Dynamic eye-protection lighting control method and device based on physiological index sensing
By using multi-sensor fusion technology and deep neural network analysis, the lighting parameters of the LED reading lamp are dynamically adjusted, solving the problem that existing intelligent lighting systems cannot fully perceive user needs. This enables personalized and intelligent lighting control, improving users' visual comfort and health.
Patent Information
- Application Number
- CN202411870105.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-18
- Publication Date
- 2026-11-17
AI Technical Summary
Existing intelligent lighting systems cannot fully and accurately perceive users' real-time physiological state and behavioral characteristics. Their lighting adjustment strategies are simple and difficult to adapt to complex and ever-changing actual needs. They lack fine-grained control over the lighting area and fail to fully consider the impact of long-term use on visual health.
Employing multi-sensor fusion technology, the system collects the user's heart rate and breathing information through photoelectric pulse sensors and respiration sensors. Combined with camera recognition of reading behavior, it uses deep neural networks to analyze the user's fatigue state and concentration, dynamically adjusting the lighting parameters of the LED reading lamp to achieve regional brightness and color temperature control. It also uses an intelligent anti-flicker algorithm to eliminate flicker.
It enables personalized, real-time response to users' lighting needs, improves visual comfort, reduces eye fatigue, enhances work efficiency, saves energy, and protects users' vision health.
Smart Images

Figure FT_1 
Figure FT_2 
Figure FT_3
Abstract
Description
Technical Field
[0001] This invention relates to the field of lighting control technology, and more specifically, to an adaptive desk lamp lighting control method and system based on multi-sensor fusion. Background Technology
[0002] With the transformation of work and study methods in modern society, people increasingly need to read and work for extended periods under artificial lighting. This prolonged exposure to artificial light inevitably affects the human visual system, leading to eye fatigue, decreased vision, and other problems. Therefore, providing a healthy and comfortable lighting environment has become an important research topic in the field of lighting technology.
[0003] Traditional lighting technologies primarily focus on basic parameters such as the brightness and color temperature of the light source. With technological advancements, intelligent lighting products with adjustable brightness and color temperature have emerged on the market. These products typically allow users to manually adjust lighting parameters or automatically adjust the lighting environment according to a preset schedule. However, this approach ignores individual differences and real-time changes in needs, failing to truly meet the dynamic lighting requirements of users.
[0004] In recent years, some researchers have begun to explore the integration of physiological sensing technologies into lighting control systems. For example, some systems assess a user's fatigue level by detecting their heart rate or skin conductance and adjust lighting parameters accordingly. Other systems attempt to use eye-tracking technology to analyze a user's reading behavior, thereby providing more precise lighting control. However, these methods often focus on only a single physiological indicator or behavioral characteristic, making it difficult to comprehensively and accurately reflect the user's actual needs.
[0005] Furthermore, most existing smart lighting systems employ a uniform lighting adjustment strategy, neglecting the differences in lighting needs across different areas. For example, during reading, the user's focal point and surrounding areas have different lighting requirements. A uniform adjustment method struggles to accommodate these subtle differences, potentially leading to problems such as localized glare or insufficient illumination.
[0006] In summary, the existing technology has the following main problems: 1. Unable to fully and accurately perceive the user's real-time physiological state and behavioral characteristics; 2. The lighting adjustment strategy is too simplistic and cannot adapt to complex and ever-changing actual needs; 3. Lack of precise control over the lighting area makes it impossible to provide optimal local lighting effects; 4. The long-term effects on visual health were not adequately considered. Summary of the Invention
[0007] To address the aforementioned issues, this invention proposes a dynamic eye-protection lighting control method and device based on physiological indicator perception. This invention aims to achieve real-time, personalized adjustment of the lighting environment through multi-dimensional physiological indicator perception and behavioral analysis, combined with intelligent algorithms and refined lighting control, thereby providing a more comfortable and healthy lighting experience.
[0008] This invention provides an adaptive desk lamp lighting control method based on multi-sensor fusion, comprising: The acquisition steps include: The user's heart rate information is collected using a photoelectric pulse sensor; The user's breathing information is collected through a breathing sensor; Use a camera to identify the user's reading behavior; The processing steps include: Based on the heart rate information and the breathing information, a deep neural network is used to analyze the user's heart rate and breathing, obtain the user's fatigue state, and output a fatigue level value. Based on the reading behavior, the user's focus level is obtained and a focus level value is output; The LED reading lamp is dynamically controlled based on the fatigue level value and the concentration level value. Output steps, including: Based on the dynamic control results, adjust the lighting parameters of the LED reading lamp.
[0009] Preferably, the processing step of analyzing the user's heart rate and respiration using a deep neural network specifically includes: For different heart rate and respiratory characteristics, filtering and downsampling are used to preprocess the heart rate and respiratory information to make the heart rate and respiratory characteristic data have a unified format; A convolutional neural network is used to classify preprocessed heart rate and respiratory feature data to obtain the user's fatigue state; Based on the classified heart rate and respiratory status, vectors are used to represent the data and normalized to the [0,1] interval to obtain the quantified fatigue status value.
[0010] Preferably, the step of acquiring the user's focus level and outputting the focus level value specifically includes: By monitoring users in real time, information about their visual focus and the text content they are reading can be obtained; Based on the visual focus information and the reading text content information, a deep neural network is used to evaluate the user's level of focus and obtain the user's focus level.
[0011] Preferably, the dynamic control of the LED reading lamp specifically includes: Based on the relationship curve between user fatigue level and focus level, and combined with the real-time fatigue level F and real-time focus level C during reading, the dynamically changing thresholds FH and CH are obtained. The real-time fatigue level value F is compared with the real-time focus value C and the threshold FH and CH to obtain the final control threshold K; Based on the final control threshold K, and combined with the user's visual focus position while reading, the brightness and color temperature of different lighting areas are dynamically adjusted. An intelligent anti-flicker algorithm is used to eliminate LED flickering.
[0012] Preferably, during the acquisition of the heart rate signal, the photoelectric pulse sensor is attached to the tip of the user's index finger, and the sensor acquires data once every 10 seconds, with 100 data points; during the acquisition of the respiratory signal, the respiratory sensor is attached to the user's chest, and the sensor acquires data once every 1 second, with 100 data points.
[0013] Preferably, the acquisition of the user's visual focus information and reading text content information specifically includes: By recognizing eye images, locating the eye region, and calculating the location of the visual focus; Use NLP (Natural Language Processing) technology to extract text content from camera footage; The visual focus and the text content are combined to determine the relative position of the user and the text based on changes in the user's visual focus.
[0014] Preferably, obtaining the dynamically changing thresholds FH and CH specifically includes: Establish a sample database of fatigue level and focus level values, and train the sample data through machine learning to obtain the relationship curve between fatigue level and focus level values; By combining the current real-time fatigue level value F and the real-time focus value C, the fatigue level threshold FH and focus threshold CH are dynamically obtained.
[0015] Preferably, the process of obtaining the final control threshold K includes: The real-time fatigue level value F and the real-time focus value C are divided into three levels: mild fatigue and focus: (0, 0.33); slight fatigue and focus: (0.33, 0.66); fatigue and focus: (0.66, 1); The fatigue threshold FH and the focus threshold CH are divided into three levels: mild fatigue and focus: (0, 0.33); slight fatigue and focus: (0.33, 0.66); fatigue and focus: (0.66, 1); Combining the two threshold levels, the final fatigue threshold FH and focus threshold CH are obtained as follows: Combining FH and CH, we obtain the final control threshold K: .
[0016] Preferably, the dynamic adjustment of brightness and color temperature in different lighting areas specifically includes: Based on the user's visual focus position and the text center position, calculate the horizontal position difference VL and the vertical position difference VV between the two. Substitute the ratio VL / VV and the final control threshold K into the following formula: Among them, rand The value is a random number between 0 and 1, and e is the base of the natural logarithm. Based on the user's visual focus position, the RGB value corresponding to the area is dynamically adjusted according to the proportional coefficient, thereby adjusting the color temperature and brightness of that area on the reading lamp.
[0017] A dynamic eye-protection lighting control device based on physiological index sensing, the device comprising: Photoelectric pulse sensor and respiration sensor used to monitor users' physiological indicators; A camera used to identify a user's reading behavior; A controller for adjusting the lighting of an LED desk lamp; LED reading desk lamp for eye-friendly lighting; The device employs the dynamic eye-protection lighting control method based on physiological index perception.
[0018] The beneficial effects of this invention are mainly reflected in the following aspects: The method of this invention comprehensively analyzes multiple physiological and behavioral indicators such as the user's heart rate, respiration, and gaze focus, enabling a more comprehensive and accurate assessment of the user's real-time state and needs. Based on this information, this invention employs an innovative dynamic threshold control algorithm that can flexibly adjust lighting parameters according to the user's fatigue level and concentration. Furthermore, this invention introduces regionalized precise lighting control, which can dynamically adjust the brightness and color temperature of different areas based on the user's gaze focus position, providing optimal local lighting effects.
[0019] This multi-dimensional perception and refined control method not only significantly improves user visual comfort but also effectively reduces eye fatigue and enhances work efficiency. In the long term, the method of this invention helps protect users' visual health and reduces vision problems caused by improper lighting. Simultaneously, the intelligent control strategy of this invention also achieves better energy-saving effects, reducing energy consumption while providing a high-quality lighting experience.
[0020] In summary, this invention, through its innovative technical solution, achieves personalized, intelligent, and health-oriented lighting control, providing a novel approach and method for addressing existing problems in lighting technology. This dynamic eye-protection lighting control method based on physiological indicator perception is not only technologically groundbreaking but also has broad application prospects, expected to have a positive impact on multiple fields such as office, education, and healthcare, creating a more comfortable and healthy visual environment for users. Attached Figure Description
[0021] Figure 1 This is a flowchart of the method of the present invention.
[0022] Figure 2 This is a logic block diagram of the deep neural network analysis of the present invention.
[0023] Figure 3 This is a logic block diagram of the attention analysis of the present invention.
[0024] Figure 4 This is a logic block diagram of the dynamic control of the present invention. Detailed Implementation
[0025] Please refer to Figure 1-4 This invention provides a dynamic eye-protection lighting control method and device based on physiological indicator perception. This method can monitor the user's physiological state and reading behavior in real time and dynamically adjust lighting parameters accordingly to provide a more comfortable and eye-protection reading environment. The invention will now be described in detail with reference to specific embodiments.
[0026] First, the method of this invention includes an acquisition step, a processing step, and an output step. In the acquisition step, this method collects the user's heart rate information through a photoelectric pulse sensor 1, collects the user's breathing information through a respiratory sensor 2, and identifies the user's reading behavior through a camera 3. The use of these sensors enables this method to comprehensively acquire the user's physiological state and behavioral information, providing a rich data foundation for subsequent analysis and control.
[0027] In the processing steps, this method first analyzes the user's heart rate and respiration using a deep neural network based on the acquired heart rate and respiration information. The deep neural network model used here has been trained on a large number of samples and can effectively extract features related to fatigue state from physiological data. Through this analysis, this method can obtain the user's fatigue state and output a fatigue level value. Preferably, this fatigue level value is represented by a value between 0 and 1, where 0 represents the least fatigued state and 1 represents the most fatigued state.
[0028] Next, this method acquires the user's attention level and outputs an attention value based on the user's reading behavior detected by the camera. In this step, the invention employs a combination of computer vision and natural language processing techniques. Specifically, by analyzing the user's eye movement trajectory, blink frequency, and other visual features, combined with semantic analysis of the text content being read, this method can accurately assess the user's level of attention. Similarly, the attention value is represented by a number between 0 and 1, where 1 represents the most focused state and 0 represents the least focused state.
[0029] After obtaining the fatigue level and concentration level values, this method dynamically controls the LED reading lamp 4 based on these two values. This control process employs an innovative dynamic threshold control algorithm, which can flexibly adjust the lighting parameters according to the user's real-time state. Specifically, this method first obtains dynamically changing thresholds FH and CH based on the relationship curve between the user's fatigue level and concentration level values, combined with the user's real-time fatigue level value F and real-time concentration level value C during reading. The relationship curve here is derived from statistical analysis of a large amount of user data and can effectively reflect the mutual influence between fatigue level and concentration level.
[0030] Then, this method compares the real-time fatigue level value F with the real-time focus value C and the threshold FH with CH to obtain the final control threshold K.
[0031] Based on the obtained final control threshold K, this method dynamically adjusts the brightness and color temperature of different lighting areas by considering the user's visual focus position while reading. This precise, localized control is a key feature of this invention, providing users with a more comfortable and personalized lighting environment. For example, when the user's gaze is focused on a certain area, the brightness of that area will be appropriately increased, while the brightness of other areas will be reduced accordingly to minimize eye strain.
[0032] Furthermore, this method employs an intelligent anti-flicker algorithm to eliminate LED flicker. This algorithm monitors the LED's output waveform in real time and automatically adjusts the parameters of the drive circuit when it detects flicker that may cause visual fatigue, thereby achieving a smooth light output. This feature is especially important for users who read for extended periods, as it effectively reduces eye strain.
[0033] In the output step, this method adjusts the lighting parameters of the LED reading lamp based on the results of the dynamic control described above. These parameters mainly include brightness, color temperature, and light distribution. It is worth noting that this adjustment is gradual, avoiding sudden changes in light that could cause discomfort to the user.
[0034] The method of the present invention, through the above steps, realizes intelligent and personalized adjustment of the reading environment, which can effectively improve the user's reading comfort, reduce visual fatigue, and is of great significance for protecting the user's eye health.
[0035] Next, some key steps in the method of the present invention will be described in more detail.
[0036] In analyzing a user's heart rate and respiration using deep neural networks, this method first preprocesses different heart rate and respiration features. This step includes filtering and downsampling, the purpose of which is to remove noise from the signal and to give the heart rate and respiration feature data a uniform format for easier subsequent analysis and processing.
[0037] Specifically, for heart rate signals, this method uses a bandpass filter with a filtering range of 0.5Hz to 3Hz, which covers the frequency range of normal human heart rates. For respiratory signals, a low-pass filter with a cutoff frequency of 0.5Hz is used to remove high-frequency noise. During downsampling, this method uniformly adjusts the sampling rate of both heart rate and respiratory signals to 100Hz. This sampling rate ensures signal integrity while effectively reducing data volume and improving processing efficiency.
[0038] The preprocessed data is then fed into a convolutional neural network for classification. This network consists of three convolutional layers, three pooling layers, and two fully connected layers. Each convolutional layer uses 32 3x3 convolutional kernels with ReLU activation. The pooling layers use 2x2 max pooling. The final fully connected layer outputs a value between 0 and 1, representing the user's fatigue level.
[0039] The network was trained using a large amount of labeled data, which came from user tests in a laboratory environment. Each user's heart rate and respiratory data under different fatigue states were recorded and labeled. The training process used the cross-entropy loss function and the Adam optimizer, with a learning rate of 0.001, a batch size of 64, and 100 training epochs.
[0040] In this way, this method can infer a user's fatigue level relatively accurately from heart rate and respiratory data. For example, when a user's heart rate variability decreases and their respiratory rate slows down, it usually indicates a higher level of fatigue.
[0041] To acquire user attention, this method primarily analyzes the user's visual focus information and the content of the text being read. First, the user's visual focus information is obtained through real-time monitoring. In this process, the method uses a deep learning-based eye-tracking algorithm, which can locate the user's eye position and gaze direction in real time at a frame rate of 30fps with an accuracy of 0.5°.
[0042] Simultaneously, this method also acquires information about the text content the user is reading. It employs a combination of Optical Character Recognition (OCR) and Natural Language Processing (NLP) techniques. OCR is used to extract text content from images captured by the camera, while NLP is used to analyze the semantic complexity and information density of this text.
[0043] Based on the acquired visual focus information and text content information, this method uses a deep neural network to evaluate the user's level of focus. The network's input includes the user's gaze movement trajectory, dwell time, and semantic features of the text, and the output is a numerical value between 0 and 1, representing the user's level of focus.
[0044] For example, when a user's gaze moves smoothly across the text at a speed that matches the text's complexity, it usually indicates a high level of concentration. Conversely, if a user's gaze jumps frequently or lingers on non-text areas for extended periods, it may indicate a lower level of concentration.
[0045] Through the above method, this invention can comprehensively and accurately perceive the user's physiological state and reading behavior, providing a reliable basis for subsequent lighting control. This perception method based on multi-dimensional physiological indicators, compared to traditional fixed lighting or simple brightness adjustment, can more accurately meet the user's lighting needs in different states, thereby effectively improving reading comfort and reducing visual fatigue. Next, this invention will describe in detail the specific process of dynamically controlling the LED reading lamp. In a preferred embodiment of this invention, the dynamic control process mainly includes the following steps: First, this method derives dynamically changing thresholds FH and CH based on the relationship curve between user fatigue level and focus level, combined with the user's real-time fatigue level F and real-time focus level C during reading. The relationship curve is derived through statistical analysis of a large amount of user data and effectively reflects the mutual influence between fatigue level and focus level. Specifically, this method employs an adaptive threshold algorithm, which can be expressed as: Where F represents the real-time fatigue level and C represents the real-time focus level. The purpose of this algorithm design is to ensure the lighting effect while avoiding excessively frequent changes in light that could disturb the user.
[0046] Next, this method compares the real-time fatigue level value F with the real-time focus value C and the threshold FH with CH to obtain the final control threshold K. This process can be expressed by the following formula: This formula is designed to take into account the combined effects of fatigue and concentration on lighting needs, achieving a good balance between the two.
[0047] After obtaining the final control threshold K, this method dynamically adjusts the brightness and color temperature of different lighting areas based on the user's visual focus position while reading. This precise, localized control is a key feature of this invention, providing users with a more comfortable and personalized lighting environment.
[0048] Specifically, this method first calculates the horizontal position difference VL and the vertical position difference VV between the user's visual focus and the text center. Then, it combines the ratio of these two values, VL / VV, with the final control threshold K to obtain a lighting adjustment coefficient. This process can be expressed by the following formula: Here, rand() is a random number between 0 and 1, and e is the base of the natural logarithm. This formula introduces a degree of randomness to avoid overly mechanical lighting changes and to provide users with a more natural lighting experience.
[0049] Based on the adjustment coefficients calculated above, this method dynamically adjusts the RGB values of each area of the LED reading lamp, thereby achieving precise control over color temperature and brightness. For example, when the user's gaze is focused on a certain area, the lighting brightness of that area will be appropriately increased, and the color temperature will be slightly decreased (warmed) to provide a more comfortable reading environment.
[0050] Furthermore, the method of this invention also employs an intelligent anti-flicker algorithm to eliminate LED flicker. This algorithm monitors the output waveform of the LED in real time, and automatically adjusts the parameters of the driving circuit when flicker that may cause visual fatigue is detected, thereby achieving a smooth light output effect. Specifically, this method uses high-frequency driving technology based on PWM (Pulse Width Modulation) to increase the driving frequency of the LED to over 20kHz, far exceeding the frequency range perceptible to the human eye, thus effectively eliminating visible flicker.
[0051] In another embodiment of the invention, to collect user physiological data more accurately, the method optimizes the acquisition process of heart rate and respiratory signals. Specifically, during the heart rate signal acquisition process, a photoelectric pulse sensor is attached to the tip of the user's index finger, and the sensor collects data every 10 seconds, collecting 100 data points each time. This acquisition method can effectively capture subtle changes in heart rate without interfering with the user's reading behavior.
[0052] For respiratory signal acquisition, this method involves attaching a respiratory sensor to the user's chest, with the sensor acquiring data once per second, resulting in 100 data points. This high-frequency sampling accurately records the user's breathing patterns, including important indicators such as respiratory rate, depth, and regularity.
[0053] The method of this invention has also been significantly optimized in terms of acquiring the user's visual focus information and reading text content information. First, by recognizing eye images, the eye region is located, and the location of the visual focus is calculated. Here, a deep learning-based eye-tracking algorithm is used. This algorithm can accurately locate the user's gaze direction under various lighting conditions by analyzing features such as the shape of the eyeball, the position of the pupil, and corneal reflection.
[0054] Furthermore, this method also utilizes advanced NLP (Natural Language Processing) technology to extract reading text content from the camera image. This process first converts the text in the image into actual text using Optical Character Recognition (OCR) technology, and then uses a natural language processing model to analyze the semantic complexity, topic content, and other features of the text. Combining this information with the user's visual focus information allows for a more comprehensive assessment of the user's reading status and level of concentration.
[0055] After obtaining information about visual focus and the content of the text being read, this method combines and analyzes these two pieces of information to determine the user's relative position to the text based on changes in the user's visual focus. Specifically, when the visual focus is at the center of the text area, it is determined that the user is focused on reading; when the visual focus is between the center of the text area and the user's eye level, it is determined that the user is scanning the text. This detailed behavioral analysis can provide a more accurate basis for lighting control.
[0056] Through the above optimizations and improvements, the method of this invention can more accurately perceive the user's physiological state and reading behavior, thereby providing more personalized and intelligent lighting control, effectively improving the user's reading experience, reducing visual fatigue, and protecting eye health. In a preferred embodiment of this invention, in order to more accurately determine the dynamically changing thresholds FH and CH, this method employs an adaptive algorithm based on machine learning. Specifically, a sample database of fatigue level values and focus level values is first established. This database contains a large amount of physiological indicator data of users under different fatigue and focus states, as well as the corresponding optimal lighting parameters.
[0057] By training these sample data using machine learning, this method yields a model that accurately describes the relationship between fatigue level and focus level. This model employs the random forest algorithm, which effectively handles non-linear relationships and multi-dimensional features. During training, cross-validation was used to avoid overfitting and ensure the model's generalization ability.
[0058] Preferably, in practical applications, this method combines the current real-time fatigue level value F and the real-time focus level value C to dynamically obtain the fatigue level threshold FH and the focus level threshold CH. This dynamic adjustment mechanism enables the system to respond quickly based on the user's real-time status, providing more personalized lighting control.
[0059] After obtaining the dynamic thresholds FH and CH, the method of this invention further refines the calculation process of the final control threshold K. Specifically, this method divides the real-time fatigue level value F and the real-time focus value C into three levels: 1. Mild fatigue and focus: (0, 0.33); 2. Mild fatigue and focus: (0.33, 0.66); 3. Fatigue and focus: (0.66, 1); Similarly, the fatigue threshold (FH) and focus threshold (CH) are also divided into three corresponding levels. This classification method allows the system to more finely distinguish the user's state, thereby providing more precise lighting control.
[0060] Based on the above classification, this method calculates the final fatigue threshold FH and focus threshold CH using the following formulas: Where F represents the real-time fatigue level and C represents the real-time focus level. These two formulas are designed to balance fatigue and focus, ensuring appropriate lighting parameters are provided under various conditions. Finally, this method combines FH and CH to obtain the final control threshold K. This formula further optimizes the precision of lighting control, enabling the system to provide gentler lighting when the user is more fatigued and brighter lighting when the user is more focused. In practical applications, the method of this invention also refines the brightness and color temperature adjustments for different lighting areas. First, based on the user's visual focus position and the text center position, the horizontal position difference VL and the vertical position difference VV between the two are calculated. Then, the ratio VL / N and the final control threshold K are substituted into the following formula. Among them, rand The formula introduces a degree of randomness, where is a random number between 0 and 1, and e is the base of the natural logarithm. This aims to prevent lighting changes from becoming too mechanical and to provide users with a more natural lighting experience.
[0061] Based on the calculated ratios, this method dynamically adjusts the RGB values of each area of the LED reading lamp. For example, when the VL / VV ratio is large, it indicates that the user's gaze is far from the center of the text. The system will appropriately reduce the brightness of the surrounding areas and increase the brightness of the central area to guide the user's gaze back to the center of the text. Simultaneously, the system will also adjust the color temperature according to the final control threshold K, providing warmer light when the user is fatigued and cooler light when the user is focused.
[0062] The method of this invention, through this refined area control, can not only provide a more comfortable reading environment, but also guide the user's reading behavior to a certain extent and improve reading efficiency.
[0063] Finally, the present invention also provides a dynamic eye-protection lighting control device based on physiological indicator perception. This device includes a photoelectric pulse sensor 1 and a respiration sensor 2 for monitoring the user's physiological indicators, a camera 3 for recognizing the user's reading behavior, a controller for adjusting the LED desk lamp lighting, and an LED reading desk lamp 4 for eye-protection lighting.
[0064] In this device, the photoelectric pulse sensor 1 is preferably a reflective photoplethysmography (PPG) sensor, which can measure heart rate by detecting changes in blood flow on the skin surface. The respiration sensor 2 is a piezoelectric sensor, which can measure respiratory rate by detecting minute deformations in the chest cavity. Both sensors are characterized by high sensitivity and low power consumption, making them suitable for long-term continuous monitoring.
[0065] Camera 3 employs a high-resolution CMOS sensor, combined with a wide-angle lens, enabling it to simultaneously capture the user's facial expressions and the content of the reading material. The controller utilizes a high-performance microprocessor, capable of processing data from various sensors in real time and rapidly calculating the optimal lighting parameters according to the algorithm of this invention.
[0066] The LED reading lamp 4 features a multi-zone independent control design, with each zone equipped with an LED light source that allows for adjustable color temperature and brightness. This design enables the device to precisely control the lighting in different zones based on the user's reading behavior and physiological state, providing an optimal reading environment.
[0067] In summary, the method and apparatus of this invention, by combining advanced sensing technology, artificial intelligence algorithms, and refined lighting control, achieve dynamic eye-protection lighting based on physiological indicators. This not only significantly improves user reading comfort and reduces visual fatigue, but also improves users' reading habits to a certain extent, which is of great significance for protecting eye health.
[0068] To verify the superiority of the dynamic eye-protection lighting control method and device based on physiological index perception of the present invention, a series of comparative experiments were designed. The embodiments employed the method of the present invention, while the comparative examples used a traditional fixed lighting scheme. The experiments were conducted in a simulated office environment for four weeks, with participants working eight hours a day. A total of 30 volunteers, aged 25-45, participated in the experiment; all were regular office workers.
[0069] In this embodiment, the dynamic eye-protection lighting control system of the present invention is employed. The system includes a photoelectric pulse sensor, a respiration sensor, a high-definition camera, an intelligent controller, and an adjustable LED reading lamp. The system dynamically adjusts the lighting parameters based on real-time collected user physiological indicators and reading behavior.
[0070] In the comparison, a traditional fixed lighting scheme was used. This scheme uses standard office LED lighting with a fixed color temperature of 4000K and a fixed brightness of 500 lux, without any dynamic adjustment.
[0071] The following key metrics were selected for testing: 1. Visual Comfort (VCL): A subjective rating method is used, ranging from 1 to 10 points. The higher the score, the higher the comfort level.
[0072] 2. Eye Fatigue Index (EFI): Calculated by measuring the amplitude and speed of accommodation, ranging from 0 to 100. The lower the score, the less fatigued the patient.
[0073] 3. Work Efficiency (WE): Measured through standardized reading and data processing tasks, it is calculated by combining the speed and accuracy of task completion, with a baseline value of 100.
[0074] 4. Energy Consumption (EC): Records the actual electricity consumption of the lighting system, in kWh.
[0075] The test results are shown in the table below: index Example Comparative Example Increase ratio VCL 8.7 6.2 40.3% EFI 28 62 54.8% WE 118 103 14.6% EC 0.42 0.56 25.0% The test results show that the method of the present invention has significant advantages in all indicators: Visual comfort (VCL) improved by 40.3%, demonstrating that the dynamic dimming system of this invention can better adapt to user needs and provide a more comfortable reading environment. Many subjects reported that the dynamically adjusted light felt natural and non-glaring, especially maintaining a high level of comfort even after long periods of work.
[0076] The eye fatigue index (EFI) decreased by 54.8%, a very significant improvement. A lower EFI means that the user's eyes can maintain good accommodation ability even after prolonged work. This is likely because the system of this invention can adjust the light in a timely manner according to the user's fatigue level, reducing excessive eye strain.
[0077] Work efficiency (WE) improved by 14.6%, a considerable figure. This improvement likely stems from a better visual environment that reduces eye strain, allowing users to focus more intently. Additionally, dynamically adjusted lighting may have helped users maintain better alertness.
[0078] Energy consumption (EC) was reduced by 25.0%, indicating that the present invention not only improves the user experience but also has significant energy-saving effects. This energy saving may stem from the system's ability to adjust light intensity according to actual needs, avoiding unnecessary energy waste.
[0079] A comprehensive analysis of this data leads to the conclusion that the dynamic eye-protection lighting control method and device based on physiological indicator perception of the present invention demonstrates significant advantages in improving visual comfort, reducing eye fatigue, enhancing work efficiency, and saving energy. These improvements not only enhance the user's work experience and efficiency in the short term but may also have a positive impact on the user's eye health in the long run.
[0080] It is worth noting that the advantages of this invention may become more apparent after prolonged use. Over time, user fatigue may accumulate under traditional fixed lighting schemes, while the dynamic adjustment system of this invention can continuously provide the most suitable lighting environment, potentially mitigating the accumulation of fatigue.
[0081] Furthermore, the system of this invention exhibits excellent adaptability. During experiments, it was observed that the system can be personalized according to the physiological characteristics and work habits of different users. This personalized lighting solution may be the key to the superiority of this invention over traditional solutions.
[0082] In summary, these experimental results strongly support the superiority of this invention and demonstrate the enormous potential of the dynamic eye-protection lighting control method and device based on physiological indicator perception in practical applications. In the future, this intelligent lighting system is expected to be widely used in various fields such as offices, education, and healthcare, creating a healthier and more comfortable visual environment for users.
[0083] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A dynamic eye-protection lighting control method based on physiological indicator perception, characterized in that, include: The acquisition steps include: The user's heart rate information is collected using a photoelectric pulse sensor; The user's breathing information is collected through a breathing sensor; Use a camera to identify the user's reading behavior; The processing steps include: Based on the heart rate information and the breathing information, a deep neural network is used to analyze the user's heart rate and breathing, obtain the user's fatigue state, and output a fatigue level value. Based on the reading behavior, the user's focus level is obtained and a focus level value is output; The LED reading lamp is dynamically controlled based on the fatigue level value and the concentration level value. Output steps, including: Based on the dynamic control results, adjust the lighting parameters of the LED reading lamp.
2. The dynamic eye-protection lighting control method based on physiological index perception according to claim 1, characterized in that, The processing steps that utilize deep neural networks to analyze the user's heart rate and respiration specifically include: For different heart rate and respiratory characteristics, filtering and downsampling are used to preprocess the heart rate and respiratory information to make the heart rate and respiratory characteristic data have a unified format; A convolutional neural network is used to classify preprocessed heart rate and respiratory feature data to obtain the user's fatigue state; Based on the classified heart rate and respiratory status, vectors are used to represent the data and normalized to the [0,1] interval to obtain the quantified fatigue status value.
3. The dynamic eye-protection lighting control method based on physiological index perception according to claim 1, characterized in that, The process of acquiring the user's focus level and outputting the focus level value specifically includes: By monitoring users in real time, information about their visual focus and the text content they are reading can be obtained; Based on the visual focus information and the reading text content information, a deep neural network is used to evaluate the user's level of focus and obtain the user's focus level.
4. The dynamic eye-protection lighting control method based on physiological index perception according to claim 1, characterized in that, The dynamic control of the LED reading lamp specifically includes: Based on the relationship curve between user fatigue level and concentration level, and combined with the real-time fatigue level F and real-time concentration level C during reading, the dynamically changing thresholds FH and CH are obtained. The real-time fatigue level value F is compared with the real-time focus value C and the threshold FH and CH to obtain the final control threshold K; Based on the final control threshold K, and combined with the user's visual focus position while reading, the brightness and color temperature of different lighting areas are dynamically adjusted. An intelligent anti-flicker algorithm is used to eliminate LED flickering.
5. The dynamic eye-protection lighting control method based on physiological index perception according to claim 2, characterized in that, During the acquisition of the heart rate signal, the photoelectric pulse sensor is attached to the tip of the user's index finger. The sensor acquires data once every 10 seconds, with 100 data points. During the acquisition of the respiratory signal, the respiratory sensor is attached to the user's chest. The sensor acquires data once every 1 second, with 100 data points.
6. The dynamic eye-protection lighting control method based on physiological index perception according to claim 3, characterized in that, The acquisition of the user's visual focus information and reading text content information specifically includes: By recognizing eye images, the eye region is located, and the location of the visual focus is calculated; Use NLP (Natural Language Processing) technology to extract text content from camera footage; The visual focus and the text content are combined to determine the relative position of the user and the text based on changes in the user's visual focus.
7. The dynamic eye-protection lighting control method based on physiological index perception according to claim 4, characterized in that, The dynamically changing thresholds FH and CH specifically include: Establish a sample database of fatigue level and focus level values, and train the sample data through machine learning to obtain the relationship curve between fatigue level and focus level values; By combining the current real-time fatigue level value F and the real-time focus value C, the fatigue level threshold FH and focus threshold CH are dynamically obtained.
8. The dynamic eye-protection lighting control method based on physiological index perception according to claim 7, characterized in that, The process of obtaining the final control threshold K includes: The real-time fatigue level value F and the real-time focus value C are divided into three levels: mild fatigue and focus: (0, 0.33); slight fatigue and focus: (0.33, 0.66); fatigue and focus: (0.66, 1). The fatigue threshold FH and the focus threshold CH are divided into three levels: mild fatigue and focus: (0, 0.33); slight fatigue and focus: (0.33, 0.66); fatigue and focus: (0.66, 1). Combining the two threshold levels, the final fatigue threshold FH and focus threshold CH are obtained as follows: Combining FH and CH, we obtain the final control threshold K: 。 9. The dynamic eye-protection lighting control method based on physiological index perception according to claim 4, characterized in that, The dynamic adjustment of brightness and color temperature in different lighting areas specifically includes: Based on the user's visual focus position and the text center position, calculate the horizontal position difference VL and the vertical position difference VV between the two. Substitute the ratio VL / VV and the final control threshold K into the following formula: Among them, rand The value is a random number between 0 and 1, and e is the base of the natural logarithm. Based on the user's visual focus position, the RGB value corresponding to the area is dynamically adjusted according to the proportional coefficient, thereby adjusting the color temperature and brightness of that area on the reading lamp.
10. A dynamic eye-protection lighting control device based on physiological indicator perception, characterized in that, The device includes: Photoelectric pulse sensor and respiration sensor used to monitor users' physiological indicators; A camera used to identify a user's reading behavior; A controller for adjusting the lighting of an LED desk lamp; LED reading desk lamp for eye-friendly lighting; The device employs the dynamic eye-protection lighting control method based on physiological index perception as described in any one of claims 1-9.
