Desk control system based on smart home system

Abstract

The invention discloses a desk control system based on an intelligent home system, and relates to the technical field of intelligent desks. Comprising a functional hardware module, a data acquisition module, a sitting posture comparison module, a fatigue detection module, an alarm reminding module and a feedback suggestion module, according to the system, by collecting user images and sensor data, the sitting posture state is monitored in real time and compared with a sitting posture model library, and graded vibration and voice reminding are carried out on bad sitting postures; by analyzing the eye aspect ratio, the mouth aspect ratio and the nose tip coordinate fluctuation, the fatigue state of the user is comprehensively judged, the fatigue coefficient is calculated, and the intelligent home environment is adjusted in a linkage mode according to the fatigue level; and personalized rest suggestions and efficient time period references can be provided according to the use habits of the user. Accurate monitoring, intelligent reminding and home environment adaptive adjustment of the sitting posture and the fatigue state are achieved, user health is effectively promoted, and the use experience is improved.

Description

Technical Field

[0001] This invention relates to the field of smart desk technology, specifically a desk control system based on a smart home system. Background Technology

[0002] As an important tool for daily work and study, the traditional functions of desks are no longer sufficient to meet the increasingly diverse needs of users. Although electric desks with single functions such as height adjustment and lighting control have appeared on the market, these functions often require manual operation by the user, lacking intelligence and initiative.

[0003] Existing smart desks lack real-time monitoring and correction of user posture, and cannot identify the user's work or study status or proactively intervene. Furthermore, existing smart desks lack data analysis and personalized suggestions based on individual user habits, and their level of intelligence needs improvement. Therefore, we provide a desk control system based on a smart home system. Summary of the Invention

[0004] The purpose of this invention is to provide a desk control system based on a smart home system to solve the problems in the background art.

[0005] This invention can be achieved through the following technical solution: a desk control system based on a smart home system, comprising: The functional hardware module includes hardware facilities such as height-adjustable tables and chairs, adjustable angle work panels, camera units, and sensor units, which are used to provide functional implementation and data acquisition; The data acquisition module is used to acquire user images and build user data entries in the database based on the image data. It also acquires images and video streams of the user's sitting posture while using the desk. The sitting posture comparison module processes the real-time captured user sitting posture images, marks key joint nodes, calculates the angles of each key joint of the user, and compares and matches them with the data in the sitting posture model library to output the current sitting posture status. The fatigue detection module is used to analyze the user's facial video stream, and to identify fatigue-related behaviors and assess the fatigue level by downsampling the state of the eyes, mouth and micro-movements of the head. The alarm reminder module alerts the user via voice or sensory vibration based on the poor sitting posture and fatigue level obtained from the posture comparison module and fatigue detection module.

[0006] A further technical improvement of the present invention is that: the sitting posture model library includes three states: poor sitting posture, good sitting posture, and resting sitting posture; the data acquisition frequency switches accordingly when different sitting postures are identified and matched. When the current sitting posture is determined to be poor sitting posture, the data acquisition frequency is increased, and when it is determined to be resting sitting posture, the data acquisition frequency is decreased.

[0007] A further technical improvement of the present invention is that: the posture comparison module matches the joint angles of the user's posture with the desired angles in the posture model library by calculating the angles, and the matching conditions are: ; in, For the corresponding joint groups in the sitting posture model library From the perspective of expectations, The corresponding joint group for the user's sitting posture The actual measured angle, This is a preset allowable angle deviation threshold.

[0008] A further technical improvement of the present invention is that the fatigue detection module identifies fatigue behavior: The eye aspect ratio is calculated to identify the open and closed state of each frame of the image, and the duration of eye closure is monitored to distinguish between blinking behavior and resting behavior with eyes closed. Calculate the aspect ratio of the mouth; if it exceeds the set value and continues for a certain period of time, it is determined to be a yawning behavior. By analyzing the temporal changes in the distance projected from the tip of a user's nose to the desktop, we can identify napping behaviors with periodic small fluctuations.

[0009] A further technical improvement of the present invention is that the fatigue detection module determines that the user is in a state of fatigue based on the following criteria: The system determines that a user is in a state of fatigue when it detects any two of the following behaviors within a set time-sliding window: closing their eyes to rest, yawning, and dozing off.

[0010] A further technical improvement of this invention lies in: the fatigue detection module calculates the fatigue coefficient within a time-slip window to assess the fatigue level. The fatigue coefficient is positively correlated with the fatigue level, and its assessment formula is as follows: ; in, This refers to the cumulative duration of resting with eyes closed. The number of yawns The cumulative duration of napping behavior, Indicates the fatigue coefficient. These represent the weight coefficients of the three corresponding behaviors, and , This indicates the width of the time-slide window. This indicates that a threshold is set for the number of times a yawn occurs.

[0011] A further technical improvement of the present invention is that the alarm reminder module provides reminders for the user's poor posture in two stages. When the duration of poor posture exceeds the first-level set threshold, the desktop vibrates to remind the user. When the cumulative duration of non-continuous poor posture within a certain period exceeds the second-level set threshold, the user is reminded to adjust their posture via voice and the impact of poor posture is explained. The alarm reminder module reminds users to rest via voice broadcast and provides different suggested rest durations based on fatigue levels. It also proactively asks whether users want to switch the current environment status and adapts the smart home devices associated with the room to the rest and relaxation scenario.

[0012] A further technical improvement of the present invention is that the system also includes a feedback and suggestion module, which statistically analyzes the user's efficient usage periods and durations based on the user's usage habits, and provides the user with optimal work time references and dynamic rest reminder settings.

[0013] A further technical improvement of the present invention is that the feedback suggestion module will extract images from the video stream to generate moments of focus and moments of fatigue and send them to the user.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention compares the user's real-time sitting posture with the joint angles in three-dimensional coordinates using a predefined sitting posture model library, thereby accurately determining whether the user is in a poor sitting posture state. It also provides effective reminders to the user through a graded reminder mechanism, while avoiding excessive interference to the user.

[0015] 2. This invention integrates multiple features such as eye, mouth, and head micro-movements to determine fatigue status, with a more comprehensive recognition dimension and high accuracy. It can not only issue reminders, but also actively link with smart home systems to change the environment, forcing users to get out of fatigue status, realizing a closed loop from "perception" to "intervention", and providing more thorough protection for user health.

[0016] 3. This invention analyzes users' most efficient usage periods and durations to provide data-supported suggestions for optimal work periods and dynamic rest plans, enabling the system to adapt to different users' usage habits and continuously optimize the user experience. At the same time, by capturing images such as "moments of focus" and "moments of fatigue" and sending them to users, the product's interactivity and fun are increased, enhancing user stickiness. Attached Figure Description

[0017] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0018] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0019] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0020] Please see Figure 1 As shown, the desk control system based on the smart home system includes a functional hardware module, a data acquisition module, a posture comparison module, an alarm reminder module, an interactive control module, a fatigue detection module, a feedback suggestion module, and a database. The functional hardware modules of the desk include a height-adjustable base, an adjustable angle worktable, a controllable lighting unit, a camera unit, and a sensor unit. The sensor unit includes an infrared distance sensor and a pressure sensor. The chair that comes with the desk is also equipped with a pressure sensor and a height-adjustable base. The power components for the height-adjustable base and the adjustable angle worktable are electric actuators.

[0021] When a user uses the desk for the first time, the camera unit captures the user's facial and full-body image information. The data acquisition module calculates the user's height based on the full-body image information, the height of the retractable camera, and the distance between the desk and the user obtained by the infrared distance sensor, according to a proportional conversion relationship. The database is then used to create an information entry for the user, and the user's facial features and voiceprint features are recorded in the corresponding entry. After the user sits down at the desk, the posture comparison module marks the user's eyes, head, shoulders, elbows, wrists, waist, and knees based on the acquired full-body image information. It then compares the angles of each joint in the user's sitting posture with the posture model library in the database. Specifically: The sitting posture model library determines the joint angles for different sitting postures through literature review and data training. During posture matching, the relative coordinates of each joint in the human body are collected by a camera unit. The coordinate system is a virtual three-dimensional coordinate system constructed by the desk with a fixed anchor point as the origin. Based on the relative coordinates, the joint angle between two adjacent joints is calculated. Each joint angle is compared with the sitting posture model library in turn. If the comparison is successful, the comparison result is output; if the comparison fails, no result is output. A sitting posture with joint angles satisfying the following formula can be successfully matched with the sitting posture model library: ;in, For the corresponding joint groups in the sitting posture model library From the perspective of expectations, The corresponding joint group for the user's sitting posture The actual measured angle, This is a preset allowable angle deviation threshold.

[0022] Furthermore, the sitting posture model library is divided into three sitting posture states: poor sitting posture, good sitting posture, and resting sitting posture. Each sitting posture state includes a variety of similar sitting postures.

[0023] Users can control the working angle of the adjustable workboard, the brightness of the controllable lights, and the height of the desk and chair through voice control or mobile terminal via the interactive control module. After adjusting the corresponding parameters, the interactive control module stores these parameters in the database and generates the user's preference settings. During the user's use of the desk, the data acquisition module collects images of the user's sitting posture at a certain frequency and transmits them to the sitting posture comparison module. The sitting posture comparison module processes and calculates the images and outputs the comparison results to the alarm reminder module. More specifically, the posture comparison module compares the current sitting posture with the sitting posture model library in the database to obtain the current sitting posture status; when the sitting posture status is poor sitting posture, the sampling frequency is increased; when the sitting posture status is resting sitting posture, the sampling frequency is decreased; otherwise, the current sampling frequency is maintained.

[0024] The alarm reminder module has two levels of reminders for sitting posture. When the sitting posture is poor at a certain time, a timer is started. When the duration of poor sitting posture exceeds the first-level set threshold, the desktop will vibrate slightly to remind the user. When the cumulative duration of non-continuous poor sitting posture exceeds the second-level set threshold within a certain period of time, the user will be reminded to adjust their sitting posture through voice and the adverse effects of poor sitting posture will be explained.

[0025] The data acquisition module also tracks and captures the user's face through the camera and sends the video stream to the fatigue detection module. The fatigue detection module accurately locates key points such as the eyes, mouth, and nose of the face in the captured video frames. It can directly use the 68-point face landmark model of the Dlib library for marking, downsamples the video stream image, delineates the eye and mouth contour areas in each downsampled frame image, and calculates the aspect ratio of the eyes and the aspect ratio of the mouth respectively. When the eyes are open, the ratio of eyelid width to the distance between the corners of the eyes, i.e., the aspect ratio of the eyes, remains basically unchanged. However, when the eyes are closed, the aspect ratio of the eyes decreases significantly and approaches zero. A threshold for the closed-eye state is set. When the aspect ratio of the eyes in a certain frame is less than the threshold, the aspect ratio of the eyes in the subsequent frames is continuously compared. When the aspect ratio of the eyes is not less than the threshold for the first time in the subsequent frames, it means that the user's eyes are open in this frame. The timestamp difference between this frame and the previous frame where the eyes were first closed is calculated. This timestamp difference is compared with the blink determination time * 1.5. When the timestamp difference is less than the blink determination time * 1.5, the closed-eye action is determined to be in the blinking behavior process; otherwise, it is determined to be the closed-eye rest behavior. Similarly, the fatigue detection module calculates the ratio of the distance between the center of the upper and lower lips to the distance between the corners of the mouth, i.e., the aspect ratio of the mouth. When speaking or reading, the aspect ratio of the mouth is small and fluctuates frequently. When yawning, the aspect ratio of the mouth is large and relatively stable. Therefore, when the aspect ratio of the mouth exceeds the set threshold and remains above the set threshold for a certain period of time, it is determined to be a yawning behavior.

[0026] In addition, the fatigue detection module also obtains the coordinates of the nose tip in each frame of the image, constructs the projection of the nose tip coordinates onto the desktop in the virtual coordinate system, calculates the distance between the two points, constructs a time-series distance scatter plot, and uses a smooth curve to connect the scatter points to obtain a distance curve. When the curve segment of the distance curve shows continuous, small-amplitude periodic fluctuations, the duration of the curve segment is recorded. When the duration exceeds the threshold, it is determined to be a drowsy behavior. If the fatigue detection module detects any two of the three behaviors—closing eyes to rest, yawning, and dozing off—within the same time period, it considers the user to be in a state of fatigue. A time-sliding window with a fixed duration and a sliding step of 2 (1 minute) is set for each time-sliding action. After each sliding step, the cumulative duration of the aforementioned eye-closing rest behavior within the window is calculated. Number of yawns and the cumulative duration of napping. And assess the level of fatigue exhibited during this period, specifically: The evaluation formula is: ; in, Indicates the fatigue coefficient. These represent the weight coefficients of the three corresponding behaviors, and , This indicates the width of the time-slide window. This indicates that a threshold is set for the number of yawning behaviors; The value of is between (0,1), and the fatigue level increases with the increase of the fatigue coefficient. The fatigue detection module outputs different suggested rest durations to the alarm reminder module based on the fatigue level. The higher the fatigue level, the longer the suggested rest duration. The alarm reminder reminds the user to rest through voice broadcast and actively asks whether to switch the current environment status. It adapts the smart home devices in the room, such as air conditioners, lights, or speakers, to the rest and relaxation scenario. If the user does not operate, it will switch by default, forcing the user to get out of the current work or study state and protecting the user's physical and mental health.

[0027] Furthermore, the feedback and suggestion module records users' desk usage habits during the week or on weekends, such as usage time and duration per session. It also statistically analyzes the user's efficient usage time during different usage periods without triggering posture and fatigue reminders, thereby determining the user's optimal work or study time for reference. At the same time, it calculates the average efficient usage time for the corresponding period and dynamically sets rest reminders to encourage users to take breaks, ensuring that users can reasonably arrange their rest time and maintain efficient work or study. The feedback and suggestion module will also extract images from the video stream to generate moments of focus and moments of fatigue and send them to the user, improving the user experience and making it more fun.

[0028] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A desk control system based on a smart home system, characterized in that, include: The functional hardware module includes hardware facilities such as height-adjustable tables and chairs, adjustable angle work panels, camera units, and sensor units, which are used to provide functional implementation and data acquisition; The data acquisition module is used to acquire user images and build user data entries in the database based on the image data. It also acquires images and video streams of the user's sitting posture while using the desk. The sitting posture comparison module processes the real-time captured user sitting posture images, marks key joint nodes, calculates the angles of each key joint of the user, and compares and matches them with the data in the sitting posture model library to output the current sitting posture status. The fatigue detection module is used to analyze the user's facial video stream, and to identify fatigue-related behaviors and assess the fatigue level by downsampling the state of the eyes, mouth and micro-movements of the head. The alarm reminder module alerts the user via voice or sensory vibration based on the poor sitting posture and fatigue level obtained from the posture comparison module and fatigue detection module.

2. The desk control system based on a smart home system according to claim 1, characterized in that, The sitting posture model library includes three states: poor sitting posture, good sitting posture, and resting sitting posture. The data collection frequency switches accordingly when different sitting postures are identified and matched. When the current sitting posture is determined to be poor sitting posture, the data collection frequency is increased, and when it is determined to be resting sitting posture, the data collection frequency is decreased.

3. The desk control system based on a smart home system according to claim 1, characterized in that, The sitting posture comparison module matches the user's sitting posture joint angles with the desired angles in the sitting posture model library. The matching conditions are as follows: ； in, For the joint groups corresponding to the sitting posture in the sitting posture model library From the perspective of expectations, The corresponding joint group for the user's sitting posture The actual measured angle, This is a preset allowable angle deviation threshold.

4. The desk control system based on a smart home system according to claim 1, characterized in that, The fatigue detection module identifies fatigue behaviors: The eye aspect ratio is calculated to identify the open and closed state of each frame of the image, and the duration of eye closure is monitored to distinguish between blinking behavior and resting behavior with eyes closed. Calculate the aspect ratio of the mouth; if it exceeds the set value and continues for a certain period of time, it is determined to be a yawning behavior. By analyzing the temporal changes in the distance projected from the tip of a user's nose to the desktop, we can identify napping behaviors with periodic small fluctuations.

5. The desk control system based on a smart home system according to claim 4, characterized in that, The fatigue detection module determines that the user is in a state of fatigue based on the following criteria: The system determines that a user is in a state of fatigue when it detects any two of the following behaviors within a set time-sliding window: closing their eyes to rest, yawning, and dozing off.

6. The desk control system based on a smart home system according to claim 5, characterized in that, The fatigue detection module calculates the fatigue coefficient within a time-slip window to assess the fatigue level. The fatigue coefficient is positively correlated with the fatigue level, and the assessment formula is as follows: ; in, The cumulative duration of resting with eyes closed. The number of yawns The cumulative duration of napping behavior, Indicates the fatigue coefficient. These represent the weight coefficients of the three corresponding behaviors, and , This indicates the width of the time-slide window. This indicates that a threshold is set for the number of times a yawn occurs.

7. The desk control system based on a smart home system according to claim 1, characterized in that, The alarm reminder module provides reminders for poor posture in two stages. When the duration of poor posture exceeds the first-level set threshold, the desktop vibrates to remind the user. When the cumulative duration of non-continuous poor posture within a certain period exceeds the second-level set threshold, the user is reminded to adjust their posture via voice and the effects of poor posture are explained. The alarm reminder module reminds users to rest via voice broadcast and provides different suggested rest durations based on fatigue levels. It also proactively asks whether users want to switch the current environment status and adapts the smart home devices associated with the room to the rest and relaxation scenario.

8. The desk control system based on a smart home system according to claim 1, characterized in that, It also includes a feedback and suggestion module, which statistically analyzes the user's efficient usage periods and durations based on the user's usage habits, and provides the user with optimal work time references and dynamic rest reminder settings.

9. The desk control system based on a smart home system according to claim 8, characterized in that, The feedback suggestion module will extract images from the video stream to generate moments of focus and moments of fatigue, which will then be sent to the user.

Images