Intelligent control method, system and storage medium for vehicle-mounted atmosphere light

Abstract

This application relates to the field of vehicle control technology and discloses an intelligent control method, system, and storage medium for vehicle-mounted ambient lighting. The method includes: collecting external environmental data, vehicle occupant status data, and vehicle operation data through a data acquisition module; a hue setting module determining an initial hue path based on occupant status scores and a hue strategy; a brightness setting module determining initial brightness based on occupant age type, external environmental data, and a brightness mapping table; a frequency setting module setting an initial frequency based on vehicle operating conditions and a frequency mapping table; and an ambient lighting control module generating control instructions based on the initial hue path, brightness, and frequency, and controlling the ambient lighting based on the control instructions. This application improves the intelligence, personalization, and scenario-based nature of ambient lighting control, enhancing the user experience.

Description

Technical Field

[0001] The present application relates to the field of vehicle control technology, and in particular to an intelligent control method, system and storage medium for vehicle-mounted ambient light. Background Art

[0002] With the continuous advancement of automotive technology, in-vehicle ambient lighting has gradually gained attention as a key feature that enhances the driving experience. Traditional in-vehicle ambient lighting is primarily controlled through manual adjustment or preset modes, and lacks intelligent adaptation to the occupants' actual state and environmental changes. In recent years, the development of artificial intelligence, sensor technology, and machine learning has made intelligent control of in-vehicle ambient lighting possible.

[0003] A similar prior art application is Chinese patent application publication number CN119497281A, which discloses an intelligent control method, device, electronic device, and storage medium for vehicle ambient lighting. The method acquires temperature data collected from the vehicle's external ambient temperature using multiple temperature sensors; preprocesses the temperature data to obtain a composite external temperature; determines the color and brightness values ​​corresponding to the ambient lighting based on the composite external temperature using a preset mapping relationship; and transmits the corresponding color and brightness values ​​to an ambient lighting controller, which then drives the ambient lighting to adjust its color and brightness in response to changes in the vehicle's external ambient temperature. This method controls the ambient lighting based on the external environment, fails to address user needs, and exhibits a low degree of personalization. Another Chinese patent application, publication number CN119364604A, discloses a method and control system for controlling in-vehicle ambient lighting. This method uses an in-vehicle image acquisition device to capture image data about the vehicle's occupants and, based on the image data, obtains basic information about the occupants. This basic information is input into a trained prediction model to obtain the occupants' preference information, which includes their respective color preferences. An algorithm is used to obtain designated color information based on the preference information, which includes multiple colors determined by RGB values. A cloud service generates lighting control commands based on the designated color information and sends them to the vehicle's central control system, allowing the in-vehicle ambient lighting to adjust based on the age and gender distribution of the occupants. This method controls the ambient lighting based on the static characteristics and preferences of the occupants and lacks dynamic response to the real-time status of the occupants.

[0004] Therefore, it is an urgent problem to provide an intelligent control method, system and storage medium for vehicle-mounted ambient light to achieve automation, personalization and scenario-based adjustment of ambient light and improve driving experience and comfort. Summary of the Invention

[0005] The present application provides an intelligent control method, system and storage medium for vehicle-mounted ambient light.

[0006] In a first aspect, the present application provides a method for intelligently controlling a vehicle-mounted ambient light, the method comprising:

[0007] Extracting a data acquisition module for acquiring external environment data, first user status data of vehicle occupants, and vehicle operation data;

[0008] A hue setting module is provided to analyze the first user status data, identify the first status score of each member, determine the hue change path based on the status identification result and the preset hue strategy, and set the hue change path as the initial hue path;

[0009] A brightness setting module is provided to analyze the first user status data, identify the age type of each member, and determine the initial brightness based on the age identification result, external environment data and a preset brightness mapping table;

[0010] Provide a frequency setting mode to analyze vehicle operation data, identify vehicle operating conditions, and determine the initial frequency of the hue change path based on the operating condition identification results and a preset frequency mapping table;

[0011] An ambient light control module is provided for generating a control instruction based on an initial hue path, an initial brightness and an initial frequency, and controlling the ambient light based on the control instruction.

[0012] In conjunction with the first aspect, in a first implementation of the first aspect of the present application, the first user status data includes a facial image, first physiological data, and second physiological data, and analyzing the first user status data to identify the first status score of each member includes:

[0013] Extract any member, analyze the facial image, first physiological data, and second physiological data corresponding to any member, respectively, to obtain a first member status, a second member status, and a third member status, wherein the member status includes a status type and a level corresponding to the status type, and the status type includes negative, normal, and positive;

[0014] Based on the first preset rule table, adjusting the first member state with reference to the second member state and the third member state, respectively, to generate a first reference state and a second reference state;

[0015] Search the first state score mapping table, obtain the first score and the second score corresponding to the first reference state and the second reference state respectively, perform weighted average on the first score and the second score, obtain the first state score corresponding to any member, and after traversing all members, obtain the first state score of each member.

[0016] In combination with the first aspect, in a second implementation of the first aspect of the present application, analyzing the facial image to obtain the first member status includes:

[0017] Performing facial pose estimation on the facial image to obtain a first pose angle, performing key point detection on the facial image to obtain first key point information, and detecting a first eigenvector of each facial organ based on the first key point information;

[0018] Obtaining a standard facial image corresponding to any member, controlling the standard facial image to rotate based on a first posture angle, obtaining a reference facial image, performing key point detection on the reference facial image, obtaining second key point information, and detecting a second eigenvector of each facial organ based on the second key point information;

[0019] Comparing the first eigenvector with the second eigenvector, and marking each state type based on the comparison result and the second preset rule table, wherein the marking includes forward marking and reverse marking;

[0020] Based on the annotation results, the net value of each status type is counted, the status type corresponding to the maximum net value is defined as the pending status type, and the second status score mapping table is searched based on the pending status type and the maximum net value to obtain the first member status.

[0021] In combination with the first aspect, in a third implementation of the first aspect of the present application, the standard facial image is a facial image when the state type is normal and the second posture angle is zero, wherein the second posture angle is the facial posture angle of the standard facial image.

[0022] In combination with the first aspect, in a fourth implementation of the first aspect of the present application, determining the hue change path based on the state recognition result and the preset hue strategy includes:

[0023] Step 11: Obtain the minimum score and the maximum score in the first state score, and compare the minimum score and the maximum score with a preset state range;

[0024] Step 12: Determine whether the minimum score is within a first preset range. If not, proceed to step 13. If so, determine that the ambient light control is in the first mode. Then, determine whether the historical hue storage module stores a first hue sequence corresponding to the minimum score. If so, set the first hue sequence as the hue change path. If not, obtain a second hue sequence corresponding to the minimum score based on the first mode and the hue strategy, and set the second hue sequence as the hue change path.

[0025] Step 13: Determine whether the maximum score is within a second preset range. If not, proceed to step 14. If so, determine that the ambient light control is in the first mode. Then, based on the first mode and the hue strategy, set the third hue sequence corresponding to the maximum score as the hue change path.

[0026] Step 14: Determine that the ambient light control is in the second mode, obtain the display image of the vehicle's central control display screen, extract the main color other than black in the display image, and set the fourth hue sequence corresponding to the main color as the hue change path based on the second mode and the hue strategy.

[0027] In combination with the first aspect, in a fifth implementation of the first aspect of the present application, an atmosphere light adjustment module is further provided, which adjusts the atmosphere light when the minimum score is within a first preset range, including:

[0028] Step 121: After the ambient light operates for a preset time according to the control instruction, second user status data of the member corresponding to the minimum score is obtained, and the second user status data is analyzed to obtain a second status score;

[0029] Step 122: Determine whether the second state score is within a third preset range. If not, proceed to step 123. If so, continue to control the ambient light based on the control instruction, and store the hue change path contained in the control instruction in correspondence with the minimum score in the historical hue storage module, wherein the third preset range is between the first preset range and the second preset range.

[0030] Step 123: Calculate the difference between the second state score and the minimum score, define it as the adjustment value, adjust the initial hue path, initial brightness and initial frequency based on the adjustment value, generate a new control instruction, control the atmosphere light based on the new control instruction, and then return to step 121.

[0031] In conjunction with the first aspect, in a sixth implementation of the first aspect of the present application, the first user status data includes a facial image, and analyzing the first user status data to identify the age type of each member includes:

[0032] Step 21: extract any member and perform facial analysis on any member to obtain multiple skin appearance feature images, wherein the skin appearance feature image is an image that retains any appearance feature and eliminates other appearance features;

[0033] Step 22: extract any skin appearance feature image, define the appearance feature of any skin appearance feature image as a first appearance feature, extract a representative image corresponding to the first appearance feature, compare any skin appearance feature image with the representative image, perform positive or negative adjustment on the age stage corresponding to the representative image based on the comparison result, and extract a new representative image based on the adjusted age stage. Repeat step 22 until the number of negative adjustment points is greater than or equal to a preset value or the age stage tends to be stable, and calculate the estimated age corresponding to the first appearance feature based on the age stage at the negative adjustment point, wherein the negative adjustment point is the point at which the adjustment direction changes;

[0034] Step 23: After calculating the estimated ages corresponding to all appearance features, perform a weighted average of all estimated ages to obtain the final estimated age of any member, and obtain the age type of any member based on the final estimated age and a preset age type mapping table.

[0035] In combination with the first aspect, in a seventh implementation of the first aspect of the present application, determining the initial brightness based on the age recognition result, the external environment data, and the preset brightness mapping table includes:

[0036] Step 31: Determine whether there is a member whose age type is the first preset type. If not, proceed to step 32. If so, determine the initial brightness based on the brightness mapping table, referring to the first preset type and external environment data respectively.

[0037] Step 32: Determine whether there is a member whose age type is the second preset type. If not, proceed to step 33. If so, determine the initial brightness based on the brightness mapping table, referring to the second preset type and external environment data respectively.

[0038] Step 33: Determine the initial brightness based on the brightness mapping table, with reference to the third preset type and external environment data.

[0039] In a second aspect, the present application provides an intelligent control system for a vehicle-mounted ambient light, the system comprising:

[0040] A data acquisition module, used to acquire external environment data, first user status data of vehicle occupants, and vehicle operation data;

[0041] a hue setting module, configured to analyze the first user status data, identify the first status score of each member, determine a hue change path based on the status identification result and a preset hue strategy, and set the hue change path as the initial hue path;

[0042] a brightness setting module, configured to analyze the first user status data, identify the age type of each member, and determine an initial brightness based on the age identification result, external environment data, and a preset brightness mapping table;

[0043] Frequency setting mode is used to analyze vehicle operation data, identify vehicle operating conditions, and determine the initial frequency of the hue change path based on the operating condition identification results and a preset frequency mapping table;

[0044] The ambient light control module is used to generate control instructions according to the initial hue path, initial brightness and initial frequency, and control the ambient light based on the control instructions.

[0045] A third aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned intelligent control method for a vehicle-mounted ambient light.

[0046] Compared with the prior art, the beneficial effects of the technical solution of this application are at least as follows:

[0047] 1. The data acquisition module acquires external environment data, first-user status data of vehicle occupants, and vehicle operation data, and combines it with the hue setting module, brightness setting module, and frequency setting module to dynamically adjust the hue, brightness, and frequency of the ambient light. Through multi-dimensional data fusion, it can fully perceive the status of the driver and passengers, provide the most suitable lighting effects for the drivers and passengers in different scenarios, make the control of the ambient light more intelligent and personalized, and enhance the comfort and experience of the drivers and passengers.

[0048] 2. By analyzing the first-user status data of the passengers in the car, it can accurately identify the emotional state and age type of each member, and generate corresponding hue change paths, brightness and frequency based on preset strategies. It responds to the status and needs of different members through personalized adjustment methods, improving the psychological comfort and driving safety of drivers and passengers.

[0049] 3. Through dynamic adjustment and adaptive learning, the hue, brightness and frequency of the ambient light are corrected and optimized based on real-time feedback of user status data to provide ambient light settings that better meet user needs. The optimized parameters are stored in the historical hue storage module, which can better adapt to the needs of different users and scene changes, further improving the system's intelligence level and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 A schematic diagram of an embodiment of an intelligent control method for a vehicle-mounted ambient light in an embodiment of the present application;

[0052] Figure 2 This is a schematic diagram of an embodiment of a method for determining a hue change path in an embodiment of the present application;

[0053] Figure 3 A schematic diagram of an embodiment of an atmosphere light adjustment method in an embodiment of the present application;

[0054] Figure 4 This is a schematic diagram of an embodiment of an intelligent control system for a vehicle-mounted ambient light in an embodiment of the present application. DETAILED DESCRIPTION

[0055] The embodiments of the present application provide an intelligent control method, system and storage medium for vehicle-mounted ambient lighting. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices.

[0056] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In an embodiment of the present application, an intelligent control method for a vehicle ambient light includes:

[0057] A data acquisition module is extracted to acquire external environment data, first user status data of vehicle occupants, and vehicle operation data.

[0058] Specifically, external environment data, first user status data of vehicle occupants and vehicle operation data are collected in real time through various sensors and devices (such as cameras, light sensors, wearable devices, vehicle sensors, etc.). External environment data includes light intensity, time (day, night), etc., first user status data includes character images, tone, heart rate, pulse, etc., and vehicle operation data includes vehicle speed, acceleration, steering angle, etc.

[0059] A hue setting module is provided to analyze the first user status data, identify the first status score of each member, determine the hue change path based on the status identification result and the preset hue strategy, and set the hue change path as the initial hue path.

[0060] Specifically, hue describes the basic type of color, its position on the color wheel, typically expressed as an angle or numerical value. Common color wheel designs include 12-hue and 24-hue wheels. For example, a 12-hue color wheel has three primary colors: red, yellow, and blue; three intermediate colors: orange (red + yellow), green (yellow + blue), and violet (red + blue); and six secondary colors: red-orange, yellow-orange, yellow-green, blue-green, blue-violet, and red-violet. Together, these hues form a complete color wheel, with seamless transitions between adjacent colors on the wheel, without distinct boundaries.

[0061] Specifically, by analyzing the first user status data of the vehicle occupants and analyzing each occupant's status score, the system determines their current emotional state and corresponding level. These states are typically categorized as negative, normal, and positive. For example, a score above a certain threshold indicates a positive state, while a score below a certain threshold indicates a negative state. A hue strategy is a pre-defined set of rules that defines the hue change path corresponding to different user status scores. These rules are typically formulated based on principles of psychology and color science. For example, in a positive state, the system might select warm tones (such as red, orange, and yellow) to create a warm atmosphere; in a negative state, it might select cool tones (such as blue, purple, and green) to provide a soothing effect; and in a normal state, it might select neutral tones (such as white and light yellow).

[0062] Based on the identified state type and the preset hue strategy, an initial hue change path is generated. For example, you can choose to gradually change from one color to another, or cycle between several colors, so that the vehicle's ambient lighting can provide a more comfortable visual environment that is more in line with the user's emotional needs. This personalized lighting adjustment not only effectively relieves fatigue and improves mood, but also has an entertainment effect, further enhancing the in-car atmosphere and driving experience.

[0063] A brightness setting module is provided for analyzing the first user status data, identifying the age type of each member, and determining the initial brightness based on the age identification result, external environment data and a preset brightness mapping table.

[0064] For example, age categories can be set based on age, such as children, teenagers, adults, and seniors. Passengers of different age groups have different brightness requirements. For example, children may be more sensitive to high brightness and require lower brightness to reduce irritation, while seniors may require higher brightness to ensure visual clarity. Furthermore, the intensity of external light directly impacts the required brightness inside the vehicle. For example, during the day or in bright sunlight, higher brightness is required to ensure visibility, while at night or in low light conditions, lower brightness is required to avoid glare and reduce visual distraction for the driver.

[0065] The brightness mapping table is a pre-defined table that maps age categories and external environment data to specific brightness values. For example, at night when there are children in the vehicle, the brightness is set to a lower value (such as 20%); during the day when all passengers are adults, the brightness is set to a higher value (such as 60%). By dynamically adjusting the ambient light brightness based on the age category of the vehicle occupants and the external environment, the system provides a comfortable visual experience for the driver and passengers, reducing eye fatigue and discomfort.

[0066] A frequency setting mode is provided to analyze vehicle operation data, identify vehicle operating conditions, and determine the initial frequency of the hue change path based on the operating condition identification results and a preset frequency mapping table.

[0067] Specifically, the vehicle operating conditions include low-speed operating conditions, normal operating conditions, and high-speed operating conditions.

[0068] When the vehicle is traveling at high speeds, if the ambient light's flashing frequency is too low, the flickering light may create a visual dissonance with the vehicle's rapid motion, causing dizziness or discomfort to the driver and passengers. Furthermore, low-frequency flickering may be perceived as a "jumping" light at high speeds, easily distracting the driver and even causing visual fatigue. At low speeds, if the ambient light's flashing frequency is too high, it may exceed the human eye's comfortable perception range, resulting in an overly strong visual "flickering sensation." Furthermore, excessively high flashing frequencies may induce dizziness, especially when the occupants are relatively still, where this visual stimulation is more pronounced. Therefore, when driving at low speed (0-30 km / h): the flashing frequency should be kept within a low but stable range (for example, 1-2 flashes per second) to avoid strong visual stimulation; when driving at medium speed (30-60 km / h): the flashing frequency can be appropriately increased (for example, 2-3 flashes per second) to enhance driving feedback and avoid visual fatigue; when driving at high speed (above 60 km / h): the flashing frequency should be further increased (for example, 3-5 flashes per second) to ensure the coordination of light flashing and vehicle movement, and to avoid visual discomfort caused by low-frequency flashing.

[0069] A frequency mapping table is a pre-defined set of rules that defines the corresponding hue change frequency under different vehicle operating conditions. Based on the identified vehicle operating condition and the preset frequency mapping table, the ambient light flashing frequency that matches the current operating condition is determined. The flashing frequency can be increased at high speeds to remind drivers and passengers that the vehicle is accelerating, while the flashing frequency is reduced at constant speeds to reduce visual distraction. This effectively avoids dizziness or visual discomfort caused by a mismatch between vehicle speed and flashing frequency, providing a more soothing visual environment and helping to enhance driving immersion, perception of vehicle status, and safety.

[0070] An ambient light control module is provided for generating a control instruction based on an initial hue path, an initial brightness and an initial frequency, and controlling the ambient light based on the control instruction.

[0071] The ambient light control module receives initial parameters from other modules, converts the hue, brightness and frequency parameters generated by the front-end module into specific lighting control instructions, and drives the ambient light hardware equipment to work to achieve the corresponding lighting effects.

[0072] In a specific embodiment, the first user status data includes a facial image, first physiological data, and second physiological data. Analyzing the first user status data to identify the first status score of each member includes:

[0073] (1) Extract any member, analyze the facial image, first physiological data and second physiological data corresponding to any member respectively, and obtain the first member status, the second member status and the third member status, wherein the member status includes the status type and the level corresponding to the status type, and the status type includes negative, normal and positive.

[0074] (2) Based on the first preset rule table, the first member state is adjusted with reference to the second member state and the third member state, respectively, to generate a first reference state and a second reference state.

[0075] (3) Search the first state score mapping table, obtain the first score and the second score corresponding to the first reference state and the second reference state respectively, perform weighted average of the first score and the second score, obtain the first state score corresponding to any member, and after traversing all members, obtain the first state score of each member.

[0076] Specifically, facial expressions can be subjective and easily disguised. When facial expressions are limited or restrained, it can be difficult to discern a person's true feelings. Physiological signals (such as heart rate, skin conductance, and body temperature) are generally difficult to subjectively control or disguise, and can provide real-time information about emotional state. Facial expressions primarily reflect an individual's outward emotional expression, while physiological signals reflect their internal physiological reactions. Both provide information on different dimensions of emotional state. Combining the two can verify and complement each other, thereby improving the accuracy and reliability of emotion recognition.

[0077] Exemplarily, the physiological data include indicators such as tone, heart rate, pulse, respiratory rate, body temperature, and skin conductance. Preferably, the tone is set as the first physiological data and the pulse is set as the second physiological data.

[0078] The first preset rule table is a set of predefined rules for adjusting the status type and / or the level corresponding to the status type of the first member status of any member according to the second member status or the third member status of any member. Among them, the levels corresponding to negative, normal and positive include high, medium and low. The higher the level corresponding to positive or normal, the higher the degree of positivity or normality. The higher the level of negative, the higher the degree of negativity. For example, if the status type of the first member status is positive and the level is high, and the status type of the second member status is positive and the level is low, then the adjusted first reference status is a status type of positive and a level of medium; if the status type of the first member status is normal and the level is low, and the status type of the second member status is negative and the level is medium, then the adjusted first reference status is a status type of negative and a low level.

[0079] The first state score mapping table is used to map the adjusted member states into specific scores, which are used to quantify the member states for subsequent weighted average calculation and ambient lighting control strategy formulation. For example, a positive state score ranges from 70 to 100, indicating a positive and happy state; a normal state score ranges from 40 to 69, indicating a calm and neutral state; and a negative state score ranges from 0 to 39, indicating a negative, tired, or stressed state. Based on this state score classification, different state types are further divided into different levels. For example, a positive state score ranges from 90 to 100 for high, 80 to 89 for medium, and 70 to 79 for low; a normal state score ranges from 60 to 69 for high, 50 to 59 for medium, and 40 to 49 for low; and a negative state score ranges from 0 to 19 for high, 20 to 29 for medium, and 30 to 39 for low. To simplify calculations, a fixed value can be set for each level of each state type. For example, the positive state level is: high is 95 points, medium is 85 points, and low is 75 points; the normal state level is: high is 65 points, medium is 55 points, and low is 45 points; the negative state level is: high is 10 points, medium is 25 points, and low is 35 points. The above example uses a percentage system as an example, and it can also be set to a ten-point system. The specific setting is based on the experience of those skilled in the art or according to the actual application scenario, and the embodiments of this application are not limited to this.

[0080] Specifically, different weights can be assigned based on the reliability, importance, and recognition accuracy of different data sources. A single data source can be biased due to various factors (such as measurement error and environmental interference). Weighted averaging not only takes into account the importance and reliability of different data sources, but also reduces bias from a single source through multi-dimensional data fusion, improving the accuracy of condition assessments.

[0081] In a specific embodiment, analyzing the facial image to obtain the first member status includes:

[0082] (1) Performing facial pose estimation on the facial image to obtain the first pose angle, and performing key point detection on the facial image to obtain the first key point information, and detecting the first eigenvector of each facial organ based on the first key point information.

[0083] (2) Obtain a standard facial image corresponding to any member, control the standard facial image to rotate based on the first posture angle, obtain a reference facial image, perform key point detection on the reference facial image, obtain second key point information, and detect a second eigenvector of each facial organ based on the second key point information.

[0084] (3) Compare the first eigenvector and the second eigenvector, and mark each state type based on the comparison result and the second preset rule table, including forward marking and reverse marking.

[0085] (4) Based on the annotation results, the net value of annotations for each state type is calculated, and the state type corresponding to the maximum net value of annotations is defined as the pending state type. Based on the pending state type and the maximum net value of annotations, the second state score mapping table is searched to obtain the first member state.

[0086] In a specific embodiment, the standard facial image is a facial image when the state type is normal and the second posture angle is zero, wherein the second posture angle is the facial posture angle of the standard facial image.

[0087] Specifically, when capturing images inside a vehicle, the captured facial image of the passenger may not necessarily be a standard frontal image of the person looking directly at the camera. To improve the accuracy of subsequent facial state estimation based on the annotated facial image, the passenger's first posture angle is preferentially calculated to determine the passenger's facial orientation. The annotated facial image serves as a benchmark for comparison with the current facial image. To eliminate recognition bias caused by varying facial postures and thereby improve the accuracy and reliability of emotional state recognition, the posture angle of the standard facial image is adjusted based on the first posture angle rotation before state recognition, ensuring that it aligns with the posture angle of the passenger's facial image.

[0088] During key point detection, key points are detected for each facial feature (eyes, nose, mouth, eyebrows, chin, etc.). For example, these key points include the inner and outer corners of the eyes, upper and lower eyelids, the beginning, peak, and tail of the eyebrows, the tip of the nose, both nostrils, both corners of the mouth, the middle point of the labial arch, and the chin. After key point detection, a feature vector is generated for each facial feature, where each element represents a specific measurement value. For example, for eyebrows, the corresponding feature vector is [eyebrow length, eyebrow spacing, eyebrow tilt angle, eyebrow arc, and eyebrow lift degree (vertical distance between the peak and tail of the eyebrow)]; for the mouth, the corresponding feature vector is [mouth width, vertical distance between the mouth corners and the center of the mouth, and distance between the upper and lower lips].

[0089] The second eigenvector of each facial organ of the reference facial image is used as a standard, and the first eigenvector of each facial organ of the member facial image is compared with it to analyze the changes of the facial organs, and further mark each state type based on the changes and a second preset rule table. State types include negative, normal and positive. For example, for the feature vector corresponding to the eyebrows, if the distance between the eyebrows decreases, the negative state type is positively labeled, +1, and the normal state type and the positive state type are negatively labeled, -1; if the eyebrow inclination angle decreases, the positive state type is positively labeled, +1, and the negative state type is negatively labeled, -1; if the eyebrow arc increases, the positive state type is positively labeled, +1, and the negative state type is negatively labeled, -1; if the eyebrow upward degree increases, the positive state type is positively labeled, +1, and the negative state type is negatively labeled, -1; for the feature vector corresponding to the mouth, if the mouth width becomes larger, the positive state type is positively labeled, +1; if the vertical distance of the mouth corner relative to the mouth center increases, the positive state type is positively labeled, +1, and the negative state type is negatively labeled, -1; if the distance between the upper and lower lips increases, the positive state type is positively labeled, +1, and the negative state type is negatively labeled, -1.

[0090] The second status score mapping table is a preset reference table used to map the analyzed net value to a specific status type and further differentiate the status types based on the size of the net value. The principle is similar to the first status score mapping table and will not be repeated here.

[0091] By comparing the features of a facial image with a reference image, changes in key facial features are detected. Further forward and backward annotation is used to quantify the likelihood of specific states, improving the accuracy and precision of facial image state estimation. Furthermore, compared to machine learning, this technical solution is highly interpretable and customizable. Relying on predefined rules and a small number of feature points, it does not require extensive training data and requires minimal computation.

[0092] In a specific embodiment, determining the hue change path based on the state recognition result and the preset hue strategy includes:

[0093] Step 11: Obtain the minimum score and the maximum score in the first state score, and compare the minimum score and the maximum score with a preset state range.

[0094] Step 12: Determine whether the minimum score is within the first preset range. If not, proceed to step 13. If so, determine that the ambient light control is in the first mode. Then determine whether the historical hue storage module stores a first hue sequence corresponding to the minimum score. If so, set the first hue sequence as the hue change path. If not, obtain the second hue sequence corresponding to the minimum score based on the first mode and the hue strategy, and set the second hue sequence as the hue change path.

[0095] Step 13: Determine whether the maximum score is within the second preset range. If not, proceed to step 14. If so, determine that the ambient light control is in the first mode, and then set the third hue sequence corresponding to the maximum score as the hue change path based on the first mode and the hue strategy.

[0096] Step 14: Determine that the ambient light control is in the second mode, obtain the display image of the vehicle's central control display screen, extract the main color other than black in the display image, and set the fourth hue sequence corresponding to the main color as the hue change path based on the second mode and the hue strategy.

[0097] See also Figure 2 , which is a schematic diagram of an embodiment of the method for determining the hue change path in the embodiment of the present application. Specifically, the first state score corresponds to the state type and the level corresponding to the state type. The smaller the score, the more negative the state, and the larger the score, the more positive the state. According to the range of the minimum score and the maximum score, it is determined which mode the ambient light control should be in (the first mode (based on the member setting) or the second mode (based on the vehicle setting)), and the corresponding hue sequence is selected as the hue change path to respond to the state of the occupants in the car.

[0098] Negative states have a significant impact on driving safety and ride comfort, and are highly prevalent and urgent, so priority is given to responding to negative states. When the minimum score falls within the first preset range, it indicates that a member of the vehicle is in a negative state. A hue sequence (blue → purple → green, light blue → pink → off-white, or dark blue → silver gray → light gray, etc.) designed to soothe the user's emotions is set as the hue change path to produce a calming effect. The historical hue storage module is capable of recording valid hue sequences. If the current state matches a historical state, the historical sequence is used directly to avoid repeated calculations and improve system efficiency. If there is no matching historical sequence, a new hue sequence is generated according to a preset hue strategy. There can be multiple historical sequences that match the current state. Preferably, the member corresponding to the negative state is identified, and the historical sequence corresponding to the member is extracted based on the identification result.

[0099] When the maximum score is within the second preset range, it indicates that there are members in a positive state in the car. In response to the emotional state of the members, a cheerful hue sequence (yellow→orange→red, light yellow→pink→lavender or gold→champagne→white, etc.) is provided as a hue change path to enhance pleasure and vitality.

[0100] The central control display is a focal point within the vehicle's interior, and its displayed content (such as navigation maps and multimedia information) significantly impacts the driver's and passengers' visual experience. When the ambient lighting control is in the second mode, a hue transition path is set based on the primary color (other than black) in the central control display's image. This helps coordinate and unify visual elements within the vehicle, avoiding visual clashes or disharmony, improving the overall driving experience, and reducing visual distractions. For example, if the central control display's primary color is green, the hue transition path is set from green to yellow-green to light green.

[0101] According to the technical solution of the present invention, the hue change path of the ambient light is dynamically adjusted according to the status type of the occupants in the car, realizing a highly intelligent and personalized adjustment method of the ambient light, providing the driver and passengers with a more comfortable and pleasant in-car environment.

[0102] In a specific embodiment, the method further provides an atmosphere light adjustment module, which adjusts the atmosphere light when the minimum score is within a first preset range, including:

[0103] Step 121 : After the ambient light operates for a preset time according to the control instruction, the second user status data of the member corresponding to the minimum score is obtained, and the second user status data is analyzed to obtain a second status score.

[0104] Step 122: determine whether the second state score is within the third preset range. If not, proceed to step 123. If so, continue to control the ambient light based on the control instruction, and at the same time store the hue change path contained in the control instruction in the historical hue storage module corresponding to the minimum score, wherein the third preset range is between the first preset range and the second preset range.

[0105] Step 123: Calculate the difference between the second state score and the minimum score, define it as the adjustment value, adjust the initial hue path, initial brightness and initial frequency based on the adjustment value, generate a new control instruction, control the atmosphere light based on the new control instruction, and then return to step 121.

[0106] See also Figure 3, which is a schematic diagram of an embodiment of the atmosphere lighting adjustment method in the embodiments of the present application. Specifically, after the atmosphere lighting operates for a period of time according to the initial control instructions, the lighting effect is evaluated to determine whether it is effective. The preset time is set to ensure that the lighting adjustment has sufficient time to affect the mood of the guests. This setting is based on the experience of those skilled in the art or actual application scenarios and is not limited by the embodiments of the present application. For example, the preset time is 15 seconds, 30 seconds, or 1 minute.

[0107] If the second state score is within the third preset range, it means that the member state has calmed down and the current lighting settings are valid. The system stores the current hue path in the historical hue storage module, which can be reused in similar states later to improve control efficiency.

[0108] If the second state score is not within the third preset range, the current lighting effect is further determined based on the adjustment value, and the initial hue path, initial brightness, and initial frequency are adjusted based on the adjustment value. For example, if the adjustment value is greater than a preset threshold, the current control strategy is effective, and the current hue path, brightness, and frequency continue to be used to control the ambient lighting. If the adjustment value is less than the preset threshold but greater than 0, the current lighting control strategy is effective, but the effect is not ideal. In this case, the brightness and / or frequency can be fine-tuned to enhance the lighting control effect. For example, if the current brightness is low, the brightness can be increased appropriately. If the current brightness is already high, the dynamic range of the brightness can be adjusted so that it changes gradually over a certain period of time rather than being fixed at a single value. If the adjustment value is less than 0, the current lighting control strategy is ineffective and may even have a negative impact on mood regulation. In this case, comprehensive adjustments to the hue path, brightness, and frequency are necessary. For example, if the initial hue path is "green → yellow-green → light green", it can be changed to "blue → dark blue → light blue". If the current brightness is high, the brightness can be reduced appropriately. If the current flicker frequency is high, the flicker frequency can be reduced.

[0109] The technical solution of the present invention provides vehicle occupants with ambient lighting that suits their status and preferences through dynamic adaptability and personalized adjustment, thereby enhancing the comfort and safety of the driver and passengers. Furthermore, through closed-loop control and historical data storage, it can continuously learn and optimize, thereby improving the intelligence level and control efficiency of the system.

[0110] In a specific embodiment, the first user status data includes a facial image, and analyzing the first user status data to identify the age type of each member includes:

[0111] Step 21: extract any member, perform skin appearance feature analysis on the facial image of any member, and obtain multiple skin appearance feature images, wherein the skin appearance feature image is an image that retains any appearance feature and eliminates other appearance features.

[0112] Step 22: extract any skin appearance feature image, define the appearance feature of any skin appearance feature image as a first appearance feature, extract a representative image corresponding to the first appearance feature, compare any skin appearance feature image with the representative image, make positive or negative adjustments to the age stage corresponding to the representative image according to the comparison result, and extract a new representative image based on the adjusted age stage, repeat step 22 until the number of reverse adjustment points is greater than or equal to the preset value or the age stage tends to be stable, calculate the estimated age corresponding to the first appearance feature based on the age stage at the reverse adjustment point, wherein the reverse adjustment point is the point when the adjustment direction changes.

[0113] Step 23: After calculating the estimated ages corresponding to all appearance features, perform a weighted average of all estimated ages to obtain the final estimated age of any member, and obtain the age type of any member based on the final estimated age and a preset age type mapping table.

[0114] Specifically, skin appearance features include wrinkles (nasolabial folds, forehead wrinkles and / or crow's feet), pigmentation spots (age spots, freckles and / or chloasma) and / or skin firmness. A skin appearance feature image containing only wrinkle features, pigmentation spot features and / or skin firmness features can be obtained from a facial image through techniques such as adaptive threshold processing. That is, the skin appearance feature image is an image that eliminates other features and retains only one feature, separating and highlighting one feature while ignoring other interfering features, which helps to more accurately analyze the impact of this feature on age estimation.

[0115] Representative images are extracted and analyzed from a large amount of sample data, reflecting typical features of different age groups. These images serve as a reference standard and are compared with the skin appearance characteristic images of the passengers in the vehicle. This comparison allows the similarity of specific facial features (such as wrinkles, pigmentation, and / or skin firmness) to the standard images, thereby inferring the passenger's age type.

[0116] Exemplarily, any skin appearance feature image is extracted and first compared with the representative image corresponding to 30 years old. If the skin appearance feature image is older than the 30-year-old representative image, a positive adjustment is performed, and a representative image corresponding to 35 years old is further extracted. If the positive adjustment is continued, when the representative image corresponding to 47 years old is extracted, the skin appearance feature image is younger than the 47-year-old representative image, a reverse adjustment is performed (at this time, 47 years old is the reverse adjustment point), and a representative image corresponding to 45 years old is extracted. If the reverse adjustment is continued, when the representative image corresponding to 43 years old is extracted, the skin appearance feature image is older than the 43-year-old representative image, a positive adjustment is performed (at this time, 43 years old is the reverse adjustment point). This process is repeated until the adjustment becomes stable or reaches a preset condition (such as the number of reverse adjustment points reaches a threshold), and the average age at the reverse adjustment point is used as the estimated age corresponding to the above-mentioned appearance feature.

[0117] The estimated ages corresponding to all appearance features are weighted averaged to obtain the member's final estimated age. This final estimated age is then combined with a pre-set age type mapping table to determine the member's age type (e.g., child, adolescent, adult, or elderly). Weights can be assigned based on the importance of the features.

[0118] The appearance characteristics of the skin can provide different information about age, but there may be certain errors or limitations when estimating age based on each feature alone. By combining the estimation results of multiple features, the errors caused by a single feature can be reduced, the age type of the members can be estimated more accurately, and the accuracy of the overall estimation can be improved.

[0119] In a specific embodiment, determining the initial brightness based on the age recognition result, the external environment data, and a preset brightness mapping table includes:

[0120] Step 31: Determine whether there is a member whose age type is the first preset type. If not, proceed to step 32. If so, determine the initial brightness based on the brightness mapping table, refer to the first preset type and external environment data respectively.

[0121] Step 32: Determine whether there is a member whose age type is the second preset type. If not, proceed to step 33. If so, determine the initial brightness based on the brightness mapping table, refer to the second preset type and external environment data respectively.

[0122] Step 33: Determine the initial brightness based on the brightness mapping table, with reference to the third preset type and external environment data.

[0123] Specifically, the first preset category includes children, the second includes seniors, and the third includes teenagers and adults. Children's eyes are more sensitive to light, and stronger light can damage their vision. Therefore, brightness adjustment for children is prioritized to ensure a comfortable and safe light level for them. Seniors' vision typically declines with age, and their ability to adapt to light is relatively weaker. Stronger light can cause visual fatigue or discomfort, so brightness adjustment is prioritized to ensure their comfort. Teenagers and adults have relatively common visual needs. If there are no children or seniors in the vehicle, the system will determine brightness based on the default settings (for teenagers and adults) combined with external environmental data.

[0124] The technical solution of the present invention reflects the concern and care for members with special needs by prioritizing the adjustment of the brightness of children and the elderly. It not only conforms to the principle of humanization, but also effectively improves the overall comfort and safety of the in-vehicle environment. It can also ensure that the lighting settings can adapt to different environmental conditions, improve visual comfort and safety, and optimize the user experience.

[0125] The above describes an intelligent control method of a vehicle-mounted atmosphere light in an embodiment of the present application. The following describes an intelligent control system of a vehicle-mounted atmosphere light in an embodiment of the present application. Figure 4 In one embodiment of the present application, an intelligent control system for a vehicle-mounted ambient light includes:

[0126] The data acquisition module 10 is used to acquire external environment data, first user status data of vehicle occupants, and vehicle operation data.

[0127] The hue setting module 20 is used to analyze the first user status data, identify the first status score of each member, determine the hue change path based on the status identification result and the preset hue strategy, and set the hue change path as the initial hue path.

[0128] The brightness setting module 30 is used to analyze the first user status data, identify the age type of each member, and determine the initial brightness based on the age identification result, external environment data and a preset brightness mapping table.

[0129] The frequency setting mode 40 is used to analyze the vehicle operation data, identify the vehicle operating condition, and determine the initial frequency of the hue change path based on the operating condition identification result and a preset frequency mapping table.

[0130] The ambient light control module 50 is configured to generate a control instruction according to the initial hue path, the initial brightness, and the initial frequency, and control the ambient light based on the control instruction.

[0131] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the computer executes the steps of the intelligent control method for a vehicle-mounted ambient light.

[0132] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0133] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

[0134] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An intelligent control method for vehicle-mounted ambient light, characterized in that: The method comprises: Extracting a data acquisition module for acquiring external environment data, first user status data of vehicle occupants, and vehicle operation data; Providing a hue setting module for analyzing the first user status data, identifying the first status score of each member, determining a hue change path based on the status identification result and a preset hue strategy, and setting the hue change path as an initial hue path; providing a brightness setting module for analyzing the first user status data, identifying the age type of each member, and determining an initial brightness based on the age identification result, the external environment data, and a preset brightness mapping table; Providing a frequency setting module for analyzing the vehicle operation data, identifying the vehicle operating condition, and determining the initial frequency of the hue change path based on the operating condition identification result and a preset frequency mapping table; Providing an ambient light control module, for generating a control instruction based on the initial hue path, the initial brightness and the initial frequency, and controlling the ambient light based on the control instruction; The first user status data includes a facial image, first physiological data, and second physiological data. Analyzing the first user status data to identify a first status score for each member includes: Extract any member, analyze the facial image, the first physiological data, and the second physiological data corresponding to any member, respectively, to obtain a first member status, a second member status, and a third member status, wherein the member status includes a status type and a level corresponding to the status type, and the status type includes negative, normal, and positive; Based on a first preset rule table, adjusting the first member state with reference to the second member state and the third member state, respectively, to generate a first reference state and a second reference state; Search the first state score mapping table, obtain the first score and the second score corresponding to the first reference state and the second reference state respectively, perform weighted average on the first score and the second score, obtain the first state score corresponding to any of the members, and after traversing all members, obtain the first state score of each member.

2. The intelligent control method of vehicle-mounted ambient light according to claim 1, characterized in that: Analyzing the facial image to obtain the first member status includes: Performing facial pose estimation on the facial image to obtain a first pose angle, performing key point detection on the facial image to obtain first key point information, and detecting a first eigenvector of each facial organ based on the first key point information; Obtaining a standard facial image corresponding to any one of the members, controlling the standard facial image to rotate based on the first posture angle, obtaining a reference facial image, performing key point detection on the reference facial image, obtaining second key point information, and detecting a second eigenvector of each facial organ based on the second key point information; Comparing the first eigenvector and the second eigenvector, and marking each state type based on the comparison result and the second preset rule table, wherein the marking includes forward marking and reverse marking; Based on the annotation results, the annotated net value of each state type is counted, the state type corresponding to the maximum annotated net value is defined as the pending state type, and the second state score mapping table is searched based on the pending state type and the maximum annotated net value to obtain the first member state.

3. The intelligent control method of vehicle-mounted ambient light according to claim 2, characterized in that: The standard facial image is a facial image when the state type is normal and the second posture angle is zero, wherein the second posture angle is the facial posture angle of the standard facial image.

4. The intelligent control method of vehicle-mounted ambient light according to claim 1, characterized in that: The determining of the hue change path based on the state recognition result and the preset hue strategy includes: Step 11: Obtain the minimum score and the maximum score in the first status score, and compare the minimum score and the maximum score with a preset status range; Step 12: Determine whether the minimum score is within a first preset range. If not, proceed to step 13. If so, determine that the ambient light control is in the first mode. Then, determine whether the historical hue storage module stores a first hue sequence corresponding to the minimum score. If so, set the first hue sequence as the hue change path. If not, obtain a second hue sequence corresponding to the minimum score based on the first mode and the hue strategy, and set the second hue sequence as the hue change path. Step 13: Determine whether the maximum score is within a second preset range. If not, proceed to step 14. If so, determine that the ambient light control is in the first mode. Then, based on the first mode and the hue strategy, set a third hue sequence corresponding to the maximum score as the hue change path. Step 14: Determine that the ambient light control is in the second mode, obtain a display image of the vehicle's central control display screen, extract a primary color other than black in the display image, and set a fourth hue sequence corresponding to the primary color as the hue change path based on the second mode and the hue strategy.

5. The intelligent control method of vehicle-mounted ambient light according to claim 4, characterized in that: An atmosphere light adjustment module is also provided, which adjusts the atmosphere light when the minimum score is within the first preset range, including: Step 121: After the ambient light operates for a preset time according to the control instruction, second user status data of the member corresponding to the minimum score is obtained, and the second user status data is analyzed to obtain a second status score; Step 122: Determine whether the second state score is within a third preset range. If not, proceed to step 123. If so, continue to control the ambient light based on the control instruction, and store the hue change path contained in the control instruction in the historical hue storage module in correspondence with the minimum score, wherein the third preset range is between the first preset range and the second preset range. Step 123: Calculate the difference between the second state score and the minimum score, define it as an adjustment value, adjust the initial hue path, the initial brightness and the initial frequency based on the adjustment value, generate a new control instruction, control the atmosphere light based on the new control instruction, and then return to step 121.

6. The intelligent control method of vehicle-mounted ambient light according to claim 1, characterized in that: The first user status data includes a facial image, and analyzing the first user status data to identify the age type of each member includes: Step 21: extract any member, perform face recognition on the facial image of any member, and obtain multiple skin appearance feature images, wherein the skin appearance feature image is an image that retains any appearance feature and eliminates other appearance features; Step 22: extract any skin appearance feature image, define the appearance feature of any of the skin appearance feature images as a first appearance feature, extract a representative image corresponding to the first appearance feature, compare any of the skin appearance feature images with the representative image, perform positive or negative adjustment on the age stage corresponding to the representative image according to the comparison result, and extract a new representative image based on the adjusted age stage, repeat step 22 until the number of negative adjustment points is greater than or equal to a preset value or the age stage tends to be stable, and calculate the estimated age corresponding to the first appearance feature based on the age stage at the negative adjustment point, wherein the negative adjustment point is the point where the adjustment direction changes; Step 23: After calculating the estimated ages corresponding to all appearance features, perform a weighted average of all estimated ages to obtain a final estimated age of any of the members, and obtain the age type of any of the members based on the final estimated age and a preset age type mapping table.

7. The intelligent control method of vehicle ambient light according to claim 1, characterized in that: Determining the initial brightness based on the age recognition result, the external environment data, and a preset brightness mapping table includes: Step 31: Determine whether there is a member of the first preset age type. If not, proceed to step 32. If yes, determine the initial brightness based on the brightness mapping table, referencing the first preset age type and the external environment data. Step 32: Determine whether there is a member of the second preset age type. If not, proceed to step 33. If yes, determine the initial brightness based on the brightness mapping table, referencing the second preset age type and the external environment data. Step 33: Determine the initial brightness based on the brightness mapping table and with reference to a third preset type and the external environment data.

8. An intelligent control system for vehicle-mounted ambient light, characterized in that: The system comprises: A data acquisition module, used to acquire external environment data, first user status data of vehicle occupants, and vehicle operation data; a hue setting module, configured to analyze the first user status data, identify the first status score of each member, determine a hue change path based on the status identification result and a preset hue strategy, and set the hue change path as an initial hue path; a brightness setting module, configured to analyze the first user status data, identify the age type of each member, and determine an initial brightness based on the age identification result, the external environment data, and a preset brightness mapping table; a frequency setting module, configured to analyze the vehicle operation data, identify the vehicle operating condition, and determine the initial frequency of the hue change path based on the operating condition identification result and a preset frequency mapping table; an ambient light control module, configured to generate a control instruction according to the initial hue path, the initial brightness, and the initial frequency, and control the ambient light based on the control instruction; The first user status data includes a facial image, first physiological data, and second physiological data. Analyzing the first user status data to identify a first status score for each member includes: Extract any member, analyze the facial image, the first physiological data, and the second physiological data corresponding to any member, respectively, to obtain a first member status, a second member status, and a third member status, wherein the member status includes a status type and a level corresponding to the status type, and the status type includes negative, normal, and positive; Based on a first preset rule table, adjusting the first member state with reference to the second member state and the third member state, respectively, to generate a first reference state and a second reference state; Search the first state score mapping table, obtain the first score and the second score corresponding to the first reference state and the second reference state respectively, perform weighted average on the first score and the second score, obtain the first state score corresponding to any of the members, and after traversing all members, obtain the first state score of each member.

9. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, an intelligent control method for a vehicle ambient light according to any one of claims 1 to 7 is implemented.

Images