Self-adaptive dimming system and method for front windshield of vehicle based on pupil detection

Abstract

The invention discloses a self-adaptive dimming system and method for a front windshield of a vehicle based on pupil detection, and belongs to the field of automobile intelligent glass and driver monitoring. The system monitors the ambient light intensity in real time through the photosensitive module, and extracts the pupil diameter and the change rate of a driver in combination with the near-infrared image acquisition module, so as to judge the visual comfort state of the driver. After a gaze area on the windshield is accurately positioned by using a sight tracking algorithm, the system drives the pixelated electrochromic layer to realize local, rapid and multi-gray-scale light transmittance adjustment. The core mechanism is to establish a negative feedback adjustment model based on pupil physiological reaction and light transmittance, and calculate and match the optimal transparency in real time. According to the method, manual intervention is not needed, the instantaneous blindness phenomenon caused by entering and exiting of the tunnel or strong light irradiation can be effectively relieved, and the driving safety and driving comfort in the complex light and shadow environment are remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of automotive smart glass and driver monitoring technology, specifically relating to an adaptive dimming system and method for vehicle windshield based on the driver's pupil state. Background Technology

[0002] As automobiles become increasingly intelligent, driving safety and comfort have become paramount concerns. In daily driving, drastic changes in lighting conditions are a significant factor affecting road safety. For example, when a vehicle is driving towards a strong sunset or rapidly entering or exiting a tunnel, ambient brightness can change by orders of magnitude within milliseconds. While the human eye's pupil has the ability to automatically adjust the amount of light entering, this physiological adjustment process is somewhat delayed. This can cause drivers to experience temporary "momentary blindness" or severe visual discomfort, greatly increasing the risk of traffic accidents.

[0003] In existing technologies, traditional sun visors require manual operation, which is not only slow to respond but also obstructs most of the driver's view, posing a safety hazard. Commercially available photochromic glasses or full-body photochromic glass typically rely on external light sensors for adjustment, failing to perceive the driver's actual visual perception. Furthermore, their slow overall color-changing response can easily lead to an overly dark interior, affecting the observation of the dashboard or other road conditions. Therefore, there is an urgent need for an intelligent system capable of sensing changes in human pupil size and achieving rapid and precise localized dimming. Summary of the Invention

[0004] The purpose of this invention is to overcome the aforementioned problems in the prior art and provide an adaptive dimming system and method for vehicle windshields based on pupil detection. This system can directly utilize the physiological feedback of the human eye to light—pupil changes—as the basis for dimming control, achieving closed-loop control of "darkening when the human eye feels glaring and brightening when the human eye feels dark," and ensuring clear visibility for the driver through local pixelation dimming technology.

[0005] To achieve the above objectives, the embodiments of the present invention adopt the following technical solutions:

[0006] An adaptive dimming system for vehicle windshield based on pupil detection includes: a photosensitive module, a near-infrared image acquisition module, a data processing module, a dimming control module, and a feedback adjustment module.

[0007] The photosensitive module continuously detects the ambient light intensity in front of the vehicle to sense the characteristic signals of alternating brightness or sudden changes in light intensity in real time. When a sudden surge in light intensity is detected that exceeds the initial brightness value, it is determined that the vehicle has entered a "strong light environment". When a sudden drop in light intensity is detected that the vehicle has entered an "extremely dark environment", the subsequent image acquisition and dimming intervention process is immediately activated. If the ambient light intensity is in a preset intermediate balance range and there are no sudden changes, the current light transmission state is maintained and a smoothing filtering strategy is executed.

[0008] The near-infrared image acquisition module is activated by the photosensitive module and uses near-infrared spectral imaging technology to accurately capture the pupil outline even when wearing sunglasses or in nighttime environments. The data processing module is connected to the image acquisition module and uses image algorithms to calculate the pupil diameter and its rate of change in real time. By analyzing the physiological response of the pupil, the driver's adaptability to the current lighting conditions can be further verified.

[0009] The data processing module has a built-in gaze tracking model. By combining the collected facial feature point coordinates, it constructs a driver-windshield spatial mapping model, accurately calculates the area where the driver's gaze falls on the windshield, and thus provides the execution module with specific local pixelated dimming coordinates.

[0010] The dimming control module dynamically calculates the required glass transmittance based on the pupil's contraction or dilation speed, and controls the electrochromic pixel units in the landing area to undergo oxidation-reduction reactions, thereby changing the color depth.

[0011] The feedback adjustment module continuously monitors the pupil state and ambient light after dimming. If it finds that the pupil stops contracting violently and remains within the reference diameter range, it determines that the adjustment is effective and locks the current transmittance or slowly recovers with the environment.

[0012] Preferably, the pupil change rate mentioned in step S3 The calculation formula is: ,in, For the current moment The diameter of the pupil, The pupil diameter at the previous sampling time. The sampling time interval is defined by this formula. This formula allows for precise quantification of the human eye's physiological response speed to changes in light intensity.

[0013] Preferably, in step S6, the target transmittance The calculation uses a dynamic compensation algorithm, and the specific formula is as follows: ;in, Given the current light transmittance of the glass, The main gain coefficient, This is the ambient light weighting factor. When the system determines a "glare state" (i.e., ...), ... When the pupil constricts sharply (during a rapid vasoconstriction), the system reduces light transmittance, making... The system tends towards low transmittance; when the system is determined to be in a "dark adaptation state" (i.e., ... When the pupil dilates rapidly, the system increases light transmittance, making... It quickly recovered to 100%.

[0014] Preferably, the pixelated electrochromic layer of the execution module is driven by a matrix. After the main viewing area is determined by the line-of-sight mapping algorithm, the dimming control module uses a Gaussian blur distribution strategy to control the surrounding pixels so that the light transmittance of the central area is the lowest (or the highest) and smoothly transitions to the surrounding area, avoiding the formation of obvious hard borders in the field of vision that would interfere with the driver's vision.

[0015] Preferably, the data processing module includes a preset database of driver comfort pupil ranges. This database is established through adaptive learning using machine learning algorithms during the initial driving phase, setting the comfort range as follows: ,in Record the average pupil diameter of a specific driver under standard ambient light to enable personalized dimming.

[0016] The present invention has the following beneficial effects:

[0017] 1. This invention utilizes the physiological changes in the human pupil as a direct control source for dimming, constructing a biofeedback closed loop between the pupil and the windshield. Compared to traditional passive adjustment relying solely on external light sensors, this system can capture the driver's actual visual perception, effectively solving the problem of mismatch between sensor data and actual human visual perception. Especially when entering or exiting tunnels or encountering direct sunlight, it can respond to drastic changes in the pupil within milliseconds, greatly mitigating the "momentary blindness" phenomenon and improving visual comfort.

[0018] 2. This invention employs pixelated matrix dimming and Gaussian blur feathering technology to achieve precise local dimming. Unlike full-screen tinted glass or sunglasses, this system only reduces the light transmittance of the area where the driver's line of sight falls, while retaining high light transmittance in the rest of the windshield. This allows the driver to clearly perceive road conditions and dashboard information through peripheral vision while avoiding glare, significantly improving driving safety and eliminating the "blind spot" hazard caused by traditional shading methods.

[0019] 3. This invention incorporates machine learning and personalized comfort zone settings, enabling it to adapt to the physiological characteristics of different drivers. The system continuously monitors pupil recovery after dimming (feedback adjustment module). If the pupil fails to return to its comfort zone, it automatically performs a secondary fine-tuning. This dynamic adaptive adjustment mechanism ensures that the system consistently provides the optimal visual environment under varying lighting conditions and driver states, enhancing the system's intelligence and user experience. Attached Figure Description

[0020] Figure 1 This is a framework diagram of a vehicle windshield adaptive dimming system according to an embodiment of the present invention;

[0021] Figure 2 This is a flowchart of a control method according to an embodiment of the present invention; Detailed Implementation

[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0023] See Figure 1 As shown, a vehicle windshield adaptive dimming system based on pupil detection includes:

[0024] The system includes a photosensitive module, a near-infrared image acquisition module, a data processing module, a dimming control module, and a feedback module.

[0025] See Figure 2 The flowchart below illustrates an adaptive dimming system and method for a vehicle windshield based on pupil detection, according to the present invention. The intelligent control method includes the following steps:

[0026] After the vehicle starts, the system performs a self-check and activates the photosensor module to continuously detect ambient light intensity. When it detects characteristic signals of alternating light and dark or sudden changes in light intensity, it triggers the NIR camera to capture images of the driver's face at a high frame rate of 60fps. Because it uses near-infrared light, it will not interfere with the driver's vision.

[0027] The data processing module first crops the region of interest in the image, locates the eye, extracts the pupil edge using an edge fitting algorithm, fits an ellipse equation, and calculates the major axis as the pupil diameter. Simultaneously, the system calculates the rate of change of pupil diameter. The sampling interval is To eliminate noise, a moving average filter is applied to the most recent 5 frames of data, and then a judgment is made. Has the preset threshold for activating dimming been reached?

[0028] The system presets two physiological thresholds: the contraction alert threshold. and expanding the warning threshold ;when At that time, it was determined to be "bright light glare," and the human eye was undergoing stress-induced pupil constriction; when When the light source is in a dark environment, the eye is considered to be dilating its pupils to obtain light; otherwise, it is considered to be in a comfortable / steady state.

[0029] Combining eye position and head posture, a 3D gaze estimation algorithm is used, assuming the coordinates of the eye center in the vehicle coordinate system are... The line-of-sight vector is The equation of the windshield surface is: Solve for the line-of-sight ray. With curved surfaces intersection The intersection point That is, the driver's gaze point on the windshield, which the system determines. Centered on, with radius as The circular area is used as the target area for dimming.

[0030] The dimming control module determines the dimming level based on the intensity of the pupil's response. When facing backlight or exiting a tunnel, the pupil constricts sharply, and the system rapidly dims. The light transmittance of the point area; for a natural visual effect, the edges of the area are treated with Gaussian feathering; when entering the tunnel, the pupil dilates rapidly, and the system quickly... The light transmittance of the point and its surrounding area is increased to the maximum (bleached state), eliminating the "black hole effect".

[0031] The system continuously monitors and adjusts the settings. If the light is adjusted, It rapidly approaches 0, and At the baseline value If it is nearby, the adjustment is successful; if If the light is still too small, it means the light is still too strong. The system should continue to reduce the light transmittance or increase the dimming area.

[0032] This invention utilizes pixelated control to change the light transmittance only in the driver's field of vision area, while retaining the high light transmittance of the rest of the windshield area. This allows the driver to avoid glare while still being able to perceive road conditions through surrounding vision, making it safer than full-screen color-changing glass.

[0033] It will be apparent to those skilled in the art that this invention is not limited to the details of the specific embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or core characteristics. Therefore, from any perspective, the embodiments should be considered exemplary rather than restrictive. The scope of protection of this invention is defined by the appended claims rather than the foregoing description. Thus, all modifications, equivalent substitutions, and improvements falling within the meaning and scope of the equivalent elements of the claims are intended to be included within this invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A vehicle windshield adaptive dimming system based on pupil detection, comprising: The system comprises a photosensitive module, a near-infrared image acquisition module, a data processing module, a dimming control module, and a feedback adjustment module. The photosensitive module continuously detects ambient light ahead, triggering the system to initiate subsequent image acquisition and detection procedures upon sensing alternating light and dark conditions or sudden changes in light intensity. The near-infrared image acquisition module, activated by the photosensitive module, acquires real-time images of the driver's face and eyes; this module is a near-infrared camera and infrared supplementary light installed on the vehicle's dashboard or inside the A-pillar. The data processing module, connected to the near-infrared image acquisition module, includes a pupil state analysis algorithm and a gaze mapping algorithm. The pupil state analysis algorithm calculates pupil diameter and pupil change. The system measures the speed and acceleration of light to assess the degree to which the human eye adapts to the current ambient light. The gaze mapping algorithm is used to construct the spatial geometric relationship between the driver's eye center and the windshield, determining the driver's primary viewing area on the windshield. The dimming control module consists of a pixelated electrochromic layer and a control unit integrated into the windshield. The electrochromic layer is divided into M×N independently controllable dimming pixel units. It is connected to the data processing module to generate control signals based on the processing results, adjusting the transmittance of pixel units in specific areas of the execution module. The feedback adjustment module monitors the recovery of pupil diameter and ambient light after dimming, forming a closed-loop control until the pupil state enters a preset comfort range.

2. A method for adaptive dimming of a vehicle windshield based on pupil detection, employing the system described in claim 1, characterized in that, Includes the following steps: S1: The system starts and initializes all pixels of the windshield to a fully transparent state. Then, the photosensor module continuously detects the ambient light intensity in front of the vehicle to perceive the characteristic signals of alternating light and dark or sudden changes in light intensity in real time. S2: When the photosensor module detects that the ambient light signal meets the trigger condition, the near-infrared image acquisition module is immediately activated, capturing real-time images of the driver's face and eyes using the onboard near-infrared camera and infrared supplementary lighting. S3: The data processing module receives the images acquired in S2, extracts the pupil contour and determines the head posture through image preprocessing and edge detection algorithms, and calculates the pupil diameter in real time. and pupillary change rate S4: The system according to Determine if there has been a sudden change in the current lighting environment; if If the light intensity is below the contraction threshold (strong light stimulation), it is considered a "dazzling state"; if If the expansion threshold is greater than the threshold (entering darkness), it is determined to be in "dark adaptation state"; if Within the threshold range, it is determined to be in a "steady state"; S5: If it is in a non-steady state, the gaze mapping algorithm calculates the coordinates of the intersection point between the driver's gaze and the windshield based on the position of both eyes and the head posture. Determine the dimming area centered on that point; S6: Detect the absolute brightness of the surrounding environment. It also performs threshold determination and collects ambient brightness data. ;like If the brightness is below the low-brightness threshold, the system determines it as an "extremely dark environment," meeting the dimming intervention conditions; if (If the brightness is higher than the initial value for a relatively bright environment), the system determines it as a "relatively bright environment," meeting the dimming intervention conditions; if (If it is in the intermediate transition zone), then the current transmittance remains unchanged and a smooth transition strategy is implemented; S7: After the dimming intervention conditions are met, the dimming control module adjusts the dimming control module according to the pupil change rate. Calculate the target transmittance And drive the corresponding pixel unit of the execution module to reach the target transmittance within a preset time; S8: Feedback adjustment module continuously monitors pupil diameter. If the pupil diameter does not return to the preset comfortable range due to changes... Then, the transmittance is finely adjusted a second time.

3. The method according to claim 2, characterized in that, The pupillary change rate mentioned in step S3 The calculation formula is: ;in, For the current moment The diameter of the pupil, The pupil diameter at the previous sampling time. This represents the sampling time interval.

4. The method according to claim 2, characterized in that, Target transmittance in step S7 The calculation uses a dynamic compensation algorithm, and the specific formula is as follows: ;in, Given the current light transmittance of the glass, The main gain coefficient, The ambient light weighting factor; when the system determines it to be a "glare state" (pupils constrict rapidly), When this happens, the system reduces the light transmittance, making... The light transmittance tends to be low; when the system determines that it is in a "dark adaptation state" (the pupil dilates rapidly), When this happens, the system increases light transmittance, making... It quickly recovered to 100%.

5. The system according to claim 1, characterized in that, The pixelated electrochromic layer of the execution module is driven by a matrix. After determining the main viewing area, the dimming control module uses a Gaussian blur distribution strategy to control the surrounding pixels, so that the light transmittance of the central area is the lowest (or the highest) and smoothly transitions to the surrounding area, avoiding the formation of obvious hard borders in the field of vision that would interfere with the driver's vision.

6. The system according to claim 1, characterized in that, The data processing module includes a pre-defined database of driver comfort pupil ranges. This database is established through adaptive learning using machine learning algorithms during the initial driving phase, setting the comfort range as follows: Record the average pupil diameter of the specific driver under standard ambient light.

Images