Intelligent light adjusting system based on facial emotion and environment recognition technology

Abstract

According to the intelligent light adjusting system based on the facial emotion and environment recognition technology, the Internet of Things technology is used for sensing weather environment changes, meanwhile, light adjustment is carried out in combination with recognized emotions, and compared with an existing emotion light adjusting system, the intelligent light adjusting system has the advantages that sensing and adjustment are more accurate, and the user experience is improved. The facial emotion recognition of the EmoRepLKNet provides excellent emotion recognition accuracy; secondly, environmental factors are considered, and light adjustment can be more flexible and real-time in combination with perception of weather environment changes; in addition, the natural light intensity can be adjusted by controlling the closing degree of the curtain, and the light adjusting dimensionality is improved. Compared with the prior art, the system has the advantages that the user experience is enhanced, and light adjustment can be more personalized by combining emotional response of the user and environmental perception data.

Description

Technical Field

[0001] The invention belongs to the field of artificial intelligence and the Internet of Things, and specifically relates to an intelligent light adjustment system based on facial emotion and environment recognition technology. Background Art

[0002] With the continuous development of artificial intelligence, the Internet of Things, and smart hardware, emotion perception and environmental regulation technology has gradually become an important application in the field of smart home and health management. Emotion and mental health have a significant impact on people's daily life, work efficiency, and overall well-being. Therefore, how to use smart devices to improve emotional state has become a hot topic of research.

[0003] In the field of smart home, the use of smart sensors and adaptive systems to adjust the environment (such as temperature, light, and sound) has been widely studied and applied. Among them, light, as a controllable environmental factor, has been proven to have an important regulatory effect on human emotions. By adjusting parameters such as light brightness, color temperature, and color, smart lighting systems can effectively influence people's emotional responses. Therefore, the technology of adjusting lighting based on emotional changes has great market potential, especially in scenarios such as homes, offices, hospitals, and psychotherapy.

[0004] Although the technology of adjusting lighting based on mood changes has made some progress, there are still some technical challenges and shortcomings in practical applications. The specific problems are as follows:

[0005] (1) Accuracy of emotion perception: Currently, most emotion perception systems rely on single environmental data (such as facial expressions or voice emotions), which results in low accuracy of emotion recognition in complex environments, especially when multiple environments are intertwined. Therefore, they lack consideration of other environmental factors that affect emotions.

[0006] (2) Limitations of emotion regulation effects: Although studies have shown that lighting can affect emotions, the significance and sustainability of existing intelligent lighting systems in regulating emotions are still relatively poor, especially in complex environments (such as high-pressure work environments, social interactions, etc.). Simple lighting adjustments often fail to effectively improve emotional states. At the same time, most existing technologies only rely on a single dimension of lighting (such as brightness and color temperature), lack more levels of adjustment, and fail to fully utilize the deep regulatory effects of multi-sensory stimulation on emotions.

[0007] (3) Individual differences in emotion models: Everyone has different emotional responses and physiological characteristics. Most existing technologies are based on universal emotion models and cannot customize emotion recognition and regulation for different individuals. For users with different cultures, personalities, or physiological characteristics, the universality of existing technologies is poor.

[0008] In summary, although the technology of adjusting lighting based on emotional changes has made certain progress, there are still challenges such as low accuracy of emotion perception, low effect of emotion regulation, and insufficient consideration of individual differences. Summary of the invention

[0009] The purpose of the present invention is to provide an intelligent light adjustment system based on facial emotion and environmental recognition technology, which uses Internet of Things technology to perceive weather and environmental changes, and adjusts the light in combination with the recognized emotions. At the same time, combined with the user's emotional response and environmental perception data, the light adjustment can be more personalized.

[0010] An intelligent light adjustment system based on facial emotion and environment recognition technology, comprising a central processing unit and a personalization module, a light control module, an emotion perception module, and an environment perception module respectively connected thereto;

[0011] The emotion recognition module analyzes the user's physiological parameters and identifies the emotional state, while the environmental perception module monitors the external weather conditions in real time. The central processing unit calculates the optimal lighting parameters based on the environmental conditions and emotional state. The light control module then automatically adjusts the light settings and the opening and closing status of the curtains to optimize the indoor light based on the calculation results of the central processing unit. In this process, the user can view the current emotional state, light settings, and curtain status through the personalized customization interface and make personalized adjustments.

[0012] Furthermore, the emotion recognition module adopts the facial emotion recognition network structure EmoRepLKNet based on the UniRepLKNet framework.

[0013] Furthermore, in the emotion recognition module, the user's facial image data is captured in real time by the camera, and the collected raw data is preprocessed. The preprocessed data is then input into EmoRepLKNet for emotion recognition. The facial image data is processed through a series of convolutional layers, RPS Blocks, and downsampling and upsampling operations in EmoRepLKNet to extract facial emotion features. The extracted features are classified using a fully connected layer to identify the user's emotional state. The identified emotional state will match the preset emotion and light setting correspondence, which is adjusted according to the user's personal preferences.

[0014] Furthermore, the environmental perception module includes a temperature and humidity sensor, a light sensor and an air pressure sensor, which monitor the external weather conditions of light intensity, temperature and humidity in real time, and transmit the perception data to the central processing module in real time through a wireless communication protocol.

[0015] Furthermore, in the environmental perception module, light intensity, temperature, and humidity are mapped to emotional results.

[0016] Furthermore, the central processing unit is responsible for coordinating the work of each module and calculating the optimal lighting parameters based on real-time data; the central processing unit collects necessary data from the emotion recognition module and the environmental perception module, and at the same time extracts the user's historical lighting habits and preference settings from the historical database in order to better understand the user's needs.

[0017] Furthermore, the calculation steps of the lighting parameters in the central processing unit include: based on the comprehensive analysis results of emotions and environment, the central processing unit will calculate the optimal lighting parameters, including brightness, color temperature and curtain control; among them, the brightness adjustment calculation takes into account the ambient light intensity and the user's emotional state to determine the appropriate indoor light brightness; the color temperature adjustment calculation combines the ambient temperature and emotional state to adjust the warm and cool tones of the light; the curtain control calculation optimizes the use of natural light and controls the indoor light by adjusting the degree of opening and closing of the curtains; after calculating the optimal lighting parameters, the central processing unit will generate corresponding instructions and send them to the light control module.

[0018] Furthermore, the central processing unit collects user feedback to evaluate satisfaction with current lighting settings; users provide feedback through the user interface to express their satisfaction with current lighting effects or make adjustment suggestions; the system learns based on this feedback and continuously optimizes emotion recognition algorithms and lighting adjustment strategies.

[0019] Furthermore, the light control module adjusts the indoor light by adjusting the light brightness, color temperature and natural light intensity according to the instructions generated by the central processing unit.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] (1) More accurate perception and adjustment:

[0022] First, EmoRepLKNet's facial emotion recognition provides superior emotion recognition accuracy. Secondly, taking environmental factors into consideration, traditional emotional lighting adjustment systems usually rely on the user's emotions to adjust the brightness and tone of the lights. Combining the perception of weather and environmental changes can make lighting adjustment more flexible and real-time. For example, in rainy weather or low temperature environments, the lights can be adjusted to warm tones or higher brightness to enhance comfort and mood. In addition, the degree of curtain closure can be controlled to adjust the intensity of natural light, improving the dimension of light adjustment.

[0023] (2) Enhance user experience:

[0024] Personalized customization, combined with the user's emotional response and environmental perception data, lighting adjustment can be more personalized. For example, some users may prefer dim light on a rainy day, while some users may want to feel brighter and warmer under the same weather conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 4 is a structural block diagram of an intelligent light adjustment system in an embodiment of the present invention.

[0026] Figure 2 4 is a camera position diagram of the intelligent light adjustment system in an embodiment of the present invention.

[0027] Figure 3 FIG. 4 is a diagram showing the sensor locations of the intelligent light adjustment system in an embodiment of the present invention.

[0028] Figure 4 FIG. 4 is a schematic diagram of a user interface of an intelligent light adjustment system in an embodiment of the present invention.

[0029] In the picture, 1-camera, 2-air pressure sensor, 3-light sensor, 4-temperature and humidity sensor, 5-ceiling light, 6-light strip. DETAILED DESCRIPTION

[0030] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings.

[0031] The purpose of the present invention is to provide a light adjustment system that combines environmental factors and emotion recognition. Figure 1 The central processing unit, personalization module, light control module, emotion perception module and environment perception module shown in the figure. The system includes an emotion recognition module that analyzes the user's facial expression parameters to identify the user's emotional state, a real-time monitoring indoor environment perception module, a central processing module that coordinates various modules, a light control module that automatically adjusts the color, brightness and color temperature of the light and controls the opening and closing of the curtains, and a module that allows the user to make personalized adjustments.

[0032] The specific implementation method is that the emotion recognition module first analyzes the user's physiological parameters and identifies the emotional state, while the environmental perception module monitors the external weather and other environmental conditions in real time. Secondly, the central processing unit calculates the optimal lighting parameters based on the environmental conditions and emotional state. Then, the light control module automatically adjusts the light settings and the opening and closing status of the curtains according to the calculation results of the central processing unit to optimize the indoor light. In this process, the user can also view the current emotional state, light settings and curtain status through the personalized customization interface, and make personalized adjustments.

[0033] Compared with the prior art, the intelligent light adjustment system of the present invention can provide a more personalized and comfortable lighting environment and enhance the user experience. By real-time monitoring and responding to the external environment and user emotions, the system can create a more harmonious and comfortable living environment.

[0034] As for the emotion recognition module, the present invention uses the existing EmoRepLKNet as the emotion recognition module in the intelligent lighting system. EmoRepLKNet is a facial emotion recognition network structure based on the universal large-core convolutional neural network UniRepLKNet framework. Its main task is to analyze the user's facial expression and identify the emotional state. This module is a key component of the system because it directly affects the accuracy and responsiveness of lighting adjustment.

[0035] Among them, the masked polarized self-attention module (RPS Block) is an important component of EmoRepLKNet, which combines the Unet structure and the polarized self-attention mechanism. The Unet structure helps to accurately locate the key features in the image, while the polarized self-attention mechanism can highlight or suppress features, similar to the filtering of light by an optical lens, which can improve contrast and detail perception. This mechanism is very suitable for emotion recognition, because subtle changes in facial expressions are often key indicators of emotional state.

[0036] In actual operation, the system captures the user's facial image data in real time through the integrated camera, and preprocesses the collected raw data to improve the accuracy of subsequent analysis. The preprocessing process is shown in Table 1:

[0037] Table 1 Description of preprocessing steps

[0038]

[0039] When analyzing the data later, the preprocessed data will be input into EmoRepLKNet for emotion recognition. The preprocessed facial image data is input into EmoRepLKNet, and the network extracts facial emotion features through a series of convolutional layers, RPS Blocks, and downsampling and upsampling operations. The extracted features are classified using a fully connected layer to identify the user's emotional state. The identified emotional state will be matched with the preset correspondence between emotions and lighting settings. The default matching relationship is shown in Table 2, but this correspondence can be adjusted according to the user's personal preferences.

[0040] Table 2 Emotion and light adjustment mapping table

[0041]

[0042]

[0043] EmoRepLKNet's facial emotion recognition accuracy on the FER2013 dataset reached 76.20%. This result not only surpasses the existing best model, but also significantly improves the accuracy of facial emotion recognition compared to niRepLKNet. At the same time, experiments were conducted on the single-label part of the RAF-DB dataset, and the result was 89.67% accuracy. Camera position Figure 2 shown.

[0044] In addition to emotional state, the system also needs to consider the impact of external weather conditions on lighting. The environmental perception module will monitor external weather conditions such as light intensity, temperature, humidity, etc. in real time.

[0045] The environment perception module uses multiple environment perception sensors to transmit the perception data to the central processing module in real time through wireless communication protocols for analysis and processing. The sensors included are temperature and humidity sensors, light sensors, and air pressure sensors.

[0046] (1) Temperature and humidity sensor:

[0047] The sensor detects the temperature and humidity data in the air in real time through the built-in temperature probe and humidity sensing element. The human body will have different reactions when it is in different temperatures and humidities. For example, in high temperature and high humidity conditions, the human body often feels particularly uncomfortable, and the mood is prone to become irritable, anxious, or even angry. When people are in a low temperature and low humidity environment, they often have dry skin and difficulty breathing, which in turn affects their mood.

[0048] (2) Light sensor:

[0049] The light sensor senses the surrounding light intensity through the photosensitive element. According to the change of light intensity, it outputs the corresponding electrical signal. When the external light becomes dim, the system can adjust the indoor light appropriately through this sensor.

[0050] (3) Air pressure sensor:

[0051] The air pressure sensor measures the current air pressure level through the built-in pressure probe, and predicts weather trends by sensing changes in atmospheric pressure. When people are in different air pressure environments, their emotions will change accordingly. For example, under long-term low pressure conditions caused by dark, gloomy weather and stormy weather, people tend to develop symptoms of depression or low mood. And low pressure is often accompanied by unstable weather conditions, such as heavy rain or thunderstorms, which can also cause uneasiness and anxiety. In contrast to low pressure, high pressure is usually associated with clear, dry and stable weather conditions. When people are in a high-pressure environment, their emotions become pleasant and relaxed. And because there is no trouble brought by gloomy weather, many people will behave positively and optimistically in a high-pressure environment. The specific environmental conditions and emotional mapping table are shown in Table 3. Figure 3 Shown are suggested sensor locations in a room layout.

[0052] Table 3 Environmental conditions and emotion mapping table

[0053]

[0054]

[0055] The emotion recognition results and environmental monitoring data finally obtained by environmental perception and facial emotion recognition need to be transmitted to the central processing system via Wi-Fi, and then the mapping of specific light colors and emotional results is implemented in the central processor.

[0056] The central processor is responsible for coordinating the work of each module and calculating the optimal lighting parameters based on real-time data.

[0057] The specific processing process first carries out data collection and preprocessing. The central processing unit collects necessary data from the emotion recognition module and the environmental perception module. At the same time, the system also extracts the user's historical lighting habits and preference settings from the historical database to better understand the user's needs.

[0058] After data preprocessing, the central processing unit will conduct a comprehensive analysis of emotions and environment. Emotional state analysis processes the collected emotional data through a deep learning model to identify the user's current emotional state. Users' lighting needs may vary greatly under different emotional states. At the same time, an environmental impact assessment is performed to analyze how external environmental factors affect users' emotions and lighting needs.

[0059] Next, the lighting parameters are calculated and optimized. Based on the comprehensive analysis results of emotions and environment, the central processing unit will calculate the optimal lighting parameters, including brightness, color temperature and curtain control. Among them, the brightness adjustment calculation will take into account the ambient light intensity and the user's emotional state to determine the appropriate indoor light brightness. The color temperature adjustment calculation combines the ambient temperature and emotional state to adjust the warm and cold tones of the light. The curtain control calculation optimizes the use of natural light and controls the indoor light by adjusting the degree of opening and closing of the curtains. Once the optimal lighting parameters are calculated, the central processing unit will generate the corresponding instructions and send them to the light control module. The specific comprehensive state control is shown in Table 4:

[0060] Table 4 Comprehensive control mapping table

[0061]

[0062]

[0063] At the same time, the central processing unit will collect user feedback to evaluate the satisfaction of the current lighting settings. Users can provide feedback through the user interface to express their satisfaction with the current lighting effect or make suggestions for adjustments. The system will learn based on this feedback and continuously optimize the emotion recognition algorithm and lighting adjustment strategy.

[0064] Kalman is used to fuse data from multiple sensors during the data processing phase. The Kalman filter algorithm is mainly divided into prediction and update. The input data of the Kalman filter algorithm includes environmental data collected in real time by temperature and humidity sensors, light sensors, and air pressure sensors, as well as emotional state data output by the emotion recognition module. The algorithm output is the optimal environmental state estimate of the system at any given time step, including the optimal estimated values ​​of parameters such as ambient light intensity, temperature, humidity, and the associated emotional state estimate.

[0065] Define the initial state estimate and the initial state covariance matrix P0.

[0066] Determine the state transfer matrix F, the measurement matrix H, the process noise covariance matrix Q, and the measurement noise covariance matrix R.

[0067] Prediction stage:

[0068] Use the state transition equation to predict the state at the next moment: Where k: represents the time step, which is used to distinguish the states and measurement data at different times.

[0069] Predict the state covariance matrix for the next moment: P k|k-1 =FP k-1|k-1 F T +Q.

[0070] Update phase:

[0071] Calculate the Kalman gain K k :K k =P k|k-1 H T (HP k|k-1 H T +R) -1 .

[0072] Update the state estimate using the new measurement data: where z k Represents the measurement data at the kth time step, including the environmental data collected by the sensor and the emotion recognition results.

[0073] Update the covariance matrix of the state estimate: P k|k =(IK k H)P k|k-1 .

[0074] These outputs provide the system with the optimal estimate of the environmental state at any given time step, and through data fusion of the Kalman filter, the system can perceive changes in the external environment more accurately.

[0075] When optimizing lighting decisions, the Q-learning algorithm is used to determine the appropriate light control instructions. The algorithm inputs sensor data fused by Kalman filter, facial emotion recognition data and user personalized selection data.

[0076] Define the state space, action space, and Q-value table.

[0077] Set the learning rate α, discount factor γ, and exploration strategy (such as ε-greedy strategy).

[0078] Learning process:

[0079] State: The agent observes the current state S t .

[0080] Action selection: Select an action based on the current strategy: A t .

[0081] Execute action: The agent executes the action and moves to the new state S t+1 and receive reward R t+1 .

[0082] Q value update:

[0083] Q(S t ,A t )←Q(S t ,A t )+α[R t+1 +γmax a Q(St+1 ,a)-Q(S t ,A t )].

[0084] Among them, Q(s,a) is the Q value of taking action a in state s. α is the learning rate, which determines the influence of new information on the Q value; γ is the discount factor, which determines the influence of future rewards on the current Q value; max a Q(s',a') means finding the maximum Q value for all possible actions a' in the new state s'; s: current state; a: action selected in the current state s, including adjusting light brightness, color temperature, etc.

[0085] Repeat: The agent repeats this process until it reaches a terminal state.

[0086] Over time, the Q-value table will gradually converge to the optimal value, and the agent can choose the best lighting decision based on the current Q-value table.

[0087] In this system, the Kalman filter algorithm is used to fuse data from multiple environmental perception sensors, such as temperature and humidity sensors, light sensors, and air pressure sensors, to obtain a more accurate estimate of the environmental state. The algorithm uses the state transfer matrix and measurement matrix, as well as the process noise and measurement noise covariance matrix, through the prediction and update stages to optimize the prediction of the environmental state and adjust these predictions based on the actual sensor readings. At the same time, the Q-learning algorithm is used in the central processing module to determine the best lighting control action in response to changes in the environment and emotional state. The algorithm learns and updates the expected utility of each state-action pair by defining the state space, action space, and Q value table, and setting the learning rate, discount factor, and exploration strategy. The system observes the current state, selects and executes the action, and then updates the Q value based on the received reward and the expected utility of the next state. Therefore, the combined use of the Kalman filter algorithm and the Q-learning algorithm can help the system make better decisions in a dynamically changing environment and provide a more personalized and comfortable lighting environment.

[0088] According to the instructions generated by the central processing unit, indoor light adjustment is achieved by adjusting the light brightness, color temperature and natural light intensity.

[0089] When adjusting the light brightness, according to the instructions, for example, when the user is in a relaxed state, the light control module will gradually reduce the power of the LED or fluorescent light source after receiving the instructions, so that the light is soft and not dazzling. The adjustment process is gradual to avoid disturbing the user with sudden brightness changes. The light brightness will slowly decrease until it reaches a preset brightness level, which is usually lower than that during daily activities to create a more comfortable and peaceful environment. On the contrary, when the user is in an excited state, the light control module receives instructions and gradually increases the brightness of the light, making the light brighter and more vibrant. Such brightness levels can help increase the user's alertness and excitement, thereby better matching the user's current activities and emotional state.

[0090] The adjustment of color temperature is achieved by changing the color of the light emitted by the light source, usually using LED bulbs or lamps with adjustable color temperature. For example, in a relaxed state, after receiving the instruction, the light control module will adjust the color temperature to a warmer range of tones, such as 2700K to 3000K, so that the light presents a warm yellow or orange tone. Warm-toned light helps to reduce the output of blue light, reduce stimulation to the user's eyes, and create a relaxing and soothing atmosphere, helping users relieve stress and enjoy a quiet leisure time. In an excited state, after receiving the instruction, the light control module will adjust the color temperature to a cooler tone, such as 5000K to 6500K, so that the light presents a brighter white. Cool-toned light contains more blue light components, which can improve the user's alertness and attention, and enhance vitality and wakefulness.

[0091] The intensity of natural light is adjusted by adjusting the degree of closure of the curtains to change the intensity of natural light entering the room. The curtains can be closed to reduce interference from the external environment and provide a more private and quiet environment. The curtains can also be opened to allow more natural light to enter the room, increasing the sense of openness and vitality of the space. The intensity of natural light can also be adjusted according to different times of the day to supplement or balance artificial lighting and create the most suitable lighting environment. For specific adjustment relationships, refer to Table 4.

[0092] The system also provides the function of personalized adjustment. Based on the idea of ​​user interaction, the system provides a user interface that allows users to set the correspondence between their emotions and light colors, as well as manually adjust the lighting settings. This interactivity enables users to customize the lighting environment according to their needs and preferences. Users can observe the emotions and weather conditions observed by the system in real time. When they are not satisfied with the lighting, they can enter the user modification interface below. Users can decide the color temperature and light intensity they want, as well as the degree of opening and closing of the curtains. After the user's selection, feedback is given to the central processing unit, and the system can make personalized adjustments more accurately based on this.

[0093] The system will provide an intuitive color picker, and users can select colors by sliding or clicking the color panel. When selecting a color, the interface will display a preview of the selected color in real time, including the color display effect at different brightness and color temperature, to help users select colors more accurately. At the same time, the system will remember the colors configured by the user for each emotion. When the user selects the same emotion again, the system will automatically apply the previously saved color settings.

[0094] Secondly, in terms of brightness and color temperature control, users can adjust the lighting effects by sliding the brightness and color temperature sliders. The brightness slider controls the brightness of the light, while the color temperature slider controls the cool and warm tones of the light. At the same time, the system provides several preset brightness and color temperature modes, such as "Morning Mode" (higher brightness, cooler color temperature) and "Night Mode" (lower brightness, warmer color temperature), and users can switch to these modes with one click.

[0095] Finally, in terms of environmental factors, real-time data is displayed, including weather (sunny, cloudy, rainy, snowy, etc.), temperature, humidity and other information. This information is displayed through charts or dynamic icons to intuitively reflect the current environmental conditions. The system gives recommended lighting adjustment plans based on the current environmental status. For example, on cloudy days or in dimly lit environments, the system recommends increasing the brightness of indoor lights to enhance the user's comfort and emotional state. When the external environment changes significantly (such as sudden rain or sunset), the system will remind the user through the notification bar and make corresponding lighting adjustments.

[0096] The system will adjust the parameters of the emotion recognition algorithm based on user feedback and behavior patterns to improve recognition accuracy. This feedback mechanism is the key to the system's continuous learning and optimization, ensuring that the system can adapt to user changes and provide more personalized services.

[0097] The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiment. Any equivalent modifications or changes made by ordinary technicians in this field based on the contents disclosed by the present invention should be included in the protection scope recorded in the claims.

Claims

1. An intelligent light adjustment system based on facial emotion and environment recognition technology, characterized in that: The system includes a central processing unit and a personalization module, a light control module, an emotion perception module, and an environment perception module respectively connected thereto; The emotion recognition module analyzes the user's physiological parameters and identifies the emotional state, while the environmental perception module monitors the external weather conditions in real time. The central processing unit calculates the optimal lighting parameters based on the environmental conditions and emotional state. The light control module then automatically adjusts the light settings and the opening and closing status of the curtains to optimize the indoor light based on the calculation results of the central processing unit. In this process, the user can view the current emotional state, light settings, and curtain status through the personalized customization interface and make personalized adjustments.

2. The intelligent light adjustment system based on facial emotion and environment recognition technology according to claim 1, characterized in that: The emotion recognition module adopts the facial emotion recognition network structure EmoRepLKNet based on the UniRepLKNet framework.

3. The intelligent light adjustment system based on facial emotion and environment recognition technology according to claim 2, characterized in that: In the emotion recognition module, the user's facial image data is captured in real time through the camera, and the collected raw data is preprocessed. The preprocessed data will then be input into EmoRepLKNet for emotion recognition. The facial image data is processed through a series of convolutional layers, RPS Blocks, and downsampling and upsampling operations in EmoRepLKNet to extract facial emotion features. The extracted features are classified using a fully connected layer to identify the user's emotional state. The identified emotional state will match the preset emotion and light setting correspondence, which is adjusted according to the user's personal preferences.

4. The intelligent light adjustment system based on facial emotion and environment recognition technology according to claim 1, characterized in that: The environmental perception module includes temperature and humidity sensors, light sensors and air pressure sensors, which monitor the external light intensity, temperature, humidity and weather conditions in real time, and transmit the perception data to the central processing module in real time through wireless communication protocols.

5. The intelligent light adjustment system based on facial emotion and environment recognition technology according to claim 4, characterized in that: In the environmental perception module, light intensity, temperature, and humidity are mapped to emotional results.

6. The intelligent light adjustment system based on facial emotion and environment recognition technology according to claim 1, characterized in that: The central processing unit is responsible for coordinating the work of each module and calculating the optimal lighting parameters based on real-time data; the central processing unit collects necessary data from the emotion recognition module and the environmental perception module, and at the same time extracts the user's historical lighting habits and preference settings from the historical database to better understand the user's needs.

7. The intelligent light adjustment system based on facial emotion and environment recognition technology according to claim 6, characterized in that: The calculation steps of lighting parameters in the central processing unit include: based on the comprehensive analysis results of emotions and environment, the central processing unit will calculate the optimal lighting parameters, including brightness, color temperature and curtain control; among them, the brightness adjustment calculation takes into account the ambient light intensity and the user's emotional state to determine the appropriate indoor light brightness; the color temperature adjustment calculation combines the ambient temperature and emotional state to adjust the warm and cool tones of the light; the curtain control calculation optimizes the use of natural light and controls the indoor light by adjusting the degree of opening and closing of the curtains; after calculating the optimal lighting parameters, the central processing unit will generate corresponding instructions and send them to the light control module.

8. The intelligent light adjustment system based on facial emotion and environment recognition technology according to claim 7, characterized in that: The central processing unit collects user feedback to evaluate satisfaction with the current lighting settings; users provide feedback through the user interface to express their satisfaction with the current lighting effects or make suggestions for adjustments; The system will learn based on this feedback and continuously optimize the emotion recognition algorithm and lighting adjustment strategy.

9. The intelligent light adjustment system based on facial emotion and environment recognition technology according to claim 1, characterized in that: The light control module adjusts the indoor light by adjusting the light brightness, color temperature and natural light intensity according to the instructions generated by the central processing unit.

Images