A polarizing optical lens glare cancellation system for vehicle security

Abstract

This invention relates to the field of optical engineering technology, specifically disclosing a polarized optical lens glare elimination system for vehicle-mounted security. The system includes: a reference polarization module, which constructs a controllable environment simulation test platform to collect reflected light data under different incident angles and water film thicknesses, calculates the degree of polarization and phase difference using the Stokes vector method, and constructs a quasi-polarized light characteristic library; a composite filtering module, which collects real-time reflected light data and compares it with the quasi-polarized light characteristic library to determine elliptically polarized light, and corrects it to linearly polarized light through phase compensation; a direction calibration module, which calculates the optical path deviation angle through a real-time direction calibration module and dynamically adjusts the filtering direction; a glare recognition module, which performs light intensity analysis on the corrected light to determine glare residue, identifies micro-glare spot areas through polarization degree calculation; and a grayscale correction module, which extracts the average grayscale value of the normal area and corrects the grayscale of the spot pixels, thereby achieving dynamic elimination of glare from reflected light on wet road surfaces.

Description

Technical Field

[0001] This invention relates to the field of optical engineering technology, and more specifically to a polarizing optical lens glare elimination system for vehicle-mounted security. Background Technology

[0002] Vehicle cameras are the core sensing components of vehicle security and driver assistance systems. Their imaging clarity directly affects the accuracy of algorithm recognition and the reliability of system decision-making. Polarized optical lenses are key components for eliminating road surface reflection glare. Existing technologies are mostly designed for ideal scenarios such as calm roads and water surfaces, and can effectively filter regularly polarized light. However, they have obvious shortcomings in complex and wet road conditions.

[0003] With the widespread use of hydrophobic coatings for automobiles, coating residues easily adhere to road surfaces. After rain, these residues combine with water to form a microscopically irregular oil film. When exposed to light, this film produces quasi-polarized light, somewhere between specular and diffuse reflection. This light contains elliptically polarized components, has no uniform polarization direction, and exhibits an asymmetrical phase distribution, making it completely incompatible with the filtering logic of standard polarizing lenses. Current technologies lack research and data support on the characteristics of this type of quasi-polarized light, and lack phase compensation and real-time polarization direction calibration functions. Glare filtering is incomplete, leaving residual micro-glare. These light spots increase image noise, interfere with algorithms' recognition of lane lines and obstacles, and can easily lead to algorithmic misjudgments under extreme road conditions, making it difficult to meet the glare elimination requirements of complex road conditions.

[0004] Therefore, the present invention provides a polarized optical lens glare elimination system for vehicle security. Summary of the Invention

[0005] The purpose of this invention is to provide a polarized optical lens glare elimination system for vehicle security, in order to solve the aforementioned background problems.

[0006] The objective of this invention can be achieved through the following technical solution: a polarized optical lens glare elimination system for vehicle security, comprising the following modules: Reference polarization module: Collects effective reflected light data under different incident angles and water film thicknesses, calculates the degree of polarization and phase difference based on the effective reflected light data, and constructs a quasi-polarized light characteristic library; Composite filtering module: Collects real-time reflected light to extract current scene parameters, compares them with the quasi-polarized light characteristic library to determine whether it is elliptically polarized light. If it is, it performs phase compensation analysis to construct a composite polarization filtering structure to adjust the elliptically polarized light into primary correction light. Direction calibration module: Obtain the current phase difference of the real-time reflected light, perform offset filtering analysis to obtain the optical path deviation angle, and determine whether the filtering direction of the composite polarization filter structure meets the requirements based on the optical path deviation angle. If it does not meet the requirements, fine-tune it based on the optical path deviation angle. Glare Recognition Module: Acquires the first-correction light and performs light intensity analysis to obtain light intensity uniformity, and determines whether there is residual glare in the first-correction light; if so, it performs polarization calculation to obtain the light spot performance value and identifies the micro-glare spot area; Grayscale correction module: Extract the normal area without glare from the micro-glare spot area, calculate the normal grayscale mean, and perform grayscale correction on the pixels in the micro-glare spot area based on the normal grayscale mean to obtain the grayscale correction value.

[0007] Furthermore, the process of obtaining the degree of polarization and phase difference is as follows: The effective light data of the reflected light were obtained and the total light intensity S0, the vertical polarization component S1, the diagonal polarization component S2, and the circular polarization component S3 were obtained by Stokes vector analysis. S1, S2, and S3 are squared and summed. The square root of the sum is then calculated, and the ratio of the square root to S0 is used to calculate the degree of polarization. The phase difference is obtained by performing arctangent analysis based on S2 and S3.

[0008] Furthermore, the method for performing arctangent analysis is as follows: The phase difference is calculated by inputting the diagonal and circular polarization components of the reflected light into the arctangent function.

[0009] Furthermore, the process of performing phase compensation analysis is as follows: The real-time reflected light is determined to be elliptically polarized light, and the current phase difference is obtained; Obtain the visible light center wavelength and the refractive index of the compensation material. Multiply the visible light center wavelength by a times the refractive index of the compensation material. Calculate the ratio of the current phase difference to π. Multiply the results of the product calculation and the ratio calculation to obtain the phase compensation layer thickness.

[0010] Furthermore, the process of determining that the real-time reflected light is elliptically polarized light is as follows: The current scene parameters of the real-time reflected light are obtained and matched with the quasi-polarized light characteristic library to obtain scenes of the same type and mark them as reference mapping scenes; The number of times linearly polarized light, elliptically polarized light, and circularly polarized light appear in the reference mapping scene are counted respectively, and the ratios are calculated with the total number of reference mapping scenes to obtain the proportions of linearly polarized light, elliptically polarized light, and circularly polarized light respectively. If the proportion of elliptically polarized light is at its maximum, then the real-time reflected light will be determined to be elliptically polarized light.

[0011] Furthermore, the optical path deviation angle is obtained as follows: Obtain the current phase difference of the real-time reflected light and extract the standard phase difference of the composite polarization filter structure; Obtain the polarization direction conversion coefficient, calculate the arctangent function of the current phase difference and the standard phase difference, and multiply the calculation result with the polarization direction conversion coefficient to obtain the optical path deviation angle.

[0012] Furthermore, the process of determining whether glare remains after a correction is as follows: The intensity uniformity of light is obtained by acquiring a corrected light source and performing light intensity analysis. Obtain the glare retention threshold. If the light intensity uniformity is less than the glare retention threshold, it is determined that there is glare retention in the first correction light.

[0013] Furthermore, the process of performing light intensity analysis is as follows: Background interference removal is performed on the corrected light, and the light data of the effective area of ​​the road surface is selected by image segmentation algorithm. Extract the light intensity value of each pixel from the light data of the effective area of ​​the road surface, obtain the minimum light intensity value and the maximum light intensity value, and calculate the light intensity uniformity of the effective area of ​​the road surface by the ratio of the minimum light intensity value to the maximum light intensity value.

[0014] Furthermore, the process of obtaining the light spot performance value is as follows: Obtain the Stokes vector parameters corresponding to each pixel in the effective area of ​​the road surface, and calculate the pixel polarization degree corresponding to each pixel. The average polarization degree of all pixels in the effective area of ​​the road surface is calculated by averaging the polarization degree of each pixel. The polarization degree deviation value is calculated by the absolute difference between the polarization degree of each pixel and the average polarization degree. After normalizing the polarization deviation value and light intensity value corresponding to each pixel, the light spot performance value is obtained by multiplying them.

[0015] Furthermore, the grayscale correction value is obtained as follows: Obtain the glare grayscale value, the normal grayscale mean value, and the grayscale correction factor; The difference between the glare grayscale value and the normal grayscale mean is calculated. The result of the difference calculation is multiplied by the grayscale correction coefficient to obtain the correction feature value. The difference between the normal grayscale mean and the correction feature value is calculated to obtain the grayscale correction value of the pixel in the micro-glare spot area.

[0016] The beneficial effects of this invention are as follows: 1. Construct a controllable environment simulation test platform to collect effective reflected light data under different incident angles and water film thicknesses. Calculate the degree of polarization and phase difference based on the effective reflected light data, and construct a quasi-polarized light characteristic library. The controllable environment simulation test platform and quasi-polarized light characteristic library enable the acquisition of polarized light data under controllable conditions with multiple incident angles and water film thicknesses. Collect real-time reflected light to extract current scene parameters, compare them with the quasi-polarized light characteristic library to determine if it is elliptically polarized. If it is elliptically polarized, perform phase compensation analysis to construct a composite polarization filter structure, adjusting the current elliptically polarized light to a first-correction light. Quickly identify elliptically polarized light and complete targeted adjustment and standardization correction. Obtain the current phase difference of real-time reflected light and perform offset filtering analysis to obtain the optical path deviation angle. Adjust the filtering direction of the composite polarization filter structure based on the optical path deviation angle. Dynamically adjust the filtering direction of the composite polarization filter structure to achieve adaptation and dynamic control of the polarization filtering direction, significantly improving the targeting and accuracy of polarization filtering.

[0017] 2. Obtain the light intensity uniformity by performing light intensity analysis on the first correction light and determine whether there is residual glare in the first correction light. If so, perform polarization calculation to obtain the light spot performance value and identify the micro-glare spot area. It can detect the glare residue problem of the first correction light and locate the micro-glare spot, providing a clear target basis for subsequent micro-glare elimination. Based on the micro-glare spot area, extract the spotless normal area and perform gray-scale mean calculation to obtain the normal gray-scale mean. Based on the normal gray-scale mean, perform gray-scale correction on the glare gray-scale value of the pixels in the micro-glare spot area to obtain the gray-scale correction value. It can achieve gray-scale correction of micro-glare spots with the normal area as the benchmark, effectively eliminate the influence of micro-glare, and improve the uniformity of the final light effect and visual effect. Attached Figure Description

[0018] The invention will now be further described with reference to the accompanying drawings.

[0019] Figure 1 This is a functional block diagram of a polarizing optical lens glare elimination system for vehicle security in this invention; Figure 2 This is a logic diagram for determining whether the filtering direction of the composite polarization filter structure meets the requirements in this invention. Figure 3 This is a flowchart illustrating the steps of a method for eliminating glare from polarized optical lenses used in vehicle security, as described in this invention. Detailed Implementation

[0020] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments. Example 1

[0021] Please see Figures 1-2As shown, a polarizing optical lens glare cancellation system for vehicle security includes: Reference polarization module: Collects effective reflected light data under different incident angles and water film thicknesses, calculates the degree of polarization and phase difference based on the effective reflected light data, and constructs a quasi-polarized light characteristic library; The process of collecting effective reflected light data under different incident angles and water film thicknesses, and calculating the degree of polarization and phase difference based on the effective reflected light data, is as follows: We obtained mainstream hydrophobic coating products from the market, simulated the residual state of the road surface, and sprayed a certain amount of deionized water to form a water film, restoring the wet road surface environment after rain. An adjustable-angle parallel light source is set to simulate natural light incidence, and a polarized light detection array is deployed to simultaneously collect effective light data (light intensity data and polarization state information) of the reflected light. It should be noted that the incident angle of the simulated natural light ranges from 15° to 75°, corresponding to different solar altitude angles during vehicle movement; For each type of coating sample, under different incident angles and different water film thicknesses, N sets of reflected light and effective light data of the reflected light were collected to ensure that the samples cover the polarization light variation characteristics under complex road conditions. Preferably, the range of different water film thicknesses is 0.1mm-0.5mm, and N is 100; The total light intensity S0, the vertical polarization component S1, the diagonal polarization component S2, and the circular polarization component S3 were obtained by Stokes vector analysis based on the effective light data of reflected light. S1, S2, and S3 are squared and summed. The square root of the sum is then calculated, and the ratio of the square root to S0 is used to calculate the degree of polarization. Understandably, the physical meaning of polarization degree is: it describes the proportion of polarized light component in the total light intensity of a beam of light, intuitively reflecting the degree of polarization of light; when the polarization degree is 0, it means that the beam of light is natural light, and its light vibration direction is completely randomly distributed, with no polarization characteristics; when the polarization degree is 1, it means that the beam of light is completely polarized light (including linearly polarized light, elliptically polarized light, and circularly polarized light), and the light vibration direction shows a regular arrangement pattern; while between 0 and 1, the beam of light is quasi-polarized light, meaning that it is a mixture of completely polarized light and natural light. The closer the polarization degree value is to 1, the higher the proportion of polarized light component, and the more significant the polarization characteristics of the light. The reflected light is determined to be quasi-polarized based on the degree of polarization of the effective reflected light data; The phase difference is calculated by inputting the diagonal polarization component and the circular polarization component of the quasi-polarized light into the arctangent function; It should be noted that the phase difference ranges from -π to π. When the phase difference is 0 or ±π, the reflected light is linearly polarized. When the phase difference is ±π / 2 and the amplitudes of the two orthogonal directions are equal, the reflected light is circularly polarized. When the phase difference is any other value, the reflected light is elliptically polarized. The process of constructing the quasi-polarized light property library is as follows: A quasi-polarized light characteristic library is constructed by associating and storing the degree of polarization and phase difference corresponding to different incident angles and different water film thicknesses. Composite filtering module: Collects real-time reflected light to extract current scene parameters, compares them with the quasi-polarized light characteristic library to determine whether it is elliptically polarized light. If it is, it performs phase compensation analysis to construct a composite polarization filtering structure to adjust the elliptically polarized light into primary correction light. The process of collecting real-time reflected light to extract current scene parameters and comparing them with a quasi-polarized light characteristic library to determine whether it is elliptically polarized light is as follows: Based on real-time reflected light data of different hydrophobic coating residues on wet road surfaces after rain collected during driving, the current polarization degree and current phase difference are calculated. It should be noted that the real-time reflected light data includes the Stokes vector parameters of the reflected light, the incident angle, and the water film thickness; The real-time reflected light data is compared with the quasi-polarized light characteristic library, specifically: The current scene parameters (incident angle, water film thickness) of the real-time reflected light data are matched with the quasi-polarized light characteristic library to obtain scenes of the same type and mark them as reference mapping scenes; The number of times linearly polarized light, elliptically polarized light, and circularly polarized light appear in the reference mapping scene are counted respectively, and the ratios are calculated with the total number of reference mapping scenes to obtain the proportions of linearly polarized light, elliptically polarized light, and circularly polarized light respectively. The proportions of linearly polarized light, elliptically polarized light, and circularly polarized light are sorted in reverse order. If the proportion of elliptically polarized light is the largest, then the completely polarized light contained in the real-time reflected light is marked as elliptically polarized light. It should be noted that the real-time reflected light is quasi-polarized light, which is composed of a part of fully polarized light and a part of natural light. Among them, fully polarized light includes three forms: linearly polarized light, circularly polarized light, and elliptically polarized light. If the condition is met, then phase compensation analysis is performed to construct a composite polarization filter structure. The process of adjusting the elliptically polarized light into primary corrected light is as follows: Obtain the center wavelength of visible light and the refractive index of the compensation material, and obtain the current phase difference of elliptically polarized light; Preferably, the visible light center wavelength is 550nm and the refractive index of the compensation material is 1.5; The phase compensation layer thickness is obtained by multiplying the visible light center wavelength by a times the refractive index of the compensation material, calculating the ratio of the current phase difference to π, and multiplying the results of the product calculation and the ratio calculation. Preferably, a is 4; A phase compensation layer is fabricated on a polarizing optical lens using micro-nano imprinting technology, which corrects elliptically polarized light into linearly polarized light. Mark the current linearly polarized light as the first correction light; It should be noted that the reason for correcting elliptically polarized light to the current linearly polarized light is that elliptically polarized light has an elliptical vibration trajectory, which is formed by the superposition of two orthogonal linearly polarized components with a fixed phase difference. Not only does it have no clear and unified polarization direction, but its phase distribution is also asymmetrical, which does not match the filtering logic of standard polarizing optical lenses. However, linearly polarized light has a clear and single polarization direction, which is perfectly compatible with the filtering logic of standard polarizing optical lenses and can greatly improve the filtering effect. Direction calibration module: Obtain the current phase difference of the real-time reflected light, perform offset filtering analysis to obtain the optical path deviation angle, and determine whether the filtering direction of the composite polarization filter structure meets the requirements based on the optical path deviation angle. If it does not meet the requirements, fine-tune it based on the optical path deviation angle. The process of obtaining the optical path deviation angle by acquiring the current phase difference of the real-time reflected light and performing offset filtering analysis is as follows: A real-time orientation calibration module is constructed, which includes a high-speed acquisition unit, a real-time computing core, and an angle driving unit. Specifically: It should be noted that the high-speed acquisition unit is deployed on the light-emitting side of the composite polarization filter structure and works synchronously with the image sensor of the vehicle camera in the same optical path to capture real-time reflected light; the real-time computing core is equipped with a lightweight computing chip and establishes a local high-speed data interaction channel with the quasi-polarized light characteristic library to avoid the latency problem of cloud transmission; the micro piezoelectric ceramic element of the angle driving unit is flexibly connected to the polarizer frame of the composite polarization filter structure without gaps, which can realize the full-angle adjustment of the polarizer from 0 to 90°, and the adjustment step is preferably 0.01° to meet the calibration requirements of micro-angle offset; Obtain the current phase difference of the real-time reflected light and extract the standard phase difference of the composite polarization filter structure; Among them, the standard phase difference is the theoretical reference value determined by simulation and optimization during the integrated optical design stage of the composite polarization filter structure; The optical path deviation angle is obtained by calculating the arctangent function of the current phase difference and the standard phase difference, and multiplying the calculation result with the polarization direction conversion coefficient. It is understood that the polarization direction conversion coefficient is the conversion coefficient between the phase difference and the polarization direction. Preferably, k is 0.02° / rad. The polarization direction conversion coefficient is obtained by fitting sample data from 100 sets of quasi-polarized light characteristic libraries, which can realize the reasonable conversion of phase difference value to polarization direction angle offset. The optical path deviation angle ranges from -5° to +5°, and the positive and negative values ​​correspond to the clockwise and counterclockwise offset of the polarization direction relative to the preset direction, respectively. This range covers the extreme values ​​of polarization direction offset of the light reflected by the hydrophobic coating residue on the wet road surface. The process of determining whether the filtering direction of the composite polarization filter structure meets the requirements based on the optical path deviation angle, and if not, fine-tuning based on the optical path deviation angle, is as follows: In some embodiments, the absolute value of the calculated optical path deviation angle is compared with a calibration threshold; Preferably, the calibration threshold is 0.5°; If the absolute value of the optical path deviation angle is less than or equal to the calibration threshold, it is determined that the deviation between the current polarization direction of the real-time reflected light and the filtering direction of the composite polarization filter structure meets the filtering requirements, and the composite polarization filter structure keeps the current angle unchanged. If the absolute value of the optical path deviation angle is greater than the calibration threshold, it is determined that the current polarization direction of the real-time reflected light does not meet the requirements of the filtering direction of the composite polarization filter structure, and the filtering direction of the composite polarization filter structure needs to be finely adjusted according to the calculated optical path deviation angle. Those skilled in the art will understand that the purpose of constructing the real-time orientation calibration module is as follows: Although the composite polarization filter structure can correct elliptically polarized light, the dynamic changes in parameters such as the incident angle and water film thickness in the vehicle environment can cause the polarization direction to shift, making it impossible for the fixed filter structure to align; the construction of the real-time orientation calibration module enables the composite polarization filter structure to change synchronously with the polarization direction of the light, ensuring that filtering can always be performed. The technical solution of this embodiment is as follows: A controllable environment simulation test platform is constructed to collect effective reflected light data under different incident angles and water film thicknesses. The degree of polarization and phase difference are calculated based on the effective reflected light data, and a quasi-polarized light characteristic library is constructed. The construction of the controllable environment simulation test platform and the establishment of the quasi-polarized light characteristic library enable the acquisition of polarized light data under controllable conditions with multiple incident angles and water film thicknesses. Real-time reflected light is collected to extract current scene parameters, which are compared with the quasi-polarized light characteristic library to determine whether it is elliptically polarized light. If it is elliptically polarized light, phase compensation analysis is performed to construct a composite polarization filter structure, adjusting the current elliptically polarized light to a first-correction light. Elliptically polarized light is quickly identified and targeted adjustment and standardization correction are completed. The current phase difference of the real-time reflected light is obtained, and offset filtering analysis is performed to obtain the optical path deviation angle. The filtering direction of the composite polarization filter structure is adjusted according to the optical path deviation angle. Dynamic adjustment of the filtering direction of the composite polarization filter structure enables adaptation and dynamic control of the polarization filtering direction, significantly improving the targeting and accuracy of polarization filtering. Example 2

[0022] Please see Figure 1 As shown, a polarizing optical lens glare cancellation system for vehicle security includes: Glare Recognition Module: Acquires the first-correction light and performs light intensity analysis to obtain light intensity uniformity, and determines whether there is residual glare in the first-correction light; if so, it performs polarization calculation to obtain the light spot performance value and identifies the micro-glare spot area; The process of obtaining light intensity uniformity by acquiring the first-correction light and performing light intensity analysis to determine whether there is residual glare in the first-correction light is as follows: Background interference removal is performed on the corrected light. The light data of the effective area of ​​the road surface (total light intensity distribution matrix, pixel-level Stokes vector parameters and pixel two-dimensional spatial coordinates) is filtered out by image segmentation algorithm to remove light signal interference from non-road backgrounds such as sky, road guardrails, and surrounding buildings. The minimum and maximum light intensity values ​​of pixels within the effective area of ​​the road surface are obtained, and the light intensity uniformity of the effective area of ​​the road surface is calculated by the ratio of the minimum and maximum light intensity values. The value of light intensity uniformity ranges from 0 to 1. The closer the light intensity uniformity value is to 1, the more uniform the light intensity distribution of the reflected light from the road surface after filtering, and the less obvious the highlight area on the road surface. Compare light intensity uniformity with glare retention threshold; Preferably, the glare retention threshold is 0.85; If the light intensity uniformity is greater than or equal to the glare residue threshold, it is determined that there is no glare residue after the first correction, and no further analysis is required. If the light intensity uniformity is less than the glare residue threshold, it is determined that there is glare residue in the first correction light, and further polarization analysis is needed to identify the specific location and characteristics of the micro-glare spot. If it exists, polarization calculation is performed to obtain the light spot performance value. The process of identifying the micro-glare light spot region is as follows: If it is determined that there is residual glare from the first correction light, the effective area of ​​the road surface is divided into pixel-level regions, and the Stokes vector parameters corresponding to each pixel are obtained. The pixel polarization degree of each pixel is calculated based on the Stokes vector parameters corresponding to each pixel. The average polarization degree of all pixels in the effective area of ​​the road surface is calculated by averaging the polarization degree of each pixel. The polarization degree deviation value is calculated by the absolute difference between the polarization degree of each pixel and the average polarization degree. After normalizing the polarization degree deviation value and light intensity value corresponding to each pixel, the light spot performance value is obtained by multiplying them. In some embodiments, the spot performance value is compared with the spot performance threshold; If the spot performance value is greater than or equal to the spot performance threshold, then the pixel is marked as a micro-glare spot pixel. If the spot performance value is less than the spot performance threshold, then no processing is performed on that pixel. All micro-glare spot pixels are acquired and eight-neighbor connected component analysis is performed. Adjacent micro-glare spot pixels are merged into regions to form a complete micro-glare spot region. At the same time, the core feature parameters of each micro-glare spot region are recorded, including the center coordinates of the spot, the pixel coverage area, the peak light intensity and the extreme value of polarization. Grayscale correction module: Extract the normal area without glare from the micro-glare spot area, calculate the grayscale mean to obtain the normal grayscale mean, and perform grayscale correction on the pixels in the micro-glare spot area based on the normal grayscale mean to obtain the grayscale correction value. The process of extracting the normal area without glare based on the micro-glare spot area and calculating the normal gray-scale mean is as follows: The micro-glare spot area is fully shielded using masking technology, and the remaining unshielded area in the effective area of ​​the road surface is marked as a normal area without glare spots; Obtain the original grayscale values ​​of all pixels within the normal area without light spots, and construct the original grayscale set; The original grayscale set is subjected to outlier removal processing to remove pixels with grayscale values ​​greater than L1 or less than L2, resulting in a clean grayscale set after outlier removal. Preferably, L1 is 250 and L2 is 5; The arithmetic mean of the pure grayscale set is used to calculate the normal grayscale mean of the normal region without light spots; The process of obtaining grayscale correction value pairs by correcting the grayscale of pixels in the micro-glare spot area based on the normal grayscale mean is as follows: The difference between the glare grayscale value and the normal grayscale mean is calculated. The result of the difference calculation is multiplied by the grayscale correction coefficient to obtain the correction feature value. The difference between the normal grayscale mean and the correction feature value is calculated to obtain the grayscale correction value of the pixel in the micro-glare spot area. Preferably, the grayscale correction factor is 0.7; It should be noted that this correction rule is based on the self-light intensity characteristics of the micro-glare spot area for adaptive adjustment. The more significant the glare feature of a pixel, the greater the correction range, and the smaller the correction range of a pixel with weak glare feature. After correction, the gray value always matches the gray value of the normal area without glare spots. This not only eliminates the micro-glare bright spots, but also maintains the original features of the road surface texture by preserving the pixel gray-level gradient, ensuring that the on-board computer vision algorithm can effectively extract key information such as lane lines, markings, and obstacles on the road surface. The technical solution of this embodiment is as follows: First, the light intensity of the corrected light is analyzed to obtain the light intensity uniformity, and it is determined whether there is residual glare in the first corrected light. If so, polarization calculation is performed to obtain the light spot performance value, and the micro-glare spot area is identified. This method can detect the residual glare problem of the first corrected light and locate the micro-glare spot, providing a clear target basis for subsequent micro-glare elimination. Based on the micro-glare spot area, a normal area without light spots is extracted, and the gray-scale mean is calculated to obtain the normal gray-scale mean. Based on the normal gray-scale mean, the gray-scale value of the glare gray-scale of the pixels in the micro-glare spot area is corrected to obtain the gray-scale correction value. This method can achieve gray-scale correction of the micro-glare spot using the normal area as a reference, effectively eliminating the influence of micro-glare and improving the uniformity and visual effect of the final light effect. Example 3

[0023] Please see Figure 3 As shown, a method for eliminating glare from polarized optical lenses used in vehicle security systems includes the following steps: Step 1: Collect effective reflected light data under different incident angles and water film thicknesses, calculate the degree of polarization and phase difference based on the effective reflected light data, and construct a quasi-polarized light characteristic library; Step 2: Collect real-time reflected light to extract current scene parameters, compare them with the quasi-polarized light characteristic library to determine whether it is elliptically polarized light. If it is, perform phase compensation analysis to construct a composite polarization filter structure and adjust the elliptically polarized light into a primary correction light. Step 3: Obtain the current phase difference of the real-time reflected light and perform offset filtering analysis to obtain the optical path deviation angle. Based on the optical path deviation angle, determine whether the filtering direction of the composite polarization filter structure meets the requirements. If not, make fine adjustments based on the optical path deviation angle. Step 4: Obtain the first correction light and perform light intensity analysis to obtain the light intensity uniformity, and determine whether there is residual glare in the first correction light; if so, perform polarization calculation to obtain the light spot performance value and identify the micro-glare spot area. Step 5: Extract the normal area without glare from the micro-glare spot area, calculate the normal gray-scale mean, and perform gray-scale correction on the pixels in the micro-glare spot area based on the normal gray-scale mean to obtain the gray-scale correction value.

[0024] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A polarizing optical lens glare cancellation system for vehicle-mounted security, characterized in that: Includes the following modules: Reference polarization module: Collects effective reflected light data under different incident angles and water film thicknesses, calculates the degree of polarization and phase difference based on the effective reflected light data, and constructs a quasi-polarized light characteristic library; Composite filtering module: Collects real-time reflected light to extract current scene parameters, compares them with the quasi-polarized light characteristic library to determine whether it is elliptically polarized light. If it is, it performs phase compensation analysis to construct a composite polarization filtering structure to adjust the elliptically polarized light into primary correction light. Direction calibration module: Obtain the current phase difference of the real-time reflected light, perform offset filtering analysis to obtain the optical path deviation angle, and determine whether the filtering direction of the composite polarization filter structure meets the requirements based on the optical path deviation angle. If it does not meet the requirements, fine-tune it based on the optical path deviation angle. Glare Recognition Module: Acquires the first-correction light and performs light intensity analysis to obtain light intensity uniformity, and determines whether there is residual glare in the first-correction light; if so, it performs polarization calculation to obtain the light spot performance value and identifies the micro-glare spot area; Grayscale correction module: Extract the normal area without glare from the micro-glare spot area, calculate the normal grayscale mean, and perform grayscale correction on the pixels in the micro-glare spot area based on the normal grayscale mean to obtain the grayscale correction value.

2. The polarized optical lens glare elimination system for vehicle-mounted security as described in claim 1, characterized in that: The process of obtaining the degree of polarization and phase difference is as follows: The effective light data of the reflected light were obtained and the total light intensity S0, the vertical polarization component S1, the diagonal polarization component S2, and the circular polarization component S3 were obtained by Stokes vector analysis. S1, S2, and S3 are squared and then summed. The square root of the sum is calculated, and the ratio of the square root to S0 is used to calculate the degree of polarization. The phase difference is obtained by performing arctangent analysis based on S2 and S3.

3. A polarizing optical lens glare elimination system for vehicle-mounted security as described in claim 2, characterized in that: The method for performing arctangent analysis is as follows: The phase difference is calculated by inputting the diagonal and circular polarization components of the reflected light into the arctangent function.

4. A polarizing optical lens glare elimination system for vehicle-mounted security as described in claim 1, characterized in that: The process of performing phase compensation analysis is as follows: The real-time reflected light is determined to be elliptically polarized light, and the current phase difference is obtained; Obtain the visible light center wavelength and the refractive index of the compensation material. Multiply the visible light center wavelength by a times the refractive index of the compensation material. Calculate the ratio of the current phase difference to π. Multiply the results of the product calculation and the ratio calculation to obtain the phase compensation layer thickness.

5. A polarizing optical lens glare elimination system for vehicle-mounted security as described in claim 4, characterized in that: The process of determining whether the real-time reflected light is elliptically polarized light is as follows: The current scene parameters of the real-time reflected light are obtained and matched with the quasi-polarized light characteristic library to obtain scenes of the same type and mark them as reference mapping scenes; The number of times linearly polarized light, elliptically polarized light, and circularly polarized light appear in the reference mapping scene are counted respectively, and the ratios are calculated with the total number of reference mapping scenes to obtain the proportions of linearly polarized light, elliptically polarized light, and circularly polarized light respectively. If the proportion of elliptically polarized light is at its maximum, then the real-time reflected light will be determined to be elliptically polarized light.

6. A polarizing optical lens glare elimination system for vehicle-mounted security as described in claim 1, characterized in that: The optical path deviation angle is obtained as follows: Obtain the current phase difference of the real-time reflected light and extract the standard phase difference of the composite polarization filter structure; Obtain the polarization direction conversion coefficient, calculate the arctangent function of the current phase difference and the standard phase difference, and multiply the calculation result with the polarization direction conversion coefficient to obtain the optical path deviation angle.

7. A polarizing optical lens glare elimination system for vehicle-mounted security as described in claim 1, characterized in that: The process of determining whether glare remains after a single correction is as follows: The intensity uniformity of light is obtained by acquiring a corrected light source and performing light intensity analysis. Obtain the glare retention threshold. If the light intensity uniformity is less than the glare retention threshold, it is determined that there is glare retention in the first correction light.

8. A polarizing optical lens glare elimination system for vehicle-mounted security as described in claim 7, characterized in that: The process of performing light intensity analysis is as follows: Background interference removal is performed on the corrected light, and the light data of the effective area of ​​the road surface is selected by image segmentation algorithm. Extract the light intensity value of each pixel from the light data of the effective area of ​​the road surface, obtain the minimum light intensity value and the maximum light intensity value, and calculate the light intensity uniformity of the effective area of ​​the road surface by the ratio of the minimum light intensity value to the maximum light intensity value.

9. A polarizing optical lens glare elimination system for vehicle-mounted security as described in claim 1, characterized in that: The process of obtaining the light spot performance value is as follows: Obtain the Stokes vector parameters corresponding to each pixel in the effective area of ​​the road surface, and calculate the pixel polarization degree corresponding to each pixel. The average polarization degree of all pixels in the effective area of ​​the road surface is calculated by averaging the polarization degree of each pixel. The polarization degree deviation value is calculated by the absolute difference between the polarization degree of each pixel and the average polarization degree. After normalizing the polarization deviation value and light intensity value corresponding to each pixel, the light spot performance value is obtained by multiplying them.

10. A polarizing optical lens glare elimination system for vehicle-mounted security as described in claim 1, characterized in that: The grayscale correction value is obtained as follows: Obtain the glare grayscale value, the normal grayscale mean value, and the grayscale correction factor; The difference between the glare grayscale value and the normal grayscale mean is calculated. The result of the difference calculation is multiplied by the grayscale correction coefficient to obtain the correction feature value. The difference between the normal grayscale mean and the correction feature value is calculated to obtain the grayscale correction value of the pixel in the micro-glare spot area.

Images