Light supplementing control system and method for vehicle-mounted cosmetic mirror

Abstract

The invention relates to the technical field of vehicle-mounted cosmetic mirrors, in particular to a light supplementing control system of a vehicle-mounted cosmetic mirror, and the system comprises an array type light sensing module which is used for collecting an illumination intensity signal; the judgment module is used for receiving the illumination intensity signal collected by the array type light sensing module and generating a starting instruction when the illumination intensity signal changes suddenly; in the starting process, an illumination intensity signal is collected, a dynamic safety threshold value of the illumination intensity of the make-up mirror light supplementing lamp is set according to the intensity of the illumination intensity signal, and a corresponding light supplementing scheme is drawn up; after the brightness is stable, the collected illumination intensity signal is converted into a user operation signal to be analyzed, and a light supplementing scheme is adjusted according to a preset rule; and the execution module is used for receiving the starting instruction and then accurately controlling the illumination intensity of the make-up mirror light supplementing lamp according to the dynamic safety threshold value of the illumination intensity of the make-up mirror. According to the invention, a light supplementing scheme can be drawn up, automatic adjustment of light supplementing parameters is realized, and manual adjustment requirements of passengers are met.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle-mounted makeup mirrors, and particularly to a supplementary lighting control system and method for a vehicle-mounted makeup mirror. Background Art

[0002] Through the evolution of social culture, makeup plays a very important role in people's lives. Makeup not only reflects one's own beauty but is also an indispensable presence in formal occasions. With the continuous improvement of consumers' requirements for automotive interiors and comfort, as an important accessory to enhance the driving and riding experience, the functionality of vehicle-mounted makeup mirrors has received increasing attention.

[0003] Traditional vehicle-mounted makeup mirrors usually only come with simple lighting devices, such as LED lights or fluorescent lights, to provide basic lighting functions. However, these lighting devices often cannot be automatically adjusted according to the ambient light and user needs, resulting in poor makeup application effects for users and affecting the user experience.

[0004] In summary, traditional vehicle-mounted makeup mirrors usually only have simple lighting devices, which are difficult to be automatically adjusted according to the ambient light and user needs, and the problem of poor makeup application effects for users has become a difficult problem that urgently needs to be solved in this field. Therefore, it is necessary to propose a supplementary lighting control system and method for a vehicle-mounted makeup mirror. Summary of the Invention

[0005] To solve the above problems, the present invention provides a supplementary lighting control system and method for a vehicle-mounted makeup mirror, which can formulate a supplementary lighting plan, realize automatic adjustment of supplementary lighting parameters, and meet the manual adjustment needs of passengers.

[0006] To achieve the above object, the technical solution of the present invention is as follows: A supplementary lighting control system for a vehicle-mounted makeup mirror, comprising: An array-type photosensitive module for collecting light intensity signals; A judgment module for receiving the light intensity signals collected by the array-type photosensitive module, generating a start instruction when a sudden change occurs in the light intensity signals; during startup, collecting light intensity signals, setting a dynamic safety threshold for the light intensity of the makeup mirror supplementary lighting according to the intensity of the light intensity signals, and formulating a corresponding supplementary lighting plan; after the brightness is stable, converting the collected light intensity signals into user operation signals for analysis and adjusting the supplementary lighting plan according to preset rules; An execution module for receiving the start instruction, and then precisely controlling the light intensity of the makeup mirror supplementary lighting according to the dynamic safety threshold of the makeup mirror light intensity, or executing the supplementary lighting plan according to the user operation signals.

[0007] Further, the manner in which the judgment module generates the start instruction is as follows: When the array light-sensing module moves from the folded state to the opened state along with the makeup mirror, the judgment module monitors in real time the light intensity signal collected by the array light-sensing module; assume that n continuously collected light intensity values form a sequence , where n is the preset number of samples. In a normal daytime environment, if the inequality is satisfied, where k is the starting index, m is the length of the preset continuous comparison window, is the daytime start threshold constant, the judgment module immediately generates a start instruction; while in a nighttime environment, due to the weak change in light intensity, a classification model based on deep learning is adopted. This model is pre-trained using a large number of sample data containing different light change scenarios at night. The input of the model is the light intensity value sequence I, and the output is the start decision value D. When , where is the nighttime start threshold, the judgment module generates a start instruction.

[0008] Furthermore, during the startup process, the judgment module performs the following operations to set the dynamic safety threshold and formulate a supplementary lighting plan: Assume that the light intensity signal sequence continuously collected by the array light-sensing module is , where t is the number of sampling moments during the startup process; first, divide the light intensity signal sequence into multiple sub-sequences with equal time lengths, and each sub-sequence contains m sampling values, that is: .

[0009] Furthermore, for each sub-sequence , calculate its average light intensity ; then, use the weighted moving average method to process these average light intensity values to obtain the comprehensive light intensity index , where is the preset weight coefficient, and it satisfies . The weight coefficient can be set according to the importance of different time stages where the sub-sequences are located. The weight of the sub-sequences in the initial stage of startup is relatively low, and the weight is higher closer to the current moment to highlight the influence of recent light changes.

[0010] Furthermore, based on the comprehensive light intensity index I 综合 , set the dynamic safety threshold T 安全 , and adopt the linear regression model T 安全 = aI 综合 + b, where a and b are coefficients obtained by fitting a large amount of experimental data, aiming to give a reasonable range of safety thresholds according to the comprehensive situation of environmental light, ensuring that the brightness of the supplementary light can not only meet the makeup needs but also not cause visual interference to the driver.

[0011] Further, when formulating the supplementary lighting scheme, a supplementary lighting decision-making model based on a neural network is constructed. The input layer of this model receives the comprehensive light intensity index I 综合 , the current vehicle speed signal v, which is assumed to be obtained from the vehicle-mounted system, and the current supplementary lighting state s of the vanity mirror. The hidden layer uses multiple neurons for feature extraction and operation, and the output layer outputs the supplementary lighting adjustment strategy , where represents the adjustment instruction for a certain attribute of the supplementary light, and k is the number of adjustment attributes.

[0012] Further, after the brightness is stabilized, the judgment module performs the following operations: Let the light intensity signal sequence collected by the array-type photosensitive module be , where m is the number of sampling points; by setting a time window , the sequence Q is divided into multiple subsequences , where , n is the number of windows, and each subsequence corresponds to a potential user operation time period.

[0013] Further, for each subsequence , calculate its standard deviation , where , if there is a subsequence that satisfies , it is determined that there is a user operation during this time period, and the user instruction recognition process is started.

[0014] Further, the user instruction recognition adopts a model based on a convolutional neural network. The input is the subsequence matrix M containing the characteristics of the light intensity change. The convolutional layer extracts the local characteristics of the light intensity change, the pooling layer performs dimensionality reduction, and the fully connected layer performs feature fusion. The output is the user instruction category vector , where represents different user instruction types, such as brightness increase, brightness decrease, color temperature switching. Through the softmax function for normalization, so that , to determine the most likely user instruction; After the user instruction is recognized, the supplementary lighting scheme is adjusted according to the preset rules. Let the parameter vector of the current supplementary lighting scheme be , corresponding to the brightness and color temperature attributes of the supplementary light. According to the recognized user instruction category, an adjustment matrix A is constructed. If the user instruction is brightness increase, the corresponding row elements of A are the preset brightness increment values and other adjustment amounts. Through matrix multiplication The parameter vector of the adjusted supplementary lighting scheme is obtained , so as to realize the smooth and intelligent control of the brightness of the vanity mirror supplementary light.

[0015] The technical principle of the above scheme is as follows: The perception module first perceives the current driving speed of the vehicle, the light intensity of the current environment, the lighting of the passenger's face by the vanity mirror in the current environment, the passenger's dressing style, the passenger's skin color, and the progress of the passenger's makeup. Subsequently, the judgment module sets a dynamic safety threshold for the vanity mirror lighting intensity based on the current driving speed of the vehicle and formulates a supplementary lighting scheme based on a deep learning model. Finally, the execution module executes the supplementary lighting scheme and compares the vanity mirror lighting intensity with the set dynamic threshold, and adjusts the lighting intensity of the vanity mirror based on the comparison result. When the passenger needs to adjust the parameters of the vanity mirror light, the lighting can be manually adjusted through the adjustment module.

[0016] The beneficial effects of adopting the above scheme are as follows: The supplementary lighting control system of this in-vehicle vanity mirror mainly consists of an array photosensing module, a judgment module, and an execution module working together. Its core principle is to accurately collect light intensity signals through the array photosensing module, the judgment module conducts intelligent analysis and decision-making based on these signals, and finally the execution module executes the corresponding supplementary lighting operation to achieve intelligent control of the vanity mirror supplementary light.

[0017] In this invention, by perceiving the driving speed of the vehicle and setting a dynamic safety threshold for the vanity mirror lighting intensity accordingly, the lighting intensity of the vanity mirror can be intelligently adjusted during vehicle driving, reducing the possibility of driver distraction or blocked vision caused by excessive light, thereby improving driving safety.

[0018] In this invention, by using a deep learning model and combining the light intensity of the current environment, the lighting of the passenger's face by the vanity mirror, the passenger's dressing style, and skin color, a personalized supplementary lighting scheme can be formulated. This customized supplementary lighting not only improves the makeup effect but also enables the passenger to maintain the best makeup effect under different lighting conditions.

[0019] The array photosensing module continuously collects light intensity signals. When the vanity mirror is in different states, such as being folded to opened, or when the user operates after the brightness is stable, this module can keenly capture the changes in light intensity and output these changes in the form of a signal sequence. This is the basic data source for subsequent judgment and control of the entire system. It can automatically judge whether to start the supplementary lighting system according to different environmental light conditions (such as day and night), and dynamically adjust the brightness of the supplementary light during the startup process, achieving a high degree of intelligence. For example, in a night environment, the startup conditions can be accurately judged through a deep learning model, avoiding problems such as misjudgment or untimely startup caused by weak light changes.

[0020] The judgment module receives the light intensity signals collected by the array light-sensing module. When a sudden change occurs in the light intensity signals, the judgment module will, according to the preset rules and models, determine whether the startup conditions are met. For example, in the daytime environment, by calculating the change in light intensity and comparing it with the daytime startup threshold constant; in the nighttime environment, using a classification model based on deep learning, and comparing the startup decision value output by the model with the nighttime startup threshold to determine whether to generate a startup instruction. After the brightness stabilizes, the system can recognize the user's operation instructions, such as increasing brightness, decreasing brightness, or color temperature switching, etc., and adjust the fill light scheme according to these instructions, meeting the personalized needs of users. Different users can flexibly adjust the lighting effect of the dressing mirror according to their makeup habits and preferences. During the entire control process, the system adopts a variety of mathematical models and algorithms, such as the weighted moving average method, linear regression model, and convolutional neural network model, etc., to achieve smooth and intelligent control of the brightness of the fill light of the dressing mirror. It avoids the discomfort caused to users by sudden changes in brightness, and at the same time can more accurately adapt to changes in the environment and user needs.

[0021] During the startup process, the judgment module further processes the collected light intensity signals. It divides the signal sequence into subsequences, calculates the average light intensity, and then obtains the comprehensive light intensity index through the weighted moving average method. Based on this index, a dynamic safety threshold for the light intensity of the fill light of the dressing mirror is set using a linear regression model, and a corresponding fill light scheme is drawn up.

[0022] After the brightness stabilizes, the judgment module continues to analyze the collected light intensity signals. It judges whether there is a user operation by calculating the standard deviation of the subsequences. If there is, it uses a convolutional neural network model to identify the category of the user instruction. According to the recognition result, combined with the preset rules and adjustment matrix, the parameter vector of the current fill light scheme is adjusted, thus realizing the dynamic optimization of the fill light scheme. By setting the dynamic safety threshold of the dressing mirror light intensity, it ensures that the brightness of the fill light will not interfere with the driver's line of sight, improving the user's makeup experience while ensuring driving safety. For example, during vehicle driving, the system will comprehensively adjust the fill light scheme according to factors such as vehicle speed, etc., to avoid the influence of too bright or too dark light on the driver's attention.

[0023] The execution module receives the startup instruction and the fill light scheme adjustment instruction, and precisely controls the light intensity of the fill light of the dressing mirror according to the dynamic safety threshold of the dressing mirror light intensity, ensuring that the brightness and other attributes of the fill light meet the system's decision and the user's needs. In summary, the present invention realizes the effect of being able to draw up a fill light scheme, realizing the automatic adjustment of fill light parameters, and meeting the manual adjustment needs of passengers. Description of the Drawings

[0025] Figure 1This is the logic block diagram of the supplementary light control system for the in-vehicle makeup mirror of the present invention. Specific Embodiments

[0026] The following is a further detailed description through specific embodiments: As shown in the embodiment Figure 1 A supplementary light control system for an in-vehicle makeup mirror includes: An array-type photosensitive module for collecting light intensity signals; A judgment module for receiving the light intensity signals collected by the array-type photosensitive module, generating a start instruction when the light intensity signal mutates; during the startup process, collecting the light intensity signals, setting a dynamic safety threshold for the light intensity of the makeup mirror supplementary light according to the intensity of the light intensity signals, and formulating a corresponding supplementary light scheme; after the brightness stabilizes, converting the collected light intensity signals into user operation signals for analysis and adjusting the supplementary light scheme according to preset rules; An execution module for receiving the start instruction, and then precisely controlling the light intensity of the makeup mirror supplementary light according to the dynamic safety threshold of the makeup mirror light intensity, or executing the supplementary light scheme according to the user operation signal.

[0027] During specific use: The array-type photosensitive module selects a suitable light sensor, arranges multiple sensors in an array form, and installs them around the in-vehicle makeup mirror to ensure that the light intensity signals around the makeup mirror can be comprehensively and accurately collected. The sensor should have high sensitivity and fast response characteristics to capture subtle changes in light intensity in real time.

[0028] The judgment module uses a high-performance microprocessor as the core of the judgment module (8155 or a chip with higher capabilities). This module needs to have sufficient computing power to handle complex mathematical operations and model calculations, such as operations of deep learning models and neural network models. At the same time, a corresponding storage device is equipped to store preset parameters, model data, experimental data, etc.

[0029] The execution module mainly consists of a drive circuit and a supplementary light. The drive circuit is responsible for receiving the instructions sent by the judgment module and precisely controlling attributes such as the brightness and color temperature of the supplementary light according to the instructions. The supplementary light should select a light source with high color rendering and low power consumption to provide high-quality makeup lighting effects.

[0030] During the specific use process, assume a female car owner, Xiao Li, is driving to a picnic in the suburbs on a sunny weekend. When she sits in the driver's seat and wants to touch up her makeup before reaching the destination, she opens the in-vehicle makeup mirror.

[0031] When the array photosensitive module moves from the folded state to the opened state along with the makeup mirror, the judgment module monitors in real time the light intensity signal collected by the array photosensitive module. Since it is daytime, the judgment module uses the daytime startup judgment mechanism to analyze the collected sequence of light intensity values. Suppose that n continuously collected light intensity values form a sequence , where n is the preset sample quantity. In a normal daytime environment, because it is noon and the sunlight is sufficient, the light outside the vehicle changes significantly. If the inequality is satisfied, where k is the starting index, m is the preset length of the continuous comparison window, is the daytime startup threshold constant, the judgment module immediately generates a startup instruction; while in a night environment, due to the weak change in light intensity, a classification model based on deep learning is adopted. This model is pre-trained with a large number of sample data containing different light change scenarios at night. The input of the model is the sequence I of light intensity values, and the output is the startup decision value D. When , where is the night startup threshold, the judgment module generates a startup instruction. After receiving the startup instruction, the execution module immediately turns on the fill light of the makeup mirror.

[0032] During the startup process, the judgment module performs the following operations to set the dynamic safety threshold and formulate the fill light scheme: The sequence of light intensity signals continuously collected by the array photosensitive module is , where t is the number of sampling moments during the startup process; First, divide the sequence of light intensity signals into multiple subsequences , each subsequence contains m sampling values, that is: .

[0033] For each subsequence , calculate its average light intensity ; Then, use the weighted moving average method to process these average light intensity values to obtain the comprehensive light intensity index , where is the preset weight coefficient, and it satisfies . The weight coefficient can be set according to the importance of different time stages where the subsequences are located. The weight of the subsequences in the initial stage of startup is relatively low, and the weight is higher closer to the current moment to highlight the influence of recent light changes.

[0034] Based on the comprehensive light intensity index I 综合 set the dynamic safety threshold T 安全 , adopt the linear regression model T 安全 = aI 综合+ b, where a = 0.5 and b = 10 (obtained by fitting a large amount of experimental data), aiming to give a reasonable range of safety thresholds according to the comprehensive environmental light conditions to ensure that the brightness of the fill light can meet the makeup needs without causing visual interference to the driver.

[0035] When formulating the fill light scheme, a fill light decision-making model based on a neural network is constructed. The input layer of this model receives the comprehensive light intensity index I 综合 , the current vehicle speed signal v, which is assumed to be obtainable from the in-vehicle system, and the current fill light state s of the makeup mirror. The hidden layer uses multiple neurons for feature extraction and calculation, and the output layer outputs the fill light adjustment strategy. , where represents the adjustment instruction for a certain attribute of the fill light, and k is the number of adjustment attributes. The execution module adjusts the brightness of the fill light to an appropriate level according to this strategy to adapt to the in-vehicle light environment and avoid being too bright or too dark.

[0036] When Xiao Li finishes her preliminary makeup and wants to fine-tune the brightness of the fill light, her hand makes an occlusion action in front of the makeup mirror, causing a change in the light intensity. The judgment module performs the following operations: Let the light intensity signal sequence collected by the array-type photosensitive module be , where m is the number of sampling points; by setting a time window , the sequence Q is divided into multiple subsequences , where , n is the number of windows, and each subsequence corresponds to a potential user operation time period.

[0037] For each subsequence , calculate its standard deviation , where , because Xiao Li's hand movement makes the subsequence satisfy , it is determined that there is a user operation during this time period, and the user instruction recognition process is started.

[0038] User instruction recognition uses a model based on a convolutional neural network. The input is the subsequence matrix M containing the characteristics of the light intensity change. The convolutional layer extracts the local characteristics of the light intensity change, the pooling layer performs dimensionality reduction, and the fully connected layer performs feature fusion. The output is the user instruction category vector , where represents different user instruction types, such as brightness increase, brightness decrease, color temperature switching. Through the softmax function for normalization, it is made that , to determine the most likely user instruction. If it is determined that Xiao Li's instruction is to increase the brightness.

[0039] After identifying the user instruction, adjust the fill light scheme according to a preset rule. Let the parameter vector of the current fill light scheme be , corresponding to the brightness and color temperature attributes of the fill light. According to the identified user instruction category, construct an adjustment matrix A. If the user instruction is to increase the brightness, the corresponding row elements of A are preset adjustment amounts such as the brightness increment value. Through matrix multiplication Obtain the adjusted parameter vector of the fill light scheme . By increasing the brightness of the fill light, Xiao Li obtained a fill light effect that better met his own needs and successfully completed the improvement of his makeup, welcoming the upcoming picnic party in a perfect state.

[0040] Obviously, the above embodiments are merely examples given for clear illustration and are not limitations on the implementation manners. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to enumerate all the implementation manners here. And the obvious changes or modifications derived therefrom are still within the protection scope of the present invention.

Claims

1. A supplementary lighting control system for a vehicle-mounted makeup mirror, characterized in that Including: An array-type photosensitive module for collecting light intensity signals; A judgment module for receiving the light intensity signals collected by the array-type photosensitive module and generating a start instruction when the light intensity signals mutate; During the startup process, collect light intensity signals, set a dynamic safety threshold for the light intensity of the makeup mirror fill light according to the intensity of the light intensity signals, and formulate a corresponding fill light scheme; after the brightness is stable, convert the collected light intensity signals into user operation signals for analysis and adjust the fill light scheme according to preset rules; An execution module for receiving the start instruction and then precisely controlling the light intensity of the makeup mirror fill light according to the dynamic safety threshold of the makeup mirror light intensity, or executing the fill light scheme according to the user operation signal.

2. The supplementary lighting control system of the in-vehicle makeup mirror according to claim 1, wherein The judgment module generates the start instruction in the following manner: When the array - type photosensitive module moves from the folded state to the open state with the makeup mirror, the judgment module monitors the light intensity signal collected by the array - type photosensitive module in real time. Let n continuously collected light intensity values form a sequence , where n is the preset number of samples. In a normal daytime environment, if the inequality is satisfied, where k is the starting index and m is the preset length of the continuous comparison window, is the daytime start - up threshold constant, the judgment module immediately generates a start - up instruction. In a nighttime environment, due to the weak change in light intensity, a classification model based on deep learning is used. This model is pre - trained with a large number of sample data containing different nighttime light change scenarios. The input of the model is the light intensity value sequence I, and the output is the start - up decision value D. When , where is the nighttime start - up threshold, the judgment module generates a start - up instruction.

3. The supplementary light control system of the in-vehicle makeup mirror according to claim 2, characterized in that, During the startup process, the judgment module performs the following operations to set the dynamic safety threshold and formulate the fill light scheme: Let the sequence of light intensity signals collected in real time by the array - type photosensitive module be , where \(t\) is the number of sampling moments during the startup process; First, divide the sequence of light intensity signals into multiple subsequences with equal time lengths , and each subsequence contains \(m\) sampling values, that is: 。 4. The supplementary light control system of the in-vehicle makeup mirror according to claim 3, characterized in that, For each subsequence , calculate its average light intensity ; then, use the weighted moving average method to process these average light intensity values to obtain a comprehensive light intensity index , where is a preset weight coefficient and satisfies . The weight coefficient can be set according to the importance of the time stages of different subsequences. The weight of the subsequence in the initial stage of startup is relatively low, and the weight is higher the closer it is to the current moment, so as to highlight the influence of recent light changes.

5. The supplementary lighting control system of the in-vehicle makeup mirror according to claim 4, characterized in that, Based on the comprehensive light intensity index I 综合 Set the dynamic safety threshold T 安全 , and use the linear regression model T 安全 = aI 综合 + b, where a and b are coefficients obtained by fitting a large amount of experimental data, aiming to give a reasonable range of safety thresholds according to the comprehensive environmental light conditions, ensuring that the brightness of the supplementary light can not only meet the makeup needs but also will not cause visual interference to the driver.

6. The supplementary lighting control system of the vehicle-mounted makeup mirror according to claim 5, wherein When formulating the supplementary lighting plan, a supplementary lighting decision-making model based on a neural network is constructed. The input layer of this model receives the comprehensive light intensity index I 综合 , the current vehicle speed signal v, which is assumed to be obtained from the vehicle-mounted system, and the current supplementary lighting state s of the vanity mirror. The hidden layer uses multiple neurons for feature extraction and operation, and the output layer outputs the supplementary lighting adjustment strategy , where represents the adjustment instruction for a certain attribute of the supplementary light, and k is the number of adjusted attributes.

7. The supplementary light control system of the in-vehicle makeup mirror according to claim 6, characterized in that, After the brightness is stable, the judgment module performs the following operations: Let the sequence of light intensity signals collected by the array photosensitive module be , where m is the number of sampling points; by setting a time window , the sequence Q is divided into multiple subsequences , where , n is the number of windows, and each subsequence corresponds to a potential user operation time period.

8. The supplementary light control system of the in-vehicle makeup mirror according to claim 7, characterized in that, For each subsequence , calculate its standard deviation , where , if there exists a subsequence that satisfies , it is determined that there is a user operation during this time period, and the user instruction recognition process is started.

9. The supplementary lighting control system of the in-vehicle makeup mirror according to claim 8, characterized in that, The user instruction recognition uses a model based on a convolutional neural network. The input is a subsequence matrix M containing the characteristics of the light intensity change. The convolutional layer extracts the local features of the light intensity change, the pooling layer performs dimensionality reduction, and the fully connected layer performs feature fusion. The output is a user instruction category vector , where represents different types of user instructions, such as brightness increase, brightness decrease, and color temperature switching. Through the softmax function for normalization, to determine the most likely user instruction; After recognizing the user instruction, adjust the fill light scheme according to the preset rules. Let the parameter vector of the current fill light scheme be , corresponding to the brightness and color temperature attributes of the fill light. According to the recognized user instruction category, construct an adjustment matrix A. If the user instruction is to increase the brightness, the corresponding row elements of A are preset adjustment amounts such as the brightness increment value. Through matrix multiplication Obtain the adjusted fill light scheme parameter vector , thus realizing smooth and intelligent control of the brightness of the makeup mirror fill light.

10. A method for controlling supplementary lighting of an in-vehicle makeup mirror, characterized in that, A fill light control system for a vehicle-mounted makeup mirror as described in any one of claims 1-9 is adopted.

Images