A DLP car lamp adaptive lighting mode switching method and system

Abstract

The present application relates to the technical field of adaptive lighting, in particular to a DLP vehicle lamp adaptive lighting mode switching method and system, the method comprising the following steps: collecting front image and attitude data, calculating light shape boundary gradient, dark area stray light leakage and bright area illumination decay, determining lens contamination and inverting the maximum diffuse scattering angle, calculating boundary retreat width, micromirror gradual change duty cycle and compensation driving current, checking anti-dazzle parameters according to light spot feedback, establishing feedforward mapping in association with mileage, temperature and wiper state and executing mode switching.In the present application, the scattering and absorption deviation caused by lens contamination is converted into a degradation index to uniformly constrain dark area leakage suppression and bright area illumination recovery by means of the diffuse angle, the boundary retreat, gradual change bandwidth and lateral spread angle are corrected in combination with light spot feedback, the anti-dazzle parameters are converged with the projection state, the feedforward reference is formed by means of vehicle working condition association, the opposite glare is reduced and effective illumination is maintained, and the mode switching adaptability and stability are improved.

Description

Technical Field

[0001] This invention relates to the field of adaptive lighting technology, and in particular to a method and system for switching adaptive lighting modes in DLP vehicle lights. Background Technology

[0002] The field of adaptive lighting technology encompasses the core aspect of dynamically adjusting the distribution characteristics of light sources based on changes in the external environment. This technology involves acquiring information about the road conditions and target locations around the vehicle using external environment acquisition equipment, and adjusting the illumination angle, brightness level, and light pattern profile of the underlying light source to provide a dynamic lighting beam that matches the current driving scenario. Traditional DLP vehicle headlight adaptive lighting mode switching methods refer to the technical aspects of changing the projected lighting pattern of digital light processing headlights under different driving conditions. Traditional methods typically employ a front-mounted complementary metal-oxide-semiconductor (CMOS) image sensor to continuously acquire image frames of the road ahead and transmit the image data. Inside the main control chip, the edge detection operator extracts the physical boundary coordinates of the lane line pixels and oncoming vehicles in the image. Based on the coordinate difference, the straight-line distance and relative deflection angle of the target object are calculated. Then, the micromirror flip state matrix that matches the current distance and angle values ​​is retrieved from the pre-burned light pattern mapping two-dimensional table. Finally, the corresponding pulse width modulation voltage signal is output to directly drive the digital micromirror device array, controlling the micron-level aluminum reflective mirrors at specific positions inside the array to flip at specific mechanical angles along the diagonal hinges. By changing the reflection path of the physical mirror, the pixel-level light spot arrangement of the light-emitting diodes projected onto the actual road surface is switched.

[0003] Existing DLP headlights switch micromirror states based on the image ahead and the preset light pattern mapping relationship. The control logic uses the projection reference under clean lens conditions as a reference. After dust, oil film and fine particulate contamination layer form on the lens surface, light is diffused, scattered and absorbed, the cutoff line between light and dark gradually becomes blurred, light energy in the bright area leaks into the dark area, stray light in the dark area is enhanced and causes oncoming glare, and the effective illuminance below the cutoff line is simultaneously reduced. Electronic self-test and environmental perception may still determine that the output is normal, the anti-glare state deviates from the control command, and the lighting mode switching is difficult to take into account both oncoming anti-glare and the visibility of the road surface of the vehicle. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and propose a DLP vehicle headlight adaptive lighting mode switching method and system.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a DLP vehicle headlight adaptive lighting mode switching method, comprising the following steps:

[0006] S1: Based on the digital image of the front field of view and the data from the vehicle attitude sensor, a set of mapped pixels is obtained, and the smoothness of the gradient of the projection light shape boundary, the amount of stray light leakage in the dark area of ​​the lens, and the amount of effective illuminance attenuation in the bright area of ​​the lens are calculated.

[0007] S2: Calculate the leakage value deviation between the stray light leakage amount in the dark area of ​​the lens and the preset safety anti-glare threshold. Use the leakage value deviation as an indicator to evaluate the degree of pollution. Combine the numerical evolution characteristics of the gradient smoothness of the projection light shape boundary and the effective illuminance attenuation amount in the bright area of ​​the lens with the preset lens pollution degradation conditions, mark the pollution degradation judgment status, establish an optical degradation inverse filter approximation model, and calculate the maximum diffuse scattering angle.

[0008] S3: Based on the maximum dispersion scattering angle, the width of the light shape boundary retreat compensation pixel and the micromirror drive duty cycle after gradient assignment are obtained. Combined with the global illuminance compensation coefficient determined based on the effective illuminance attenuation of the lens bright area, the amplification duty cycle value is obtained. Based on the difference between the amplification duty cycle value and the preset rated output upper limit, the illuminance compensation drive current is obtained, and the corrected low beam anti-glare execution parameters are established.

[0009] S4: Call the modified low beam anti-glare execution parameters to extract the optical feedback value of the road surface spot to obtain the recalculated attenuation index. Based on the recalculated attenuation index, adjust the width of the light shape boundary retreat compensation pixel, the width of the driving gradient band and the light shape side illumination spread angle to establish the verification anti-glare parameters. Combine the dynamic tracking coordinates to generate the convergent anti-glare compensation parameters.

[0010] S5: Obtain the current vehicle's cumulative mileage, ambient temperature characteristics, and wiper operating status, and input the convergent anti-glare compensation parameters together with the current vehicle's cumulative mileage, ambient temperature characteristics, and wiper operating status into an associated mapping table. Call the associated mapping table to obtain the baseline feedforward input and perform adaptive lighting mode switching.

[0011] The present invention is improved in that the specific steps of S1 are as follows:

[0012] S111: Acquire digital images of the forward field of view and vehicle relative distance and relative speed data. Use a multi-source data fusion algorithm to combine the two-dimensional pixel coordinates of the digital images of the forward field of view with the relative distance and relative speed data to extract three-dimensional coordinate positions. Acquire vehicle posture sensor data and external parameter matrix. Calculate angle compensation deviation based on the three-dimensional coordinate positions, vehicle posture sensor data, and external parameter matrix. Establish a spatial mapping matrix. Call the spatial mapping matrix to map the target pixel region of the digital images of the forward field of view to the micromirror control array to obtain a mapped pixel set.

[0013] S112: Based on the mapped pixel set, scan the grayscale value along the direction perpendicular to the preset ideal light shape cutoff line, calculate the brightness difference between adjacent pixels, obtain the gradient features of the projected light shape boundary, extract the optical angular resolution span of the gradient features of the projected light shape boundary, and obtain the smoothness of the gradient of the projected light shape boundary.

[0014] S113: Obtain the average gray value of the mapped pixel set in the target dark area and the preset photometric calibration parameters. Calculate the illuminance conversion value based on the average gray value and the photometric calibration parameters to obtain the stray light leakage in the dark area of ​​the lens. Extract the average illuminance estimate of the illuminated bright area of ​​the mapped pixel set in the core road surface. Call the preset calibration illuminance benchmark to calculate the illuminance difference between the calibration illuminance benchmark and the average illuminance estimate to obtain the effective illuminance attenuation in the bright area of ​​the lens.

[0015] The present invention is improved in that the specific steps of S2 are as follows:

[0016] S211: Collect vehicle height sensor data and suspension displacement sensor data. Under the condition that the vehicle driving posture is in a stable state based on the vehicle height sensor data and the suspension displacement sensor data, call the preset safety anti-glare threshold, calculate the difference between the stray light leakage amount in the dark area of ​​the lens and the safety anti-glare threshold, and obtain the leakage value deviation. Under the condition that the leakage value deviation exceeds the safety limit, combine the numerical evolution characteristics of the gradient smoothness of the projection light shape boundary and the degradation condition of the effective illuminance attenuation in the bright area of ​​the lens, and mark the pollution degradation judgment status indicator.

[0017] S212: Based on the pollution degradation judgment status identifier, the influence of the pollution adhesion layer on the lens surface on the diffusion of the headlight beam is regarded as a spatial probability distribution. The spatial distribution ratio is extracted according to the gradient smoothness of the projected light shape boundary and the diffusion radius distribution weight is calculated. The absorption conversion ratio is extracted according to the effective illuminance attenuation of the lens bright area and the light energy absorption conversion coefficient is calculated. Combining the diffusion radius distribution weight, the light energy absorption conversion coefficient and the preset two-dimensional point diffusion function analytical correlation scattering parameters, an optical degradation inverse filtering approximation model is established.

[0018] S213: The degree of physical optical link contamination degradation is analyzed by reverse analysis using the optical degradation inverse filtering approximation model. The extended boundary mapping value is calculated by combining the internal model boundary conversion parameters of the optical degradation inverse filtering approximation model, and the maximum diffusion scattering angle is obtained.

[0019] The present invention is improved in that the process of obtaining the maximum dispersion scattering angle is specifically as follows:

[0020] Call the preset baseline smoothness, calculate the difference between the gradient smoothness of the projected light shape boundary and the baseline smoothness, call the preset maximum tolerance limit, and determine the diffusion radius distribution weight based on the ratio of the difference to the maximum tolerance limit;

[0021] The attenuation ratio is calculated by calling the preset total luminous flux reference and the ratio of the effective illuminance attenuation in the bright area of ​​the lens to the total luminous flux reference. The value that matches the attenuation ratio is extracted as the light energy absorption conversion coefficient based on the preset mapping table.

[0022] The variance and amplitude of the two-dimensional point spread function are replaced by the dispersion radius distribution weight and the light energy absorption conversion coefficient, respectively, to obtain the associated scattering parameters. The optical degradation inverse filtering approximation model is constructed based on the two-dimensional point spread function after replacing the parameter values.

[0023] Extract the boundary coordinates that have attenuated to a preset ratio in the optical degradation inverse filtering approximation model, and calculate the distance from the origin to the boundary coordinates to obtain the degree of physical optical link contamination degradation;

[0024] The extended boundary mapping value is obtained by multiplying the degree of contamination and degradation of the physical optical link by the scaling factor in the internal model boundary conversion parameter of the optical degradation inverse filtering approximation model, and the extended boundary mapping value is set as the maximum diffusion scattering angle.

[0025] The present invention is improved in that the specific steps of S3 are as follows:

[0026] S311: Call the preset vehicle light space mapping resolution, calculate the spatial pixel offset based on the maximum diffusion angle and the vehicle light space mapping resolution to obtain the light shape boundary clearance compensation pixel width, and extend the anti-glare dark area boundary in the projected light shape to the coordinate side of the high brightness illumination area based on the light shape boundary clearance compensation pixel width to establish a light shape reserved clearance boundary. Define a brightness gradient zone area in the light shape high brightness transition area adjacent to the light shape reserved clearance boundary, calculate the gradient gray scale ratio between the edge position near the light shape reserved clearance boundary and the core bright area position using a non-linear smooth curve, and assign a transition gradient value to the micromirror drive duty cycle based on the gradient gray scale ratio to obtain the micromirror drive duty cycle after gradient assignment.

[0027] S312: Call the preset light shape calibration illuminance reference, calculate the illuminance compensation ratio based on the calibration illuminance reference and the effective illuminance attenuation of the lens bright area to obtain the global illuminance compensation coefficient, amplify the micromirror drive duty cycle after the gradient assignment in the core road lighting bright area corresponding to the mapped pixel set according to the global illuminance compensation coefficient, obtain the amplified duty cycle value, calculate the difference between the amplified duty cycle value and the preset rated output upper limit to obtain the residual illuminance gap, increase the headlight source drive current according to the residual illuminance gap to obtain the illuminance compensation drive current;

[0028] S313: Combine the pixel width of the light shape boundary retreat compensation, the value of the magnification duty cycle, and the illuminance compensation drive current to establish the corrected low beam anti-glare execution parameters.

[0029] The present invention is improved in that the specific steps of S4 are as follows:

[0030] S411: Call the corrected low beam anti-glare execution parameters, output the micromirror control matrix to drive the headlight beam projection, call the forward-looking camera to capture the current field of view during the continuous detection period, extract the road surface light spot optical feedback value in the current field of view, and recalculate the stray light leakage of the lens dark area and the road illuminance attenuation amplitude in the lens state according to the road surface light spot optical feedback value to obtain the recalculated attenuation index.

[0031] S412: Based on the recalculated attenuation index, under the condition that the stray light leakage in the dark area of ​​the lens is greater than the safety anti-glare threshold, calculate the difference between the stray light leakage in the dark area of ​​the lens and the safety anti-glare threshold to obtain the leakage excess value. Increase the width of the light shape boundary retreat compensation pixel according to the leakage excess value, and calculate the boundary gradient difference between the current frame and the previous frame to obtain the boundary change value. Increase the width of the driving gradient band according to the boundary change value. Under the condition that the road surface illuminance attenuation is lower than the compensation available lower limit and the total output power reaches the thermal protection upper limit, adjust the local light energy to narrow the light shape side illumination angle, and integrate the adjusted light shape boundary retreat compensation pixel width, driving gradient band width and light shape side illumination angle to establish the verification anti-glare parameters.

[0032] S413: Obtain the offset value of the angle change of the oncoming vehicle during the meeting process, calculate the dynamic tracking coordinates based on the offset value, and generate convergent anti-glare compensation parameters by superimposing the verification anti-glare parameters on the boundary of the light-shaped anti-glare dark area of ​​the position movement based on the dynamic tracking coordinates.

[0033] The present invention is improved in that the process of establishing and verifying the anti-glare parameters is specifically as follows:

[0034] The preset upper limit of the oncoming visual safety illuminance is used as the safety anti-glare threshold, and the difference between the stray light leakage in the dark area of ​​the lens and the safety anti-glare threshold is calculated to obtain the leakage tolerance value.

[0035] The preset optical dot spread rate is used as the backoff gain coefficient. The leakage excess value is multiplied by the backoff gain coefficient to obtain the pixel increment. The pixel increment is then superimposed on the optical boundary backoff compensation pixel width.

[0036] Extract the grayscale slope of the newly added pixel area corresponding to the increase in the width of the light-shaped boundary retreat compensation pixel to establish the boundary change value, and expand the width of the driving gradient band according to the boundary change value;

[0037] The preset standard minimum road surface illuminance index is used as the lower limit of the compensation available, and the preset power critical point corresponding to the highest temperature that the vehicle lamp can withstand is obtained as the upper limit of the thermal protection.

[0038] The lateral edge pixel columns of the micromirror control array are closed one by one, and the light energy is allocated and transferred to the central region of the projected light pattern. When the road surface illuminance attenuation is lower than the lower limit of the compensation available, the lateral illumination spread angle of the narrowed light pattern is determined.

[0039] The anti-glare parameters are established by integrating the pixel width of the light shape boundary retreat compensation after the pixel increment is integrated, the width of the driving gradient band after the value is increased, and the light shape side illumination angle after the value is narrowed.

[0040] The present invention is improved in that the specific steps of S5 are as follows:

[0041] S511: Calculate the signal-to-noise ratio of the image data of the current field of view. Under the condition that the signal-to-noise ratio of the image data is lower than the preset judgment threshold and causes the closed-loop control link to break, freeze the partition correction adjustment of the micromirror control array and output the basic safety low beam mode command to trigger the degradation protection. Obtain the vehicle's cumulative mileage, ambient temperature characteristics and wiper operating status. Record the convergent anti-glare compensation parameters together with the cumulative mileage, ambient temperature characteristics and wiper operating status into a non-volatile storage device and establish an association mapping table.

[0042] S512: Identify oncoming vehicles during the subsequent driving period, combine the current cumulative driving mileage with the ambient temperature characteristics and the wiper operating status to match the context conditions, call the association mapping table to extract historical state parameters, and obtain the baseline feedforward input;

[0043] S513: Based on the reference feedforward input, reconstruct the anti-glare shape control command, drive the micromirror control array to deflect according to the anti-glare shape control command, and perform adaptive illumination mode switching.

[0044] The present invention is improved in that the process of retrieving historical state parameters by calling the associated mapping table is specifically as follows:

[0045] The lowest effective resolution baseline of the vehicle-mounted visual perception device under extreme recognition conditions is extracted to set the judgment threshold.

[0046] Calculate the absolute difference between the current cumulative mileage and the historical cumulative mileage recorded in the associated mapping table;

[0047] Calculate the absolute temperature difference between the current ambient temperature feature and the historical ambient temperature features recorded in the association mapping table;

[0048] The current wiper operating status is compared with the operating speed of the historical wiper operating status recorded in the associated mapping table.

[0049] The preset physical pollution stability tolerance range and the preset lens deformation stability tolerance range are invoked. Under the conditions that the absolute difference in mileage is less than the physical pollution stability tolerance range, the absolute difference in temperature is less than the lens deformation stability tolerance range, and the consistency of the operating gear is exactly the same, the context conditions are confirmed to be successfully matched.

[0050] Extract the convergent anti-glare compensation parameters associated with the successfully matched items in the association mapping table, set the convergent anti-glare compensation parameters as historical state parameters, and establish a benchmark feedforward input.

[0051] A DLP vehicle headlight adaptive lighting mode switching system is provided, the system being used to implement the aforementioned DLP vehicle headlight adaptive lighting mode switching method, the system comprising:

[0052] The multi-source state perception module derives a mapped pixel set based on the forward field of view digital image and vehicle attitude sensor data, and calculates the gradient smoothness of the projected light shape boundary, the amount of stray light leakage in the dark area of ​​the lens, and the amount of effective illuminance attenuation in the bright area of ​​the lens.

[0053] The physical degradation assessment module calculates the leakage value deviation between the stray light leakage amount in the dark area of ​​the lens and the preset safety anti-glare threshold. The leakage value deviation is used as an indicator to assess the degree of contamination. Combined with the numerical evolution characteristics of the gradient smoothness of the projection light shape boundary and the effective illuminance attenuation in the bright area of ​​the lens, it is compared with the preset lens contamination degradation conditions, and the contamination degradation judgment status is marked. An optical degradation inverse filtering approximation model is established, and the maximum diffuse scattering angle is calculated.

[0054] The partition compensation calculation module calculates the pixel width of the light shape boundary retreat compensation and the micromirror drive duty cycle after the gradient assignment based on the maximum dispersion scattering angle. It also calculates the amplification duty cycle value by combining the global illuminance compensation coefficient determined based on the effective illuminance attenuation of the lens bright area. Finally, it calculates the illuminance compensation drive current based on the difference between the amplification duty cycle value and the preset rated output upper limit, and establishes the corrected low beam anti-glare execution parameters.

[0055] The closed-loop verification and tracking module calls the modified low beam anti-glare execution parameters to extract the optical feedback value of the road surface spot and obtain the recalculated attenuation index. Based on the recalculated attenuation index, it adjusts the width of the light shape boundary retreat compensation pixel, the width of the driving gradient band and the light shape side illumination spread angle to establish the verification anti-glare parameters. It then combines the dynamic tracking coordinates to generate the convergent anti-glare compensation parameters.

[0056] The memory feedforward scheduling module obtains the current vehicle's cumulative mileage, ambient temperature characteristics, and wiper operating status. It then records the convergent anti-glare compensation parameters along with the current vehicle's cumulative mileage, ambient temperature characteristics, and wiper operating status into an associated mapping table. The module calls the associated mapping table to obtain the baseline feedforward input and performs adaptive lighting mode switching.

[0057] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0058] In this invention, a mapped pixel set is constructed based on image and vehicle posture data. The gradient of the light shape boundary, stray light leakage in dark areas, and illuminance attenuation in bright areas are jointly analyzed. The scattering and absorption deviations caused by lens contamination are transformed into degradation indicators. The contamination state is determined by combining the anti-glare threshold and evolution characteristics, and the maximum diffuse scattering angle is inverted. Based on this, the boundary setback width, micromirror gradient duty cycle, and illuminance compensation driving amount are obtained, so that the suppression of leakage in dark areas and the restoration of illuminance in bright areas form a unified constraint. The boundary width, gradient bandwidth, and lateral expansion angle are corrected by road surface light spot feedback. The convergence parameters are correlated with mileage, temperature, and wiper status to provide a feedforward reference for switching to adapt to lens degradation and environmental changes, reduce oncoming glare, and maintain road lighting. Attached Figure Description

[0059] Figure 1 This is a flowchart of the method of the present invention;

[0060] Figure 2 This is a flowchart of S1 of the present invention;

[0061] Figure 3 This is a flowchart of S2 of the present invention;

[0062] Figure 4 This is a flowchart of S3 of the present invention;

[0063] Figure 5This is a flowchart of S4 of the present invention;

[0064] Figure 6 This is a flowchart of S5 of the present invention;

[0065] Figure 7 This is a system module relationship diagram of the present invention. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0067] This embodiment provides a DLP vehicle headlight adaptive lighting mode switching method and system. During nighttime vehicle operation, when the vehicle is using DLP headlights to form a low-beam anti-glare pattern, a forward-facing camera continuously acquires digital images of the forward field of view corresponding to the road, lane edges, oncoming vehicles, and road surface light spots. Simultaneously, a vehicle attitude sensor outputs pitch, roll, and height change data. The headlight beam projection state is continuously adjusted based on vehicle attitude, oncoming vehicle position, lens contamination status, and road surface illuminance feedback. This embodiment includes a DLP vehicle headlight adaptive lighting mode switching method and a DLP vehicle headlight adaptive lighting mode switching system associated with this method. The method embodiment serves as a detailed implementation example, while the system embodiment illustrates the corresponding relationship between each functional module and the aforementioned method flow.

[0068] In actual vehicle operation, dust and water adhesion to the lens surface, changes in optical state caused by temperature, and environmental conditions corresponding to windshield wiper operation can cause dark area leakage, blunting of the cutoff line, and attenuation of illuminance in bright areas in the headlight projection. If the lighting mode is switched solely based on a fixed low beam template, the lack of compensation for pollution degradation between the anti-glare dark area and the position of oncoming vehicles will lead to a mismatch between the projection boundary and the actual optical degradation. This embodiment incorporates the forward field of view digital image, vehicle posture data, dark area leakage, bright area attenuation, boundary smoothing changes, and historical correlation status into the same method flow, enabling the low beam anti-glare execution parameters to form a continuous data closed loop between detection, compensation, feedback verification, and feedforward invocation.

[0069] Method Implementation Examples

[0070] Please see Figure 1 and Figure 2S1: Based on the forward field-of-view digital image and vehicle attitude sensor data, a mapped pixel set is derived, and the gradient smoothness of the projected light shape boundary, the amount of stray light leakage in the dark area of ​​the lens, and the amount of effective illuminance attenuation in the bright area of ​​the lens are calculated. The forward field-of-view digital image refers to the road scene pixel data formed by the vehicle's forward-looking imaging data, carrying the grayscale distribution of oncoming vehicles, road surface light spots, bright and dark cutoff areas, and target dark areas. The vehicle attitude sensor data refers to the attitude data related to vehicle pitch, roll, height changes, and suspension status, used to correct the spatial offset between the image coordinates and the headlight projection coordinates. The mapped pixel set refers to the set of pixels on the DLP micromirror control array after the target pixel region in the forward field-of-view digital image has been jointly located by distance, velocity, attitude, and external parameters. Its fields include image pixel position, spatial coordinate position, micromirror array position, target region category, time state, and effectiveness state. The gradient smoothness of the projected light shape boundary refers to the width of the brightness transition and the continuity of grayscale changes observed along the normal direction of the ideal bright and dark cutoff line, used to determine whether the light shape boundary has diffused due to lens contamination. The stray light leakage in the dark zone of the lens refers to the non-target illumination illuminance conversion result introduced by pollution scattering or optical degradation within the anti-glare dark zone, and is used for comparison with the safe anti-glare threshold. The effective illuminance attenuation in the bright zone of the lens refers to the illuminance decrease state of the core road lighting bright zone relative to the calibrated illuminance reference, and is used to subsequently determine the direction of global compensation and drive current adjustment.

[0071] Please see Figure 2 S111: Acquires digital images of the forward field of view and vehicle relative distance and speed data. Time-aligns the image frames, distance data, and speed data, and removes data with unbalanced exposure, missing target areas, duplicate frames, and incomplete attitude recordings. For the retained data, the target pixel area is first determined based on the two-dimensional pixel coordinates in the forward field of view digital image, and then the corresponding three-dimensional coordinate position is generated by combining the vehicle relative distance and speed data. The external parameter matrix refers to the calibration data used to record the correspondence between the forward imaging coordinates, headlight projection coordinates, and vehicle body reference coordinates. It originates from the optical calibration and sensor installation calibration after vehicle assembly. Angle compensation deviation refers to the projection coordinate offset caused by changes in vehicle pitch, roll, and height, determined based on the three-dimensional coordinate position, vehicle attitude sensor data, and the external parameter matrix. The spatial mapping matrix is ​​a mapping record that converts the target pixel area into the position of the micromirror control array, carrying image coordinates, projection angle, micromirror position, and attitude compensation state. By calling the spatial mapping matrix, the target pixel region in the digital image of the front field of view is mapped to the micromirror control array to obtain the mapped pixel set, and the mapped pixel set is output to S112 and S113 for further processing.

[0072] Please see Figure 2S112: Based on the mapped pixel set, determine the scanning direction perpendicular to the preset ideal light shape cutoff line. The ideal light shape cutoff line refers to the light-dark boundary reference formed under the conditions of clean lens, completed vehicle attitude correction, and effective photometric calibration, which is derived from the factory light shape calibration record of the vehicle headlight. Read the gray-level changes of adjacent pixels along this scanning direction to form the projected light shape boundary gradient feature. The projected light shape boundary gradient feature refers to the change queue of continuous pixel gray-level when transitioning from dark area to bright area, carrying the gray-level, boundary area, and effective scanning state of each scanning position. If there are insufficient consecutive frames, the most recent effective boundary gradient feature is used as the reference for the start-up stage; if there is no historical effective boundary gradient feature, the calibration gradient corresponding to the ideal light shape cutoff line is called as the initial reference. Then, the optical angular resolution span of the projected light shape boundary gradient feature is extracted, converted into the projected light shape boundary gradient smoothness, and this smoothness is used as the input field for S2 to judge contamination degradation.

[0073] Please see Figure 2 S113: Obtain the grayscale distribution of the mapped pixel set within the target dark area and perform validity screening on the pixels in the target dark area. The target dark area refers to the projection area that should be avoided by oncoming vehicles or sensitive observation areas in low beam anti-glare control, and its boundary is jointly determined by the mapped pixel set and the position of oncoming vehicles. Validity screening uses consistent timestamps, unsaturated pixels, unobstructed target area, and image noise within the allowable threshold as conditions for entering subsequent processing. Photometric calibration parameters refer to the calibrated correspondence fields between image grayscale and actual illuminance, derived from vehicle lamp optical calibration records and forward imaging calibration records. Based on the average grayscale state of the retained pixels within the target dark area and the photometric calibration parameters, the illuminance conversion value is obtained and determined as the stray light leakage amount in the lens dark area. Subsequently, the average illuminance estimate of the illuminated bright area of ​​the core road surface is extracted from the mapped pixel set. The illuminated bright area of ​​the core road surface refers to the projection area that undertakes the main road lighting function in low beam mode. The preset illuminance benchmark is invoked, and the current illuminance estimate of the core road lighting bright area is compared with the illuminance benchmark with the same caliber to determine the effective illuminance attenuation of the lens bright area, and then output to S2 and S3.

[0074] In this embodiment, the target pixel region is processed by combining the forward field of view digital image, relative motion data, vehicle posture data and photometric calibration parameters, so that the road spot state on the image side can be converted into a set of mapped pixels that can be called on the micromirror control array side, thereby providing input under the same coordinate caliber for subsequent pollution degradation judgment, boundary compensation and illuminance compensation.

[0075] Please see Figure 1 and Figure 3S2: Calculate the leakage value deviation between the stray light leakage in the dark area of ​​the lens and the preset safety anti-glare threshold. Use this leakage value deviation as an indicator of the degree of contamination. Combined with the numerical evolution characteristics of the gradient smoothness of the projected light shape boundary and the attenuation of the effective illuminance in the bright area of ​​the lens, compare it with the preset lens contamination degradation conditions, mark the contamination degradation judgment status, establish an optical degradation inverse filtering approximation model, and calculate the maximum diffuse scattering angle. The safety anti-glare threshold refers to the configurable judgment field corresponding to the upper limit of the oncoming visual safety illuminance, derived from the vehicle anti-glare photometric calibration record and regulatory adaptation calibration record, used to determine whether the dark area leakage has entered a state requiring compensation. The leakage value deviation refers to the record of the stray light leakage in the dark area of ​​the lens exceeding, approaching, or falling below the safety anti-glare threshold, used to assess the degree of impact of lens contamination on the anti-glare area. The contamination degradation judgment status indicator is a status field classifying the current lens state, including at least textual states such as no degradation triggered, suspected degradation, and confirmed degradation. Its output is used to determine whether to establish an optical degradation inverse filtering approximation model. The optical degradation inverse filtering approximation model is a pollution degradation inverse analytical model established based on the boundary diffusion, bright area absorption, and two-dimensional point spread function calibration relationship. Its inputs are the diffusion radius distribution weights, light energy absorption conversion coefficients, and associated scattering parameters, and its output is the maximum diffusion scattering angle. The maximum diffusion scattering angle refers to the maximum angular mapping result of the beam boundary spreading towards the anti-glare dark area under lens pollution conditions, and is used for S3 calculation of boundary retreat compensation.

[0076] Please see Figure 3S211: Collect data from the vehicle height sensor and suspension displacement sensor, and first determine whether the vehicle's driving posture is in a stable state. A stable state means that both the changes in vehicle height and suspension displacement are within the calibrated allowable state, and this state is derived from the vehicle posture calibration table and suspension state record. If the vehicle is in a non-stable state, the current dark area leakage judgment is marked as a temporarily suspended state, and the pollution degradation judgment state identifier formed in the previous stable cycle is used; if there is no previous stable cycle state, the non-triggered degradation is used as the start-up stage state. If the vehicle is in a stable state, the safety anti-glare threshold is called, and the amount of stray light leakage in the lens dark area is compared with the safety anti-glare threshold with the same caliber to obtain the leakage value deviation. Then, the direction of change, continuity of change, and deviation state of the gradient smoothness of the projected light shape boundary in continuous effective frames are combined with the deviation state between the effective illuminance attenuation of the lens bright area and the calibrated illuminance benchmark, and compared according to the preset lens pollution degradation conditions. Lens contamination degradation conditions are derived from vehicle optical calibration records under clean, contaminated, and rain-covered lens conditions. The judgment order is as follows: first confirm the vehicle's attitude is valid; then confirm the dark area leakage state; next, confirm the boundary smooth change state; and finally, confirm the bright area attenuation state. When all the above conditions are met, the contamination degradation judgment status is marked as confirmed degradation; when the conditions are not fully met but a continuous trend exists, it is marked as suspected degradation; when none of the conditions are triggered, it is marked as non-triggered degradation.

[0077] Please see Figure 3 S212: Based on the pollution degradation status indicator, when the status is confirmed degradation or suspected degradation, the influence of the pollution adhesion layer on the lens surface on the headlight beam dispersion is regarded as a spatial probability distribution. The spatial distribution proportion is extracted according to the gradient smoothness of the projected light shape boundary, and the dispersion radius distribution weight is calculated, denoted as... The absorption conversion ratio is extracted based on the effective illuminance attenuation in the bright area of ​​the lens, and the light energy absorption conversion coefficient is calculated and denoted as . .

[0078] The preset two-dimensional point spread function uses a two-dimensional Gaussian distribution model to characterize the physical scattering properties of the light beam, and its basic mathematical form is:

[0079] ;

[0080] in, The coordinates of the two-dimensional plane of the micromirror control array are centered at the center of the projected beam. The center amplitude, The variance is taken into account. This is combined with the dispersion radius distribution weights. With light energy absorption conversion coefficient Replace the variance and magnitude in the two-dimensional point spread function mentioned above, that is, let , The specific formula for obtaining the associated scattering parameters and establishing the approximate model of optical degradation inverse filtering is as follows:

[0081] ;

[0082] If the pollution degradation status is marked as "degradation not triggered", the model will not be updated, and the status of the previous effective model or the clean calibration model will be used as the subsequent reference.

[0083] Please see Figure 3 S213: The degree of contamination and degradation of the physical optical link is analyzed using an inverse filtering approximation model for optical degradation. The degradation level is extracted from the inverse filtering approximation model to a predetermined ratio (with a set boundary energy attenuation threshold). The boundary coordinates of the origin are given. Due to the circular symmetry of the model, the distance from the origin to the boundary coordinates is assumed to be... ,Right now ,make The degree of contamination degradation of the physical optical link is obtained by reverse engineering. The formula is:

[0084] ;

[0085] Subsequently, the internal model boundary conversion parameters of the optical degradation inverse filtering approximation model are combined (setting a fixed conversion ratio coefficient from the micromirror pixel array to the actual spatial projection angle as ). Calculate the extended boundary mapping values. Determine the degree of contamination and degradation of the physical optical link. Multiply by conversion factor The extended boundary mapping value is obtained, and the extended boundary mapping value is set as the maximum dispersion scattering angle, denoted as . The final quantitative conversion formula is:

[0086] ;

[0087] The maximum diffuse scattering angle is then output to S3 to determine the boundary setback compensation of the anti-glare dark area.

[0088] In one embodiment, the process of obtaining the maximum diffuse scattering angle further includes: calling a preset reference smoothness, comparing the deviation state of the projected light shape boundary gradient smoothness with the reference smoothness, and determining the diffusion level of the diffusion radius distribution weight in combination with the maximum tolerance limit; calling a preset total luminous flux reference, converting the effective illuminance attenuation of the lens bright area into a bright area attenuation level, and extracting the corresponding light energy absorption conversion coefficient from the mapping table; updating the diffusion width level and center light energy level in the two-dimensional point diffusion function with the diffusion radius distribution weight and the light energy absorption conversion coefficient, respectively, to form associated scattering parameters; then extracting the attenuation boundary coordinate state by the optical degradation inverse filtering approximation model, and generating extended boundary mapping values ​​according to the internal model boundary conversion parameters, finally obtaining the maximum diffuse scattering angle. All of the above fields are generated from calibration records, previous image processing results, or lookup table rules, and are updated under the same pollution degradation state record.

[0089] In this embodiment, by first confirming the validity of the vehicle's posture and then comparing dark area leakage, boundary smoothing changes, and bright area attenuation, the pollution degradation judgment status indicator is generated only under the same valid data caliber, thereby avoiding false triggering caused by vehicle bumps or image anomalies. By converting boundary diffusion and light energy absorption into model fields respectively, the maximum diffuse scattering angle can be used as a unified input for subsequent boundary retreat and micromirror duty cycle gradation processing.

[0090] Please see Figure 1 and Figure 4 S3: Based on the maximum dispersion scattering angle, the pixel width for light shape boundary retreat compensation and the micromirror drive duty cycle after gradient assignment are obtained. Combined with the global illuminance compensation coefficient determined based on the effective illuminance attenuation of the lens bright area, the amplified duty cycle value is obtained. The illuminance compensation drive current is then calculated based on the difference between the amplified duty cycle value and the preset rated output upper limit, establishing the corrected low-beam anti-glare execution parameters. The pixel width for light shape boundary retreat compensation refers to the width of the anti-glare dark area boundary that needs to be reserved on the micromirror array side after converting the maximum dispersion scattering angle, used to control the dark area boundary to retreat towards the high-brightness illumination area. The micromirror drive duty cycle refers to the driving ratio state of the DLP micromirror in forming different grayscale light shapes within the projection cycle, which is recorded in this embodiment using a grayscale level field and a transition area position field. The global illuminance compensation coefficient refers to the bright area compensation level field determined based on the effective illuminance attenuation of the lens bright area, used to amplify the micromirror drive duty cycle of the core road lighting bright area. The amplified duty cycle value refers to the micromirror drive duty cycle state after global illuminance compensation. Illuminance compensation drive current refers to the current adjustment field output to the headlight drive side when the amplified duty cycle value is still insufficient to meet the residual illuminance gap. Corrected low beam anti-glare execution parameters refer to the set of low beam anti-glare control parameters composed of boundary yield, gradual duty cycle, amplified duty cycle, and drive current.

[0091] Please see Figure 4S311: The preset headlight spatial mapping resolution is invoked to convert the maximum dispersion scattering angle into a spatial pixel offset state on the micromirror array side, and the beam shape boundary clearance compensation pixel width is obtained. The headlight spatial mapping resolution refers to the calibration correspondence field between the projection angle change and the micromirror array pixel position change, derived from the DLP headlight projection calibration record. Based on the beam shape boundary clearance compensation pixel width, the anti-glare dark area boundary in the projected beam shape is shifted and extended towards the coordinate side of the high-brightness illumination area, establishing a beam shape reserved clearance boundary. The beam shape reserved clearance boundary refers to the compensation boundary used to prevent scattered light from entering the observation area of ​​oncoming vehicles under pollution dispersion conditions. Subsequently, a brightness gradient zone area is defined within the beam shape high-brightness transition zone adjacent to the beam shape reserved clearance boundary. The brightness gradient zone area refers to the pixel area used to smoothly transition the micromirror drive duty cycle from the edge of the anti-glare dark area to the core bright area. The nonlinear smooth curve refers to the brightness transition rule preset in the calibration record; its input is the position level of the pixel within the gradient zone relative to the beam shape reserved clearance boundary, and its output is the gradient grayscale ratio. The duty cycle of the micromirror drive is assigned a transitional gradient value based on the gradient grayscale ratio, and the micromirror drive duty cycle after the gradient assignment is obtained and output to S312.

[0092] Please see Figure 4 S312: The preset light pattern calibration illuminance benchmark is invoked. This benchmark is compared with the effective illuminance attenuation of the lens bright area using the same caliber to determine the illuminance compensation ratio and obtain the global illuminance compensation coefficient. The light pattern calibration illuminance benchmark originates from the calibration record of the core road lighting bright area under lens clean conditions. The illuminance compensation ratio is recorded using a compensation level field; the closer the bright area attenuation is to the strong attenuation level, the closer the compensation level corresponding to the global illuminance compensation coefficient is to the high compensation level. Based on the global illuminance compensation coefficient, the duty cycle of the micromirror drive after gradual assignment within the core road lighting bright area corresponding to the mapped pixel set is amplified to obtain the amplified duty cycle value. The preset rated output upper limit refers to the calibration upper limit state that the micromirror drive and light source output are allowed to reach, derived from the vehicle lamp thermal management calibration record and the light source drive calibration record. The amplified duty cycle value is compared with the preset rated output upper limit. If the amplified duty cycle value does not reach the output upper limit, the amplified duty cycle value is used as the main execution field for illuminance compensation. If there is still a residual illuminance gap after the amplified duty cycle value reaches the output upper limit, the headlight source drive current is increased according to the level of the residual illuminance gap to obtain the illuminance compensation drive current, and the current adjustment field is output to S313.

[0093] Please see Figure 4S313: The pixel width of the converged light-shaped boundary retreat compensation, the amplified duty cycle value, and the illuminance compensation drive current are used to establish corrected low-beam anti-glare execution parameters. These execution parameters are recorded in boundary control, transition control, bright area compensation, and light source compensation fields. The boundary control field indicates the retreat position of the anti-glare dark area boundary; the transition control field indicates the change state of the micromirror drive duty cycle within the brightness gradient zone; the bright area compensation field indicates the amplified duty cycle state of the core road lighting bright area; and the light source compensation field indicates the drive current adjustment state. If the previous pollution degradation judgment status is marked as suspected degradation, the corrected low-beam anti-glare execution parameters are output with a conservative retreat state; if degradation is confirmed, a complete retreat and illuminance compensation state are output; if degradation is not triggered, the calibrated low-beam anti-glare execution parameters are used and the uncompensated state is recorded. The corrected low-beam anti-glare execution parameters then proceed to S4 for projection feedback verification.

[0094] In this embodiment, by converting the maximum diffusion angle into the pixel width of the light shape boundary retreat compensation, and setting a brightness gradient band in the area adjacent to the retreat boundary, the boundary of the anti-glare dark area still retains the avoidance space under the state of pollution diffusion; by converting the effective illuminance attenuation of the lens bright area into the global illuminance compensation coefficient and the driving current adjustment state, the core road lighting bright area still maintains the usable lighting state after the anti-glare retreat, thereby forming a low beam execution parameter that takes into account both the anti-glare boundary and the road lighting.

[0095] Please see Figure 1 and Figure 5 S4: The corrected low-beam anti-glare execution parameters are invoked to extract the optical feedback values ​​of the road surface spot, resulting in a recalculated attenuation index. Based on this recalculated attenuation index, the pixel width of the beam boundary retreat compensation, the width of the driving gradient band, and the beam side illumination angle are adjusted to establish verification anti-glare parameters. Converged anti-glare compensation parameters are then generated by combining the dynamic tracking coordinates. The road surface spot optical feedback values ​​refer to the feedback data formed by the road surface spot, dark area boundary, and bright area illuminance status in the current field of view after the corrected low-beam anti-glare execution parameters are invoked. Its fields include dark area grayscale status, bright area illuminance estimation status, boundary transition status, and effective frame status. The recalculated attenuation index refers to the combined status of the lens dark area stray light leakage and road illuminance attenuation amplitude, obtained again based on the road surface spot optical feedback values, used to determine whether the aforementioned corrected parameters have converged. The verification anti-glare parameters refer to the combined parameters of boundary retreat, gradient band width, and side illumination angle, adjusted again based on the feedback results. Dynamic tracking coordinates refer to the moving coordinate state corresponding to the change in the angle of the oncoming vehicle during oncoming traffic, used to move the boundary of the anti-glare dark area to follow the position of the oncoming vehicle. Converging anti-glare compensation parameters refer to the final compensation parameters formed after feedback verification and dynamic tracking superposition, used for S5 recording and subsequent feedforward calls.

[0096] Please see Figure 5S411: Calls the corrected low-beam anti-glare execution parameters and outputs the micromirror control matrix to drive the headlight beam projection. The micromirror control matrix refers to the micromirror drive record organized according to the mapped pixel set, boundary retreat state, gradual duty cycle state, and bright area compensation state, used to control the micromirror deflection state corresponding to each mapped pixel. During the continuous detection period, the forward-looking camera is called to capture the current field of view and the validity of the current field of view is determined. The current field of view refers to the road lighting feedback image obtained after the corrected low-beam anti-glare execution parameters are applied. If the image has strong reflective occlusion, missing dark area targets, discontinuous timestamps, or field of view noise exceeding the judgment threshold, the frame is marked as an invalid feedback frame, and the road surface spot optical feedback value of the previous valid feedback frame is used; if there are not enough consecutive valid frames, the most recent valid feedback state and the corrected low-beam anti-glare execution parameters are used together as transient feedback. For valid feedback frames, the road surface spot optical feedback value in the current field of view is extracted, and the lens dark area stray light leakage and road surface illuminance attenuation amplitude under lens state are recalculated to obtain the recalculated attenuation index.

[0097] Please see Figure 5 S412: Based on the recalculated attenuation index, when the stray light leakage in the dark area of ​​the lens exceeds the safety anti-glare threshold, the out-of-tolerance state of the stray light leakage in the dark area of ​​the lens relative to the safety anti-glare threshold is recorded as a leakage out-of-tolerance value, and the beam boundary setback compensation pixel width is increased according to the leakage out-of-tolerance level. The upper limit of the oncoming visual safety illuminance is used as the safety anti-glare threshold, which is derived from the anti-glare calibration record. The optical dot spread rate is used as the setback gain coefficient, which is derived from the lens contamination scattering calibration record and is used to convert the leakage out-of-tolerance state into a pixel increment level. After the pixel increment is superimposed on the beam boundary setback compensation pixel width, the grayscale slope of the newly added pixel area is extracted, and the boundary gradient state of the current frame is compared with that of the previous frame to obtain the boundary change value. The boundary change value is used to determine whether the gradient band needs to be expanded. The closer the boundary change is to the abrupt state, the closer the expansion level corresponding to the driving gradient band width is to the strong smoothing level. The lower limit of compensation refers to the calibration field corresponding to the standard minimum road illuminance index, and the upper limit of thermal protection refers to the power critical state field corresponding to the highest temperature that the vehicle lamp can withstand. If the road surface illuminance attenuation is below the lower limit of compensation and the total output power reaches the upper limit of thermal protection, the lateral edge pixel columns of the micromirror control array are shut down column by column, and the corresponding light energy is adjusted to the central region of the projected beam pattern to determine the lateral illumination angle of the narrowed beam pattern. Subsequently, the adjusted beam pattern boundary setback compensation pixel width, driving gradient band width, and beam pattern lateral illumination angle are integrated to establish the verification anti-glare parameters.

[0098] Please see Figure 5S413: Obtain the offset value of the oncoming vehicle's angle change during the meeting process. This offset value comes from the position change of the oncoming vehicle's light area or vehicle body target area in the forward field of view digital image within consecutive valid frames, and enters the dynamic tracking process after vehicle body attitude correction. The dynamic tracking coordinates are jointly determined by the oncoming vehicle's offset direction, offset continuity, and the position of the anti-glare dark area in the mapped pixel set, and are used to represent the movement position of the anti-glare dark area boundary in the current field of view. Based on the dynamic tracking coordinates, the calibration anti-glare parameters are superimposed on the moving light-shaped anti-glare dark area boundary. If the oncoming vehicle target is briefly lost, the previous valid dynamic tracking coordinates are maintained and the boundary movement amplitude is reduced; if the target is re-identified, the current identification position is smoothly connected with the previous valid coordinates. Finally, convergent anti-glare compensation parameters are generated and output to S5 for association recording and feedforward retrieval.

[0099] In one embodiment, the process of establishing the verification anti-glare parameters further includes: calling a preset upper limit of the oncoming visual safety illuminance as the safety anti-glare threshold, comparing the amount of stray light leakage in the dark area of ​​the lens with the safety anti-glare threshold and determining the leakage excess state; calling a preset optical dot spread rate as the yield gain coefficient, converting the leakage excess state into a pixel increment level, and superimposing it on the light shape boundary yield compensation pixel width; extracting the grayscale slope of the newly added pixel area to establish the boundary change value, and expanding the driving gradient band width according to the boundary change value; calling a preset standard minimum road surface illuminance index as the compensation usable lower limit, obtaining the power critical point corresponding to the highest temperature that the vehicle lamp can withstand as the thermal protection upper limit; when the road surface illuminance attenuation is lower than the compensation usable lower limit and the total output power is in the thermal protection upper limit state, closing the lateral edge pixel column column by column and adjusting the light energy to the center area of ​​the projected light shape, and determining the narrowed light shape lateral illumination spread angle; finally integrating the light shape boundary yield compensation pixel width after superimposing pixel increments, the expanded driving gradient band width, and the narrowed light shape lateral illumination spread angle to establish the verification anti-glare parameters.

[0100] In this embodiment, by re-collecting and recalculating the attenuation index of the road surface spot state after the low beam anti-glare execution parameters are projected, the boundary setback, gradient strip width, and lateral illumination angle can be continuously adjusted based on actual feedback. By converting the oncoming vehicle angle change into dynamic tracking coordinates, the verification anti-glare parameters are updated as the oncoming vehicle position moves, thereby forming convergent anti-glare compensation parameters and reducing the deviation between the fixed compensation boundary and the dynamic vehicle position.

[0101] Please see Figure 1 and Figure 6S5: Obtain the current vehicle's cumulative mileage, ambient temperature characteristics, and wiper operating status. Record the convergent anti-glare compensation parameters along with these data into a mapping table. Use this mapping table to obtain the baseline feedforward input and execute the adaptive lighting mode switch. The cumulative mileage refers to the vehicle's accumulated driving status field since recording began, used to describe the conditions for lens contamination accumulation. Ambient temperature characteristics refer to the vehicle's external temperature and temperature status fields related to lens optical deformation, used to describe the impact of temperature on the lens and projected light pattern. Wiper operating status refers to the current wiper setting and start / stop status field, used to describe the correlation between rainwater, sewage, or cleaning activities and the lens contamination status. The mapping table is a data table that binds the convergent anti-glare compensation parameters with the cumulative mileage, ambient temperature characteristics, and wiper operating status. Its fields include context conditions, historical status parameters, compensation parameter status, validity status, and call status. Reference feedforward input refers to the initial anti-glare compensation parameters that are called in advance based on the historical matching status during the subsequent driving period. These parameters are used to generate a safe low beam control command before the closed-loop feedback stabilizes.

[0102] Please see Figure 6 S511: Calculate the signal-to-noise ratio (SNR) of the image data in the current field of view. The image data SNR refers to the quality state of the effective image components used to identify road surface glare and the boundary of the anti-glare dark area in the current field of view relative to the noise components. The signal components are formed by road surface glare, the cutoff line between light and dark areas, and the target area of ​​oncoming vehicles. The noise components are formed by rain obstruction, overexposure reflection, motion blur, and image grain interference. The judgment threshold is derived from the lowest effective resolution baseline of the vehicle's visual perception under extreme recognition conditions. If the image data SNR is lower than the judgment threshold and causes a break in the closed-loop control link, the partition correction adjustment of the micromirror control array is frozen, and a basic safety low beam mode command is output to trigger degradation protection. The basic safety low beam mode command refers to the calibrated low beam anti-glare command called when the feedback link is unavailable, used to maintain the basic avoidance state of the anti-glare dark area. Subsequently, the vehicle's cumulative mileage, ambient temperature characteristics, and wiper operating status are acquired. The converged anti-glare compensation parameters, along with the aforementioned context conditions, are entered into a non-volatile recording area to establish an association mapping table. If the signal-to-noise ratio of the image data meets the judgment threshold, the convergence anti-glare compensation parameters and context conditions are recorded normally, and the recording status is marked as callable.

[0103] Please see Figure 6S512: Identify oncoming vehicles during subsequent driving periods and match context conditions based on the current cumulative mileage, ambient temperature characteristics, and wiper operating status. During matching, first, the lowest effective resolution threshold under extreme recognition conditions of the vehicle's visual perception is extracted as the judgment threshold; then, the current cumulative mileage is compared with the historical cumulative mileage recorded in the association mapping table to determine the mileage difference; subsequently, the current ambient temperature characteristics are compared with the historical ambient temperature characteristics recorded in the association mapping table to determine the temperature difference; finally, the consistency of the current wiper operating status with the historical wiper operating status is compared. The physical pollution stability tolerance range is derived from the calibration record of lens pollution changes with cumulative driving, and the lens deformation stability tolerance range is derived from the lens temperature deformation calibration record. When the mileage difference falls within the physical pollution stability tolerance range, the temperature difference falls within the lens deformation stability tolerance range, and the operating status is completely consistent, the context condition matching is confirmed as successful. Then, the convergent anti-glare compensation parameters associated with the successfully matched items in the association mapping table are extracted, set as historical state parameters, and a baseline feedforward input is established. If multiple matching items exist, the historical state parameter whose validity state is closer to the current image quality state and whose state is not invalid will be called first. If no matching item exists, the current calibrated low beam anti-glare parameter will be used as the reference feedforward input, and the closed-loop verification will be waited for to regenerate the convergent anti-glare compensation parameter.

[0104] Please see Figure 6 S513: Reconstructing the anti-glare beam pattern control command based on the reference feedforward input. The anti-glare beam pattern control command refers to the control record used to drive the micromirror control array to deflect and form the low beam anti-glare beam pattern. Its fields include the anti-glare dark area boundary position, beam pattern boundary retreat state, gradient band width state, core bright area compensation state, side illumination spread angle state, and execution state. The micromirror control array is driven to deflect according to the anti-glare beam pattern control command, so that when the low beam beam pattern detects an oncoming vehicle, it first calls the reference feedforward input to form the initial anti-glare boundary, and then continues to be corrected by the real-time feedback process formed by S1 to S4. When the closed-loop feedback returns to stability, the converged anti-glare compensation parameters formed in the current cycle replace the reference feedforward input as the new execution reference, and the associated mapping table is updated when the context conditions meet the recording requirements, completing the adaptive lighting mode switching.

[0105] In this embodiment, by establishing a mapping table between the converged anti-glare compensation parameters and the cumulative driving mileage, ambient temperature characteristics, and wiper operating status, the anti-glare control under similar pollution and environmental conditions can first call historical state parameters to form a reference feedforward input. By freezing the partition correction and outputting the basic safe low beam mode command when the image data signal-to-noise ratio is lower than the judgment threshold, an executable degradation protection path is still retained when the closed-loop control link is broken, thereby ensuring the continuity of data flow between feedforward call, closed-loop verification, and abnormal degradation.

[0106] System-related Implementation Examples

[0107] Please see Figure 7 System-related embodiment: This embodiment also provides a DLP vehicle headlight adaptive lighting mode switching system, which is used to implement the above-mentioned DLP vehicle headlight adaptive lighting mode switching method. The system includes a multi-source state perception module, a physical degradation assessment module, a partition compensation calculation module, a closed-loop verification and tracking module, and a memory feedforward scheduling module. The multi-source state perception module executes S1, receiving the forward field of view digital image, vehicle relative distance data, relative speed data, vehicle attitude sensor data, and external parameter matrix, and outputs the mapped pixel set, the gradient smoothness of the projected light shape boundary, the stray light leakage in the dark area of ​​the lens, and the effective illuminance attenuation in the bright area of ​​the lens. The physical degradation assessment module executes S2, receiving the boundary, dark area, and bright area state fields output by the multi-source state perception module, and outputting the pollution degradation judgment status indicator and the maximum diffusion scattering angle according to the safety anti-glare threshold, lens pollution degradation conditions, two-dimensional point spread function calibration relationship, and internal model boundary conversion parameters. The partition compensation calculation module, corresponding to S3, receives the maximum diffusion scattering angle and the effective illuminance attenuation in the lens bright area, and outputs the pixel width for light shape boundary setback compensation, the micromirror drive duty cycle after gradient assignment, the magnified duty cycle value, the illuminance compensation drive current, and the corrected low beam anti-glare execution parameters. The closed-loop verification and tracking module, corresponding to S4, receives the corrected low beam anti-glare execution parameters and the optical feedback value of the road surface spot in the current field of view, and outputs the recalculated attenuation index, verification anti-glare parameters, dynamic tracking coordinates, and converged anti-glare compensation parameters. The memory feedforward scheduling module, corresponding to S5, receives the converged anti-glare compensation parameters, cumulative mileage, ambient temperature characteristics, and wiper operating status, establishes an association mapping table, and outputs the reference feedforward input and anti-glare light shape control command when an oncoming vehicle is subsequently detected.

[0108] In the system-associated embodiment, the data transmission direction between the multi-source state perception module, physical degradation assessment module, partition compensation calculation module, closed-loop verification and tracking module, and memory feedforward scheduling module is consistent with the sequence of S1 to S5 in the aforementioned method embodiment. Each module only undertakes the input reception, state generation, and result output functions in its corresponding step, and does not repeatedly undertake the judgment or compensation tasks of other modules. Through this module relationship, the mapped pixel set, pollution degradation judgment state identifier, maximum dispersion scattering angle, corrected low beam anti-glare execution parameters, convergent anti-glare compensation parameters, and reference feedforward input in the aforementioned method can form corresponding transmission links in the system structure, thereby enabling the DLP vehicle headlight adaptive lighting mode switching method to be collaboratively implemented by various modules of the system.

[0109] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for switching adaptive lighting modes in DLP vehicle lights, characterized in that, Includes the following steps: S1: Based on the digital image of the front field of view and the data from the vehicle attitude sensor, a set of mapped pixels is obtained, and the smoothness of the gradient of the projection light shape boundary, the amount of stray light leakage in the dark area of ​​the lens, and the amount of effective illuminance attenuation in the bright area of ​​the lens are calculated. S2: Calculate the leakage value deviation between the stray light leakage amount in the dark area of ​​the lens and the preset safety anti-glare threshold. Use the leakage value deviation as an indicator to evaluate the degree of pollution. Combine the numerical evolution characteristics of the gradient smoothness of the projection light shape boundary and the effective illuminance attenuation amount in the bright area of ​​the lens with the preset lens pollution degradation conditions, mark the pollution degradation judgment status, establish an optical degradation inverse filter approximation model, and calculate the maximum diffuse scattering angle. S3: Based on the maximum dispersion scattering angle, the width of the light shape boundary retreat compensation pixel and the micromirror drive duty cycle after gradient assignment are obtained. Combined with the global illuminance compensation coefficient determined based on the effective illuminance attenuation of the lens bright area, the amplification duty cycle value is obtained. Based on the difference between the amplification duty cycle value and the preset rated output upper limit, the illuminance compensation drive current is obtained, and the corrected low beam anti-glare execution parameters are established. S4: Call the modified low beam anti-glare execution parameters to extract the optical feedback value of the road surface spot to obtain the recalculated attenuation index. Based on the recalculated attenuation index, adjust the width of the light shape boundary retreat compensation pixel, the width of the driving gradient band and the light shape side illumination spread angle to establish the verification anti-glare parameters. Combine the dynamic tracking coordinates to generate the convergent anti-glare compensation parameters. S5: Obtain the current vehicle's cumulative mileage, ambient temperature characteristics, and wiper operating status, and input the convergent anti-glare compensation parameters together with the current vehicle's cumulative mileage, ambient temperature characteristics, and wiper operating status into an associated mapping table. Call the associated mapping table to obtain the baseline feedforward input and perform adaptive lighting mode switching.

2. The DLP vehicle headlight adaptive lighting mode switching method according to claim 1, characterized in that, The specific steps of S1 are as follows: S111: Acquire digital images of the forward field of view and vehicle relative distance and relative speed data. Use a multi-source data fusion algorithm to combine the two-dimensional pixel coordinates of the digital images of the forward field of view with the relative distance and relative speed data to extract three-dimensional coordinate positions. Acquire vehicle posture sensor data and external parameter matrix. Calculate angle compensation deviation based on the three-dimensional coordinate positions, vehicle posture sensor data, and external parameter matrix. Establish a spatial mapping matrix. Call the spatial mapping matrix to map the target pixel region of the digital images of the forward field of view to the micromirror control array to obtain a mapped pixel set. S112: Based on the mapped pixel set, scan the grayscale value along the direction perpendicular to the preset ideal light shape cutoff line, calculate the brightness difference between adjacent pixels, obtain the gradient features of the projected light shape boundary, extract the optical angular resolution span of the gradient features of the projected light shape boundary, and obtain the smoothness of the gradient of the projected light shape boundary. S113: Obtain the average gray value of the mapped pixel set in the target dark area and the preset photometric calibration parameters. Calculate the illuminance conversion value based on the average gray value and the photometric calibration parameters to obtain the stray light leakage in the dark area of ​​the lens. Extract the average illuminance estimate of the illuminated bright area of ​​the mapped pixel set in the core road surface. Call the preset calibration illuminance benchmark to calculate the illuminance difference between the calibration illuminance benchmark and the average illuminance estimate to obtain the effective illuminance attenuation in the bright area of ​​the lens.

3. The DLP vehicle headlight adaptive lighting mode switching method according to claim 1, characterized in that, The specific steps of S2 are as follows: S211: Collect vehicle height sensor data and suspension displacement sensor data. Under the condition that the vehicle driving posture is in a stable state based on the vehicle height sensor data and the suspension displacement sensor data, call the preset safety anti-glare threshold, calculate the difference between the stray light leakage amount in the dark area of ​​the lens and the safety anti-glare threshold, and obtain the leakage value deviation. Under the condition that the leakage value deviation exceeds the safety limit, combine the numerical evolution characteristics of the gradient smoothness of the projection light shape boundary and the degradation condition of the effective illuminance attenuation in the bright area of ​​the lens, and mark the pollution degradation judgment status indicator. S212: Based on the pollution degradation judgment status identifier, the influence of the pollution adhesion layer on the lens surface on the diffusion of the headlight beam is regarded as a spatial probability distribution. The spatial distribution ratio is extracted according to the gradient smoothness of the projected light shape boundary and the diffusion radius distribution weight is calculated. The absorption conversion ratio is extracted according to the effective illuminance attenuation of the lens bright area and the light energy absorption conversion coefficient is calculated. Combining the diffusion radius distribution weight, the light energy absorption conversion coefficient and the preset two-dimensional point diffusion function analytical correlation scattering parameters, an optical degradation inverse filtering approximation model is established. S213: The degree of physical optical link contamination degradation is analyzed by reverse analysis using the optical degradation inverse filtering approximation model. The extended boundary mapping value is calculated by combining the internal model boundary conversion parameters of the optical degradation inverse filtering approximation model, and the maximum diffusion scattering angle is obtained.

4. The DLP vehicle headlight adaptive lighting mode switching method according to claim 3, characterized in that, The process of obtaining the maximum dispersion scattering angle is as follows: Call the preset baseline smoothness, calculate the difference between the gradient smoothness of the projected light shape boundary and the baseline smoothness, call the preset maximum tolerance limit, and determine the diffusion radius distribution weight based on the ratio of the difference to the maximum tolerance limit; The attenuation ratio is calculated by calling the preset total luminous flux reference and the ratio of the effective illuminance attenuation in the bright area of ​​the lens to the total luminous flux reference. The value that matches the attenuation ratio is extracted as the light energy absorption conversion coefficient based on the preset mapping table. The variance and amplitude of the two-dimensional point spread function are replaced by the dispersion radius distribution weight and the light energy absorption conversion coefficient, respectively, to obtain the associated scattering parameters. The optical degradation inverse filtering approximation model is constructed based on the two-dimensional point spread function after replacing the parameter values. Extract the boundary coordinates that have attenuated to a preset ratio in the optical degradation inverse filtering approximation model, and calculate the distance from the origin to the boundary coordinates to obtain the degree of physical optical link contamination degradation; The extended boundary mapping value is obtained by multiplying the degree of contamination and degradation of the physical optical link by the scaling factor in the internal model boundary conversion parameter of the optical degradation inverse filtering approximation model, and the extended boundary mapping value is set as the maximum diffusion scattering angle.

5. The DLP vehicle headlight adaptive lighting mode switching method according to claim 1, characterized in that, The specific steps of S3 are as follows: S311: Call the preset vehicle light space mapping resolution, calculate the spatial pixel offset based on the maximum diffusion angle and the vehicle light space mapping resolution to obtain the light shape boundary clearance compensation pixel width, and extend the anti-glare dark area boundary in the projected light shape to the coordinate side of the high brightness illumination area based on the light shape boundary clearance compensation pixel width to establish a light shape reserved clearance boundary. Define a brightness gradient zone area in the light shape high brightness transition area adjacent to the light shape reserved clearance boundary, calculate the gradient gray scale ratio between the edge position near the light shape reserved clearance boundary and the core bright area position using a non-linear smooth curve, and assign a transition gradient value to the micromirror drive duty cycle based on the gradient gray scale ratio to obtain the micromirror drive duty cycle after gradient assignment. S312: Call the preset light shape calibration illuminance reference, calculate the illuminance compensation ratio based on the calibration illuminance reference and the effective illuminance attenuation of the lens bright area to obtain the global illuminance compensation coefficient, amplify the micromirror drive duty cycle after the gradient assignment in the core road lighting bright area corresponding to the mapped pixel set according to the global illuminance compensation coefficient, obtain the amplified duty cycle value, calculate the difference between the amplified duty cycle value and the preset rated output upper limit to obtain the residual illuminance gap, increase the headlight source drive current according to the residual illuminance gap to obtain the illuminance compensation drive current; S313: Combine the pixel width of the light shape boundary retreat compensation, the value of the magnification duty cycle, and the illuminance compensation drive current to establish the corrected low beam anti-glare execution parameters.

6. The DLP vehicle headlight adaptive lighting mode switching method according to claim 1, characterized in that, The specific steps of S4 are as follows: S411: Call the corrected low beam anti-glare execution parameters, output the micromirror control matrix to drive the headlight beam projection, call the forward-looking camera to capture the current field of view during the continuous detection period, extract the road surface light spot optical feedback value in the current field of view, and recalculate the stray light leakage of the lens dark area and the road illuminance attenuation amplitude in the lens state according to the road surface light spot optical feedback value to obtain the recalculated attenuation index. S412: Based on the recalculated attenuation index, under the condition that the stray light leakage in the dark area of ​​the lens is greater than the safety anti-glare threshold, calculate the difference between the stray light leakage in the dark area of ​​the lens and the safety anti-glare threshold to obtain the leakage excess value. Increase the width of the light shape boundary retreat compensation pixel according to the leakage excess value, and calculate the boundary gradient difference between the current frame and the previous frame to obtain the boundary change value. Increase the width of the driving gradient band according to the boundary change value. Under the condition that the road surface illuminance attenuation is lower than the compensation available lower limit and the total output power reaches the thermal protection upper limit, adjust the local light energy to narrow the light shape side illumination angle, and integrate the adjusted light shape boundary retreat compensation pixel width, driving gradient band width and light shape side illumination angle to establish the verification anti-glare parameters. S413: Obtain the offset value of the angle change of the oncoming vehicle during the meeting process, calculate the dynamic tracking coordinates based on the offset value, and generate convergent anti-glare compensation parameters by superimposing the verification anti-glare parameters on the boundary of the light-shaped anti-glare dark area of ​​the position movement based on the dynamic tracking coordinates.

7. The DLP vehicle headlight adaptive lighting mode switching method according to claim 6, characterized in that, The process of establishing and verifying anti-glare parameters is as follows: The preset upper limit of the oncoming visual safety illuminance is called as the safety anti-glare threshold, and the difference between the stray light leakage in the dark area of ​​the lens and the safety anti-glare threshold is calculated to obtain the leakage tolerance value. The preset optical dot spread rate is used as the backoff gain coefficient. The leakage excess value is multiplied by the backoff gain coefficient to obtain the pixel increment. The pixel increment is then superimposed on the optical boundary backoff compensation pixel width. Extract the grayscale slope of the newly added pixel area corresponding to the increase in the width of the light-shaped boundary retreat compensation pixel to establish the boundary change value, and expand the width of the driving gradient band according to the boundary change value; The preset standard minimum road surface illuminance index is used as the lower limit of the compensation available, and the preset power critical point corresponding to the highest temperature that the vehicle lamp can withstand is obtained as the upper limit of the thermal protection. The lateral edge pixel columns of the micromirror control array are closed one by one, and the light energy is allocated and transferred to the central region of the projected light pattern. When the road surface illuminance attenuation is lower than the lower limit of the compensation available, the lateral illumination spread angle of the narrowed light pattern is determined. The anti-glare parameters are established by integrating the pixel width of the light shape boundary retreat compensation after the pixel increment is integrated, the width of the driving gradient band after the value is increased, and the light shape side illumination angle after the value is narrowed.

8. The DLP vehicle headlight adaptive lighting mode switching method according to claim 1, characterized in that, The specific steps of S5 are as follows: S511: Calculate the signal-to-noise ratio of the image data of the current field of view. Under the condition that the signal-to-noise ratio of the image data is lower than the preset judgment threshold and causes the closed-loop control link to break, freeze the partition correction adjustment of the micromirror control array and output the basic safety low beam mode command to trigger the degradation protection. Obtain the vehicle's cumulative mileage, ambient temperature characteristics and wiper operating status. Record the convergent anti-glare compensation parameters together with the cumulative mileage, ambient temperature characteristics and wiper operating status into a non-volatile storage device and establish an association mapping table. S512: Identify oncoming vehicles during the subsequent driving period, combine the current cumulative driving mileage with the ambient temperature characteristics and the wiper operating status to match the context conditions, call the association mapping table to extract historical state parameters, and obtain the baseline feedforward input; S513: Based on the reference feedforward input, reconstruct the anti-glare shape control command, drive the micromirror control array to deflect according to the anti-glare shape control command, and perform adaptive illumination mode switching.

9. The DLP vehicle headlight adaptive lighting mode switching method according to claim 8, characterized in that, The process of retrieving historical state parameters by calling the associated mapping table is as follows: The lowest effective resolution baseline of the vehicle-mounted visual perception device under extreme recognition conditions is extracted to set the judgment threshold. Calculate the absolute difference between the current cumulative mileage and the historical cumulative mileage recorded in the associated mapping table; Calculate the absolute temperature difference between the current ambient temperature feature and the historical ambient temperature features recorded in the association mapping table; The current wiper operating status is compared with the operating speed of the historical wiper operating status recorded in the associated mapping table. The preset physical pollution stability tolerance range and the preset lens deformation stability tolerance range are invoked. Under the conditions that the absolute difference in mileage is less than the physical pollution stability tolerance range, the absolute difference in temperature is less than the lens deformation stability tolerance range, and the consistency of the operating gear is exactly the same, the context conditions are confirmed to be successfully matched. Extract the convergent anti-glare compensation parameters associated with the successfully matched items in the association mapping table, set the convergent anti-glare compensation parameters as historical state parameters, and establish a benchmark feedforward input.

10. A DLP vehicle headlight adaptive lighting mode switching system, characterized in that, The system is used to implement the DLP vehicle headlight adaptive lighting mode switching method as described in any one of claims 1-9, the system comprising: The multi-source state perception module derives a mapped pixel set based on the forward field of view digital image and vehicle attitude sensor data, and calculates the gradient smoothness of the projected light shape boundary, the amount of stray light leakage in the dark area of ​​the lens, and the amount of effective illuminance attenuation in the bright area of ​​the lens. The physical degradation assessment module calculates the leakage value deviation between the stray light leakage amount in the dark area of ​​the lens and the preset safety anti-glare threshold. The leakage value deviation is used as an indicator to assess the degree of contamination. Combined with the numerical evolution characteristics of the gradient smoothness of the projection light shape boundary and the effective illuminance attenuation in the bright area of ​​the lens, it is compared with the preset lens contamination degradation conditions, and the contamination degradation judgment status is marked. An optical degradation inverse filtering approximation model is established, and the maximum diffuse scattering angle is calculated. The partition compensation calculation module calculates the pixel width of the light shape boundary retreat compensation and the micromirror drive duty cycle after the gradient assignment based on the maximum dispersion scattering angle. It also calculates the amplification duty cycle value by combining the global illuminance compensation coefficient determined based on the effective illuminance attenuation of the lens bright area. Finally, it calculates the illuminance compensation drive current based on the difference between the amplification duty cycle value and the preset rated output upper limit, and establishes the corrected low beam anti-glare execution parameters. The closed-loop verification and tracking module calls the modified low beam anti-glare execution parameters to extract the optical feedback value of the road surface spot and obtain the recalculated attenuation index. Based on the recalculated attenuation index, it adjusts the width of the light shape boundary retreat compensation pixel, the width of the driving gradient band and the light shape side illumination spread angle to establish the verification anti-glare parameters. It then combines the dynamic tracking coordinates to generate the convergent anti-glare compensation parameters. The memory feedforward scheduling module obtains the current vehicle's cumulative mileage, ambient temperature characteristics, and wiper operating status. It then records the convergent anti-glare compensation parameters along with the current vehicle's cumulative mileage, ambient temperature characteristics, and wiper operating status into an associated mapping table. The module calls the associated mapping table to obtain the baseline feedforward input and performs adaptive lighting mode switching.

Images