Light source adjusting method and system

Abstract

The invention discloses a light source adjustment method and system, and relates to the technical field of light source adjustment. The light source adjustment method comprises the steps of obtaining light source adjustment data of a to-be-adjusted landscape vehicle lamp in real time, performing data adjustment analysis to obtain a vehicle condition adjustment index and an emotional response adjustment index of the to-be-adjusted landscape vehicle lamp, obtaining historical light source data set for a plurality of historical times of the to-be-adjusted landscape vehicle lamp, and performing analysis to obtain a preference adjustment index; according to the invention, intelligent light source adjustment is carried out based on the light source adjustment index of the to-be-adjusted landscape vehicle lamp, so that dynamic adjustment can be carried out according to the individual needs of a driver, and the adjustment accuracy of the to-be-adjusted landscape vehicle lamp is improved. And the landscape vehicle lamp is automatically adjusted according to different emotions, so that the arrangement of the landscape vehicle lamp is optimized, the most appropriate light source can be provided, and the driving safety and psychological comfort are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of light source adjustment, and in particular to a light source adjustment method and system. Background Art

[0002] With the rapid development of the automotive industry, the functions of cars are no longer limited to providing convenient transportation. More and more consumers are beginning to pay attention to the comfort of car interiors and driving experience. Driven by this demand, landscape lights, also known as interior ambient lights, have gradually entered the field of vision of car owners as a new type of car interior element. Landscape lights can not only enhance the visual effect of the car interior, but also provide car owners with a more comfortable and personalized driving environment through different brightness, color temperature and hue. The adjustment of interior ambient lights can significantly affect the driver's mood and psychological state, thereby improving the driving experience. Especially during long night driving, reasonable landscape light adjustment can effectively relieve the driver's visual fatigue and enhance safety.

[0003] Prior art, such as a light source adjustment method and system disclosed in a patent application with announcement number CN104105249A, includes: a start module for turning on the light source to illuminate the detection workbench; a shooting module for controlling the image capture device to shoot the reference block in real time; an acquisition module for acquiring images of the reference block captured in real time at various positions around the detection workbench; a calculation module for calculating the average brightness of each image; a judgment module for judging whether the average brightness of the acquired images is consistent; an adjustment module for adjusting the light source to make the average brightness of the images consistent; a statistics module for counting the number of pixels distributed in multiple preset brightness segments in each image; and an adjustment module for adjusting the light source to make the number of pixels in the multiple brightness segments consistent when the number of pixels in the multiple brightness segments is inconsistent. The present invention adjusts the light source to make the number of pixels in the multiple preset brightness segments of the reference object consistent, thereby achieving the goal of adjusting the light source to be completely parallel to the test platform, which provides convenience for subsequent testing.

[0004] Based on the above solution, it is found that the limitations of the existing technology include at least the following problems: the existing technology lacks a real-time response mechanism to the driver's personalized needs and emotional changes, and fails to fully integrate vehicle condition data and the driver's real-time feedback to dynamically adjust the light source, which makes the light source adjustment appear relatively simple and mechanical, and difficult to flexibly adapt to the needs under night driving conditions, which easily leads to the light source adjustment being difficult to accurately match the driver's actual needs in specific scenarios. Especially when driving at night, it is easy for the light source to be unsuitable for the driver's needs or environmental changes, which in turn affects the driver's driving experience and driving safety. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides a light source adjustment method and system, which solves the problem that the existing technology fails to adjust the light source in real time according to the driver's personalized needs and emotional changes, which easily leads to the light source adjustment being too single and difficult to flexibly adapt to night driving conditions, thereby affecting the driving experience and safety.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a light source adjustment method, comprising the following steps: acquiring light source adjustment data of the landscape headlights to be adjusted in real time, the light source adjustment data including vehicle condition data and driver's facial image data; performing data adjustment analysis on the light source adjustment data of the landscape headlights to be adjusted respectively to obtain a vehicle condition adjustment index and an emotional response adjustment index of the landscape headlights to be adjusted; simultaneously acquiring historical light source data of several historical settings of the landscape headlights to be adjusted, and analyzing to obtain a preference adjustment index of the landscape headlights to be adjusted, and performing comprehensive regulation and analysis in combination with the vehicle condition adjustment index and the emotional response adjustment index of the landscape headlights to be adjusted to obtain a light source regulation index of the landscape headlights to be adjusted; and performing intelligent light source adjustment based on the light source regulation index of the landscape headlights to be adjusted.

[0007] Furthermore, the vehicle condition data includes an interior environment adjustment index, an interior reflection index, and an interior light absorption index, and the specific steps for obtaining the vehicle condition adjustment index of the landscape headlight to be adjusted are as follows: comprehensively analyzing the interior environment adjustment index, interior reflection index, and interior light absorption index of the landscape headlight to be adjusted to obtain the interior light environment adjustment index of the landscape headlight to be adjusted; obtaining the driving speed adjustment index, steering adjustment index, and external light intensity value of the landscape headlight to be adjusted, and normalizing them with the interior light environment adjustment index; comprehensively analyzing the normalized driving speed adjustment index, steering adjustment index, external light intensity value, and interior light environment adjustment index of the landscape headlight to be adjusted to obtain the vehicle condition adjustment index of the landscape headlight to be adjusted.

[0008] Furthermore, the specific formula for calculating the vehicle condition adjustment index of the landscape headlight to be adjusted is as follows: ;in, is the vehicle condition adjustment index of the landscape headlight to be adjusted, is the normalized driving speed adjustment index of the landscape vehicle lights to be adjusted, is the speed adjustment coefficient stored in the database, is the normalized external light intensity value of the landscape headlight to be adjusted, is the external light adjustment coefficient stored in the database, is the normalized steering adjustment index of the landscape headlight to be adjusted, is the steering adjustment coefficient stored in the database, is the normalized interior light environment adjustment index of the landscape headlight to be adjusted, It is the internal and external difference adjustment coefficient stored in the database.

[0009] Furthermore, the driving facial image data specifically includes the pixel value and two-dimensional coordinates of each pixel point in the driving facial image, and the specific steps for obtaining the emotional response adjustment index of the landscape headlights to be adjusted are as follows: the pixel value and two-dimensional coordinates of each pixel point in the driving facial image of the landscape headlights to be adjusted are input into a pre-trained facial expression recognition model for predictive analysis to obtain an emotional feedback set of the landscape headlights to be adjusted, namely, an emotional perception index, an emotional response sensitivity index, and an emotional adaptation index; and a comprehensive analysis is performed on the emotional perception index, emotional response sensitivity index, and emotional adaptation index of the landscape headlights to be adjusted to obtain the emotional response adjustment index of the landscape headlights to be adjusted.

[0010] Furthermore, the specific formula for calculating the emotional response adjustment index of the landscape headlight to be adjusted is as follows: ;in, is the emotional response adjustment index of the landscape headlight to be adjusted, is the emotional perception index of the landscape headlight to be adjusted, is the perception coefficient stored in the database, is the perceptual adjustment coefficient stored in the database, is the emotional response sensitivity index of the landscape headlight to be adjusted, is the responsiveness index stored in the database, is the responsiveness adjustment index stored in the database, is the emotional adaptation index of the landscape headlight to be adjusted, is the adaptation coefficient stored in the database, is the adaptive adjustment index stored in the database, is the superposition adjustment index stored in the database, .

[0011] Furthermore, the facial expression recognition model is specifically a deep convolutional neural network, which includes an input layer, several convolutional layers, a nonlinear activation layer, a pooling layer, a flattening layer, a fully connected layer, and a regression output layer. The specific steps of obtaining the emotional feedback set of the landscape lights to be adjusted are as follows: in the input layer of the deep convolutional neural network, receiving the driving facial image data of the landscape lights to be adjusted and preprocessing it; in the convolution layer of the deep convolutional neural network, performing feature extraction processing on the preprocessed driving facial image data of the landscape lights to be adjusted to obtain a facial feature map of the landscape lights to be adjusted; in the nonlinear activation layer of the deep convolutional neural network, performing nonlinear activation processing on the facial feature map of the landscape lights to be adjusted; in the deep convolutional neural network In the pooling layer of the network, the facial feature map of the landscape headlights to be adjusted after nonlinear activation processing is subjected to dimensionality reduction processing to obtain the facial dimensionality reduction feature map of the landscape headlights to be adjusted; in the flattening layer of the deep convolutional neural network, the facial dimensionality reduction feature map of the landscape headlights to be adjusted is flattened to obtain the facial feature vector of the landscape headlights to be adjusted; in the fully connected layer of the deep convolutional neural network, the facial feature vector of the landscape headlights to be adjusted is subjected to feature fusion processing to obtain the facial complex feature vector of the landscape headlights to be adjusted; in the regression output layer of the deep convolutional neural network, the facial feature vector of the landscape headlights to be adjusted is subjected to regression prediction analysis to obtain the emotion perception index, emotion response sensitivity index, and emotion adaptation index of the landscape headlights to be adjusted, that is, the emotion feedback set.

[0012] Furthermore, the historical light source data includes historical brightness values, historical color temperature values, historical hue values, historical light source saturation values, and historical gradient speed values, and the specific steps for obtaining the preference adjustment index of the landscape headlights to be adjusted are as follows: a comprehensive analysis is performed on the historical brightness values, historical color temperature values, historical hue values, historical light source saturation values, and historical gradient speed values of each historical setting of the landscape headlights to be adjusted to obtain the historical brightness fluctuation index, historical color temperature fluctuation index, historical hue fluctuation index, historical light source saturation fluctuation index, and historical gradient speed fluctuation index of the landscape headlights to be adjusted, and standardization is performed; and a comprehensive analysis is performed on the historical brightness fluctuation index, historical color temperature fluctuation index, historical hue fluctuation index, historical light source saturation fluctuation index, and historical gradient speed fluctuation index of the landscape headlights to be adjusted after standardization to obtain the preference adjustment index of the landscape headlights to be adjusted.

[0013] Furthermore, the specific formula for calculating the light source control index of the landscape vehicle light to be adjusted is as follows: ;in, is the light source control index of the landscape vehicle light to be adjusted, is the vehicle condition adjustment index of the landscape headlight to be adjusted, is the vehicle condition adjustment coefficient in the database, is the emotional response adjustment index of the landscape headlight to be adjusted, is the emotion regulation coefficient in the database, is the preference adjustment index of the landscape headlight to be adjusted, is the preference adjustment coefficient in the database, is the gain adjustment coefficient in the database, is the composite adjustment coefficient in the database.

[0014] Furthermore, the specific steps of performing intelligent light source adjustment based on the light source control index of the landscape headlight to be adjusted are as follows: judging and analyzing the light source control index of the landscape headlight to be adjusted and the preset light source control index threshold; if the light source control index of the landscape headlight to be adjusted is lower than the preset light source control index threshold, generating a corresponding gain control signal and adjusting the landscape headlight to be adjusted; if the light source control index of the landscape headlight to be adjusted is not lower than the preset light source control index threshold, generating a corresponding reduction control signal and adjusting the landscape headlight to be adjusted.

[0015] A light source adjustment system includes an adjustment data acquisition module for acquiring light source adjustment data of a landscape vehicle light to be adjusted in real time, the light source adjustment data including vehicle condition data and driver's facial image data; an adjustment data analysis module for performing data adjustment analysis on the light source adjustment data of the landscape vehicle light to be adjusted to obtain a vehicle condition adjustment index and an emotional response adjustment index of the landscape vehicle light to be adjusted; a comprehensive adjustment analysis module for simultaneously acquiring historical light source data of several historical settings of the landscape vehicle light to be adjusted, analyzing the data to obtain a preference adjustment index of the landscape vehicle light to be adjusted, and performing comprehensive control analysis in combination with the vehicle condition adjustment index and the emotional response adjustment index of the landscape vehicle light to be adjusted to obtain a light source control index of the landscape vehicle light to be adjusted; and an intelligent light source adjustment module for performing intelligent light source adjustment based on the light source control index of the landscape vehicle light to be adjusted.

[0016] The present invention has the following beneficial effects: (1) The light source adjustment method obtains vehicle condition data and driver's facial image data in real time, and combines it with the emotional response adjustment index to dynamically adjust the driver's personalized needs and automatically adjust the landscape headlights according to different emotions, thereby improving the driver's comfort, helping the driver to stay alert, and improving the driving experience. In addition, the preference adjustment index is combined to optimize the landscape headlight settings, thereby ensuring that the most appropriate light source can be provided, thereby achieving the effect of improving driving safety and psychological comfort.

[0017] (2) This light source adjustment method can provide a more adaptive headlight adjustment method during night driving by obtaining vehicle condition data in real time and combining it with the driver's emotional feedback. In particular, when the driver's emotions fluctuate greatly, the landscape headlight light source can be adjusted in time to help the driver maintain a higher level of attention and comfort, thereby effectively reducing driving fatigue and improving the safety of night driving. When the driver is in a bad mood, the adjustment of the light source can provide a stress-relieving atmosphere, help the driver maintain a good driving state, and effectively reduce the probability of traffic accidents.

[0018] (3) The light source adjustment method combines the vehicle condition adjustment index, the emotional response adjustment index and the historical light source data, and performs comprehensive control analysis to obtain the light source control index, thereby realizing intelligent and precise adjustment of the light source, thereby adjusting the light source in real time during night driving, and then ensuring that the adjusted landscape lights match the driver's needs and preferences. Then, the adjustment ability of the lights can be utilized to avoid excessive or insufficient light source brightness, improve energy efficiency and service life, and ensure the best effect of the interior atmosphere lights during night driving.

[0019] (4) The light source adjustment system, by integrating multiple modules, acquires and analyzes vehicle condition data, driver facial image data and historical light source data in real time, so as to perform comprehensive light source adjustment, thereby automatically optimizing light source adjustment according to the driver's emotional changes and vehicle conditions, thereby achieving more flexible and precise light source adjustment, thereby avoiding unnecessary energy waste, and ensuring that the headlights can provide the best lighting effect in any situation, thereby improving the accuracy and adaptability of the system and enhancing driving comfort.

[0020] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a flow chart of a light source adjustment method of the present invention.

[0022] Figure 2 The present invention is a flowchart of the specific steps for obtaining the vehicle condition adjustment index of the landscape vehicle light to be adjusted in a light source adjustment method of the present invention.

[0023] Figure 3 This is a block diagram of a light source adjustment system of the present invention. DETAILED DESCRIPTION

[0024] See also Figure 1An embodiment of the present invention provides a technical solution: a light source adjustment method, comprising the following steps: (during night driving) obtaining light source adjustment data of a landscape headlight to be adjusted (i.e., an interior ambient light) in real time, the light source adjustment data including vehicle condition data and driver's facial image data; performing data adjustment analysis on the light source adjustment data of the landscape headlight to be adjusted to obtain a vehicle condition adjustment index and an emotional response adjustment index of the landscape headlight to be adjusted; simultaneously obtaining historical light source data of several historical settings of the landscape headlight to be adjusted, analyzing the data to obtain a preference adjustment index of the landscape headlight to be adjusted, and performing a comprehensive control analysis based on the vehicle condition adjustment index and the emotional response adjustment index of the landscape headlight to be adjusted to obtain a light source control index of the landscape headlight to be adjusted; and performing intelligent light source adjustment based on the light source control index of the landscape headlight to be adjusted.

[0025] The specific formula for calculating the light source control index of the landscape vehicle lights to be adjusted is as follows: ;in, is the light source control index of the landscape vehicle light to be adjusted, is the vehicle condition adjustment index of the landscape headlight to be adjusted, is the vehicle condition adjustment coefficient in the database, is the emotional response adjustment index of the landscape headlight to be adjusted, is the emotion regulation coefficient in the database, is the preference adjustment index of the landscape headlight to be adjusted, is the preference adjustment coefficient in the database, is the gain adjustment coefficient in the database, is the composite adjustment coefficient in the database.

[0026] It needs to be explained that the formula This item is used to adjust the interaction between the vehicle condition adjustment index and the emotional response adjustment index to avoid the light source control index being too high or too low.

[0027] 、 、 、 、 It can be obtained through the following steps: using historical data, combined with the vehicle condition adjustment index, emotional response adjustment index, and preference adjustment index, to conduct statistical regression analysis, quantify the specific impact of each factor on the light source control index, and thus fit the initial weight value; secondly, using the sensitivity analysis method, adjust the value range of each coefficient, observe its impact on the light source control evaluation results, and ensure the stability and rationality of the model.

[0028] Specifically, if Figure 2As shown, the vehicle condition data includes an interior environment adjustment index, an interior reflection index, and an interior light absorption index, and the specific steps for obtaining the vehicle condition adjustment index of the landscape headlight to be adjusted are as follows: comprehensively analyzing the interior environment adjustment index, interior reflection index, and interior light absorption index of the landscape headlight to be adjusted to obtain the interior light environment adjustment index of the landscape headlight to be adjusted; obtaining the driving speed adjustment index, steering adjustment index, and external light intensity value of the landscape headlight to be adjusted, and normalizing them with the interior light environment adjustment index; comprehensively analyzing the normalized driving speed adjustment index, steering adjustment index, external light intensity value, and interior light environment adjustment index of the landscape headlight to be adjusted to obtain the vehicle condition adjustment index of the landscape headlight to be adjusted.

[0029] Among them, the interior reflection index is the comprehensive reflective ability of the surface of each interior material (such as leather, wood, metal, etc.) in the car to light. It can be obtained by obtaining the reflectivity of each interior material and performing weighted processing. The result is this parameter, and the reflectivity of each interior material can be obtained through the optical material data table stored in the database.

[0030] The interior absorption index is the comprehensive absorption capacity of each interior material (such as leather, wood, metal, etc.) in the car to light. It can be obtained by obtaining the absorbance value of each interior material and performing weighted processing. The result is the parameter. The absorbance value of each interior material can be obtained from the optical material data table stored in the database.

[0031] The in-car environment adjustment index is the comprehensive impact of the in-car environment on the adjustment of the in-car ambient light. It can be obtained by obtaining the in-car temperature value (which can be obtained through the in-car temperature control sensor), the in-car humidity value (which can be obtained through the in-car humidity sensor), the in-car light intensity value (which can be obtained through the in-car light sensor), the in-car noise value (which can be obtained through the in-car noise sensor), the in-car temperature reference value, the in-car humidity reference value, the in-car light intensity reference value, and the in-car noise reference value, and performing ratio processing on them respectively (such as the absolute value of the difference between the in-car temperature value and the in-car temperature reference value / in-car temperature reference value), and performing weighted processing based on the ratio processing result. The result obtained is the parameter, and the in-car temperature reference value can be obtained by obtaining several historical in-car temperature values and performing average processing. The in-car humidity reference value, the in-car light intensity reference value, the in-car noise reference value and the in-car temperature reference value are obtained in the same logic.

[0032] The driving speed adjustment index is used to evaluate the impact of driving speed on the adjustment of the interior ambient light source. At higher speeds, the brightness of the interior ambient light is reduced to avoid interfering with the driver's vision. Conversely, at lower speeds, the brightness of the interior ambient light needs to be increased to improve comfort. It can be obtained by obtaining the driving speed value (which can be obtained through a speed sensor) and the driving speed reference value (that is, the speed limit of the current road, which can be obtained through GPS navigation), and performing a ratio analysis. The result is this parameter, that is, |driving speed value-driving speed reference value| / driving speed reference value.

[0033] The steering adjustment index is used to evaluate the impact of vehicle steering operations on the adjustment of the interior ambient light source. When turning, the interior ambient light needs to be adjusted to ensure the comfort of the driver. It can be obtained by obtaining the steering angle value (which can be obtained through the steering angle sensor) and the steering angular velocity value (which can be obtained through the gyroscope), and performing normalization. Based on the normalization result, weighted processing is performed, and the result obtained is this parameter.

[0034] The external light intensity value can be obtained by the vehicle's external light sensor.

[0035] The specific formulas for calculating the interior light environment adjustment index and vehicle condition adjustment index of the landscape headlights to be adjusted are as follows: ;in, is the interior light environment adjustment index of the landscape vehicle lights to be adjusted, is the in-car environment adjustment index of the landscape headlight to be adjusted, is the internal environment adjustment coefficient stored in the database, The interior reflection index of the landscape lamp to be adjusted is is the interior reflection adjustment coefficient stored in the database, The interior light absorption index of the landscape headlight to be adjusted is is the interior light absorption adjustment coefficient stored in the database, is the interaction adjustment coefficient stored in the database, is the difference adjustment coefficient stored in the database, is the vehicle condition adjustment index of the landscape headlight to be adjusted, is the normalized driving speed adjustment index of the landscape vehicle lights to be adjusted, is the speed adjustment coefficient stored in the database, is the normalized external light intensity value of the landscape headlight to be adjusted, is the external light adjustment coefficient stored in the database, is the normalized steering adjustment index of the landscape headlight to be adjusted, is the steering adjustment coefficient stored in the database, is the normalized interior light environment adjustment index of the landscape headlight to be adjusted, It is the internal and external difference adjustment coefficient stored in the database.

[0036] It needs to be explained that the formula This item is used to quantify the relative differences in interior reflectivity and light absorption index to ensure appropriate adjustment effects under different interior materials.

[0037] 、 、 、 、 It can be obtained through the following steps: Based on historical data, combined with the interior environment adjustment index, interior reflection index, and interior absorption index, statistical regression analysis is performed to quantify the specific impact of each factor on the interior light environment adjustment index, thereby fitting the initial weight value. Secondly, the sensitivity analysis method is used to adjust the value range of each coefficient and observe its impact on the evaluation results of the interior light environment adjustment to ensure the stability and rationality of the model.

[0038] In the formula This item is used to adjust the difference between the external light intensity value and the interior light environment adjustment index.

[0039] 、 、 、 It can be obtained through the following steps: Based on historical data, determine the initial impact weight of each variable (such as driving speed adjustment index, external light intensity value, steering adjustment index, and interior light environment adjustment index) on the vehicle condition adjustment index through statistical regression analysis. Then, use the sensitivity analysis method to adjust the value range of the coefficient to evaluate the stability and applicability of these parameters to the formula output. Next, further fit the weights through model optimization (such as machine learning algorithm) to ensure that the formula can accurately reflect the impact of vehicle conditions on light source adjustment.

[0040] The specific implementation of calculating the vehicle condition adjustment index of the landscape lights to be adjusted is as follows, and the following data is available: The in-vehicle environment adjustment index of the landscape headlights to be adjusted is approximately: 0.305.

[0041] The interior reflectivity index of the landscape headlight to be adjusted is approximately: 0.295.

[0042] The interior light absorption index of the landscape headlight to be adjusted is approximately: 0.669.

[0043] The internal environment adjustment coefficient stored in the database is approximately: 0.412.

[0044] The interior reflection adjustment coefficient stored in the database is approximately: 1.548.

[0045] The interior light absorption adjustment coefficient stored in the database is approximately: 0.247.

[0046] The interaction adjustment coefficient stored in the database is approximately: 0.318.

[0047] The difference adjustment coefficient stored in the database is approximately: 1.426.

[0048] Substituting the above data into the specific formula for calculating the interior light environment adjustment index of the landscape headlight to be adjusted, we obtain: The interior light environment adjustment index of the landscape headlight to be adjusted is ≈0.818.

[0049] And the driving speed adjustment index of the landscape vehicle lights to be adjusted is approximately: 0.217.

[0050] The steering adjustment index of the landscape headlights to be adjusted is approximately: 0.364.

[0051] The external light intensity value of the landscape headlight to be adjusted is approximately (unit: ux): 83.000.

[0052] And normalize it to get: The normalized interior light environment adjustment index of the landscape headlight to be adjusted is approximately 0.726.

[0053] The normalized driving speed adjustment index of the landscape vehicle lights to be adjusted is approximately 0.371.

[0054] The normalized steering adjustment index of the landscape headlight to be adjusted is approximately 0.416.

[0055] The normalized external light intensity value of the landscape headlight to be adjusted is approximately 0.467.

[0056] The speed adjustment coefficient stored in the database is approximately: 0.197.

[0057] The external light adjustment coefficient stored in the database is approximately: 0.327.

[0058] The steering adjustment coefficient stored in the database is approximately: 0.241.

[0059] The internal and external difference adjustment coefficient stored in the database is approximately: 0.683.

[0060] Substituting the above data into the specific formula for calculating the vehicle condition adjustment index of the landscape lights to be adjusted, we obtain: The vehicle condition adjustment index of the landscape headlights to be adjusted is ≈0.881.

[0061] In this implementation, through a comprehensive analysis of the in-car environment adjustment index, interior reflection index, and interior absorption index, the impact of different interior materials and environmental factors (such as temperature, humidity, noise, etc.) on the in-car light source adjustment can be comprehensively evaluated, thereby ensuring that the landscape car light source can be accurately adjusted according to specific needs in different in-car environments, thereby providing the driver with the best lighting effect and avoiding light source imbalance caused by interior materials or environmental factors. Secondly, by obtaining dynamic driving data such as driving speed, steering angle, and external light intensity, and normalizing them with the in-car light environment adjustment index, the car light source can be adjusted in real time according to changes in driving conditions. For example, when driving at high speed, the brightness of the light source will automatically decrease to reduce interference with the driver's vision, while when driving at low speed or turning, the light source will be enhanced to provide more comfort and security. Finally, through statistical regression analysis and sensitivity analysis, the impact of the interior environment adjustment index, interior reflection index and interior absorption index on the light source adjustment can be accurately quantified, and the coefficients of each parameter can be adjusted to ensure that the light source adjustment is more personalized, stable and reasonable. In addition, the model is optimized in combination with machine learning algorithms, and it automatically adapts to different vehicle conditions and driving environments based on historical data, adjusts the light source in real time, and further improves intelligence and adaptability, so as to ensure the best light source adjustment effect in various driving scenarios.

[0062] Specifically, the driving facial image data specifically includes the pixel value and two-dimensional coordinates of each pixel point in the driving facial image, and the specific steps for obtaining the emotional response adjustment index of the landscape headlights to be adjusted are as follows: the pixel value and two-dimensional coordinates of each pixel point in the driving facial image of the landscape headlights to be adjusted are input into a pre-trained facial expression recognition model for predictive analysis to obtain an emotional feedback set of the landscape headlights to be adjusted, namely, an emotional perception index (measuring the driver's current emotional intensity state), an emotional response sensitivity index (measuring the driver's reaction intensity to light changes), and an emotional adaptation index (measuring the degree of match between the current in-car lighting effect and the driver's facial emotional response); and a comprehensive analysis is performed on the emotional perception index, emotional response sensitivity index, and emotional adaptation index of the landscape headlights to be adjusted to obtain the emotional response adjustment index of the landscape headlights to be adjusted.

[0063] The specific formula for calculating the emotional response adjustment index of the landscape headlights to be adjusted is as follows: ;in, is the emotional response adjustment index of the landscape headlight to be adjusted, is the emotional perception index of the landscape headlight to be adjusted, is the perception coefficient stored in the database, is the perceptual adjustment coefficient stored in the database, is the emotional response sensitivity index of the landscape headlight to be adjusted, is the responsiveness index stored in the database, is the responsiveness adjustment index stored in the database, is the emotional adaptation index of the landscape headlight to be adjusted, is the adaptation coefficient stored in the database, is the adaptive adjustment index stored in the database, is the superposition adjustment index stored in the database, .

[0064] It needs to be explained that the formula This item is used to adjust the superposition effect of the emotion perception index, emotion response sensitivity index, and emotion adaptation index to avoid the emotion response adjustment index being too high or too low.

[0065] 、 、 It can be obtained through the following steps: reading the emotion perception index, emotion response sensitivity index, and emotion adaptation index of the landscape headlights to be adjusted, and performing sum analysis to obtain the emotion sum value, and performing a proportion analysis on the emotion perception index, emotion response sensitivity index, and emotion adaptation index of the landscape headlights to be adjusted and the emotion sum value respectively, and taking the corresponding coefficient as the proportion analysis result.

[0066] 、 、 、 It can be obtained through the following steps: Based on historical data, determine the initial impact weight of each variable (emotion perception index, emotion response sensitivity index, emotion adaptation index) on the vehicle condition adjustment index through statistical regression analysis, then use the sensitivity analysis method to adjust the value range of the coefficient to evaluate the stability and applicability of these parameters to the formula output, and then further fit the weights through model optimization (such as multi-objective optimization) to ensure that the formula can accurately reflect the impact of vehicle conditions on light source adjustment.

[0067] In this embodiment, the pixel value and two-dimensional coordinates of each pixel in the facial image are input into the emotion recognition model to analyze the driver's emotional intensity, the intensity of his reaction to light source changes, and the degree of match between the current interior lighting environment and the driver's emotions in real time. This allows the light source of the interior landscape headlights to be dynamically adjusted according to the driver's emotional state, thereby improving the driving experience and reducing the impact of emotional fluctuations on driving. Secondly, a comprehensive analysis of the emotion perception index, emotion response sensitivity index, and emotion adaptation index ensures that light source adjustment does not rely solely on static parameters but can be flexibly adjusted based on the driver's immediate emotional response, thereby improving driver comfort and reducing visual discomfort or fatigue caused by emotion mismatch, thereby improving the quality of the driving experience. Finally, through sensitivity analysis and multi-objective optimization methods, the coefficients of the emotion perception index, reaction sensitivity index, and adaptation index can be optimized to ensure that the formulas work stably in various driving environments. This allows the light source adjustment to accurately adapt to different emotional states, avoiding excessive or insufficient emotional response adjustments, and ensuring that the headlight light source adjustment remains reasonable and stable despite emotional changes.

[0068] Specifically, the facial expression recognition model is a deep convolutional neural network, which includes an input layer, several convolutional layers, a nonlinear activation layer, a pooling layer, a flattening layer, a fully connected layer, and a regression output layer. The specific steps for obtaining the emotional feedback set of the landscape headlights to be adjusted are as follows: in the input layer of the deep convolutional neural network, the driving facial image data of the landscape headlights to be adjusted is received and preprocessed; in the convolution layer of the deep convolutional neural network, the preprocessed driving facial image data of the landscape headlights to be adjusted is subjected to feature extraction processing (i.e., by sliding the convolution kernel, such as scanning the input image data with a filter, extracting features of different levels from the image, such as the shape of the mouth, the degree of closure of the eyes, etc., and generating a feature map), and obtaining the facial feature map of the landscape headlights to be adjusted; in the nonlinear activation layer of the deep convolutional neural network, the facial feature map of the landscape headlights to be adjusted is subjected to nonlinear activation processing (i.e., nonlinear factors are introduced to enable the network to learn more complex features, such as the ReLU activation function to perform nonlinear processing on the convolution output); in the pooling layer of the deep convolutional neural network, the facial feature map of the landscape headlights to be adjusted after nonlinear activation processing is subjected to nonlinear activation processing. The facial feature map is subjected to dimensionality reduction processing (i.e., the most representative part is selected from the activated feature map, such as the maximum value or average value of the local area, to reduce its dimension and spatial resolution, reduce the amount of calculation, and retain important spatial information), thereby obtaining a reduced-dimensional facial feature map of the landscape headlights to be adjusted; in the flattening layer of the deep convolutional neural network, the reduced-dimensional facial feature map of the landscape headlights to be adjusted is flattened (i.e., the two-dimensional matrix of each channel in the feature map is flattened into a one-dimensional vector, and the flattened results of all channels are connected. For the flattened results of each channel, a one-dimensional vector is obtained, and finally these vectors are merged into a longer one-dimensional vector), thereby obtaining a facial feature vector of the landscape headlights to be adjusted; in the fully connected layer of the deep convolutional neural network, the facial feature vector of the landscape headlights to be adjusted is subjected to feature fusion processing (i.e., processing with a weight matrix and an activation function, wherein each input neuron and output neuron has a weight, and the input features are fused by weighted summation, and nonlinear activation is performed to increase the nonlinear ability of the model to obtain complex features), thereby obtaining a complex facial feature vector (i.e., a high-level feature vector) of the landscape headlights to be adjusted;In the regression output layer of the deep convolutional neural network, the facial feature vector of the headlight to be adjusted undergoes regression prediction analysis (i.e., each regression branch is responsible for predicting a different index, and each regression branch is independent and can make predictions based on the input shared facial feature vector. The regression branch of each task further processes the complex facial feature vector through a fully connected layer, mapping the complex facial feature vector to a new space and calculating the corresponding index through linear combination. For example, the emotion perception index is calculated by performing matrix multiplication on the input complex facial feature vector and the weight matrix, adding a bias term, and then applying a linear activation function to obtain the output, with the output value between 0 and 1). This results in the emotion perception index, emotion response sensitivity index, and emotion adaptation index of the headlight to be adjusted, i.e., the emotion feedback set.

[0069] The input layer receives facial image data of the driver whose headlights are to be adjusted. The input includes the pixel value and two-dimensional coordinate data for each pixel. This data is passed to the convolutional layer for processing as input to the model. The input layer performs preliminary preprocessing on the input image, such as normalization (to ensure that the data range is suitable for network learning).

[0070] The convolution layer is used for feature extraction. The convolution layer will extract local features related to facial expressions, such as the shape of the mouth and the degree of eye closure. These low-level features provide the basis for subsequent emotion prediction.

[0071] The nonlinear activation layer is used to convert the features extracted by the convolutional layer into a nonlinear feature map through the activation function, so that the model can learn more complex mappings and capture the complex relationship between facial expressions and emotions.

[0072] The pooling layer is used to reduce the dimension of the feature map and retain the most important features, such as retaining important information in facial expressions (such as the height of eyebrows, the upward corners of the mouth, etc.), and remove unimportant details.

[0073] The flattening layer is used to flatten the facial dimensionality reduction feature map into a one-dimensional vector and compress the features of each pixel into a column of values. This one-dimensional vector contains all the feature information extracted from the image and provides input for the fully connected layer.

[0074] The fully connected layer is used to weight and sum the output of the flattened layer to form the final complex facial feature vector, which integrates all the information from each layer (such as emotion intensity, facial features, expression changes, etc.).

[0075] The regression output layer is used for regression analysis to predict complex facial feature vectors and output the emotion perception index, emotion response sensitivity index, emotion adaptation index, etc.

[0076] And the pre-training steps of the deep convolutional neural network are as follows: Obtain an image annotation dataset containing several groups of facial annotation images of emotional states (such as happiness, sadness, anger, surprise, etc.), and divide the image annotation dataset into a face training set and a face verification set.

[0077] Initialize the convolutional neural network, initialize the weights and biases in the network to ensure that the network can effectively learn features, initialize the convolutional layer weights (use He initialization), initialize the bias (initialize the bias term to zero), and activate the function. Use the ReLU activation function to increase the nonlinearity of the network.

[0078] Training is performed based on the facial training set, that is, the number of training cycles is set, and in each training cycle, the network weights are updated through the backpropagation algorithm.

[0079] And the processing steps of each training cycle are: Forward propagation: Each input image is processed through each layer of the network (convolutional layer, pooling layer, fully connected layer) to obtain the model's predicted value.

[0080] Calculate the loss: Calculate the value of the loss function based on the difference between the predicted result and the true label.

[0081] Backpropagation: The backpropagation algorithm calculates the gradient of the loss function with respect to the weights of each layer and updates the model parameters based on these gradients. Backpropagation uses the chain rule to propagate the error from the output layer back to the input layer and update the weights based on the gradients.

[0082] After each training cycle, an evaluation analysis is performed based on the facial validation set. The model structure or hyperparameters are adjusted by calculating the error of the validation set, such as the learning rate. Choose an appropriate learning rate. Too large a learning rate may lead to unstable training, while too small a learning rate may lead to slow convergence. A dynamic learning rate adjustment strategy or an adaptive learning rate method can be used.

[0083] If the error of the face validation set is high and the error of the face training set is low, it may be overfitting. Regularization methods (such as L2 regularization or Dropout) can be used to reduce overfitting.

[0084] If the model performance does not meet expectations, you can try adjusting the network structure (increasing / decreasing the number of layers), optimizing the algorithm, data augmentation methods, etc. to further improve the model performance.

[0085] When the training is completed and the loss and accuracy on the face validation set reach the expected standards, the training process ends and a trained network model is obtained.

[0086] In this embodiment, the deep convolutional neural network can deeply analyze changes in facial expressions through multi-layer processing of the driver's facial image data (such as feature extraction, nonlinear activation, pooling, etc.), thereby accurately capturing subtle emotional fluctuations such as emotional intensity, reaction sensitivity, and adaptability. This allows the in-vehicle lighting to be adjusted in real time based on the driver's emotional state, improving the driving experience and avoiding the discomfort caused by emotional disorders. Secondly, through predictive analysis of the regression output layer, accurate adjustments can be made for different emotional states, ensuring that the in-vehicle lighting is well matched to the driver's emotional response. Finally, through multi-layer processing of the deep convolutional neural network, the driver's facial features are comprehensively analyzed, and personalized lighting adjustment is provided for each driver. As a result, during driving, changes in headlights will not adversely affect the driver's emotions, but instead provide appropriate lighting based on the driver's needs and emotions, thereby enhancing driving safety and comfort, which is particularly important during nighttime driving. In addition, through a systematic training process (including forward propagation, loss calculation, backpropagation, etc.), combined with facial expression data, the network model can be continuously optimized, ultimately achieving accurate recognition and prediction of driver emotions.

[0087] Specifically, the historical light source data includes historical brightness values, historical color temperature values, historical hue values, historical light source saturation values, and historical gradient speed values, and the specific steps for obtaining the preference adjustment index of the landscape headlights to be adjusted are as follows: a comprehensive analysis (i.e., standard deviation processing) is performed on the historical brightness values, historical color temperature values, historical hue values, historical light source saturation values, and historical gradient speed values of each historical setting of the landscape headlights to be adjusted to obtain the historical brightness fluctuation index, historical color temperature fluctuation index, historical hue fluctuation index, historical light source saturation fluctuation index, and historical gradient speed fluctuation index of the landscape headlights to be adjusted, and standardization processing is performed (i.e., unit removal processing); and a comprehensive analysis is performed on the historical brightness fluctuation index, historical color temperature fluctuation index, historical hue fluctuation index, historical light source saturation fluctuation index, and historical gradient speed fluctuation index of the landscape headlights to be adjusted after the standardized processing to obtain the preference adjustment index of the landscape headlights to be adjusted.

[0088] Among them, historical brightness values, historical color temperature values, historical hue values, historical light source saturation values, and historical gradient speed values can all be obtained through the vehicle data recorder.

[0089] The specific formula for calculating the preference adjustment index of the landscape headlight to be adjusted is as follows: ;in, is the preference adjustment index of the landscape headlight to be adjusted, is the historical brightness fluctuation index of the landscape lights to be adjusted after normalization. is the brightness coefficient in the database, is the historical color temperature fluctuation index of the landscape vehicle lights to be adjusted after standardization. is the color temperature coefficient in the database, is the historical hue fluctuation index of the landscape vehicle lights to be adjusted after normalization. is the hue coefficient in the database, is the historical light source saturation fluctuation index of the landscape vehicle lights to be adjusted after normalization. is the saturation coefficient in the database, is the historical gradient speed fluctuation index of the landscape lights to be adjusted after normalization. is the gradient coefficient in the database, , is a natural constant and takes a value of 2.71 in this embodiment.

[0090] What needs to be explained is that 、 、 、 、 It can be obtained through the following steps: reading the historical brightness fluctuation index, historical color temperature fluctuation index, historical hue fluctuation index, historical light source saturation fluctuation index, and historical gradient speed fluctuation index of the landscape headlights to be adjusted after standardization, and performing sum analysis to obtain the preference sum value, and performing a proportion analysis on the historical brightness fluctuation index, historical color temperature fluctuation index, historical hue fluctuation index, historical light source saturation fluctuation index, and historical gradient speed fluctuation index of the landscape headlights to be adjusted after standardization with the preference sum value, and using the proportion analysis results as the corresponding parameters.

[0091] In this implementation, by analyzing historical fluctuations in multiple light source parameters such as brightness, color temperature, hue, saturation, and gradient speed, the driver's preferences for these parameters can be quantified and identified. Based on this historical data, the driver can be provided with a light source adjustment solution that best meets their needs, thereby ensuring the comfort and adaptability of the headlights in different driving scenarios and improving driver satisfaction. Secondly, by processing the standard deviation and normalizing the historical light source data, the dimensional differences between different light source parameters can be eliminated and the fluctuation of each light source parameter under different driving conditions can be accurately reflected. This allows for in-depth analysis of the driver's acceptance of light source changes, ensuring that the headlight adjustment is more in line with personal preferences, thereby avoiding overly drastic or inappropriate light source changes and improving comfort. Finally, by collecting and analyzing historical data, not only can personalized adjustments be provided for the current driving environment, but the system can also be optimized through continuous data accumulation. Therefore, in future driving processes, the light source can be further accurately adjusted based on more historical data, thereby better meeting the driver's changing needs and improving the long-term adaptability and accuracy of light source adjustment.

[0092] Specifically, the specific steps of performing intelligent light source adjustment based on the light source control index of the landscape headlight to be adjusted are as follows: the light source control index of the landscape headlight to be adjusted is judged and analyzed with a preset light source control index threshold; if the light source control index of the landscape headlight to be adjusted is lower than the preset light source control index threshold, a corresponding gain control signal is generated, and the landscape headlight to be adjusted is adjusted (via the driving circuit) (certain parameters of the headlight, such as brightness, color temperature, hue, etc., are improved to achieve a more appropriate light source adjustment state. For example, the brightness of the light source can be increased, or the color temperature and hue can be adjusted to make the light source of the headlight more in line with the driver's needs and driving environment); if the light source control index of the landscape headlight to be adjusted is not lower than the preset light source control index threshold, a corresponding reduction control signal is generated, and the landscape headlight to be adjusted is adjusted (via the driving circuit) (certain parameters of the headlight can be reduced, such as reducing brightness, reducing color temperature, or adjusting hue to avoid the light source being too glaring or interfering with the driver's vision. For example, the brightness can be reduced or the color temperature can be adjusted to make the light source of the headlight more in line with the driver's comfort needs and avoid the light source being too strong and affecting the driving experience).

[0093] In this embodiment, by real-time calculation and comparison of the light source control index of the landscape headlights to be adjusted with the preset threshold, dynamic adjustment can be made according to the current driving environment and driver needs, so that parameters such as the brightness, color temperature and hue of the headlights can always be maintained within the most appropriate range, ensuring that the light source effect in each driving scenario meets the driver's comfort needs. For example, when driving at night, the brightness of the light source can be automatically increased, while in a strong light environment, the brightness can be reduced to avoid eye fatigue. Based on intelligent light source adjustment, it can automatically adjust according to the real-time control index and preset threshold of the headlights without the need for manual operation by the driver, thereby simplifying the driver's operating steps and improving driving convenience, thereby improving user experience, thereby reducing human errors or inappropriate manual adjustments, and ensuring that the light source is always in the best state under different circumstances.

[0094] See also Figure 3An embodiment of the present invention provides a technical solution: a light source adjustment system, comprising: an adjustment data acquisition module, configured to acquire light source adjustment data of a landscape vehicle light to be adjusted in real time, the light source adjustment data including vehicle condition data and driver's facial image data; an adjustment data analysis module, configured to perform data adjustment analysis on the light source adjustment data of the landscape vehicle light to be adjusted, and obtain a vehicle condition adjustment index and an emotional response adjustment index of the landscape vehicle light to be adjusted; a comprehensive adjustment analysis module, configured to simultaneously acquire historical light source data of several historical settings of the landscape vehicle light to be adjusted, and analyze the data to obtain a preference adjustment index of the landscape vehicle light to be adjusted, and perform comprehensive control analysis based on the vehicle condition adjustment index and the emotional response adjustment index of the landscape vehicle light to be adjusted, and obtain a light source control index of the landscape vehicle light to be adjusted; and an intelligent light source adjustment module, configured to perform intelligent light source adjustment based on the light source control index of the landscape vehicle light to be adjusted.

[0095] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0096] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A light source adjustment method, characterized in that: The following steps are involved: Real-time acquisition of light source adjustment data for the landscape vehicle lights to be adjusted, wherein the light source adjustment data includes vehicle condition data and driver's facial image data; Perform data adjustment analysis on the light source adjustment data of the landscape lights to be adjusted, and obtain a vehicle condition adjustment index and an emotional response adjustment index of the landscape lights to be adjusted; At the same time, historical light source data of several historical settings of the landscape lights to be adjusted are obtained, and the preference adjustment index of the landscape lights to be adjusted is obtained by analysis. A comprehensive control analysis is performed in combination with the vehicle condition adjustment index and the emotional response adjustment index of the landscape lights to be adjusted to obtain the light source control index of the landscape lights to be adjusted; And intelligent light source adjustment is performed based on the light source control index of the landscape vehicle light to be adjusted.

2. The light source adjustment method according to claim 1, wherein: The vehicle condition data includes an interior environment adjustment index, an interior reflection index, and an interior light absorption index, and the specific steps for obtaining the vehicle condition adjustment index of the landscape headlight to be adjusted are as follows: Comprehensively analyzing the interior environment adjustment index, interior reflection index, and interior light absorption index of the landscape lamp to be adjusted to obtain the interior light environment adjustment index of the landscape lamp to be adjusted; Obtain the driving speed adjustment index, steering adjustment index, and external light intensity value of the landscape headlight to be adjusted, and normalize them with the interior light environment adjustment index; Based on the normalized driving speed adjustment index, steering adjustment index, external light intensity value, and interior light environment adjustment index of the landscape headlight to be adjusted, a comprehensive analysis is performed to obtain the vehicle condition adjustment index of the landscape headlight to be adjusted.

3. The light source adjustment method according to claim 2, wherein: The specific formula for calculating the vehicle condition adjustment index of the landscape headlights to be adjusted is as follows: ； in, is the vehicle condition adjustment index of the landscape headlight to be adjusted, 、 、 、 They are the normalized driving speed adjustment index, external light intensity value, steering adjustment index, and interior light environment adjustment index of the landscape vehicle lights to be adjusted. 、 、 、 They are the speed adjustment coefficient, external light adjustment coefficient, steering adjustment coefficient, and internal and external difference adjustment coefficient stored in the database.

4. The light source adjustment method according to claim 1, wherein: The driver's facial image data specifically includes the pixel value and two-dimensional coordinates of each pixel point in the driver's facial image, and the specific steps of obtaining the emotional response adjustment index of the landscape headlight to be adjusted are as follows: The pixel value and two-dimensional coordinates of each pixel point in the facial image of the driver of the landscape headlight to be adjusted are input into a pre-trained facial expression recognition model for prediction analysis to obtain an emotional feedback set of the landscape headlight to be adjusted, namely, an emotional perception index, an emotional response sensitivity index, and an emotional adaptation index; A comprehensive analysis is performed on the emotion perception index, emotion response sensitivity index, and emotion adaptation index of the landscape headlights to be adjusted to obtain the emotion response adjustment index of the landscape headlights to be adjusted.

5. The light source adjustment method according to claim 4, wherein: The specific formula for calculating the emotional response adjustment index of the landscape headlights to be adjusted is as follows: ； in, is the emotional response adjustment index of the landscape headlight to be adjusted, 、 、 They are the emotional perception index, emotional response sensitivity index, and emotional adaptation index of the landscape lights to be adjusted. 、 、 、 、 、 、 The following are the perception coefficient, perception adjustment coefficient, response sensitivity index, response sensitivity adjustment index, adaptation coefficient, adaptation adjustment index, and superposition adjustment index stored in the database. .

6. The light source adjustment method according to claim 4, wherein: The facial expression recognition model is specifically a deep convolutional neural network, which includes an input layer, several convolutional layers, a nonlinear activation layer, a pooling layer, a flattening layer, a fully connected layer, and a regression output layer. The specific steps of obtaining the emotional feedback set of the landscape lights to be adjusted are as follows: In the input layer of the deep convolutional neural network, the driver's face image data of the landscape headlight to be adjusted is received and preprocessed; In the convolution layer of the deep convolutional neural network, feature extraction is performed on the pre-processed facial image data of the driver of the landscape headlight to be adjusted to obtain a facial feature map of the landscape headlight to be adjusted; In the nonlinear activation layer of the deep convolutional neural network, the facial feature map of the landscape headlight to be adjusted is subjected to nonlinear activation processing; In the pooling layer of the deep convolutional neural network, the facial feature map of the landscape headlight to be adjusted after the nonlinear activation processing is subjected to dimensionality reduction processing to obtain the facial dimensionality reduction feature map of the landscape headlight to be adjusted; In the flattening layer of the deep convolutional neural network, the facial dimensionality reduction feature map of the landscape headlight to be adjusted is flattened to obtain the facial feature vector of the landscape headlight to be adjusted; In the fully connected layer of the deep convolutional neural network, the facial feature vector of the landscape headlight to be adjusted is subjected to feature fusion processing to obtain the complex facial feature vector of the landscape headlight to be adjusted; In the regression output layer of the deep convolutional neural network, regression prediction analysis is performed on the facial feature vector of the landscape headlight to be adjusted to obtain the emotion perception index, emotion response sensitivity index, and emotion adaptation index of the landscape headlight to be adjusted, that is, the emotion feedback set.

7. The light source adjustment method according to claim 1, wherein: The historical light source data includes historical brightness values, historical color temperature values, historical hue values, historical light source saturation values, and historical gradient speed values. The specific steps for obtaining the preference adjustment index of the landscape headlight to be adjusted are as follows: Comprehensively analyze the historical brightness values, historical color temperature values, historical hue values, historical light source saturation values, and historical gradient speed values of each historical setting of the landscape headlight to be adjusted, obtain the historical brightness fluctuation index, historical color temperature fluctuation index, historical hue fluctuation index, historical light source saturation fluctuation index, and historical gradient speed fluctuation index of the landscape headlight to be adjusted, and perform standardization processing; A comprehensive analysis is then performed on the historical brightness fluctuation index, historical color temperature fluctuation index, historical hue fluctuation index, historical light source saturation fluctuation index, and historical gradient speed fluctuation index of the standardized landscape headlights to be adjusted, and the preference adjustment index of the landscape headlights to be adjusted is obtained.

8. The light source adjustment method according to claim 1, wherein: The specific formula for calculating the light source control index of the landscape vehicle lights to be adjusted is as follows: ； in, is the light source control index of the landscape vehicle light to be adjusted, 、 、 They are the vehicle condition adjustment index, emotional response adjustment index, and preference adjustment index of the landscape lights to be adjusted. 、 、 、 、 They are the vehicle condition adjustment coefficient, emotion adjustment coefficient, preference adjustment coefficient, gain adjustment coefficient, and composite adjustment coefficient in the database.

9. The light source adjustment method according to claim 1, wherein: The specific steps for intelligent light source adjustment based on the light source control index of the landscape vehicle light to be adjusted are as follows: The light source control index of the landscape vehicle light to be adjusted is judged and analyzed with a preset light source control index threshold; If the light source control index of the landscape vehicle light to be adjusted is lower than the preset light source control index threshold, a corresponding gain control signal is generated, and the landscape vehicle light to be adjusted is adjusted; If the light source control index of the landscape headlight to be adjusted is not lower than the preset light source control index threshold, a corresponding reduction control signal is generated, and the landscape headlight to be adjusted is adjusted.

10. A light source adjustment system, applying the light source adjustment method according to any one of claims 1 to 9, characterized in that: include: An adjustment data acquisition module is used to acquire light source adjustment data of the landscape vehicle lights to be adjusted in real time, wherein the light source adjustment data includes vehicle condition data and driver's facial image data; An adjustment data analysis module is used to perform data adjustment analysis on the light source adjustment data of the landscape lights to be adjusted, and obtain a vehicle condition adjustment index and an emotional response adjustment index of the landscape lights to be adjusted; A comprehensive adjustment analysis module is used to simultaneously obtain historical light source data of several historical settings of the landscape lights to be adjusted, analyze and obtain the preference adjustment index of the landscape lights to be adjusted, and perform comprehensive control analysis based on the vehicle condition adjustment index and emotional response adjustment index of the landscape lights to be adjusted to obtain the light source control index of the landscape lights to be adjusted; The intelligent light source adjustment module is used to perform intelligent light source adjustment based on the light source control index of the landscape vehicle light to be adjusted.

Images