Adaptive table lamp illumination control method and system based on multi-sensor fusion

Abstract

The invention relates to the technical field of table lamp control, in particular to a self-adaptive table lamp illumination control method and system based on multi-sensor fusion, and the method comprises the steps: obtaining user eye movement data, ambient light data and user body existence data; based on the user eye movement data, user fatigue detection is carried out, and a user fatigue level is determined; adjusting the color temperature and brightness of the table lamp according to the fatigue level of the user; determining an ambient light brightness level based on the ambient light data; adjusting the color temperature of the table lamp according to the ambient light brightness level; judging whether the user is within the irradiation range of the table lamp based on the user body existence data; according to a result of the processing step, adjusting parameters of the table lamp are output, self-adaptive control over table lamp illumination is achieved, comprehensive and accurate perception of ambient light conditions and user states is achieved through the multi-sensor fusion technology, and due to multi-dimensional data collection and fusion, the accuracy of ambient light detection is improved, and the user experience is improved. And the fatigue degree and eye using habits of the user can be monitored in real time.

Description

Technical Field

[0001] The present invention relates to the field of desk lamp control technology, and more specifically, to an adaptive desk lamp lighting control method and system based on multi-sensor fusion. Background Art

[0002] As people's requirements for the quality of their working and living environments continue to increase, smart lighting technology has developed rapidly in recent years. Especially in office and study scenarios, desk lamps are one of the most commonly used lighting equipment, and their intelligence and personalization levels directly affect users' visual comfort and work efficiency. At present, some smart desk lamps with adaptive functions have appeared on the market, but these products still have many shortcomings in practical applications.

[0003] Existing adaptive desk lamp lighting control methods usually use a single ambient light sensor to detect light changes and adjust the light brightness and color temperature through preset thresholds. Although this method can respond to environmental changes to a certain extent, it is difficult to meet the personalized needs of users. For example, some users may require different lighting parameters under the same ambient light conditions, and existing technologies cannot effectively identify and adapt to such differences. In addition, the limitations of a single sensor also lead to inaccurate ambient light detection, especially in complex light environments, which can easily cause frequent fluctuations in lighting parameters and affect user experience.

[0004] On the other hand, existing technologies also have obvious deficiencies in user status perception. Most smart desk lamps cannot monitor the user's fatigue level and eye habits in real time, and therefore cannot dynamically adjust lighting parameters according to the user's actual needs. This not only affects the comfort of lighting, but may also cause health problems caused by long-term eye use. At the same time, the lack of accurate user presence detection function also causes unnecessary energy waste, which is inconsistent with the energy-saving and environmental protection concept pursued by today's society.

[0005] In addition, most of the existing adaptive lighting control algorithms are simple linear control or fixed rule control, lacking learning and optimization capabilities. This results in the system being unable to self-adjust and optimize based on the user's long-term usage habits, making it difficult to provide a truly personalized lighting experience. At the same time, these algorithms often exhibit problems such as slow response and unstable adjustment when faced with complex and changing environments and user needs, affecting the user experience.

[0006] In view of these problems existing in the prior art, there is an urgent need for an adaptive desk lamp lighting control method and system that can fully perceive the environment and user status, has intelligent learning capabilities, and can provide a personalized lighting experience. The present invention is an innovative solution proposed to address this technical requirement. Summary of the invention

[0007] The technical problem to be solved by the present invention is how to achieve high-precision environment and user status perception based on multi-sensor fusion, and realize fast, accurate and personalized lighting parameter adjustment through intelligent algorithms, so as to provide optimal visual comfort and user experience while taking into account energy efficiency.

[0008] The present invention provides an adaptive desk lamp lighting control method based on multi-sensor fusion, comprising:

[0009] The acquisition steps include:

[0010] Obtain user eye movement data, ambient light data, and user body presence data;

[0011] Processing steps include:

[0012] Based on the user eye movement data, perform user fatigue detection to determine the user fatigue level;

[0013] According to the user's fatigue level, adjust the color temperature and brightness of the desk lamp;

[0014] Based on the ambient light data, determining an ambient light brightness level;

[0015] According to the ambient light brightness level, adjusting the color temperature of the desk lamp;

[0016] Based on the user's body presence data, determining whether the user is within the illumination range of the desk lamp;

[0017] Output steps include:

[0018] According to the result of the processing step, the adjustment parameters of the desk lamp are output to realize adaptive control of the desk lamp lighting.

[0019] Preferably, the user fatigue detection specifically includes:

[0020] Obtain eye movement data;

[0021] Determining blink frequency and fixation duration based on the eye movement data;

[0022] The user fatigue level is calculated according to the blinking frequency and the gaze duration.

[0023] Preferably, the user fatigue detection further comprises:

[0024] Acquire eye movement data through the camera module;

[0025] Based on the eye movement data, tracking the human eyes in real time through a camera;

[0026] The gaze direction and position are calculated according to the multiple frames of face images taken by the camera module.

[0027] Preferably, determining the ambient light brightness level based on the ambient light data specifically includes:

[0028] Get ambient light data;

[0029] Determine the ambient light brightness level based on the collected ambient light brightness;

[0030] The color temperature is adjusted according to the ambient light brightness level.

[0031] Preferably, the determining whether the user is within the illumination range of the desk lamp based on the user's body presence data specifically includes:

[0032] Obtain user physical presence data;

[0033] Determine whether the user's body is within the effective illumination range of the desk lamp;

[0034] If not, it will automatically adjust to a lighting mode with a low color temperature.

[0035] As a preference, it also includes:

[0036] Perform time series processing on the acquired eye movement data;

[0037] Data feature extraction is performed based on the eye movement data after time series processing.

[0038] Preferably, the timing processing includes:

[0039] Time-sort the acquired real-time eye movement data;

[0040] Process data in chronological order;

[0041] Data feature extraction is performed based on time-series eye movement data, including obtaining single eye movement time, eye open time ratio and blinking frequency.

[0042] Preferably, the user fatigue level includes:

[0043] Fatigue level S1, fatigue level S2 and fatigue level S3;

[0044] Wherein, the fatigue level S1, fatigue level S2 and fatigue level S3 correspond to the preset brightness, the first color temperature and the second color temperature respectively;

[0045] The preset brightness range is 0-10lux, the first color temperature range is 2500K-3200K, the second color temperature range is 1800K-2500K, the second color temperature is lower than the first color temperature, and the first color temperature is higher than the preset brightness.

[0046] Preferably, the method for obtaining ambient light data includes:

[0047] Detect changes in ambient light through light sensors;

[0048] The brightness data of the ambient light is collected to form ambient light data.

[0049] An adaptive desk lamp lighting control system based on multi-sensor fusion for executing the method comprises:

[0050] A user reading module, used to obtain user eye movement data, ambient light data, and user body presence data;

[0051] An eye movement control module is used to detect user fatigue based on the acquired eye movement data and then adjust the color temperature and brightness;

[0052] An ambient light control module, used to determine and adjust the color temperature according to ambient light data;

[0053] A status monitoring module is used to determine whether the user's body is within the illumination range of the desk lamp based on the user's body presence data, and adjust the color temperature and brightness based on the determination result;

[0054] Adaptive dimming module, used to determine the user's fatigue level and ambient brightness level based on the acquired data, and adjust the color temperature and brightness according to the user's fatigue level and ambient brightness level

[0055] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0056] First, the present invention uses multi-sensor fusion technology to achieve comprehensive and accurate perception of ambient light conditions and user status. This multi-dimensional data collection and fusion not only improves the accuracy of ambient light detection, but also monitors the user's fatigue level and eye habits in real time. This lays a solid data foundation for subsequent intelligent control, making lighting adjustment more accurate and personalized.

[0057] Secondly, the LSTM-Attention model used in the present invention has powerful learning and optimization capabilities. This advanced algorithm can learn personalized lighting preferences from users' long-term usage data and continuously optimize control strategies. This means that the system can become more and more intelligent as the usage time increases, and more and more in line with the personalized needs of users. This adaptive learning ability is unmatched by traditional fixed rule control.

[0058] Furthermore, the method of the present invention performs well in terms of response speed and adjustment stability. Through efficient data processing and decision-making algorithms, the system can respond quickly when the ambient light changes suddenly, avoiding the delay and oscillation problems common in traditional methods. This not only improves the user experience, but also protects the user's vision health in time when the light changes suddenly.

[0059] In addition, the present invention has also achieved remarkable results in energy efficiency. Through accurate user presence detection and intelligent lighting parameter control, the system can maximize energy conservation while ensuring the lighting effect. This not only reduces the user's electricity cost, but also conforms to the energy-saving and environmental protection concept of today's society.

[0060] Most importantly, the present invention greatly improves the user's visual comfort and work efficiency. By real-time monitoring of the user's fatigue level and adjusting the lighting parameters accordingly, the system can effectively relieve the fatigue caused by long-term eye use and reduce the degree of eye fatigue. This is undoubtedly an important health protection measure for users who need to perform visual work for a long time.

[0061] In summary, the present invention not only solves many problems existing in the existing adaptive desk lamp lighting control through the organic combination of innovative technologies such as multi-sensor fusion, intelligent algorithms and personalized learning, but also achieves remarkable results in improving visual comfort, energy saving and environmental protection, personalized experience, etc. This comprehensive technological innovation has opened up a new development direction for the field of intelligent lighting and is expected to play an important role in future smart home and office environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 The present invention is a flow chart of the method.

[0063] Figure 2 The figure is a logic block diagram of the user reading module of the present invention.

[0064] Figure 3 It is a logic block diagram of the eye movement control module of the present invention.

[0065] Figure 4 It is a logic block diagram of the ambient light control module of the present invention.

[0066] Figure 5 It is a logic block diagram of the status monitoring module of the present invention.

[0067] Figure 6 It is a logic block diagram of the adaptive dimming module of the present invention. DETAILED DESCRIPTION

[0068] Please refer to Figure 1-6 The present invention provides an adaptive desk lamp lighting control method and system based on multi-sensor fusion. The method realizes intelligent and personalized desk lamp lighting control by fusing multiple sensor data, providing users with a more comfortable and healthy lighting environment.

[0069] First, the method of the present invention includes an acquisition step, a processing step and an output step. In the acquisition step, the system acquires user eye movement data, ambient light data and user body presence data. These data are the basis for realizing adaptive lighting control. Preferably, the present invention adopts a high-precision eye tracking sensor to acquire the user's eye movement data, and the sensor can accurately capture the user's blinking frequency and gaze duration. At the same time, the ambient light sensor is used to monitor the surrounding light intensity and color temperature in real time, and the human presence sensor is used to detect whether the user is within the effective illumination range of the desk lamp.

[0070] In the processing step, the method first detects user fatigue based on user eye movement data and determines the user fatigue level. The implementation of this step involves a key algorithm, which we call the fatigue assessment algorithm. The algorithm can be expressed as:

[0071]

[0072] in, is the fatigue index, is the blink frequency (times / minute), is the fixation duration (seconds), is the pupil diameter change rate, is the weight coefficient, is the weight coefficient Through a large amount of experimental data, we found that when The algorithm can most accurately reflect the user's fatigue level. Based on the calculated fatigue index, the system divides the user's fatigue level into three levels: S1 (mild fatigue), S2 (moderate fatigue) and S3 (severe fatigue).

[0073] Preferably, in one embodiment of the present invention, when At S1 level, At the time it was S2 level. This classification method can effectively distinguish different degrees of user fatigue, thus providing a basis for subsequent lighting adjustments.

[0074] This formula is used to calculate the user's fatigue index. , which is an indicator that comprehensively evaluates the user's fatigue level. The calculation of the fatigue index is based on multiple physiological signals, such as blinking frequency, fixation duration, and eye diameter change rate, which can reflect the user's mental state and attention level. It is a fatigue index, and its value range is usually [0,1], where 1 indicates extreme fatigue and 0 indicates full wakefulness. Blink frequency (times / minute), which reflects the number of times the user blinks. Studies have shown that when users feel tired, the blink frequency increases significantly. Gaze duration (seconds), which indicates the time the user keeps looking at a certain target. Long-term gaze may cause eye fatigue, so the longer the gaze duration, the higher the fatigue. The pupil diameter change rate reflects the change in pupil size. The change in pupil diameter is closely related to the user's attention and mental state. When fatigued, the pupil may dilate or contract. , β and γ are weight coefficients used to adjust the influence of different factors on fatigue. Through a large amount of experimental data, we found that when The algorithm can most accurately reflect the user's fatigue level according to the calculated fatigue index. ,The system divides the user’s fatigue level into three levels:

[0075] S1 (mild fatigue): hour;

[0076] S2 (moderate fatigue): hour;

[0077] S3 (severe fatigue): hour.

[0078] This division method can effectively distinguish different degrees of user fatigue, thus providing a basis for subsequent lighting adjustments. For example, when a user is in a state of severe fatigue, the system can automatically adjust the ambient light and color temperature to help the user recover.

[0079] Next, the system adjusts the color temperature and brightness of the desk lamp according to the user's fatigue level. For example, when the user is at level S1, the system adjusts the color temperature to a higher level (about 5000K-6500K) and keeps the brightness at a moderate level (about 300-500lux); when the user is at level S2, the color temperature will be reduced to a mild level (about 3500K-4500K) and the brightness will be slightly reduced (about 200-300lux); and when the user is at level S3, the color temperature will be further reduced to a warm tone (about 2700K-3200K) and the brightness will be reduced accordingly (about 100-200lux). This adjustment strategy is designed to reduce the user's visual fatigue and improve comfort.

[0080] At the same time, this method also determines the ambient light brightness level based on the ambient light data. The division of ambient light brightness level also adopts a simple and effective algorithm:

[0081]

[0082] in, is the ambient light brightness level, is the ambient light intensity (lux), is the calibration factor, Through a large amount of experimental data, the present invention determines the calibration coefficient , the algorithm can best match the perception characteristics of the human eye.

[0083] Based on this algorithm, this method divides the ambient light brightness level into five levels: extremely dark ( , Dark(2 , moderate (3 ,Bright and extremely bright This division method can effectively distinguish different ambient light conditions and provide an accurate basis for subsequent lighting adjustments.

[0084] This formula is used to determine the ambient light brightness level ,This is an important indicator to measure the intensity of ambient light.,The human eye's perception of light intensity changes is nonlinear,,so an appropriate mapping algorithm is needed to accurately reflect the,perception of the human eye.

[0085] The ambient light brightness level usually ranges from [0,5], where 0 represents extremely dark and 5 represents extremely bright. is the ambient light intensity (lux), which means the actual measured light intensity. and is the calibration coefficient, which is used to adjust the shape of the mapping curve. , the algorithm can best match the perception characteristics of the human eye. is a logarithmic function used to convert linear light intensity to nonlinear perceived brightness. Adding 1 is to avoid The case where the logarithmic function is undefined.

[0086] Based on this algorithm, this method divides the ambient light brightness level into five levels: Very dark: When when; dark: when when; moderate: when when; bright: when when; extremely bright: when This division method can effectively distinguish different ambient light conditions and provide an accurate basis for subsequent lighting adjustments. For example, in extremely dark environments, the system can automatically increase the light intensity, while in extremely bright environments, the system can appropriately reduce the light intensity to avoid glare.

[0087] Based on the determined ambient light level, the system will adjust the color temperature of the desk lamp accordingly. For example, in extremely dark environments, the system will adjust the color temperature to a lower level (about 2700K-3000K) to create a warm and comfortable atmosphere; in extremely bright environments, the color temperature will be adjusted to a higher level (about 5500K-6500K) to provide better visual contrast.

[0088] In addition, the method also determines whether the user is within the illumination range of the desk lamp based on the user's body presence data. This function not only saves energy but also avoids unnecessary light pollution. Specifically, when the system detects that the user has been out of the illumination range for more than a preset time (for example, 5 minutes), the desk lamp will automatically be switched to standby mode or turned off.

[0089] In the output step, the system outputs the adjustment parameters of the desk lamp according to the result of the processing step, thereby realizing adaptive control of the desk lamp lighting. These parameters include color temperature value, brightness value, switch state, etc. Preferably, the present invention adopts PWM (pulse width modulation) technology to accurately control the color temperature and brightness of the LED light source, and can achieve a color temperature resolution of 0.1K and a brightness resolution of 1lux.

[0090] Next, the specific implementation of user fatigue detection is further described in detail. In one embodiment of the present invention, this step includes acquiring eye movement data, determining blinking frequency and fixation duration based on the eye movement data, and then calculating the user fatigue level based on the blinking frequency and fixation duration.

[0091] In the process of acquiring eye movement data, the present invention preferably uses a high-speed camera (such as a camera capable of reaching 120fps) in conjunction with an infrared light source to capture the user's eye movements. This method can obtain clear eye images under various lighting conditions, thereby improving the accuracy and stability of data collection.

[0092] When determining the blinking frequency, the system uses an image processing-based algorithm. The algorithm first locates the eye area through a Haar cascade classifier, and then uses a gradient-based edge detection method to identify the position of the eyelids. When the distance between the upper and lower eyelids is less than a preset threshold (usually 20% of the eye height), the system determines it as a blink. The blinking frequency is calculated by the number of blinks per unit time.

[0093] The calculation of fixation duration is based on the change of pupil position. The system uses the Hough transform algorithm to locate the pupil center and track its position changes. When the displacement of the pupil center within a certain period of time (such as 0.5 seconds) is less than a preset threshold (such as 5 pixels), it is determined as a fixation. The fixation duration is the cumulative time of continuous fixations.

[0094] Based on the obtained blinking frequency and fixation duration, the system uses the fatigue assessment algorithm mentioned above to calculate the user's fatigue level. It is worth noting that the present invention also takes individual differences into consideration in practical applications. The system will collect basic data of the user in the early stage of use and establish a personalized fatigue model to improve the accuracy of the assessment.

[0095] In one embodiment of the present invention, user fatigue detection also includes obtaining eye movement data through a camera module, tracking human eyes in real time through a camera, and calculating gaze direction and position based on multiple frames of facial images taken by the camera module.

[0096] In this process, the present invention adopts an improved convolutional neural network (CNN) model to achieve accurate positioning of the face and eyes. The network structure of the model is as follows:

[0097] 1. Input layer: receives 224x224x3 RGB images;

[0098] 2. Convolutional layer 1: 32 3x3 convolution kernels, stride 1, ReLU activation function;

[0099] 3. Max pooling layer 1: 2x2 pooling with a step size of 2;

[0100] 4. Convolutional layer 2: 64 3x3 convolution kernels, stride 1, ReLU activation function;

[0101] 5. Max pooling layer 2: 2x2 pooling with a step size of 2;

[0102] 6. Convolutional layer 3: 128 3x3 convolution kernels, stride 1, ReLU activation function;

[0103] 7. Max pooling layer 3: 2x2 pooling with a step size of 2;

[0104] 8.Fully connected layer 1: 512 neurons, ReLU activation function;

[0105] 9. Dropout layer: the dropout rate is 0.5;

[0106] 10. Fully connected layer 2: 68 neurons (corresponding to 68 facial key points);

[0107] This model is trained using a large number of face datasets and can accurately locate facial key points, including eye contours and pupil positions, under various lighting and posture conditions.

[0108] When calculating the gaze direction and position, the present invention adopts a method based on a 3D eyeball model. First, the system establishes a simplified 3D eyeball model based on the detected eye contour and pupil position. Then, the rotation angle of the eyeball is obtained by solving the offset of the pupil center relative to the eyeball center. Finally, the eyeball rotation angle is mapped to the gaze direction and position in 3D space by combining the intrinsic and extrinsic parameters of the camera.

[0109] The advantage of this method is that it can not only accurately calculate the user's gaze point, but also estimate the depth of gaze. This is important for evaluating the user's reading behavior and fatigue level. For example, if the system detects that the user is looking at a very close distance for a long time, it will remind the user to pay attention to eye hygiene and take a break at the right time.

[0110] Through the above detailed technical implementation, the method of the present invention can comprehensively and accurately evaluate the user's fatigue state and provide a reliable basis for subsequent lighting adjustment. This not only improves the intelligent level of desk lamp lighting, but also effectively protects the user's vision health and reduces the discomfort caused by long-term eye use.

[0111] The method of the present invention achieves comprehensive perception of user status and environmental conditions by fusing multiple sensor data, thereby being able to provide more personalized and intelligent lighting services. This method is not only applicable to desk lamps, but can also be extended to other lighting scenarios, such as office lighting, classroom lighting, etc., and has broad application prospects. The method of the present invention also includes a step of determining the ambient light brightness level based on the ambient light data. This step is crucial to achieving adaptive lighting control because the ambient light conditions directly affect the user's visual comfort and lighting needs.

[0112] In one embodiment of the present invention, the step specifically includes acquiring ambient light data, determining the ambient light brightness level based on the collected ambient light brightness, and adjusting the color temperature according to the ambient light brightness level. In a preferred embodiment of the present invention, the acquisition of ambient light data uses a high-precision digital light sensor. The sensor can not only measure the intensity of ambient light, but also analyze its spectral composition, thereby more comprehensively evaluating the ambient light conditions.

[0113] A nonlinear mapping algorithm is used to determine the ambient light brightness level. This algorithm takes into account the human eye's perception of light intensity changes and can be expressed as:

[0114]

[0115] in, is the ambient light brightness level, is the ambient light intensity (lux), is the calibration factor, Through a large amount of experimental data, the present invention determines the calibration coefficient , the algorithm can best match the perception characteristics of the human eye.

[0116] Based on this algorithm, this method divides the ambient light brightness level into five levels: extremely dark ( , Dark(2 , moderate (3 ,Bright and extremely bright This division method can effectively distinguish different ambient light conditions and provide an accurate basis for subsequent lighting adjustments.

[0117] When adjusting the color temperature according to the ambient light brightness level, the present invention adopts a dynamic mapping strategy. Specifically, the adjustment of the color temperature follows the following formula:

[0118]

[0119] in, is the adjusted color temperature, is the preset minimum color temperature (such as 2700K), is the preset highest color temperature (such as 6500K), is the current ambient light brightness level, is the lowest brightness level (such as 1), For the highest brightness level (such as 5), the brightness data of the ambient light includes not only the light intensity information but also the color temperature information.

[0120] This adjustment strategy can achieve a smooth transition between color temperature and ambient light brightness, avoiding abrupt light changes and providing users with a more comfortable visual experience.

[0121] This formula is used to dynamically adjust the color temperature , to adapt to different ambient light brightness conditions. Color temperature is one of the important factors affecting visual comfort. Appropriate color temperature can improve work efficiency and sleep quality. This formula uses a linear interpolation method to calculate the current ambient light brightness level. ,Dynamically adjust color temperature .when When the color temperature is low, Also lower (warmer), when When the color temperature is high Application: This dynamic mapping strategy can automatically adjust the color temperature according to the changes in ambient light brightness to provide a more comfortable lighting environment. For example, at night or in a dim environment, the system can automatically lower the color temperature to create a warm atmosphere; in the daytime or in a bright environment, the system can increase the color temperature to provide a clearer visual experience.

[0122] Next, the method of the present invention further includes the step of determining whether the user is within the illumination range of the desk lamp based on the user's body presence data. In one embodiment of the present invention, the step specifically includes acquiring the user's body presence data, determining whether the user's body is within the effective illumination range of the desk lamp, and automatically adjusting to a low color temperature lighting mode when the user is not within the illumination range.

[0123] In one embodiment of the present invention, the acquisition of user body presence data uses infrared thermal imaging technology combined with a depth camera. This combination can accurately detect the user's presence and location under various lighting conditions. Infrared thermal imaging technology can capture the thermal radiation emitted by the user's body temperature, while the depth camera provides accurate distance information.

[0124] A fuzzy logic-based algorithm is used to determine whether the user is within the effective illumination range of the desk lamp. The algorithm takes into account multiple factors, including the distance between the user and the desk lamp, the user's position angle, and the illumination range of the desk lamp. The output of the algorithm is a value between 0 and 1, indicating the degree to which the user is within the illumination range. When this value is lower than a preset threshold (such as 0.3), the system determines that the user is not within the effective illumination range.

[0125] When the system determines that the user is not within the illumination range, it will automatically adjust the desk lamp to a low color temperature lighting mode. In a preferred embodiment of the present invention, this low color temperature mode is set to 2700K, and the brightness is reduced to 20% of the minimum value. This setting can save energy on the one hand, and on the other hand, it can create a soft ambient light for the room, avoiding the discomfort caused by complete darkness.

[0126] In addition, the method of the present invention also includes a step of performing time-series processing on the acquired eye movement data. In one embodiment of the present invention, this step is crucial for accurately assessing the user's fatigue state and eye habits. Time-series processing can capture the dynamic changes of the user's eye movement pattern, thereby providing a more comprehensive and accurate eye behavior analysis.

[0127] In one embodiment of the present invention, the timing processing uses a sliding window technique combined with Fourier transform. Specifically, the system first divides the continuously acquired eye movement data into multiple overlapping time windows at fixed time intervals (such as 10 seconds). Then, the data in each window is subjected to a fast Fourier transform (FFT) to obtain frequency domain features. This method can effectively capture the periodic changes in eye movement patterns, such as fluctuations in blink frequency.

[0128] Based on the time-series processed eye movement data, the system extracts data features. In one embodiment of the present invention, this includes obtaining the single eye movement time, the proportion of eye opening time, and the blinking frequency. In a preferred embodiment of the present invention, the following method is used to extract these features:

[0129] 1. Single eye movement time: locate each eye movement by identifying the peaks in the eye movement data, and then calculate the time interval between adjacent peaks.

[0130] 2. Proportion of eyes-open time: Use the threshold method to binarize the eye movement data, regard the part below the threshold as the eyes-closed state, and then calculate the proportion of eyes-open time to the total time.

[0131] 3. Blink frequency: In the frequency domain result of Fourier transform, identify the frequency component corresponding to the blink action, which is usually in the range of 0.1-0.5Hz.

[0132] These features can not only reflect the user's immediate eye status, but also assess the user's fatigue level and eye habits by comparing the changing trends over a long period of time. For example, if the system detects that the user's blinking frequency gradually decreases and the single eye movement time increases, this may indicate that the user is experiencing visual fatigue.

[0133] Through the above detailed technical implementation, the method of the present invention can comprehensively and accurately evaluate the ambient light conditions, user position and eye behavior, and provide a reliable basis for adaptive lighting control. This method not only improves the level of intelligence of desk lamp lighting, but also effectively protects the user's vision health and reduces the discomfort caused by long-term eye use. At the same time, through automatic adjustment and energy-saving mode, this method can also achieve effective use of energy, which meets the expectations of modern society for smart homes. Continuing to explore the technical features of the present invention in depth, in one embodiment of the present invention, the user fatigue level is divided in more detail. The present invention divides the user fatigue level into three levels: fatigue level S1, fatigue level S2 and fatigue level S3. This grading method can not only more accurately reflect the user's fatigue state, but also provide a more detailed basis for the adjustment of lighting parameters.

[0134] In a preferred embodiment of the present invention, the three fatigue levels correspond to different lighting parameter settings. Specifically, fatigue level S1 corresponds to a preset brightness, fatigue level S2 corresponds to a first color temperature, and fatigue level S3 corresponds to a second color temperature. The design of this correspondence is based on a large amount of human factor engineering research and user experience testing, and is intended to provide the most suitable lighting environment for users with different fatigue levels.

[0135] The preset brightness range is set to 0-10lux. This range may seem low, but it actually takes into account the ability of the human eye to adapt to different lighting conditions. In a low-light environment, even a small change in brightness can bring significant visual effects. For example, when the user is at fatigue level S1, the system may adjust the brightness to around 8lux, which is enough to provide a comfortable reading environment without causing additional stimulation to tired eyes.

[0136] The first color temperature range is set to 2500K-3200K. This range belongs to the transition area from warm white light to neutral white light. When the user enters fatigue level S2, the system will adjust the color temperature to this range, for example, 3000K. This warm light helps relieve eye fatigue while maintaining a certain level of alertness, which is suitable for continuing mild visual tasks.

[0137] The second color temperature range is set to 1800K-2500K. This range is very warm light, close to candlelight or natural light at sunset. When the user reaches fatigue level S3, the system will lower the color temperature to this range, such as 2200K. This extremely warm light can significantly reduce stimulation to the retina and help users relax and rest.

[0138] It is worth noting that the present invention follows an important principle in color temperature setting: the second color temperature is lower than the first color temperature, and the first color temperature is higher than the preset brightness. This progressive relationship is consistent with the physiological rhythm of the human body. As fatigue increases, the color temperature gradually decreases, simulating the change process of natural light from day to dusk and then to night, which helps to regulate the physiological rhythm of the human body and improve the physiological comfort of lighting.

[0139] Next, the present invention also proposes a method for obtaining ambient light data. In one embodiment of the present invention, the method includes detecting ambient light changes through a light sensor and collecting ambient light brightness data to form ambient light data. This step is crucial for achieving accurate adaptive lighting control.

[0140] In one embodiment of the present invention, the light sensor uses a high dynamic range (HDR) digital light sensor. This sensor can accurately measure light intensity over a wide range of 0.01 lux to 100,000 lux, covering a variety of lighting conditions from moonlight to strong sunlight. The sensor's response time is set to 50ms, which means it can quickly capture instantaneous changes in ambient light, such as clouds blocking sunlight or indoor light switches.

[0141] In order to improve the accuracy and representativeness of ambient light data, this method uses multi-point sampling technology. Specifically, multiple light sensors are installed at different positions of the desk lamp to form a sensor array. These sensors collect light data from different directions and positions, and then the final ambient light data is obtained by weighted averaging. The weight distribution takes into account the position and orientation of each sensor, as well as the user's usual usage posture.

[0142] The brightness data of ambient light includes not only light intensity information but also color temperature information. The present invention adopts a color temperature estimation algorithm based on RGB tri-color sensor. The algorithm can be expressed as:

[0143]

[0144] Where CCT is the correlated color temperature, is the coordinate on the CIE1931 chromaticity diagram, is the coordinate on the CIE1931 chromaticity diagram, is the fitting coefficient, is the fitting coefficient, is the fitting coefficient, is the fitting coefficient, is the fitting coefficient, is the fitting coefficient, This method can accurately estimate the color temperature under most natural light and artificial light conditions and provide an important reference for subsequent lighting adjustment.

[0145] This formula is used to estimate the correlated color temperature , which is an important parameter to measure the color of light sources. Color temperature not only affects visual comfort, but is also closely related to the biological rhythm of the human body. Accurate estimation of color temperature is crucial for the optimization of intelligent lighting systems.

[0146] Correlated color temperature, measured in Kelvin (K), represents the color temperature of the light source. and are the coordinates on the CIE1931 chromaticity diagram, representing the proportions of red, green and blue respectively. The CIE1931 chromaticity diagram is a standard chromaticity space model, widely used in color science and lighting engineering. , , , and , , are fitting coefficients obtained by fitting experimental data. The choice of these coefficients determines the accuracy of the estimation algorithm. The formula adopts a rational polynomial form, which can better fit the complex law of color temperature variation. In this way, the system can accurately estimate the color temperature under most natural light and artificial light conditions.

[0147] The color temperature estimation algorithm based on the RGB tricolor sensor can provide accurate color temperature information under different lighting conditions, providing an important reference for subsequent lighting adjustments. For example, in an intelligent lighting system, the system can automatically adjust the color temperature of the light source based on the estimated color temperature to provide a more comfortable lighting environment.

[0148] Finally, the present invention also proposes an adaptive desk lamp lighting control system based on multi-sensor fusion. In one embodiment of the present invention, this system is the hardware basis for implementing the above method, including a user reading module 1, an eye movement control module 2, an ambient light control module 3, a state monitoring module 4 and an adaptive dimming module 5.

[0149] The user reading module 1 is responsible for acquiring the user's eye movement data, ambient light data and user's body presence data. In a preferred embodiment of the present invention, the module includes hardware devices such as a high-speed camera, an ambient light sensor and an infrared sensor, as well as corresponding data acquisition and preprocessing software.

[0150] The eye movement control module 2 detects user fatigue based on the acquired eye movement data and adjusts the color temperature and brightness accordingly. The core of this module is a fatigue assessment algorithm based on deep learning. Through the combination of convolutional neural network (CNN) and long short-term memory network (LSTM), the algorithm can extract high-level features from eye movement data and achieve accurate fatigue assessment.

[0151] The ambient light control module 3 is responsible for determining and adjusting the color temperature according to the ambient light data. This module adopts an adaptive filtering algorithm, which can effectively eliminate noise and interference in the ambient light data and improve the stability and accuracy of color temperature adjustment.

[0152] The status monitoring module 4 determines whether the user is within the illumination range of the desk lamp based on the user's body presence data, and adjusts the color temperature and brightness based on the judgment result. This module uses an intelligent decision-making algorithm based on the Markov decision process (MDP), which can make the best adjustment decision based on user habits and energy efficiency.

[0153] The adaptive dimming module 5 is the core of the whole system. It integrates the output of each module, determines the user's fatigue level and the ambient brightness level, and adjusts the color temperature and brightness accordingly. This module adopts a multi-objective optimization algorithm, which not only ensures visual comfort, but also takes into account multiple goals such as energy efficiency and physiological rhythm, realizing truly intelligent dimming.

[0154] Through the collaborative work of these modules, the system of the present invention can achieve highly intelligent and personalized lighting control, providing users with the best visual experience and health protection. This system is not only suitable for desk lamps, but can also be extended to other indoor lighting scenes, and has broad application prospects. In order to verify the superiority of the adaptive desk lamp lighting control method based on multi-sensor fusion and its system of the present invention, the present invention conducted a series of experimental comparisons. The specific implementation schemes of the embodiments and comparative examples, as well as the corresponding test results and analysis, will be introduced in detail below.

[0155] Example 1: This example adopts the complete technical solution of the present invention, including multi-sensor fusion technologies such as eye tracking, ambient light perception and user presence detection, and an intelligent control algorithm based on the LSTM-Attention model. Specifically, this example uses a high-speed camera (200fps) for eye tracking, an HDR ambient light sensor array to detect changes in ambient light, and infrared thermal imaging technology combined with a depth camera for user presence detection. In terms of control algorithms, the fatigue assessment algorithm and ambient light brightness level division method proposed in the present invention are adopted, and personalized lighting parameter adjustment is achieved through the LSTM-Attention model.

[0156] Comparative Example 1: The comparative example adopts the traditional adaptive desk lamp lighting control method. This method only uses a single ambient light sensor to detect light changes and adjusts the light brightness and color temperature through a simple threshold method. Users can manually adjust the lighting parameters through buttons, and the system will remember the last setting and apply it the next time it is turned on. This method does not take into account the user's real-time status and personalized needs, and lacks intelligent adaptive adjustment capabilities.

[0157] Test plan: In order to comprehensively evaluate the performance of the two methods, a set of comprehensive test indicators were designed, including visual comfort, energy efficiency, user satisfaction and system response time. The test was conducted in a simulated office environment for 4 weeks, with a total of 30 volunteers participating in the test, ranging in age from 25 to 45 years old.

[0158] The test indicators and their detection methods are as follows:

[0159] 1. Visual comfort: assessed by a standardized visual fatigue questionnaire (VFQ-25), completed at the end of each workday.

[0160] 2. Energy efficiency: The total power consumption of each lamp during the test is recorded and monitored in real time through smart meters.

[0161] 3. User satisfaction: A satisfaction survey is conducted once a week using a 5-point Likert scale.

[0162] 4. System response time: Use a high-speed camera to record the time interval from when the ambient light changes to when the desk lamp is adjusted.

[0163] 5. Eye fatigue level: Use a professional eye tracker (Tobii Pro Spectrum) to measure pupil diameter changes and blinking frequency.

[0164] The test results are shown in the following table:

[0165] Test indicators Example 1 Comparative Example 1 Improvement Visual comfort (VFQ-25 score) 92.5 78.3 +18.1% Energy efficiency (kWh / month) 3.2 5.7 -43.9% User satisfaction (1-5 points) 4.7 3.5 +34.3% System response time (ms) 286 1520 -81.2% Eye fatigue level (fatigue index) 0.32 0.58 -44.8%

[0166] It can be clearly seen from the test results that the method of the present invention is significantly superior to the traditional method in all indicators. The specific analysis is as follows:

[0167] In terms of visual comfort, the method of the present invention scored as high as 92.5, which is 18.1% higher than the traditional method. This is mainly due to the multi-sensor fusion technology and intelligent control algorithm adopted by the present invention, which can monitor the user's eye movement status and fatigue level in real time and dynamically adjust the lighting parameters accordingly. Especially after working for a long time, the method of the present invention can better relieve visual fatigue and provide a continuous comfortable lighting environment.

[0168] In terms of energy efficiency, the method of the present invention consumes only 3.2kWh of electricity per month, saving 43.9% of energy compared to traditional methods. This significant energy saving effect mainly comes from the intelligent user presence detection and precise ambient light perception capabilities of the present invention. When the user leaves or the ambient light is sufficient, the system can reduce the brightness or turn off the lighting in time to avoid unnecessary energy waste.

[0169] The user satisfaction test results further highlight the advantages of the present invention, with a score of 4.7 out of 5, which is 34.3% higher than the traditional method. This shows that users can clearly feel the improvement in lighting experience brought by the present invention. Many testers reported that the present invention's method can "automatically adjust to the most comfortable brightness and color temperature" and "seems to understand my needs."

[0170] The test results of the system response time are particularly amazing. The method of the present invention can complete the lighting adjustment in just 286ms, which is 81.2% faster than the traditional method. This fast response capability is due to the high-performance sensor and efficient LSTM-Attention model used in the present invention. The fast response not only improves the user experience, but also can adjust in time when the light changes suddenly (such as clouds blocking the sun), protecting the user's vision health.

[0171] Finally, in the eye fatigue test, the method of the present invention controlled the fatigue index to 0.32, which is 44.8% lower than the traditional method. This result directly proves the excellent effect of the present invention in protecting the user's vision health. By real-time monitoring of eye movement data and precise control of lighting parameters, the method of the present invention can effectively slow down the accumulation of eye fatigue, and is particularly suitable for work scenarios that require long-term eye use.

[0172] In summary, the adaptive desk lamp lighting control method based on multi-sensor fusion and the system thereof of the present invention show significant superiority in multiple aspects such as visual comfort, energy efficiency, user satisfaction, system response speed and eye fatigue relief. These advantages not only improve the work efficiency and comfort of users, but also make important contributions to energy conservation, environmental protection and health protection. It is particularly worth mentioning that the method of the present invention performs well in long-term use, which shows that it can not only meet the immediate needs of users, but also continuously optimize through continuous learning and provide a personalized lighting experience.

[0173] Based on the above test results and analysis, it can be considered that Example 1 represents the best implementation of the present invention. It fully demonstrates the great potential of multi-sensor fusion technology and advanced control algorithms in the field of intelligent lighting, and points out the direction for the development of future smart home and office environments.

[0174] 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 in the protection scope of the present invention.

Claims

1. An adaptive desk lamp lighting control method based on multi-sensor fusion, characterized in that: include: The acquisition steps include: Obtain user eye movement data, ambient light data, and user body presence data; Processing steps include: Based on the user eye movement data, perform user fatigue detection to determine the user fatigue level; According to the user's fatigue level, adjust the color temperature and brightness of the desk lamp; Based on the ambient light data, determining an ambient light brightness level; According to the ambient light brightness level, adjusting the color temperature of the desk lamp; Based on the user's body presence data, determining whether the user is within the illumination range of the desk lamp; Output steps include: According to the result of the processing step, the adjustment parameters of the desk lamp are output to realize adaptive control of the desk lamp lighting.

2. The method according to claim 1, characterized in that: The user fatigue detection specifically includes: Obtain eye movement data; Determining blink frequency and fixation duration based on the eye movement data; The user fatigue level is calculated according to the blinking frequency and the gaze duration.

3. The method according to claim 2, characterized in that The user fatigue detection also includes: Acquire eye movement data through the camera module; Based on the eye movement data, tracking the human eyes in real time through a camera; The gaze direction and position are calculated according to the multiple frames of face images taken by the camera module.

4. The method according to claim 1, characterized in that Determining the ambient light brightness level based on the ambient light data specifically includes: Get ambient light data; Determine the ambient light brightness level based on the collected ambient light brightness; The color temperature is adjusted according to the ambient light brightness level.

5. The method according to claim 1, characterized in that The determining whether the user is within the illumination range of the desk lamp based on the user's body presence data specifically includes: Obtain user physical presence data; Determine whether the user's body is within the effective illumination range of the desk lamp; If not, it will automatically adjust to a lighting mode with a low color temperature.

6. The method according to claim 1, characterized in that Also includes: Perform time series processing on the acquired eye movement data; Data feature extraction is performed based on the eye movement data after time series processing.

7. The method according to claim 6, characterized in that The timing process includes: Time-sort the acquired real-time eye movement data; Process data in chronological order; Data feature extraction is performed based on time-series eye movement data, including obtaining single eye movement time, eye open time ratio and blinking frequency.

8. The method according to claim 1, characterized in that The user fatigue levels include: Fatigue level S1, fatigue level S2 and fatigue level S3; Wherein, the fatigue level S1, fatigue level S2 and fatigue level S3 correspond to the preset brightness, the first color temperature and the second color temperature respectively; The preset brightness range is 0-10lux, the first color temperature range is 2500K-3200K, the second color temperature range is 1800K-2500K, the second color temperature is lower than the first color temperature, and the first color temperature is higher than the preset brightness.

9. The method according to claim 1, characterized in that: Also included are methods for obtaining ambient light data, including: Detect changes in ambient light through light sensors; The brightness data of the ambient light is collected to form ambient light data.

10. An adaptive desk lamp lighting control system based on multi-sensor fusion that implements the method according to any one of claims 1 to 9, characterized in that: include: A user reading module, used to obtain user eye movement data, ambient light data, and user body presence data; An eye movement control module is used to detect user fatigue based on the acquired eye movement data and then adjust the color temperature and brightness; An ambient light control module, used to determine and adjust the color temperature according to ambient light data; A status monitoring module is used to determine whether the user's body is within the illumination range of the desk lamp based on the user's body presence data, and adjust the color temperature and brightness based on the determination result; The adaptive dimming module is used to determine the user's fatigue level and the ambient brightness level based on the acquired data, and adjust the color temperature and brightness according to the user's fatigue level and the ambient brightness level.

Images