automotive lamp module lens calibration method based on light intensity or illuminance distribution data fitting according to claim 1, characterized in that, In the step of establishing the geometric model, the pixel coordinates (u, v) are first converted into millimeter coordinates (x, y) of the light source surface: x = (u - u0)·p, y = (v0 - v)·p, Where u and v are the pixel coordinates of the light source surface, (u0, v0) are the pixel coordinates of the optical center, and p is the pixel spacing.

5. The pixel-based automotive headlight module lens calibration method based on light intensity or illuminance distribution data fitting according to claim 4, characterized in that, The geometric model maps millimeter coordinates to the angle of incidence using the following formula: r² = x² + y² s(r) = 1 + Σ(k_i · r^(2i)), i=1,…,n, where n≥1, k_i is the radial distortion coefficient, and there is at least one radial distortion coefficient. θx = atan((x / f)·s(r)) + Δθx, θy = atan((y / f)·s(r)) + Δθy, Where f is the focal length, and Δθx and Δθy are the angular offsets.

6. The pixel-based automotive lamp module lens calibration method based on light intensity or illuminance distribution data fitting according to claim 1, characterized in that, The inverse parameters of the fitting include: model focal length f, at least one radial distortion coefficient, horizontal angle offset Δθx, and vertical angle offset Δθy.

7. The pixel-based automotive headlight module lens calibration method based on light intensity or illuminance distribution data fitting according to claim 1, characterized in that, Further, it includes an iterative convergence judgment step: if the parameter change or residual does not reach the preset convergence threshold, the initial fitting value is updated and the model is refitted until convergence is achieved and the calibration result is output.

8. The pixel-based automotive headlight module lens calibration method based on light intensity or illuminance distribution data fitting according to claim 1, characterized in that, The method further includes the step of applying the calibration results to ground projection compensation: inputting a ground target map, calculating and outputting a light emission pattern that can be imported into the light source surface based on the focal length and distortion parameters of the calibration results.

9. A pixel-based automotive lamp module lens calibration method based on fitting light intensity or illuminance distribution data according to claim 8, characterized in that, The ground projection compensation step also includes performing perspective transformation, inverse square distance compensation, and cosine of incident angle compensation on the luminous pattern based on the installation height and vehicle downward angle.

10. The pixel-based automotive lamp module lens calibration method based on light intensity or illuminance distribution data fitting according to claim 1, characterized in that, The process further includes applying the calibration results to the horizontal plane mode light distribution: inputting the target illuminance map on the vertical receiving plane at a fixed distance, performing back projection calculations based on the calibrated field of view and focal length mapping relationship, and outputting the horizontal mode light source surface pattern.

11. The pixel-based automotive lamp module lens calibration method based on light intensity or illuminance distribution data fitting according to claim 1, characterized in that, The process further includes the step of applying the calibration results to simulation verification: inputting the target pattern of the light source surface, calculating the theoretical projection point position and coverage based on the calibrated focal length and distortion parameters, and outputting simulation data for evaluating the light distribution effect.

Images