A desk lamp lighting control system based on the Internet of Things
Through the IoT-based desk lamp lighting control system, combined with user behavior and ambient light information, intelligently adjust the working parameters of the desk lamp, solving the problem that existing desk lamps cannot adapt to the reflectivity of the scene, and providing personalized lighting experience and energy-saving effects.
Patent Information
- Application Number
- CN202410672126.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-28
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-05-28
AI Technical Summary
When adjusting the light, existing desk lamps cannot consider the reflectivity of different materials in the scene in a timely manner, resulting in unsuitable light and require manual adjustment by users to meet the needs of different users.
Adopt a desk lamp lighting control system based on the Internet of Things, including user perception module, light perception module, data processing module, adjustment parameter module and feedback adjustment module. By perceiving user behavior and ambient light information, intelligently adjust the working parameters of the desk lamp to provide personalized Lighting experience.
It realizes the real-time adaptation of desk lamp lighting to user behavior and environmental changes, improves user experience and system adaptability, provides a comfortable lighting environment, and helps to save energy and reduce emissions.
Smart Images

Figure CN118382182B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of lighting fixtures, and in particular to a table lamp lighting control system based on the Internet of Things. Background Art
[0002] At present, with the rapid development of electronic technology, household appliances are becoming more and more intelligent. Smart washing machines, smart rice cookers, smart induction cookers, etc. have entered thousands of households. Smart home appliances have become a development trend. Under this trend, smart home desk lamps came into being.
[0003] Currently, desk lamps on the market can recognize the user's behavioral information, adjust the light when the user approaches, and can adjust the brightness in real time by recognizing the ambient light. However, when adjusting the light, they cannot adjust the reflectivity of different materials in the corresponding scene, so that when reflection occurs, the light cannot be adjusted in time. At the same time, when adjusting the light, the user's needs must be taken into consideration to adapt to the use of different users. Summary of the invention
[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is:
[0005] As a preferred technical solution of the present invention, a desk lamp lighting control system based on the Internet of Things includes:
[0006] The user perception module is used to perceive the user's behavior information in the target environment, and the behavior information includes posture information and action information.
[0007] The light sensing module is used to determine the intensity of ambient light and reflected light in the current scene.
[0008] The data processing module is used to determine the working parameters of the desk lamp under different behavior information based on the perceived user behavior information, the intensity of the ambient light and the reflected light, and to identify the behavior analysis results corresponding to the working status of the desk lamp and the user behavior information.
[0009] The parameter adjustment module is used to adjust the working parameters of the desk lamp in the current scene according to the working state, working parameters and behavior analysis results of the desk lamp.
[0010] The feedback adjustment module is used to determine the user's demand for the light of the desk lamp according to the adjustment record of the adjustment parameter module.
[0011] The beneficial effects of the present invention are as follows: 1. The present invention determines the working parameters of the desk lamp based on the perceived user behavior information and ambient light information, identifies the working state of the desk lamp and the results of the user behavior analysis, and intelligently adjusts the working parameters of the desk lamp in the current scene based on the working state, working parameters and behavior analysis results of the desk lamp. This ensures that the desk lamp lighting can adapt to user behavior and environmental changes in real time; learns the user's demand for desk lamp lighting based on historical adjustment records, and continuously optimizes the lighting control strategy, which improves the system's adaptive ability and user satisfaction.
[0012] 2. The present invention provides users with a personalized lighting experience through sophisticated behavior analysis, intelligent parameter adjustment and feedback learning. When using a desk lamp, users no longer need to make manual adjustments. The system can automatically adjust the lighting according to the user's behavior and ambient light, thereby improving the user experience.
[0013] 3. The present invention intelligently adjusts the working parameters such as the brightness and color temperature of the desk lamp. The system not only provides a comfortable lighting environment, but also helps to save energy and reduce emissions, which is in line with the concept of green environmental protection. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] The present invention is further described below in conjunction with the accompanying drawings and embodiments.
[0015] Figure 1 It is a system schematic diagram of a desk lamp lighting control system based on the Internet of Things;
[0016] Figure 2 It is a flow chart of a data processing module of a desk lamp lighting control system based on the Internet of Things;
[0017] Figure 3 It is a flow chart of step A1 of a data processing module of a desk lamp lighting control system based on the Internet of Things;
[0018] Figure 4 It is a flow chart of a parameter adjustment module of a desk lamp lighting control system based on the Internet of Things;
[0019] Figure 5 The present invention is a flow chart of a feedback adjustment module of a desk lamp lighting control system based on the Internet of Things. DETAILED DESCRIPTION
[0020] The embodiments of the present invention are described in detail below. The embodiments described below are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. If no specific techniques or conditions are specified in the embodiments, the techniques or conditions described in the literature in the art or the product specifications are used.
[0021] See also Figure 1 , a desk lamp lighting control system based on the Internet of Things, comprising:
[0022] A user perception module is used to perceive the user's behavior information in the target environment, and the behavior information includes posture information and action information;
[0023] Light sensing module, used to determine the intensity of ambient light and reflected light in the current scene;
[0024] A data processing module, used to determine the working parameters of the desk lamp under different behavior information according to the perceived user behavior information, the intensity of the ambient light and the reflected light, and to identify the behavior analysis results corresponding to the working state of the desk lamp and the user behavior information;
[0025] A parameter adjustment module is used to adjust the working parameters of the desk lamp in the current scene according to the working state, working parameters and behavior analysis results of the desk lamp;
[0026] The feedback adjustment module is used to determine the user's demand for the light of the desk lamp according to the adjustment record of the adjustment parameter module.
[0027] The working status refers to the power status, brightness status, color temperature status and corresponding working mode of the desk lamp.
[0028] The power status indicates whether the desk lamp is connected to the power supply, the brightness status indicates the current light intensity, the color temperature status indicates whether the light is warm or cool, etc. The working mode indicates whether the desk lamp has a manual mode or an automatic mode, such as a light sensing mode, a timing mode, etc. In different working states, the working parameters of the desk lamp are displayed differently, and different working parameters need to be set to adapt to different scenes. The corresponding working parameters are expressed as: illumination, brightness, and color temperature; among them, brightness indicates the ratio of the light intensity of the light source to the area of the light source, that is, the light intensity per unit area; illumination indicates the degree to which an object is illuminated, which is the luminous flux irradiated on a unit area of a surface; color temperature indicates the color of the light, and the color temperature will be adjusted according to different scenes, such as warm yellow light, which is suitable for relaxation and rest, and cold white light, which is suitable for work or activities that require high concentration.
[0029] By identifying the corresponding working status and working parameters, the working mode of the desk lamp can be adjusted in time, so that users can get the corresponding light intensity and color temperature according to their own needs.
[0030] In one embodiment of the present invention, Figure 2 As shown, in order to improve the use effect of the smart desk lamp, the data processing module also includes the following methods:
[0031] A1, according to the user's posture information and action information, determine the user's usage tendency of the desk lamp, and based on the user's usage tendency, determine the user's behavior analysis result.
[0032] The user's behavior information is determined by identifying the user's behavior in the corresponding scene and the purpose of the current behavior, such as obtaining the user's behavior when working and reading, the user reading at the desk and not reading at the desk, obtaining these different user behavior information, and the corresponding lighting under these behavior information to determine the adjustment range of the current desk lamp after user identification, to determine the light intensity adjustment range for different user behaviors.
[0033] A2, according to the user's behavior analysis results, determine the variation range of the working parameters of the desk lamp, and determine the corresponding working state of the desk lamp based on the preset standard value and the variation range of the working parameters.
[0034] The results of behavioral analysis are used to quantify user behavior. The parameters of the desk lamp are adjusted by recognizing the user's intentions and the user's behavior in the preset mode of the desk lamp through image recognition, so that the desk lamp can make certain adjustments based on the user's behavior, reduce the light of the desk lamp being too strong or too weak, and improve adaptability to behavior.
[0035] like Figure 3 As shown, the method for obtaining the behavior analysis result in step A1 includes:
[0036] A11, determining a user action sequence based on the acquired action information, and determining a first similarity index between the user action sequence and a preset action sequence;
[0037] A12, based on the acquired posture information, determining the change angle of the user in the continuous time, determining the change of the user's viewing angle, determining the change trajectory of the corresponding viewing angle area in the target area of the desk lamp, and determining a second similarity index of the posture information about the viewing angle and the preset posture based on the change trajectory of the viewing angle area;
[0038] A13, determining a change of a corresponding position in the posture information, and obtaining a third similarity index related to the position change;
[0039] A14, determining a behavior analysis result of the user based on the first similarity index, the second similarity index, and the third similarity index.
[0040] Posture information is represented by body posture and position. For example, if the body posture is a sitting posture, when a user sits under a desk lamp to read, he or she usually adopts a more comfortable sitting posture, which may be leaning on the back of a chair or leaning forward. When choosing different sitting postures, the light directly felt by the user will also be different. At the same time, position adjustment will also make the user's intuitive feeling of the light intensity of the desk lamp different. At this time, adjusting the light according to different body postures can better meet user needs.
[0041] Action information is represented by hand movements and corresponding motion states, such as adjusting a desk lamp, flipping through books, writing documents, etc. These actions are not only obviously related to the direct use of the desk lamp, but also the user's own purpose will be different when the actions are different. In this case, the corresponding light often needs to be adjusted.
[0042] The first similarity index is an index that measures the similarity between the action sequence currently executed by the user and the preset action sequence. It is mainly used to identify the user's specific actions, such as gesture control, motion recognition, etc., and is calculated by comparing the characteristics of the user's actions (such as speed, acceleration, trajectory, etc.) with the characteristics of the preset action sequence.
[0043] The second similarity index is an index that measures the similarity between the changes in the user's posture and perspective in continuous time and the preset posture and perspective. It is mainly used to track the changes in the user's head or body posture to predict or respond to the user's intentions, such as adjusting the lights or display content in a smart home environment. It is calculated by analyzing the continuous changes in the user's posture information, especially the rotation angle of the head or body, and comparing these changes with the preset posture.
[0044] The third similarity index is an indicator that measures the similarity between the changes in specific positions (such as joints, body parts, etc.) in the user posture information and the changes in preset or benchmark positions. It is used for more refined motion analysis, such as evaluating the correctness of movement posture, identifying small movement differences, etc. The similarity is calculated by capturing the position changes of key points in the user posture and comparing them with preset or standard position change patterns.
[0045] Methods for obtaining the first similarity index include:
[0046] The acquired user action sequence is input into the neural network model, the minimum path sum between each action point between the user action sequence and the preset action sequence is determined, and the minimum path sum is used as the first similarity index corresponding to the action sequence.
[0047] Create an n×m matrix D, where n is the length of the preset action sequence P, and m is the length of the user preset action sequence U. D[i,j] will store the distance between the i-th point in sequence P and the j-th point in sequence U, where 0 <i<n,0<j<m。
[0048] For each element D[i,j] in the matrix D, calculate the distance between the i-th point in the sequence P and the j-th point in the sequence U; get the path from D[1,1] to D[n,m] that satisfies the following conditions:
[0049] Boundary conditions: The path must start from D[1,1] and end at D[n,m].
[0050] Continuity: The points on the path are continuous, that is, if the current point on the path is D[i,j], then the previous point can only be D[i-1,j], D[i,j-1] or D[i-1,j-1].
[0051] Monotonicity: The points on the path increase monotonically on the time axis, that is, they will not return to the previous points.
[0052] The total distance on the path is the sum of the distances of all points on the path. This minimum total distance is the similarity measure between the action sequence and the preset action sequence.
[0053] At this time, the calculation formula for the similarity between each user action sequence and the preset action sequence is expressed as:
[0054] DTW(i,j)=dist(i,j)+min{DTW(i-1,j),DTW(i,j-1),DTW(i-1,j-1)};
[0055] Where dist(i,j) represents the distance between the i-th point and the j-th point in the sequence of the combination of the user action sequence and the preset action sequence, and DTW(i,j) represents the minimum cumulative distance from the starting point (1,1) to the point (i,j), that is, the similarity between each user action sequence and the preset action sequence.
[0056] In order to find the logical association between multiple actions from the action information and improve the accuracy of the similarity by expressing the logical association of the action sequence, when calculating the first similarity index, the logical association features of the action sequence related to the discrete actions are also included. The logical association features express the logical association between multiple actions in the action information, such as sequence and duration; and weights are set for the corresponding logical associations.
[0057] At this time, the first similarity index is expressed as:
[0058]
[0059] Among them, DTW(i,j) represents the similarity between each user action sequence and the preset action sequence, w i is the weight corresponding to the similarity between the user action sequence and the preset action sequence, and S1 is the first similarity index.
[0060] Methods for obtaining the second similarity index include:
[0061] Obtaining the corresponding azimuth, pitch angle, and roll angle in the posture information, and determining the continuous change of the user's posture information at the position of the visual area corresponding to the azimuth, pitch angle, and roll angle;
[0062] The posture information is converted into a vector, and the similarity index between the posture information in the corresponding visual area positions is determined according to the cosine value between the vectors.
[0063] In order to make the changes in the recognized posture information more consistent with the corresponding scene, weights are set for the direction angle, pitch angle and roll angle according to the scene information, and the weighted average of the posture information and the cosine value of the preset posture with the weights of the direction angle, pitch angle and roll angle is used as the output second similarity index.
[0064] The direction angle, pitch angle and roll angle in the acquired posture information are used to represent the angle corresponding to the user's head. According to the corresponding angle, the range of the user's viewing area can be obtained, and then the second similarity index after the viewing angle comparison can be obtained.
[0065] Methods for obtaining the third similarity index include:
[0066] By capturing the position changes of key points in the user's posture information and comparing them with the preset standard position change pattern; recording the user's position change sequence in continuous time, using Euclidean distance to calculate the distance between each key point and the preset position point to quantify the similarity of the position change, and averaging the distances of all key points as the third similarity indicator output.
[0067] After obtaining the first similarity indicator, the second similarity indicator, and the third similarity indicator, the range of user observation and action is regarded as the target area, and the user's action time and action type in the target area are determined based on the obtained second similarity indicator and the third similarity indicator. The user's predicted behavior in the target area is determined based on the first similarity indicator and the second similarity indicator. The user's behavior analysis result is determined based on the user's predicted behavior, action time and action type.
[0068] For example, through the first similarity index, the user's actions can be controlled. When the user makes a specific gesture, such as spreading his hands or other gestures, the desk lamp will actively turn on the light when it recognizes that this gesture is similar to the preset turning on and adjusting the light. The second similarity index can track the user's posture and perspective changes. When the user is reading, his head gradually lowers. The desk lamp captures the changes in the user's head posture and finds that it is similar to the preset "head-down reading" posture (that is, the second similarity index is high), so it automatically adjusts the angle of the lamp head to make the light better illuminate the book. When the user changes from sitting to standing, the desk lamp recognizes that the user's body posture has changed significantly, which is similar to the preset "standing" posture, so it automatically adjusts the light to adapt to the new needs. The third similarity index can track detailed action analysis. For example, when the user is painting, the hand makes a subtle sketching movement. The desk lamp captures the position changes of the key points of the hand and finds that it is similar to the preset "painting" action mode (that is, the third similarity index is high), so it automatically adjusts the light brightness and color temperature to provide a better visual environment.
[0069] In this step, the desk lamp obtains images of user behavior through miniature cameras installed on both sides and under the desk lamp. At the same time, an infrared sensor is installed to accurately detect the user's movement and position changes, and the collected data is transmitted to the corresponding processing device of the Internet of Things, so as to obtain solutions for adjusting the light intensity according to the different behaviors of the user. At the same time, wearable devices can be used to monitor and record the user's motion status and activity patterns in real time, and analyze the user's behavioral habits and health status based on these activity patterns, and adjust the brightness of the light accordingly.
[0070] After completing the behavior analysis results of the user, in order to make the corresponding behaviors have adaptive lighting requirements, after processing the behavior analysis results, the processing method of the parameter adjustment module also includes judging the corresponding behavior information when the current light is adjusted, screening out the behavior information that changes frequently, and recording the corresponding working parameters of the desk lamp, and adjusting the light according to the difference between the behavior analysis results and the preset values.
[0071] Preferably, Figure 4 As shown, the methods for adjusting the parameter module include:
[0072] B1. Based on the determined behavior analysis result, a first preset light adjustment scheme is obtained, and a similarity between the first preset light adjustment scheme and the current light allocation scheme is determined.
[0073] B2: extracting a second preset light adjustment scheme from the preset schemes according to the similarity of the first preset light allocation scheme.
[0074] The first preset light adjustment scheme is a combination scheme of multiple behavior information and corresponding light when light is allocated, and multiple behaviors in the current scheme are associated; it is used to determine the light adjustment scheme when a certain regularity exists.
[0075] The second preset light adjustment scheme is a scheme of a plurality of unrelated random behavior information and corresponding light during light distribution, and is used to determine a light adjustment scheme under irregular circumstances.
[0076] B3, determine the illumination value of the desk lamp in the first preset lighting scheme and the second preset lighting scheme, as well as the distribution uniformity of the illumination of the desk lamp, and determine the corresponding time attenuation coefficient. The time attenuation coefficient is the rate at which the illumination value of the desk lamp decreases as time increases, and adjust the light according to the time attenuation coefficient.
[0077] Adjust the light according to the time attenuation coefficient. The way to adjust the working parameters corresponding to the light is:
[0078] Determine the duration of the user's behavior under the corresponding time attenuation coefficient. If the duration of the behavior is greater than a preset threshold, determine the user's behavior pattern, adjust the working parameters according to the behavior pattern, and restore the working parameters after the target time period.
[0079] If the duration of the behavior is less than a preset threshold, determine the size of the similarity between the first preset lighting scheme and the second preset lighting scheme; when the similarity value corresponding to the first preset lighting scheme is greater than the similarity value corresponding to the second preset lighting scheme, use the working parameters of the next behavior matching the current behavior in the first preset scheme as the adjusted working parameters; when the similarity value corresponding to the first preset lighting scheme is less than the similarity value corresponding to the second preset lighting scheme, use the average value of the working parameters in the second preset lighting scheme as the adjusted working parameter value.
[0080] At this time, the corresponding adjustment plan is identified by determining the amount of change in the light of the desk lamp at the corresponding time. If the change is frequent, it means that there is a problem with the current plan and it needs to be adjusted. If the change frequency is small, then the user is in a fixed working state or behavior mode. In this mode, almost no adjustment is required. It is only necessary to maintain the corresponding state and regularly identify whether the current state has changed to make timely adjustments. At the same time, the preset threshold of the duration is adjusted according to the user behavior. The methods that can be used are: short-time threshold: 5-15 minutes, suitable for scenarios that require rapid response to changes in user behavior, but may increase the risk of misjudgment; medium-time threshold: 15-30 minutes, suitable for most situations, and can adjust the light after the user behavior stabilizes to reduce the possibility of misjudgment; long-time threshold: more than 30 minutes, suitable for scenarios where user behavior is very stable and requires higher accuracy, such as long reading or work meetings.
[0081] Assume that a user uses a desk lamp to read in a study. Initially, the desk lamp provides moderate brightness and warm color temperature according to the first preset light scheme, which is suitable for long-term reading. But over time, the brightness of the desk lamp will gradually decrease, which is controlled by the time attenuation coefficient.
[0082] If the user reads continuously for more than 30 minutes (preset threshold), the system will determine that the user's current behavior mode is "long-term reading" and automatically adjust the desk lamp to a brightness and color temperature that is more suitable for long-term reading to reduce eye fatigue.
[0083] If the user switches to using the computer after reading for 20 minutes, the system detects the change in behavior and calculates the similarity between the current behavior and the two preset light schemes. If the similarity of the first preset light scheme (for reading) is higher than that of the second preset light scheme (which may be for computer work), the desk lamp will adjust to the working parameters of the next behavior (such as computer work) recommended in the first preset light scheme. Conversely, if the similarity of the second preset scheme is higher, the desk lamp will adjust according to the average of the working parameters in the second preset scheme.
[0084] In this way, the desk lamp can intelligently adapt to changes in user behavior and provide a more comfortable and efficient lighting environment.
[0085] The time attenuation coefficient for the setting can be expressed as:
[0086] L t =L0×e λt ;
[0087] Among them, L t is the brightness at time t, L0 is the initial brightness, and λ is the attenuation coefficient, which indicates the rate at which the brightness decays over time.
[0088] In one embodiment of the present invention, in order to improve the intelligence of light adjustment, the light intensity preferred by the user is selected by obtaining the user's light adjustment record; Figure 5 As shown, the implementation method of the feedback adjustment module is expressed as:
[0089] D1, obtain historical adjustment records, obtain points of interest in the adjustment records, identify the corresponding points of interest, and determine the number of occurrences of the points of interest, scenes, and corresponding working parameters;
[0090] D2, determining the residual between the working parameter corresponding to the point of interest and the preset threshold value in different scenarios, and determining a first compensation coefficient based on the scenario according to the change amount and change frequency of the corresponding residual value;
[0091] D3, determining the number of times the interest point corresponds to the number and duration of the corresponding scene in different scenes, and determining a second compensation coefficient based on the frequency of the interest point;
[0092] D4, determining the interest score based on the tendency based on the comparison between the first compensation coefficient and the second compensation coefficient, and adjusting the light of the desk lamp according to the interest score based on the tendency.
[0093] The first compensation coefficient may be expressed as:
[0094] C1=k1×σ1+k2×σ2;
[0095] Among them, C1 is the first compensation coefficient, k1 and k2 are adjustment coefficients used to adjust the influence of residuals, σ1 is the average value of the residuals between the working parameters corresponding to the point of interest and the preset threshold, and σ2 is the standard deviation of the residuals between the working parameters corresponding to the point of interest and the preset threshold.
[0096] The second compensation coefficient is expressed as:
[0097] F I =N I ×T;
[0098] Among them, F I is the frequency of occurrence of interest points, N I is the number of times the interest point appears, T is the duration of the interest point, and I is the number of times the interest point appears.
[0099]
[0100] Among them, C2 is the second compensation coefficient, F I1 is the frequency of the current point of interest, F I2 is the standard frequency of the occurrence of interest points, and k3 is the adjustment coefficient of the frequency of the occurrence of interest points, which is used to convert the frequency into an appropriate compensation coefficient.
[0101] The interest score based on tendency is expressed as follows: the first compensation coefficient and the second compensation coefficient are normalized so that the value range of the first compensation coefficient and the second compensation coefficient are both within [0,1], and the interest score based on tendency is determined according to the normalized first compensation coefficient and the second compensation coefficient:
[0102] τ=(C1×μ1+C2×μ2)(1+C1+C2);
[0103] Among them, τ is the interest score of the tendency, C1 is the first compensation coefficient, C2 is the second compensation coefficient, μ1 is the basic interest value corresponding to the interest point, μ2 is the scene weighting factor, which is used to adjust the interest score to reflect the importance in different scenes. The first compensation coefficient and the second compensation coefficient obtained here are normalized values. The first compensation coefficient and the second compensation coefficient are normalized by subtracting the mean of the first compensation coefficient and the second compensation coefficient from the first compensation coefficient and the second compensation coefficient, and removing the standard deviation of the first compensation coefficient and the second compensation coefficient, so that the normalized first compensation coefficient and the second compensation coefficient conform to the normal distribution.
[0104] The first compensation coefficient is biased towards recording the changes in the scene, which can reflect the system's response speed and accuracy to points of interest in different scenes; the second scene coefficient is the number of times the points of interest appear and the duration of each scene, which aims to adjust the system's response according to the user's interests and scenes. The two coefficients can be used alone or in combination to improve the system's performance and accuracy in processing points of interest.
[0105] In this step, by analyzing the historical records, the "points of interest" that the user frequently adjusts or pays special attention to are identified. These points of interest may be specific light settings, time periods, user behavior patterns, etc. At the same time, the user's handling of the desk lamp in the historical records will be identified as a separate behavior pattern; data mining or machine learning techniques are used to discover these patterns. For example, clustering algorithms can help identify frequently occurring light setting combinations, and time series analysis can reveal the user's preferences for light at different times of the day; based on the identified points of interest, the system can determine different usage scenarios (such as work, reading, leisure, etc.). For each scenario, the system can learn and record the most suitable light parameter settings and establish a The system can map scenes and light parameters; when the system detects that the user has entered a specific scene, it can automatically adjust the light to the optimized settings for that scene; it can detect ambient light information (such as light intensity, color temperature, etc.) in real time through devices such as light sensors; based on the detected ambient light information and the currently identified scene, the system can further fine-tune the light parameters to ensure that the user has the best lighting experience; for example, during the day when there is sufficient natural light, the system may reduce the brightness of the desk lamp, and in the evening or at night it may increase the brightness and adjust the color temperature to provide a comfortable reading environment, and adapt the parameters to the corresponding scene, determine the adaptation method and adjustment tendency under the current processing, and further limit the adjustment method.
[0106] In step D4, the scheme is adjusted according to the interest score as follows:
[0107] For the obtained interest scores, the part with a total interest score greater than a preset total score threshold is classified as a target interest, and the part with a total interest score less than the preset total score threshold is classified as a potential interest;
[0108] The first adjustment scheme is used for target interests, and the second adjustment scheme is used for potential interests;
[0109] The first adjustment plan includes taking the same type of adjustment record corresponding to the maximum value of the interest point score as the user's first adjustment plan;
[0110] The second adjustment plan is used to generate multiple potential feature tags for potential interests, and filter out multiple recommended plans related to the potential feature tags from a preset database according to the multiple potential feature tags. If the recommended plan is consistent with the potential feature tag, it is used as the second adjustment plan of the user according to the priority.
[0111] If they are inconsistent, the recommended solution is adjusted and the adjusted recommended solution is used as the user's second adjusted solution.
[0112] The core idea is to provide personalized recommendations based on the user's interest score. For users who have clearly expressed strong interests (target interests), more accurate recommendations are provided by digging deeper into the specific content in that area. For areas that users may be interested in but have not yet clearly expressed (potential interests), tentative recommendations are made by generating potential feature tags and filtering relevant content from the database in order to stimulate more interest in users. This dual-track recommendation strategy aims to fully cover users' known and potential interests, thereby providing richer and more personalized content recommendations to meet users' needs for different light adjustments, while also reducing energy consumption during use.
[0113] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations on the present invention. A person skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention and they are still covered by the protection scope of the present invention.
Claims
1. A desk lamp lighting control system based on the Internet of Things, characterized in that: include: A user perception module is used to perceive the user's behavior information in the target environment, and the behavior information includes posture information and action information; Light sensing module, used to determine the intensity of ambient light and reflected light in the current scene; A data processing module, used to determine the working parameters of the desk lamp under different behavior information according to the perceived user behavior information, the intensity of the ambient light and the reflected light, and to identify the behavior analysis results corresponding to the working state of the desk lamp and the user behavior information; A parameter adjustment module is used to adjust the working parameters of the desk lamp in the current scene according to the working state, working parameters and behavior analysis results of the desk lamp; A feedback adjustment module is used to determine the user's demand for the light of the desk lamp according to the adjustment record of the adjustment parameter module; The implementation of the feedback adjustment module is expressed as follows: D1, obtain historical adjustment records, obtain points of interest in the adjustment records, identify the corresponding points of interest, and determine the number of occurrences of the points of interest, scenes, and corresponding working parameters; D2, determining the residual between the working parameter corresponding to the point of interest and the preset threshold value in different scenarios, and determining a first compensation coefficient based on the scenario according to the change amount and change frequency of the corresponding residual value; D3, determining the number of times the interest point corresponds to the number and duration of the corresponding scene in different scenes, and determining a second compensation coefficient based on the frequency of the interest point; D4, determining the interest score based on the tendency based on the comparison of the first compensation coefficient and the second compensation coefficient, and adjusting the light of the desk lamp according to the interest score based on the tendency; The first compensation coefficient may be expressed as follows: ; in, is the first compensation coefficient, , is the adjustment coefficient, used to adjust the impact of the residual, is the average value of the residual between the working parameter corresponding to the point of interest and the preset threshold, is the standard deviation of the residual between the working parameter corresponding to the point of interest and the preset threshold; The second compensation coefficient is expressed as: ; in, is the frequency of occurrence of interest points, is the number of occurrences of the interest point, is the duration of the point of interest, is the number of occurrences of the interest point; ; in, is the second compensation coefficient, is the frequency of the current point of interest, is the standard frequency of occurrence of points of interest, is the adjustment coefficient of the frequency of occurrence of the interest point, used to convert the frequency into an appropriate compensation coefficient; The interest score based on tendency is expressed as follows: the first compensation coefficient and the second compensation coefficient are normalized so that the value range of the first compensation coefficient and the second compensation coefficient are both within [0,1], and the interest score based on tendency is determined according to the normalized first compensation coefficient and the second compensation coefficient: ; in, is the interest score for the tendency, is the first compensation coefficient, is the second compensation coefficient, is the basic interest value corresponding to the interest point, is a scene weighting factor, which is used to adjust the interest score to reflect the importance of different scenes. The first compensation coefficient and the second compensation coefficient obtained here are normalized values; The method of adjusting the scheme according to the interest score in step D4 is: For the obtained interest scores, the part with a total interest score greater than a preset total score threshold is classified as a target interest, and the part with a total interest score less than the preset total score threshold is classified as a potential interest; The first adjustment scheme is used for target interests, and the second adjustment scheme is used for potential interests; The first adjustment plan includes taking the same type of adjustment record corresponding to the maximum value of the interest point score as the user's first adjustment plan; The second adjustment plan is used to generate multiple potential feature tags for potential interests, and filter out multiple recommended plans related to the potential feature tags from a preset database according to the multiple potential feature tags. If the recommended plan is consistent with the potential feature tag, it is used as the second adjustment plan of the user according to the priority. If they are inconsistent, the recommended solution is adjusted and the adjusted recommended solution is used as the user's second adjusted solution.
2. According to the Internet of Things-based desk lamp lighting control system of claim 1, it is characterized in that: The data processing module also includes the following methods: A1, determining the user's usage tendency of the desk lamp according to the user's posture information and action information, and determining the user's behavior analysis result based on the user's usage tendency; A2, according to the user's behavior analysis results, determine the variation range of the working parameters of the desk lamp, and determine the corresponding working state of the desk lamp based on the preset standard value and the variation range of the working parameters.
3. According to the Internet of Things-based desk lamp lighting control system of claim 2, it is characterized in that: The method of obtaining the behavior analysis result in step A1 includes: A11, determining a user action sequence based on the acquired action information, and determining a first similarity index between the user action sequence and a preset action sequence; A12, based on the acquired posture information, determining the change angle of the user in the continuous time, determining the change of the user's viewing angle, determining the change trajectory of the corresponding viewing angle area in the target area of the desk lamp, and determining a second similarity index of the posture information about the viewing angle and the preset posture based on the change trajectory of the viewing angle area; A13, determining a change of a corresponding position in the posture information, and obtaining a third similarity index related to the position change; A14, determining a behavior analysis result of the user based on the first similarity index, the second similarity index, and the third similarity index.
4. According to the Internet of Things-based desk lamp lighting control system of claim 3, it is characterized in that: Methods for obtaining the first similarity index include: Input the acquired user action sequence into the neural network model, determine the minimum path sum between each action point between the user action sequence and the preset action sequence, and use the minimum path sum as the first similarity index corresponding to the action sequence; The second similarity indicator can be obtained by: Obtaining the corresponding azimuth, pitch angle, and roll angle in the posture information, and determining the continuous change of the user's posture information at the position of the visual area corresponding to the azimuth, pitch angle, and roll angle; The posture information is converted into a vector, and the similarity index between the posture information in the corresponding visual area positions is determined based on the cosine value between the vectors. Setting weights for the direction angle, pitch angle and roll angle according to the scene information, and taking the weighted average of the posture information and the cosine value of the preset posture and the weights of the direction angle, pitch angle and roll angle as the output second similarity index; The third similarity indicator can be obtained by: By capturing the position changes of key points in the user's posture information and comparing them with the preset standard position change pattern; recording the user's position change sequence in continuous time, using Euclidean distance to calculate the distance between each key point and the preset position point to quantify the similarity of the position change, and averaging the distances of all key points as the third similarity indicator output.
5. According to the Internet of Things-based desk lamp lighting control system of claim 3, it is characterized in that: After obtaining the first similarity indicator, the second similarity indicator, and the third similarity indicator, the range of user observation and action is regarded as the target area, and the user's action time and action type in the target area are determined based on the obtained second similarity indicator and the third similarity indicator. The user's predicted behavior in the target area is determined based on the first similarity indicator and the second similarity indicator. The user's behavior analysis result is determined based on the user's predicted behavior, action time and action type.
6. The table lamp lighting control system based on the Internet of Things according to claim 1 is characterized in that: The method for implementing the parameter adjustment module includes: B1, based on the determined behavior analysis result, obtaining a first preset light adjustment scheme, and determining the similarity between the first preset light adjustment scheme and the current light allocation scheme; B2, extracting a second preset light adjustment scheme from the preset schemes according to the similarity of the first preset light allocation scheme; B3, determine the illumination value of the desk lamp in the first preset lighting scheme and the second preset lighting scheme, as well as the distribution uniformity of the illumination of the desk lamp, and determine the corresponding time attenuation coefficient. The time attenuation coefficient is the rate at which the illumination value of the desk lamp decreases as time increases, and adjust the light according to the time attenuation coefficient.
7. The table lamp lighting control system based on the Internet of Things according to claim 6, characterized in that: The method of adjusting the light according to the time attenuation coefficient and adjusting the working parameters corresponding to the light is: Determine the duration of the user's behavior under the corresponding time attenuation coefficient. If the duration of the behavior is greater than a preset threshold, determine the user's behavior pattern, adjust the working parameters according to the behavior pattern, and restore the working parameters after the target time period. If the duration of the behavior is less than a preset threshold, determine the size of the similarity between the first preset lighting scheme and the second preset lighting scheme; when the similarity value corresponding to the first preset lighting scheme is greater than the similarity value corresponding to the second preset lighting scheme, use the working parameters of the next behavior matching the current behavior in the first preset scheme as the adjusted working parameters; when the similarity value corresponding to the first preset lighting scheme is less than the similarity value corresponding to the second preset lighting scheme, use the average value of the working parameters in the second preset lighting scheme as the adjusted working parameter value.
