A driver's anti-glare glasses for nighttime use and an anti-glare method

By integrating a miniature camera and illuminance sensor into the driver's nighttime anti-glare glasses, and combining a liquid crystal pixel array with Gaussian smoothing, pixel-level adjustment of local glare areas is achieved, solving the problem of loss of visual clarity in existing technologies and ensuring high transmittance and stability of the entire field of vision.

CN123113175APending Publication Date: 2026-11-17INST OF METROLOGY OF HEBEI PROVINCE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611556839.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-09
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing anti-glare products for drivers at night cannot achieve precise anti-glare in localized areas of strong light in complex and ever-changing nighttime lighting environments, resulting in loss of visual clarity or delayed adjustment response, making it difficult to meet the requirements for high precision, high stability, and fully automatic anti-glare.

Method used

A miniature camera and illuminance sensor are used to identify glare areas. Pixel-level adjustment is achieved through a liquid crystal pixel array. Combined with perspective transformation matrix and Gaussian smoothing, the transmittance is dynamically adjusted to reduce the light intensity in glare areas and maintain high transmittance in non-glare areas.

Benefits of technology

It achieves precise adjustment in local glare areas, ensuring clear visibility across the entire field of vision, including non-glare areas, reducing abrupt dimming boundaries and flickering, and improving visual comfort and the ability to identify road details.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123113175A_ABST
    Figure CN123113175A_ABST
Patent Text Reader

Abstract

This invention discloses an anti-glare glasses for drivers at night and an anti-glare method. The glasses include a frame, temples, a data acquisition module, a processing module, a driving module, a display module, and a power supply component. The data acquisition module acquires images of the forward field of view through a miniature camera and acquires ambient illuminance through an illuminance sensor. The display module includes liquid crystal displays that serve as left and right functional lenses, each functional lens having an array of liquid crystal pixels with independently adjustable transmittance. The processing module combines the image and illuminance signal to identify glare areas, calculates the target transmittance based on pixel brightness and brightness gradient, and generates pixel-level adjustment commands using the coordinate correspondence between the image and the left and right functional lenses. This invention maintains high transmittance in non-glare areas, thus balancing visual comfort in glare areas with the ability to identify road details in non-glare areas, reducing glare while ensuring clear vision across the entire field of view, including non-glare areas.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to anti-glare glasses for drivers at night and an anti-glare method, which can reduce the interference of local strong light glare when driving at night and ensure the clarity of the entire field of vision, including non-glare areas, and belongs to the field of protective glasses technology. Background Technology

[0002] Glare from oncoming high beams, strong urban street lighting, and reflections from water or snow on the road during nighttime driving is a significant factor contributing to visual fatigue, blurred vision, and even traffic accidents. Existing anti-glare products for drivers at night mainly fall into two categories: traditional fixed-filter anti-glare mirrors, whose transmittance remains constant and cannot adapt to the complex and ever-changing nighttime lighting environment. They are ineffective at preventing glare in bright light and can lead to insufficient visibility and loss of crucial road information in low light conditions. The other category consists of simple photoelectric sensor anti-glare products, which only allow for rough adjustment of overall transmittance and cannot precisely target areas of strong light. This can result in loss of visual clarity in non-glare areas, and the delayed adjustment response makes it difficult to quickly respond to sudden glare interference.

[0003] Currently, some anti-glare devices integrating image acquisition and dimming technologies are bulky, requiring integration with the vehicle's center console, making them inconvenient to wear and use. Furthermore, their dimming precision is insufficient, making it difficult to balance anti-glare effectiveness with visual clarity. While smart anti-glare glasses have emerged in recent years to reduce transmittance at the point of strong light incidence, some products still suffer from insufficient acquisition precision, abrupt light control boundaries, insufficient adaptive dynamic dimming capabilities, or redundant hardware, failing to meet drivers' high-precision, high-stability, and fully automatic anti-glare requirements for nighttime driving. Therefore, developing portable anti-glare glasses that can identify localized glare and dynamically adjust the transmittance of corresponding areas has practical application value. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by providing anti-glare glasses for drivers at night and an anti-glare method, so as to achieve pixel-level adjustment of local glare areas, thereby reducing glare while ensuring the clarity of the entire field of vision, including non-glare areas.

[0005] The problem described in this invention is solved by the following technical solution:

[0006] A driver's nighttime anti-glare glasses includes a frame, a left temple, a right temple, core functional components, and a power supply component. The core functional components include a data acquisition module, a processing module, a driving module, and a display module. The data acquisition module includes a miniature camera for acquiring images of the forward field of view and an illuminance sensor for acquiring ambient light. The output of the data acquisition module is connected to the processing module, which is connected to the display module via the driving module. The power supply component powers the core functional components. The display module includes liquid crystal displays that serve as left and right functional lenses, each functional lens having an array of liquid crystal pixels capable of independently adjusting transmittance. The processing module is configured to identify glare areas based on image and illuminance signals, calculate the target transmittance based on the brightness and brightness gradient of each pixel within the glare area, and generate pixel-level adjustment commands based on the coordinate correspondence between the camera image and the pixels of the left and right functional lenses. The driving module adjusts the transmittance of the corresponding pixels according to the pixel-level adjustment commands, thereby reducing the transmittance in the glare area while maintaining high transmittance in the non-glare area.

[0007] The aforementioned anti-glare glasses for drivers at night have a miniature camera mounted on the frame with the lens facing forward; an illuminance sensor mounted on the frame; and a processing module and a drive module located inside the temple. The processing module includes a microcontroller, a storage circuit, a reset circuit, and a watchdog circuit. The storage circuit stores preprocessing, glare recognition, transmittance calculation, instruction output, and low-power management programs.

[0008] The aforementioned anti-glare glasses for drivers at night use feature a liquid crystal pixel array driven by a thin-film transistor active matrix, with each pixel equipped with an independent thin-film transistor switch and storage capacitor. The functional lens comprises, from the outside to the inside, a scratch-resistant and abrasion-resistant layer, a high-transparency protective layer, a liquid crystal pixel layer, a driving electrode layer, and a substrate light-transmitting layer. The driving electrode layer comprises a pixel electrode matrix formed by a transparent conductive thin film. The processing module stores the calibration relationship between the pixel transmittance and the driving voltage, and determines the pixel driving voltage based on the calibration relationship.

[0009] The aforementioned anti-glare glasses for drivers at night include a power supply component comprising a rechargeable lithium polymer battery embedded in the rear of the left and right temples and connected in parallel, as well as a charging circuit, a protection circuit, and a power detection circuit; a charging interface is provided on the right temple, and a power indicator light and a touch power switch are provided on the left temple; the processing module is also connected to a photodetector located in the non-viewing area at the edge of the frame or functional lens, and the photodetector collects the light intensity after passing through the functional lens through a light guide tube or sampling hole to correct the driving voltage of the corresponding pixel area.

[0010] A method for preventing glare at night for drivers, the method comprising the following steps:

[0011] ① Perform coordinate and transmittance calibration: Establish the perspective transformation matrix between the image coordinates of the miniature camera and the pixel coordinates of the left and right functional lenses, and establish the calibration relationship between the transmittance of each liquid crystal pixel and the driving voltage. Store the perspective transformation matrix and calibration relationship in the processing module.

[0012] ② Acquire and preprocess two signals: continuously acquire images of the field of view in front through a miniature camera, and acquire ambient illuminance through an illuminance sensor; the processing module performs noise reduction and grayscale processing on the images, and performs low-pass filtering on the illuminance signal;

[0013] ③ Identify glare areas: Calculate the ambient brightness threshold based on the filtered ambient illuminance, and calculate the local brightness threshold based on the average gray level of the local window in the image. Take the larger of the two as the final segmentation threshold, and determine the glare area through binarization, morphological processing and contour filtering.

[0014] ④ Calculate target transmittance: Normalize the brightness and brightness gradient of each image pixel within the glare area, and calculate the glare adjustment transmittance according to the following formula:

[0015] T(u, v)=T min +(T max -T min )·(1-I norm (u, v)·(1-β·G norm (u, v))).

[0016] In the formula, T(u, v) is the transmittance of the pixel in the glare region, T min and T max These represent the minimum and maximum transmittance, respectively. norm (u, v) represents the normalized luminance, G norm (u, v) represents the normalized luminance gradient, and β is the gradient weight coefficient, with a value range of [0,1]. The transmittance of the non-glare region remains constant at T. max ;

[0017] ⑤ Determine the dimming pixels of the left and right functional lenses: Map the glare image pixels to the left and right functional lenses respectively using the corresponding perspective transformation matrix; for non-integer coordinates, distribute the target transmittance to the four adjacent functional lens pixels according to the bilinear interpolation weight; when the same functional lens pixel receives dimming contributions from multiple glare image pixels, use a weighted geometric average to synthesize the transmittance of the functional lens pixel.

[0018] ⑥ Perform smoothing: Spatial domain Gaussian smoothing is performed on the mapped glare adjustment area, and the smoothed transmittance of the current frame is time-domain weighted fusion with the output transmittance of the previous frame; the functional lens pixels in the non-glare area of ​​the current frame maintain the set high transmittance.

[0019] ⑦ Output pixel-level driving instructions: Based on the calibration relationship, the processed transmittance is converted into a driving voltage, which is then controlled by the driving module to control the corresponding liquid crystal pixels in the left and right functional lenses respectively, so that the transmittance of the glare area is reduced and the non-glare area maintains high transmittance. Steps ② to ⑦ are executed cyclically as the image and illuminance signal are updated.

[0020] In the above-mentioned anti-glare method for drivers at night, step ② involves using Gaussian filtering to remove image noise; the image is then converted to grayscale, and the calculation formulas for low-pass filtering of the glare signal are as follows:

[0021] I gray (u, v) = 0.299R + 0.587G + 0.114B

[0022] E filtered (t)=αE(t)+(1-α)E filtered (t-1)

[0023] In the formula, R, G, and B are the red, green, and blue channel luminance values ​​of the image pixels, respectively, and E(t) is the original illuminance. filtered (t) represents the current filtered output, E filtered (t-1) represents the output of the previous filter, α=0.1; the ambient brightness threshold, local brightness threshold, and final segmentation threshold in step ③ are respectively:

[0024] T env =k1·E filtered

[0025] T local (u, v)=k2·I mean (u, v)

[0026] T final (u, v)=max(T env , T local (u, v))

[0027] In the formula, k1 is the ambient brightness coefficient, k2 is the local threshold coefficient, and I mean (u, v) represents the average gray level of a local window centered at image coordinates (u, v); satisfying I gray (u, v) ≥ T final The pixels of (u, v) are set to 1, and the rest of the pixels are set to 0. The glare area is determined by morphological processing, edge extraction and contour filtering.

[0028] In the above-mentioned anti-glare method for drivers at night, step ④ uses the Sobel operator to calculate the horizontal brightness gradient G of the image. x and vertical brightness gradient G y The luminance gradient, normalized luminance, and normalized luminance gradient are calculated using the following formulas:

[0029] Brightness gradient:

[0030] Normalized brightness: It is the original brightness I of the pixel. gray The result of mapping (u, v) to the interval [0,1];

[0031] Gradient normalization:

[0032] In the formula, I max I min These represent the maximum and minimum gray levels within the glare region R, respectively, and G. max The maximum luminance gradient within this region; the upper limit of transmittance T. max =95%, lower limit of transmittance T min =5%, gradient weight coefficient β=0.3.

[0033] The above-mentioned anti-glare method for drivers at night involves the following steps: In step ①, a checkerboard calibration board is used to extract the corner points of the image, and the corresponding points of the corner point's line of sight and the pixel arrays of the left and right functional lenses are determined. The perspective transformation matrix of each functional lens is obtained using at least four sets of corresponding points. In step ⑤, the coordinate mapping relationship between the glare image and the functional lenses is as follows:

[0034]

[0035] In the formula, H is the perspective transformation matrix, (u, v) are the camera image coordinates, and (X, Y) are the mapped functional lens coordinates; mapping matrices H are established for the left and right functional lenses respectively. L and H R Let the non-integer positions obtained by mapping be (x, y), let i0=[x], i1=i0+1, j0=[y], j1=j0+1, and let d x =x- i0、d y =y-j0, the bilinear interpolation weights of the four adjacent pixels are as follows:

[0036] w 00 =(1-d x )·(1- d y )

[0037] w 10 =d x ·(1- d y )

[0038] w 01 =(1-d x )·d y

[0039] w11 =d x ·d y

[0040] The weights satisfy w 00 + w 10 + w 01 + w 11 =1,

[0041] For each frame of the image, initialize the logarithmic transmittance accumulation array T[i][j] and the weight accumulation array W[i][j], both with initial values ​​of 0; for each glare image pixel, calculate its target transmittance T. target The corresponding weights w are accumulated using the following formula:

[0042] T[i][j]←T[i][j]+w·ln(T target )

[0043] W[i][j]←W[i][j]+w,

[0044] When W[i][j] > 0:

[0045] T final [i][j]=exp(T[i][j] / W[i][j])

[0046] When W[i][j]=0:

[0047] T final [i][j]=1.0

[0048] The final transmittance sent to the driver module is:

[0049] T drive [i][j]= T base ×T final [i][j]

[0050] In the formula, T base Based on the transmittance, T final [i][j] represents the cumulative transmittance, T drive [i][j] represents the transmittance that is finally sent to the driver module.

[0051] The spatial domain Gaussian smoothing and temporal domain weighted fusion formulas in step ⑥ of the above-mentioned anti-glare method for drivers at night are as follows:

[0052]

[0053] T final (i, j, t)=γ·T smooth (i, j, t)+(1-γ)·T final (i, j, t-1)

[0054] In the formula, G(m, n) is the Gaussian kernel function, and T smooth (i, j) represents the transmittance after spatial smoothing, and T final (i, j, t) represents the output transmittance after temporal fusion, where t is the current frame and t-1 is the previous frame; the Gaussian kernel window is 3×3, k=1, σ=0.8, and the inter-frame fusion coefficient γ=0.2, with a value range of 0 < γ < 1; spatial filtering is performed only on pixels within the glare area, while the transmittance in non-glare areas remains unchanged. max .

[0055] In the above-mentioned anti-glare method for drivers at night, step ⑦ involves determining the initial driving voltage based on a pre-calibrated relationship between transmittance and driving voltage, collecting the actual emitted light intensity through the functional lens in the corresponding sampling area using a photodetector, and determining the target emitted light intensity according to the following formula:

[0056] I target = T target ×I in

[0057] In the formula, T target For the target transmittance, I in The current incident light intensity; based on the actual emitted light intensity I actual With the target emitted light intensity I target The deviation is corrected to adjust the driving voltage of the corresponding pixel until the following condition is met:

[0058] |I actual - I target | ≤ ε

[0059] In the formula, ε is the preset light intensity control accuracy threshold; when the incident light intensity changes, I is updated. in And recalculate I target .

[0060] The aforementioned anti-glare method for drivers at night calculates the CIE XYZ tristimulus values ​​before and after dimming based on the incident spectrum and the spectral transmittance of the functional lens for the traffic colors that need protection. These values ​​are then converted to CIE LAB color coordinates, and the color difference before and after dimming is evaluated using CIEDE2000 color difference. Finally, a binary search is used to find the lowest transmittance T that meets the color recognition requirements. min color ; and compare it with the glare protection transmittance T glare Synthetic, target transmittance is:

[0061] T target = max(T min color , T glare , 0.1)

[0062] In the formula, 0.1 is the lower limit of absolute transmittance; the color type that needs to be protected is determined based on the scene analysis, and the corresponding color difference threshold is selected.

[0063] The above-mentioned anti-glare method for drivers at night involves pre-measuring the transmittance of liquid crystal pixels under different wavelengths and driving parameters to establish a calibration relationship between driving parameters, wavelength, and transmittance. Under given driving parameters, the original spectral transmittance, wavelength compensation coefficient, and corrected transmittance are calculated using the following formulas:

[0064]

[0065]

[0066]

[0067] In the formula, λ is the wavelength, and I in (λ) and I out (λ) represent the incident light intensity and the emitted light intensity at that wavelength, respectively, T target (λ) represents the target spectral transmittance, K fb The closed-loop feedback coefficient is derived from the colorimetric deviation ΔE. ab The ratio to the threshold is determined; the processing module adjusts the driving parameters of the corresponding pixel according to the calibration relationship to correct its spectral response.

[0068] This invention simultaneously acquires images of the forward field of view and ambient illumination, identifies local glare based on dynamic thresholds, determines target transmittance by combining pixel brightness and brightness gradient, and implements pixel-level adjustment through the coordinate relationship calibrated by the left and right functional lenses respectively. This method reduces strong light interference in glare areas while maintaining high transmittance in non-glare areas, balancing visual comfort in glare areas with the identification of road details in non-glare areas, thereby ensuring clear vision across the entire field of view, including non-glare areas.

[0069] This invention employs bilinear interpolation, weighted geometric averaging, and spatial and temporal smoothing to ensure smoother transmittance changes at dimming boundaries and between consecutive frames, reducing blocky obstructions and dimming flicker. Through transmittance and drive voltage calibration, emitted light intensity feedback, and spectral calibration, it improves dimming consistency and reduces color distortion in traffic signals and road signs. The processing module, drive module, and power supply components are integrated inside the frame and temples for easy wear and use. Attached Figure Description

[0070] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0071] Figure 1 This is a schematic diagram of the structure of the present invention;

[0072] Figure 2This is a schematic diagram of the internal module layout of the present invention, wherein (a) is the left temple and (b) is the right temple;

[0073] Figure 3 This is an electrical connection block diagram of the present invention;

[0074] Figure 4 This is a flowchart of the process of the present invention.

[0075] The following are the labels in the diagram: 1. Frame; 2. Left temple; 3. Right temple; 4. Power indicator light; 5. Touch power switch; 6. Protection circuit; 7. Power supply component; 8. Charging interface; LCD, liquid crystal display (functional lens); U1, processing module; U2, drive module; CAM, miniature camera; ALS, illuminance sensor. Detailed Implementation

[0076] See Figures 1 to 3 The driver's nighttime anti-glare glasses of the present invention include a frame 1, a left temple 2, a right temple 3, core functional components, and a power supply component 7. The core functional components include a data acquisition module, a processing module U1, a driving module U2, and a display module. The output of the data acquisition module is connected to the processing module U1, and the processing module U1 is connected to the display module via the driving module U2. The power supply component 7 supplies power to each functional component. The display module includes liquid crystal displays (LCDs) that serve as the left and right functional lenses, respectively.

[0077] The frame 1 adopts a lightweight full-frame structure and can be made of TR90 polymer material. Anti-slip silicone sleeves are provided at the ends of the temples. A miniature camera (CAM) is mounted on the frame 1 with its lens facing forward. An illuminance sensor (ALS) is also mounted on the frame 1 to collect ambient illuminance data. The processing module U1 and the drive module U2 are located inside the temples, while the power supply components 7 are distributed at the rear of the left and right temples. The illustrated circuit uses an OV2710 image sensor, a BH1750FVI illuminance sensor, an STM32L431CCT6 microcontroller, and an SSD1963 display driver chip. Each module transmits acquired data and drive commands through corresponding interfaces.

[0078] The processing module U1 includes a microcontroller, a storage circuit, a reset circuit, and a watchdog circuit. The storage circuit stores data preprocessing, glare recognition, transmittance calculation, coordinate mapping, drive command output, and low-power management programs, as well as data such as coordinate transformation matrices and pixel calibration tables. The processing module U1 determines the position, area, and brightness of the glare area based on the two acquired signals, calculates the pixel-level target transmittance, and then generates adjustment commands for the left and right functional lenses.

[0079] Each functional lens employs a liquid crystal pixel array driven by a thin-film transistor active matrix, with each pixel having an independent thin-film transistor switch and storage capacitor. From the outside in, the functional lens is sequentially composed of a scratch-resistant layer, a high-transparency protective layer, a liquid crystal pixel layer, a driving electrode layer, and a substrate light-transmitting layer. A transparent conductive film forms the pixel electrode matrix. The driving module U2 works in conjunction with the pixel array's driving circuit to apply a driving voltage to the corresponding pixel to adjust its transmittance and color depth.

[0080] The power supply component 7 includes rechargeable lithium polymer batteries located at the rear of the left and right temples, connected in parallel. Each battery can be 250 mAh, totaling 500 mAh in parallel, with a nominal voltage of 3.7 V. The charging circuit can use a TP4056, and the protection circuit 6 can use a DW01A MOSFET in conjunction with an 8205A MOSFET to protect against overcharging, over-discharging, overcurrent, and short circuits. The charging interface 8 is located on the right temple 3, while the power indicator light 4 and touch-sensitive power switch 5 are located on the left temple 2. A power detection circuit is used to detect the remaining battery level.

[0081] I. Calibration and Control of Pixel Transmittance and Emitted Light Intensity

[0082] The transmittance of a liquid crystal pixel has a non-linear relationship with the applied voltage, T(V). By adjusting the voltage to change the deflection angle of the liquid crystal molecules, the light transmittance can be controlled within a range of 5% to 95%. Before leaving the factory, a 0-5V stepped voltage is applied to each pixel under a standard light source, and the corresponding transmittance is measured to establish a pixel-level calibration table. Temperature is collected by a thermistor to dynamically correct the driving voltage.

[0083] Using a 12-bit DAC signal, the voltage regulation accuracy is calculated using the following formula:

[0084] 5V / 4096≈1.22mV

[0085] This value represents the quantization interval of the driving voltage. The transmittance adjustment accuracy is also determined by the liquid crystal response characteristics and calibration error. The processing module U1 determines the driving voltage based on the target transmittance by looking up a table or interpolation, and the driving module U2 outputs it to the corresponding pixel.

[0086] A photodetector is placed in the non-viewing area at the edge of the frame 1 or the functional lens to obtain the light intensity transmitted through the functional lens via a light guide or sampling hole. The detector can be a silicon-based photodiode. The pixel array is divided into several sampling regions, and the transmitted light intensity of a single pixel within each region is estimated using the detection results and spatial interpolation.

[0087] By fixing the incident light intensity, adjusting the pixel transmittance and collecting the corresponding outgoing light intensity, a calibration relationship is established:

[0088]

[0089] In the formula, I in For the incident light intensity, I out Let T be the emitted light intensity and T be the pixel transmittance. The total transmittance range is calibrated for each pixel or region, and a calibration table or fitting curve is established; the polynomial fitting formula is:

[0090] I out =aT 2 +bT+c

[0091] In the formula, a, b, and c are fitting coefficients. Based on the target transmittance T... target With the current incident light intensity I in Calculate the intensity of the emitted light from the target:

[0092] I target = T target ×I in

[0093] If the incident light intensity fluctuates, update I in real time. in And recalculate I target Processing module U1 compares the actual light intensity I. actual Based on the target light intensity, the transmittance is corrected by adjusting the pixel driving parameters until the following conditions are met:

[0094] |I actual - I target | ≤ ε

[0095] In the formula, ε is the preset light intensity control accuracy threshold.

[0096] II. Coordinate Registration of Camera Image with Left and Right Functional Lenses

[0097] During factory calibration, the miniature camera CAM and the left and right functional lenses maintain the installation positions used in actual operation. A checkerboard calibration board with a grid side length of 10 mm and an array of 9×7 or 11×9 is used. Multiple images are acquired by changing the distance and orientation. The Harris or Shi-Tomasi algorithm is used to extract corner points and determine the corresponding points of the corner point lines of sight and the pixel arrays of the left and right functional lenses.

[0098] Based on the pinhole camera model, the physical corner coordinates (X) of the calibration board are determined. w , Y w Z w The relationship between the coordinates (u, v) and the camera image coordinates is as follows:

[0099]

[0100] In the formula, K is the camera intrinsic parameter matrix, R is the rotation matrix, and t is the translation vector; with the center of the calibration board as the origin of the world coordinate system, Z... w=D. The projection parameters are solved by using multiple sets of calibration images with different poses; when optical distortion exists, distortion correction is performed first, and then the pixel coordinate correspondence is established.

[0101] The mapping from camera images to functional lens pixels uses perspective transformation:

[0102]

[0103] In the formula, H is a 3×3 perspective transformation matrix, which is solved by at least four sets of corresponding points (u, v) and (X, Y). The mapping matrix H for the left functional lens is established separately. L And the mapping matrix H of the right functional lens R And store it. For parallax caused by close-range targets, distance compensation or dynamic calibration methods can be used to correct it.

[0104] Optional dynamic calibration utilizes the positional changes of the wearer's facial features such as the corners of the eyes and the root of the nose relative to the glasses to correct mapping deviations caused by glasses sliding or changes in wearing position; the positioning camera is set inside the frame, and the front camera is used for glare recognition.

[0105] Suppose that the pixel coordinates (u, v) of a certain camera, after perspective transformation, yield theoretical position coordinates (x, y) on the left functional lens, where x = i L y=j L It is a floating-point number.

[0106] Determine four adjacent integer pixels, let:

[0107] i0=[x], i1=i0+1, j0=[y], j1=j0+1

[0108] The integer coordinates of the four adjacent pixels are as follows:

[0109] P 00 = (i0, j0), P 10 =(i1, j0)P 01 =(i0, j1)P 11 = (i1, j1)

[0110] Calculate the weighting coefficients

[0111] Define d x =x - i0,d y =y-j0, then d x d y ∈[0,1).

[0112] The bilinear interpolation weights (i.e., the proportion of the camera pixel's "dimming contribution" allocated to each functional lens pixel) are:

[0113] w00 =(1-d x )·(1- d y )

[0114] w 10 =d x ·(1- d y )

[0115] w 01 =(1-d x )·d y

[0116] w 11 =d x ·d y

[0117] The weights satisfy w 00 + w 10 + w 01 + w 11 =1.

[0118] The dimming contribution of an image pixel is distributed to four adjacent actual liquid crystal pixels according to the weights mentioned above, so that the non-integer coordinate mapping changes continuously. A functional lens pixel may receive dimming contributions from multiple camera pixels, which are accumulated using a weighted geometric average.

[0119] For each frame of the image, generate a logarithmic transmittance accumulation array T[i][j] (initialized to 1.0, i.e., full transmittance) and a weight accumulation array W[i][j], both initially set to 0. The target transmittance of the pixels in the glare area is T. target , satisfying 0 < T target ≤1, for example 0.15; sum the products of its logarithm and the weights:

[0120] T[i][j]←T[i][j]+w·ln(T target )

[0121] Alternatively, a weighted average method can be used: accumulate w into the total weight W[i][j] of that pixel:

[0122] W[i][j]←W[i][j]+w.

[0123] When W[i][j] > 0:

[0124] T final [i][j]=exp(T[i][j] / W[i][j])

[0125] When W[i][j]=0:

[0126] T final [i][j]=1.0 (Full transparency)

[0127] The final transmittance sent to the driver module is:

[0128] T drive [i][j]= T base ×T final [i][j]

[0129] In the formula, T base The base transmittance can be selected according to ambient light, for example, 0.8; T final [i][j] is normally no greater than 1; if it equals 1, no additional dimming is applied. If T is taken... base =T max Pixels that do not contribute to glare maintain maximum transmittance.

[0130] III. Preprocessing of Two Signals and Glare Area Identification

[0131] See Figure 4 A miniature camera (CAM) captures images of the foreground field of view, while an illuminance sensor (ALS) collects ambient illuminance. The image signal is used to determine the location and outline of the glare area, and the illuminance signal assists in threshold segmentation. Illuminance sensors can be installed on the left and right sides of the frame, allowing for independent adjustment of the two functional lenses.

[0132] Noise reduction: Gaussian filtering is used. To remove image noise, σ = 1.0. Here, (x, y) are the coordinates of the current pixel within the convolution kernel relative to the kernel center. For example, the center coordinates of a 3×3 convolution kernel are (0, 0), the pixel above it is (0, -1), and the pixel to its right is (1, 0). σ is the standard deviation, a core parameter of Gaussian filtering that determines the smoothness of the filter. A smaller σ means the weights are more concentrated at the kernel center, resulting in a weaker smoothing effect and better preservation of image details; a larger σ means the weights are more dispersed, resulting in a stronger smoothing effect and a higher degree of image blurring.

[0133] Convert a color image to a grayscale image:

[0134] I gray (u, v) = 0.299R + 0.587G + 0.114B

[0135] In the formula, R, G, and B are the luminance values ​​of the red, green, and blue channels of a pixel, respectively, ranging from 0 to 255, and the sum of the three weights is 1. The color image is also used for subsequent colorimetric analysis.

[0136] An illuminance sensor (ALS, such as the TSL 2561) collects ambient illuminance E (in lux) at a frequency of 100Hz. The illuminance signal is filtered using a first-order low-pass filter.

[0137] E filtered (t)=αE(t)+(1-α)Efiltered (t-1)

[0138] In the formula, t is the sampling time, E(t) is the original illuminance collected at time t, and E filtered (t) represents the current filtered output, E filtered (t-1) is the output of the previous filter, with a filter coefficient α=0.1.

[0139] The ambient brightness threshold, local brightness threshold, and final segmentation threshold are as follows:

[0140] T env =k1·E filtered

[0141] T local (u, v)=k2·I mean (u, v)

[0142] T final (u, v)=max(T env , T local (u, v))

[0143] In the formula, k1 is the ambient brightness coefficient, which can be manually set or adaptively adjusted according to the scene, and is usually greater than 1. For nighttime, k1=5; k2 is the local threshold coefficient, which can be k2=3; I mean (u, v) represents the average gray level of a local window centered on the pixel to be determined, with the window being 3×3 or 5×5.

[0144] Binarize the grayscale image according to the following formula:

[0145]

[0146] The binary image is first dilated and then eroded, and the edges are extracted using the Canny algorithm. The final glare region R is determined by filtering out areas with an area less than 10 pixels or irregular shapes based on the contour area and shape.

[0147] IV. Target transmittance calculation and smooth transition

[0148] For a pixel (u, v) within the glare region R, its grayscale value is I. gray (u, v) defines the maximum brightness I within the region. max =max(I gray (u, v)∣(u, v)∈R), minimum brightness I min =min(I gray (u, v)∣(u, v)∈R).

[0149] Normalized brightness is:

[0150]

[0151] Normalized brightness is the original brightness I of the pixel. gray (u, v) is mapped to the interval [0,1].

[0152] The brightness gradient is calculated using the Sobel operator. :

[0153]

[0154]

[0155] Normalize the brightness gradient:

[0156] Gradient normalization:

[0157]

[0158] Among them, G max The maximum brightness gradient within the region, ranging from 0 to 1.

[0159] The transmittance within the glare area is calculated using the following formula:

[0160] T(u, v)=T min +(T max -T min )·(1-I norm (u, v)·(1-β·G norm (u, v))).

[0161] In the formula, T min =5%, T max =95%, β=0.3, where β is the gradient weight coefficient, with a value range of [0,1]; G norm (u, v) represents the normalized brightness gradient. Higher brightness results in lower transmittance; gradient compensation smooths out the transmittance variation in the edge region.

[0162] Maintain maximum transmittance in non-glare areas:

[0163] T(u, v)=T max =95%

[0164] Gaussian filtering is applied to the transmittance matrix of the glare area to ensure a smooth transition between adjacent pixels:

[0165]

[0166] In the formula, G(m, n) is the Gaussian kernel function, and T smooth(i, j) represents the smoothed transmittance; the Gaussian kernel window is 3×3, k=1, and the standard deviation of the Gaussian function σ=0.8. Filtering is applied only to the glare region, while the transmittance in the non-glare region remains constant (T). max .

[0167] Inter-frame fusion is used to make the transmittance change smoothly between consecutive frames:

[0168] T final (i, j, t)=γ·T smooth (i, j, t)+(1-γ)·T final (i, j, t-1)

[0169] In the formula, T final The final output transmittance is given by t, where t is the current frame and t-1 is the previous frame. γ = 0.2, with a value range of 0 < γ < 1; the larger γ is, the stronger the influence of the current frame. The processed transmittance is converted into a driving voltage and output to the corresponding functional lens pixel.

[0170] V. Traffic Color Protection and Spectral Calibration

[0171] To reduce the impact of dimming on the color recognition of traffic lights, brake lights, and road signs, pixel color depth is controlled by combining spectral calibration, color protection, and feedback calibration.

[0172] The incident light is obtained using a spectral sensor or a preset spectrum, and the tristimulus values ​​and CIE LAB color coordinates before and after dimming are calculated.

[0173]

[0174] In the formula, For the incident light intensity, Pixel transmittance; A function for matching human eye color. For reference white point, the CIE LAB color coordinates are:

[0175] L * = 116 × f (Y / Y n ) - 16

[0176] a * = 500 × [f (X / X n ) – f (Y / Y n )]

[0177] b * = 200 × [f (Y / Y n ) – f (Z / Z n )]

[0178] In the formula, f(t) is a nonlinear compression function. It is assumed that the functional lens is completely transparent, i.e. Calculate the original color; set candidate transmittance, calculate the attenuated color, and then calculate the color difference:

[0179]

[0180] The CIEDE2000 standard was used to evaluate color differences, and the maximum permissible attenuation that meets the color difference threshold, i.e., the minimum transmittance that meets the color recognition requirements, was used as the lower limit of color protection T. color The protection parameters for different traffic colors are as follows:

[0181]

[0182] Based on scene analysis, the colors to be protected and their corresponding color difference thresholds are determined. The lowest color transmittance T is then searched using a binary search method. min color ; Calculate T based on ambient light intensity and glare protection requirements glare Synthetic final target transmittance:

[0183] T target = max(T min color , T glare , 0.1)

[0184] In the formula, 0.1 is the lower limit of absolute transmittance.

[0185] Spectral calibration was performed using a monochromator outputting continuously adjustable standard wavelength light in steps of 5–10 nm, covering the range of 380–760 nm. Under fixed driving parameters, the incident and emitted light intensities at each wavelength were measured, and the original spectral transmittance was calculated.

[0186]

[0187] T of each pixel raw (λ) is fitted to a response curve, and the target spectral response curve T is determined according to the color depth requirement. target (λ), calculate the wavelength compensation coefficient:

[0188]

[0189] When T raw (λ) > T target When λ > 1, K(λ) < 1; otherwise, K(λ) > 1. Based on the chromaticity information acquired by the miniature camera CAM and the light intensity acquired by the illuminance sensor ALS, the CIE LAB color rendering deviation is calculated:

[0190]

[0191] If ΔE abIf the threshold is exceeded, for example, ΔE > 2, the compensation coefficient for the corresponding wavelength is fine-tuned by ±5% to 10%, and the corrected transmittance is:

[0192]

[0193] In the formula, K fb The closed-loop feedback coefficient is given by ΔE. ab The ratio to the threshold is determined. For adjacent pixels at the same wavelength, T... corr (λ) Use Gaussian filtering or bilinear interpolation to reduce the sudden changes in transmittance between pixels.

[0194] A parameter calibration table is established by spectral calibration under different driving voltages or currents, and the pixel driving parameters are adjusted to correct the spectral response. A tunable filter can be set in front of the pixel array to adjust the center wavelength and bandwidth. After correction, the output spectrum and colorimetric information are acquired, and feedback correction continues if the threshold requirement is not met.

[0195] The above steps are executed continuously by the processing module U1: after factory calibration, images and ambient illumination are acquired in real time, glare areas are preprocessed and identified, target transmittance with brightness gradient compensation is calculated, color protection constraints are applied as needed, target transmittance is mapped to the left and right functional lenses, spatial and temporal smoothing is performed, pixel drive commands are output, and light intensity feedback and color calibration are implemented. Through local attenuation, boundary smoothing, and high transmittance maintenance in non-glare areas, overall field of view clarity, including non-glare areas, is guaranteed.

[0196] An embodiment of the present invention is as follows:

[0197] The external wearing components are lightweight full-frame frame structures. The frame 1 is made of ultra-light polymer composite material, and the temple ends are equipped with anti-slip silicone sleeves. The overall protection level is not lower than IP54, and the weight is ≤35g.

[0198] The miniature camera CAM and the illumination sensor ALS are symmetrically mounted on the inner front of the frame 1. The power supply component 7 is a miniature rechargeable lithium polymer battery, which is embedded in the rear of the temple and equipped with a charging circuit, protection circuit and power detection circuit. The charging interface 8 is a hidden Type-C interface with overcharge, over-discharge, overcurrent and short circuit protection functions.

[0199] The miniature camera (CAM) has a frame rate of ≥30fps, a resolution of ≥1080P, a field of view covering 120°, and a built-in noise reduction circuit; the illuminance sensor (ALS) has a detection range of 0.1~100000lx, a detection accuracy of ±1lx, and directly outputs digital signals without the need for additional analog-to-digital conversion.

[0200] The core of the processing module U1 is a low-power, high-performance microcontroller equipped with a storage circuit, a reset circuit, and a watchdog circuit. It supports multi-interface communication. The software processing program includes data preprocessing, glare recognition, transmittance calculation, instruction output, and low-power management modules. It can dynamically and adaptively adjust the glare judgment threshold to achieve pixel-level parallel processing.

[0201] The display module has a curved structure with a thickness of ≤2.5mm, a light transmittance of ≥92% in the non-adjustable state, a pixel density of ≥300PPI, a single pixel size of ≤0.01mm, and adopts a multi-layer composite structure. The pixel transmittance adjustment range is 5%~95%, and there is no light leakage or image retention during adjustment.

[0202] The power supply component 7 adopts a dual-battery parallel structure with a total capacity of ≥500mAh, an output voltage of 3.7V, a normal working battery life of ≥8 hours on a single charge, a standby power consumption of ≤5mA, a working power consumption of ≤30mA, and supports fast charging within 2 hours.

[0203] The working logic of the core functional components is as follows: the acquisition module synchronously acquires images and brightness and illuminance parameters → the processing module U1 preprocesses the data, identifies glare areas, calculates pixel transmittance and generates adjustment instructions → the driving module U2 converts the adjustment instructions into driving electrical signals → the display module adjusts the transmittance and color depth of the corresponding pixels to reduce glare.

[0204] The calculation logic for pixel transmittance in processing module U1 is as follows: when the brightness is ≥8000lx, the transmittance is adjusted to 5%~15%; when the brightness is 2000~8000lx, it is adjusted to 15%~40%; when the brightness is 800~2000lx, it is adjusted to 40%~70%; when the brightness is <800lx, it is adjusted to 90%~95%, and the transmittance of adjacent pixels transitions smoothly.

[0205] The core functional components have a full-process adjustment response time of ≤15ms, a strong light reduction rate of ≥82%, and a visual clarity of ≥95% in non-glare areas.

Claims

1. A type of anti-glare glasses for drivers at night, characterized in that, The system includes a frame (1), a left temple (2), a right temple (3), core functional components, and a power supply component (7). The core functional components include an acquisition module, a processing module (U1), a driving module (U2), and a display module. The acquisition module includes a miniature camera (CAM) for acquiring images of the front field of view and an illuminance sensor (ALS) for acquiring ambient illuminance. The output of the acquisition module is connected to the processing module (U1), and the processing module (U1) is connected to the display module via the driving module (U2). The power supply component (7) supplies power to the core functional components. The display module includes liquid crystal displays (LCDs) that serve as left and right functional lenses, and each functional lens has a liquid crystal pixel array that can independently adjust the transmittance. The processing module (U1) is configured to identify glare areas based on image signals and illuminance signals, calculate the target transmittance based on the brightness and brightness gradient of each pixel in the glare area, and generate pixel-level adjustment instructions based on the coordinate correspondence between the camera image and the pixels of the left and right functional lenses. The driving module (U2) adjusts the transmittance of the corresponding pixels according to the pixel-level adjustment instructions, so that the transmittance of the glare area is reduced and the non-glare area maintains high transmittance.

2. The anti-glare glasses for drivers at night according to claim 1, characterized in that, The miniature camera (CAM) is mounted on the frame (1) with the lens facing forward; the illuminance sensor (ALS) is mounted on the frame (1); the processing module (U1) and the drive module (U2) are located inside the temple. The processing module (U1) includes a microcontroller, a storage circuit, a reset circuit and a watchdog circuit. The storage circuit stores preprocessing, glare recognition, transmittance calculation, instruction output and low power management programs.

3. The anti-glare glasses for drivers at night according to claim 1, characterized in that, The liquid crystal pixel array is driven by an active matrix of thin-film transistors, and each pixel is equipped with an independent thin-film transistor switch and a storage capacitor. The functional lens includes a scratch-resistant and wear-resistant layer, a high-transparency protective layer, a liquid crystal pixel layer, a driving electrode layer and a substrate light-transmitting layer arranged sequentially from the outside to the inside. The driving electrode layer includes a pixel electrode matrix formed by a transparent conductive film. The processing module (U1) stores the calibration relationship between the transmittance of the pixel and the driving voltage, and determines the pixel driving voltage according to the calibration relationship.

4. The anti-glare glasses for drivers at night according to claim 1, characterized in that, The power supply component (7) includes a rechargeable lithium polymer battery embedded in the rear of the left temple (2) and the right temple (3) and connected in parallel, as well as a charging circuit, a protection circuit (6) and a power detection circuit; the right temple (3) is provided with a charging interface (8), and the left temple (2) is provided with a power indicator (4) and a touch power switch (5); the processing module (U1) is also connected to a photodetector located in the non-viewing area of ​​the frame (1) or the edge of the functional lens. The photodetector collects the light intensity after passing through the functional lens through a light guide tube or sampling hole, which is used to correct the driving voltage of the corresponding pixel.

5. A method for preventing glare using the anti-glare glasses for drivers at night as described in any one of claims 1-4, characterized in that, The method includes the following steps: ① Perform coordinate and transmittance calibration: Establish the perspective transformation matrix between the image coordinates of the miniature camera (CAM) and the pixel coordinates of the left and right functional lenses, and establish the calibration relationship between the transmittance of each liquid crystal pixel and the driving voltage. Store the perspective transformation matrix and calibration relationship in the processing module (U1). ② Acquire and preprocess two signals: continuously acquire images of the front field of view through a miniature camera (CAM) and acquire ambient illuminance through an illuminance sensor (ALS); the processing module (U1) performs noise reduction and grayscale processing on the images and performs low-pass filtering on the illuminance signal; ③ Identify glare areas: Calculate the ambient brightness threshold based on the filtered ambient illuminance, and calculate the local brightness threshold based on the average gray level of the local window in the image. Take the larger of the two as the final segmentation threshold, and determine the glare area through binarization, morphological processing and contour filtering. ④ Calculate target transmittance: Normalize the brightness and brightness gradient of each image pixel within the glare area, and calculate the glare adjustment transmittance according to the following formula: T(u, v)=T min +(T max -T min )·(1-I norm (u, v)·(1-β·G norm (u, v))) In the formula, T(u, v) is the transmittance of the pixel in the glare region, T min and T max These represent the minimum and maximum transmittance, respectively. norm (u,v) represents the normalized luminance, G norm (u, v) represents the normalized luminance gradient, and β is the gradient weight coefficient, with a value range of [0,1]. The transmittance of the non-glare region remains constant at T. max ; ⑤ Determine the dimming pixels of the left and right functional lenses: Map the glare image pixels to the left and right functional lenses respectively using the corresponding perspective transformation matrix; for non-integer coordinates, distribute the target transmittance to the four adjacent functional lens pixels according to the bilinear interpolation weight; when the same functional lens pixel receives dimming contributions from multiple glare image pixels, use a weighted geometric average to synthesize the transmittance of the functional lens pixel. ⑥ Perform smoothing: Spatial domain Gaussian smoothing is performed on the mapped glare adjustment area, and the smoothed transmittance of the current frame is time-domain weighted fusion with the output transmittance of the previous frame; the functional lens pixels in the non-glare area of ​​the current frame maintain the set high transmittance. ⑦ Output pixel-level driving instructions: Based on the calibration relationship, the processed transmittance is converted into a driving voltage, which is controlled by the driving module (U2) to control the corresponding liquid crystal pixels in the left and right functional lenses respectively, so that the transmittance of the glare area is reduced and the non-glare area maintains high transmittance, and steps ② to ⑦ are executed cyclically as the image and illuminance signal are updated.

6. The anti-glare method according to claim 5, characterized in that, In step ②, Gaussian filtering is used to remove image noise; the image is converted to grayscale, and the calculation formulas for low-pass filtering of the grayscale signal are as follows: I gray (u, v)=0.299R+0.587G+0.114B E filtered (t)=αE(t)+(1-α)E filtered (t-1) In the formula, R, G, and B are the red, green, and blue channel luminance values ​​of the image pixels, respectively, and E(t) is the original illuminance. filtered (t) represents the current filtered output, E filtered (t-1) represents the output of the previous filter, α=0.1; the ambient brightness threshold, local brightness threshold, and final segmentation threshold in step ③ are respectively: T env =k1·E filtered T local (u, v)=k2·I mean (u, v) T final (u, v)=max(T env , T local (u, v)) In the formula, k1 is the ambient brightness coefficient, k2 is the local threshold coefficient, and I mean (u, v) represents the average gray level of a local window centered at image coordinates (u, v); satisfying I gray (u, v) ≥ T final The pixels of (u, v) are set to 1, and the rest of the pixels are set to 0. The glare area is determined by morphological processing, edge extraction and contour filtering.

7. The anti-glare method according to claim 5, characterized in that, In step ④, the Sobel operator is used to calculate the horizontal brightness gradient G of the image. x and vertical brightness gradient G y The luminance gradient, normalized luminance, and normalized luminance gradient are calculated using the following formulas: Brightness gradient: ; Normalized brightness: It is the original brightness I of the pixel. gray The result of mapping (u, v) to the interval [0,1]; Gradient normalization: ; In the formula, I max I min These represent the maximum and minimum gray levels within the glare region R, respectively, and G. max The maximum luminance gradient within this region; the upper limit of transmittance T. max =95%, lower limit of transmittance T min =5%, gradient weight coefficient β=0.

3.

8. The anti-glare method according to claim 5, characterized in that, In step ①, a checkerboard calibration board is used to extract the corner points of the image, and the corresponding points of the corner point lines of sight and the pixel arrays of the left and right functional lenses are determined respectively. The perspective transformation matrix of each functional lens is obtained using at least four sets of corresponding points. In step ⑤, the coordinate mapping relationship between the glare image and the functional lenses is as follows: ; In the formula, H is the perspective transformation matrix, (u, v) are the camera image coordinates, and (X, Y) are the mapped functional lens coordinates; mapping matrices H are established for the left and right functional lenses respectively. L and H R Let the non-integer positions obtained by mapping be (x, y), let i0=[x], i1=i0+1, j0=[y], j1=j0+1, and let d x =x- i0、d y =y-j0, the bilinear interpolation weights of the four adjacent pixels are as follows: w 00 =(1-d x )·(1- d y ) w 10 =d x ·(1- d y ) w 01 =(1-d x )·d y w 11 =d x ·d y The weights satisfy w 00 + w 10 + w 01 + w 11 =1, For each frame of the image, initialize the logarithmic transmittance accumulation array T[i][j] and the weight accumulation array W[i][j], both with initial values ​​of 0; for each glare image pixel, calculate its target transmittance T. target The corresponding weights w are accumulated using the following formula: T[i][j]←T[i][j]+w·ln(T target ) W[i][j]←W[i][j]+w, When W[i][j] > 0: T final [i][j]=exp(T[i][j] / W[i][j]) When W[i][j]=0: T final [i][j]=1.0 The final transmittance sent to the driver module is: T drive [i][j]= T base ×T final [i][j] In the formula, T base Based on the transmittance, T final [i][j] represents the cumulative transmittance, T drive [i][j] represents the transmittance that is finally sent to the driver module.

9. The anti-glare method according to claim 8, characterized in that, The spatial domain Gaussian smoothing and temporal domain weighted fusion formulas in step ⑥ are as follows: ; T final (i, j, t)=γ·T smooth (i, j, t)+(1-γ)·T final (i, j, t-1) In the formula, G(m, n) is the Gaussian kernel function, and T smooth (i, j) represents the transmittance after spatial smoothing, and T final (i, j, t) represents the output transmittance after temporal fusion, where t is the current frame and t−1 is the previous frame; the Gaussian kernel window is 3×3, k=1, σ=0.8, and the inter-frame fusion coefficient γ=0.2, with a value range of 0 < γ < 1; spatial filtering is performed only on pixels within the glare area, while the transmittance of non-glare areas remains unchanged at T. max .

10. The anti-glare method according to claim 5, characterized in that, In step ⑦, the initial driving voltage is determined according to the pre-calibrated relationship between transmittance and driving voltage. The actual emitted light intensity after passing through the functional lens in the corresponding sampling area is collected by a photodetector, and the target emitted light intensity is determined according to the following formula: I target = T target ×I in In the formula, T target For the target transmittance, I in The current incident light intensity; based on the actual emitted light intensity I actual With the target emitted light intensity I target The deviation is corrected to adjust the driving voltage of the corresponding pixel until the following condition is met: |I actual - I target | ≤ ε In the formula, ε is the preset light intensity control accuracy threshold; when the incident light intensity changes, I is updated. in And recalculate I target .

11. The anti-glare method according to claim 5, characterized in that, For traffic colors that need protection, the CIE XYZ tristimulus values ​​before and after dimming are calculated based on the incident spectrum and the spectral transmittance of the functional lenses. These values ​​are then converted to CIE LAB color coordinates, and the color difference before and after dimming is evaluated using CIEDE2000 color difference. The lowest transmittance T that meets the color recognition requirements is then searched using a bisection method. min color ; and compare it with the glare protection transmittance T glare Synthetic, target transmittance is: T target = max(T min color , T glare , 0.1) In the formula, 0.1 is the lower limit of absolute transmittance; Based on the scenario analysis, determine the color type that needs to be protected and select the corresponding color difference threshold.

12. The anti-glare method according to claim 11, characterized in that, The transmittance of liquid crystal pixels under different wavelengths and driving parameters is measured in advance to establish the calibration relationship between driving parameters, wavelength, and transmittance. Under given driving parameters, the original spectral transmittance, wavelength compensation coefficient, and corrected transmittance are calculated using the following formulas: ; ; ; In the formula, λ is the wavelength, and I in (λ) and I out (λ) represent the incident light intensity and the emitted light intensity at that wavelength, respectively, T target (λ) represents the target spectral transmittance, K fb The closed-loop feedback coefficient is derived from the colorimetric deviation ΔE. ab The ratio to the threshold is determined; the processing module (U1) adjusts the driving parameters of the corresponding pixel according to the calibration relationship to correct its spectral response.