A writing posture supervision method

Abstract

The present application belongs to the field of medical health, and particularly relates to a writing posture supervision method. The method is used for supervising the sitting posture, pen holding posture and writing posture of a user. The method comprises the following steps: S1: establishing a space coordinate system of a desk; S2: dividing a writing plane in the desk according to lighting effect; S3: acquiring the positions of the eyes of the user and judging whether the sitting posture of the user is correct; S4: specifying a best writing area meeting ergonomics on the desk; S5: acquiring the position of a pen nib and judging whether the position of the pen nib is reasonable; S6: generating a virtual writing frame in the best writing area; S7: calculating the projection range limit of a pen shaft in the writing frame in a normal state to generate a reasonable writing projection area; S8: establishing a virtual pen shaft model in the space coordinate system; and S9: calculating the projection position of the pen shaft to judge whether the writing posture of the user is abnormal. The present application solves the problem that the sitting posture and writing posture of a user during writing are difficult to detect, supervise and correct.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of medical health, and particularly relates to a writing posture supervision method. BACKGROUND

[0002] With the popularization of compulsory education and the development of national economy, more and more families begin to pay attention to and concern about the learning habits of children. Students are usually in a key period of growth and development, and especially for the majority of children, the sitting posture and writing posture in the writing process have an important influence on the growth and development of students. In addition, a good writing posture is also helpful for students to improve the comfort and learning efficiency of writing, and to improve the aesthetic degree and calligraphy level of the writing font of students.

[0003] In order to help adolescent students develop good sitting habits, many parents will specially purchase seats and desks with adjustable height for students to automatically adjust the height of the desk and chair according to the height of the student. However, the adjustable desk and chair can only play an auxiliary role and is a passive posture correction tool. This kind of desk and chair cannot detect and supervise the actual sitting posture and writing posture of students, so it is difficult to effectively play the required posture correction effect.

[0004] In view of the above problems, some manufacturers of intelligent learning desks are also constantly innovating and developing some products with the function of sitting posture monitoring. For example, some manufacturers develop learning desks that can judge whether the user's sitting posture is normal through the weight sensing of the seat and backrest. This kind of self-adaptive backrest chair can detect the behavior of the user in the using process, such as lying on the table, to reduce the risk of hunchback caused by improper sitting posture. However, at present, almost all schemes can only detect the sitting posture of the user, but cannot supervise the writing posture of the user in all aspects. If the user's sitting posture is correct, but the writing posture is abnormal, it will still cause strain on the user's shoulders, waist or arms, and affect the writing effect and learning efficiency of the user. SUMMARY

[0005] In order to solve the problem that the prior art cannot detect the sitting posture and writing posture of the user in the writing process in all aspects, and thus cannot play an effective supervision and correction effect, the present application provides a writing posture supervision method.

[0006] The present application adopts the following technical scheme:

[0007] A writing posture supervision method for supervising the sitting posture, pen holding posture and writing posture of a writing user. The writing posture supervision method comprises the following steps:

[0008] S1: taking the long side and short side of a horizontal rectangular desktop as X and Y axes respectively, and taking the vertical direction as Z axis, a space coordinate system is established.

[0009] S2: Obtain the lighting effect of the surface lamp at various places on the rectangular table top, and define the area in the rectangular table top that meets the lighting and illumination hygiene standards as the writing plane.

[0010] S3: Real-time acquisition of the positions of the user's two eyes in the spatial coordinate system, and judgment of whether the user's sitting posture is correct according to the relative positions of the two eyes and the rectangular table top:

[0011] (1) When the relative positions of the two eyes and the rectangular table top meet the requirements, proceed to the next step;

[0012] (2) When the relative positions of the two eyes and the rectangular table top do not meet the requirements, an early warning information representing an abnormal sitting posture is issued.

[0013] S4: Obtain the body size data of the user, and determine the body position of the user according to the positions of the user's two eyes; then specify a best writing area that meets the ergonomics in the writing plane in combination with the body size data and the body position.

[0014] S5: Real-time acquisition of the position of the pen tip when the user is writing, and making the following judgments and decisions according to the position of the pen tip of the user:

[0015] (1) When the pen tip position is within the best writing area, proceed to the next step;

[0016] (2) When the pen tip position is outside the best writing area, an early warning information representing an abnormal pen holding posture is issued.

[0017] S6: Generate a virtual writing frame in the best writing area in combination with the position of the pen tip and the pose information of the user in a sitting state.

[0018] S7: Calculate the projection range limits of the pen barrel in the writing frame in combination with the reasonable offset interval of the body during writing, and generate a corresponding writing reasonable projection area.

[0019] S8: Real-time acquisition of the position of any sampling point on the pen barrel close to the pen cap side when the user is writing, and establishment of a virtual pen barrel model in the spatial coordinate system.

[0020] S9: Take the eye position as the light source position and the rectangular table top as the projection surface; calculate the projection position of the pen barrel model on the projection surface under the light source illumination condition according to the spatial projection geometry, and make the following decisions according to the projection position:

[0021] (1) When the projection position is in the writing reasonable projection area, it is confirmed that the user's writing posture is normal;

[0022] (2) When the projection position is outside the writing reasonable projection area, it is confirmed that the user's writing posture is abnormal, and an early warning is issued.

[0023] In the technical solution provided by the application, the demarcation of the writing plane in the rectangular table top is determined by the test personnel referring to the relevant mandatory standards after testing. The test indicators include: average illuminance value, surface uniformity, color temperature and color rendering index, anti-glare, anti-blue light, and stroboscopic performance.

[0024] As a further improvement of the application, in step S3, the indicators for judging whether the user's sitting posture is correct are that:

[0025] (1) The average height value of the user's eyes from the plane of the rectangular table top is within the preset height range.

[0026] (2) The angle between the space connecting line of the user's eyes and the long side of the rectangular table top is within the preset angle range.

[0027] As a further improvement of the application, the method for dividing the optimal writing area includes the following processes:

[0028] I. Prior data generation

[0029] (a) Divide the user group into different body types based on height and BMI index.

[0030] (b) Test the maximum movement range of the pen tip on the table top of each body type user in normal writing state.

[0031] (c) Generate an external rectangle corresponding to the maximum movement range of the pen tip, so that each side of the external rectangle is parallel to each side of the rectangular table top; and record the current user's body position.

[0032] (d) The area of the above-mentioned external rectangle on one side of the body position is designated as the optimal writing area of the current body type user.

[0033] (e) A best writing area distribution mapping table is constructed, which establishes the mapping relationship between each user body type and the distribution of the optimal writing area.

[0034] II. Current user matching

[0035] (e) Obtain the height and weight of the current user, and calculate the user's BMI index.

[0036] (f) Determine the body type of the current user based on height and BMI index.

[0037] (g) According to the body type of the current user, query the optimal writing area distribution mapping table to determine the optimal writing area of the current user.

[0038] As a further improvement of the application, in step S6, the method for generating the writing frame is as follows:

[0039] S61: Establish a plane coordinate system with the direction of the line connecting the two eyes as the horizontal axis and the direction of the perpendicular foot of the pen tip to the horizontal axis as the vertical axis.

[0040] S62: Draw a predetermined size of a cross-shaped grid in the plane coordinate system with the pen tip position as the center P.

[0041] S63: Rotate the cross-shaped grid around the center P until the edges of the cross-shaped grid are parallel to the horizontal axis or the vertical axis of the plane coordinate system; at this time, the cross-shaped grid is the required writing frame.

[0042] As a further improvement of the present application, in step S7, the determination method of the reasonable writing projection area is as follows:

[0043] S01: Select people of different ages, body types, and genders as test subjects.

[0044] S02: Record the relative positions of the eyes and the pen shaft of each test subject when writing normally.

[0045] S03: Record multiple sets of sample data generated by each test subject when the body and pen tip positions are reasonably adjusted, forming a sample data set.

[0046] S04: Determine the writing frame corresponding to each set of sample data in the sample data set, and calculate the projection of the pen shaft in the writing frame in each set of sample data.

[0047] S05: Define the distribution area of all sample data projections in the test process as the reasonable writing projection area.

[0048] As a further improvement of the present application, in step S9, the judgment method of the writing posture is as follows:

[0049] I. Assuming that the collected pen tip position is P and any point on the pen shaft is Q', the vector of the pen shaft is Further, the vector corresponding to the projection of the pen shaft in the writing plane can be calculated as

[0050] II. According to the pre-generated reasonable writing projection area, determine its boundary vectors as and

[0051] III. Define the direction vector of the Z-axis as e z and make the following judgments:

[0052] (1) If the vector mixed product and the vector mixed product then it is determined that the pen shaft projection PQ is distributed in the reasonable writing projection area, and the current writing posture of the user is normal.

[0053] (2) If not satisfied: vector dot product and vector dot product Then it is determined that the pen barrel projection PQ exceeds the reasonable projection area of writing, and the current writing posture of the user is abnormal.

[0054] As a further improvement of the present application, a projection spotlight is arranged on the side corresponding to the lighting source in the lamp, which is used to project a pattern on the lower rectangular table surface, which is consistent with the area generated by the lighting source that meets the lighting and illumination hygiene standards, to prompt the user to the position of the writing plane.

[0055] As a further improvement of the present application, in steps S3, S5 and S8, the images of the corresponding regions of the user's face and hand are collected by the camera respectively; then the regions containing the eyes, pen tip and pen barrel are identified by image recognition algorithm, and finally the spatial positions of the corresponding targets are determined by the image-based target spatial position calculation method.

[0056] As a further improvement of the present application, in steps S3, S5 and S8, a multi-camera or a depth camera is used to obtain the image of the target, and the spatial position of the target is calculated according to the image.

[0057] The technical scheme provided by the present application has the following beneficial effects:

[0058] The present application designs a method for calculating the spatial position of the user's eyes and the pen held by the hand by using images, and then combining the logic of judging the user's sitting posture, pen holding posture and writing posture with the projection position of the eyes, pen tip and pen barrel, and providing a method for real-time supervision of the posture of the user during writing.

[0059] The writing posture supervision method provided by the present application has the advantages of accurate and reliable monitoring effect, strong practicability, and is suitable for large-scale popularization and application, and can overcome the problem that the existing scheme can only judge the user's sitting posture and cannot monitor whether the user's writing posture is abnormal. The writing posture supervision method provided by the present application can detect and correct the posture of the student during learning, and reduce the probability of problems such as abnormal body development, ugly handwriting and vision loss caused by abnormal writing posture in the student group. BRIEF DESCRIPTION OF DRAWINGS

[0060] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0061] Figure 1 The step flow chart of the writing posture supervision method provided in embodiment 1 of the present application.

[0062] Figure 2The spatial position diagram of the writing plane on the desktop in the embodiment 1 of the present application.

[0063] Figure 3 The spatial position diagram of the optimal writing area on the desktop in the embodiment 1 of the present application.

[0064] Figure 4 The step flow chart of the dividing method of the optimal writing area in the embodiment 1 of the present application.

[0065] Figure 5 The state diagram of the pen nib of the user in the optimal writing area in the embodiment 1 of the present application.

[0066] Figure 6 The state diagram of the pen nib of the user out of the optimal writing area in the embodiment 1 of the present application.

[0067] Figure 7 The spatial layout diagram of the writing frame, the writing plane and the optimal writing area in the embodiment 1 of the present application.

[0068] Figure 8 The step flow chart of the generating process of the writing frame in the embodiment 1 of the present application.

[0069] Figure 9 The state diagram of the pen barrel projection position in the reasonable projection area in the embodiment 1 of the present application.

[0070] Figure 10 The state diagram of the pen barrel projection position out of the reasonable projection area in the embodiment 1 of the present application.

[0071] Figure 11 The step flow chart of the generating process of the writing reasonable projection area in the embodiment 1 of the present application.

[0072] Figure 12 The step flow chart of the determining whether the writing posture of the user is abnormal according to the projection vector of the pen barrel in the embodiment 1 of the present application.

[0073] Figure 13 The structural diagram of the intelligent learning table with the posture correction function provided in the embodiment 2 of the present application.

[0074] Figure 14 The structural diagram of the lamp holder part in the intelligent learning table of the embodiment 2 of the present application.

[0075] Figure 15 The module block diagram of the control part in the intelligent learning table of the embodiment 2 of the present application.

[0076] Figure 16 The state diagram of the adjustment of the lighting area of the lamp by the projection lamp in the lamp holder debugging process of the embodiment 2 of the present application.

[0077] Figure 17 The flow chart for performing the posture supervision and correction process of the intelligent learning table in Embodiment 2 of the present application.

[0078] The figure is marked as:

[0079] 1, table; 2, chair; 3, lamp; 4, camera; 5, alarm; 31, lamp holder; 32, lighting source; 33, projection lamp; 311, vertical frame; 312, connecting frame; 313, horizontal frame. DETAILED DESCRIPTION

[0080] In order to make the purpose, technical solutions and advantages of the present application clearer and more apparent, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0081] Embodiment 1

[0082] The present embodiment provides a writing posture supervision method for supervising the sitting posture, pen holding posture and writing posture of a writing user. The writing posture supervision method provided by the present embodiment generally includes the following processes: first, the relative positions of the user's eyes and the desktop are detected, and it is determined whether the user's sitting posture is normal. Second, under the condition that the user's sitting posture is normal, the best range of pen tip on the desktop when the user holds the pen to write, i.e., the best writing area, is determined according to the user's body size data. And it is detected whether the user's actual pen tip position is in the best writing area, which is used as the basis for determining whether the user's pen holding posture is normal. Finally, it is further detected whether the spatial projection of the pen barrel relative to the eyes when the user holds the pen is within a predetermined reasonable spatial range, which is used as the basis for determining whether the user's writing posture is normal.

[0083] Specifically, as shown in Figure 1 The writing posture supervision method provided by the present embodiment includes the following steps:

[0084] S1: Taking the long side and the short side of the horizontal rectangular desktop as the X and Y axes respectively, and the vertical direction as the Z axis, a spatial coordinate system is established.

[0085] S2: The lighting effect of the surface lamp at each place on the rectangular desktop is obtained, and the area in the rectangular desktop that meets the lighting and illumination hygiene standards is defined as the writing plane. The position distribution of the writing plane is as shown in Figure 2

[0086] ​The concept of the writing plane proposed in the embodiment refers to: under the lighting condition, the lighting and illumination hygiene standard on the desktop can meet the area for normal writing needs. Generally speaking, for a specific model of lamp, after it is installed at a specified position above the desktop at a specific illumination angle, the area in the lighting range of the lamp that meets the writing needs is actually fixed; therefore, the technical personnel can determine the actual distribution range of the writing plane above the desktop through actual measurement, and prompt the user with the distribution area of the "writing plane" through corresponding indication information. Of course, for a specific model of lamp, when the relative position of the light source and the desktop and the illumination angle and other parameters are adjusted, the distribution area of the "writing plane" will also be adjusted.

[0087] The scheme provided by the embodiment sets related indexes of the area on the desktop that meets the lighting and illumination hygiene standard, with reference to two national standards, i.e., "Lighting and Illumination Hygiene Standard for Classrooms in Primary and Secondary Schools (GB 7793-2010)" and "Performance Requirements for Reading and Writing Desk Lamps (GB / T 9473-2008)". Among them, the test index items include: average illuminance value, illumination surface uniformity, color temperature and color rendering index, anti-glare performance, anti-blue light performance, and low-frequency flicker performance, etc.

[0088] In order to indicate the distribution area of the writing plane, the embodiment is provided with a projection spotlight on one side of the lighting light source in the lamp, which is used to project a pattern on the lower rectangular desktop surface that coincides with the area meeting the lighting and illumination hygiene standard generated by the lighting light source, so as to prompt the user to the position of the writing plane.

[0089] S3: Real-time collection of the positions of the user's two eyes in the spatial coordinate system, and judgment of whether the user's sitting posture is correct according to the relative position of the two eyes and the rectangular desktop:

[0090] (1) When the relative position of the two eyes and the rectangular desktop meets the requirements, the next step is entered.

[0091] (2) When the relative position of the two eyes and the rectangular desktop does not meet the requirements, an early warning information representing the abnormal sitting posture is issued;

[0092] The indexes for judging whether the user's sitting posture is correct adopted by the embodiment include the following two items, and only when the user meets the following two indexes at the same time, it is determined that the user's sitting posture is normal:

[0093] A. The average height value of the user's two eyes from the plane of the rectangular desktop is within the preset height range.

[0094] B. The angle between the spatial line connecting the user's two eyes and the long side of the rectangular desktop is within the preset angle range.

[0095] For index A, the reasonable height range can be determined in different ways. For example, a determination method based on individual differences of users (method 1) and a determination method based on physiological characteristics of users (method 2), etc. In method 1, the technician can pre-collect the vertical height of the eyes of a specified user from the desktop in the normal writing state, and then record the corresponding height value as a standard value. Then, during the user's writing process, the height value of the user's eyes is continuously detected, and it is judged whether the real-time eye height of the user deviates from the standard value within a preset fluctuation range (such as ± 35%). If yes, it is determined that the current index of the user is normal, otherwise it is determined that the current index of the user is abnormal. In method 2, the technician first tests the eye height fluctuation range of each different height or body type of the user when writing normally, and records a corresponding height interval. Then, the height or body type data of the current user is obtained, and an allowed height fluctuation range is matched for it. Next, the height value of the user's eyes is continuously detected, and it is judged whether the real-time eye height of the user is within the allowed range. If yes, it is determined that the current index is normal, otherwise it is determined that the current index is abnormal.

[0096] For index B, the preset angle range between the eyes of the user and the long side of the desktop is 45°. When the actual angle exceeds 45°, it is determined that the current index is abnormal.

[0097] S4: Obtain the body type data of the user, and determine the body position of the user according to the position of the user's eyes. Then, combine the body type data and the body position to specify an ergonomic best writing area on the writing plane. The position of the best writing area is as shown in Figure 3

[0098] In this embodiment, the reference point of the best writing area is the body position of the user. The best writing area is actually a rectangular area on the desktop with a specified area in the specified direction of the body position. Specifically, as shown in Figure 4

[0099] I. Prior data generation

[0100] (a) Divide the user group into multiple different body types in combination with the height and BMI index.

[0101] In this embodiment, the height index of the user is divided into: below 110 cm, 110-125 cm, 125-140 cm, 140-155 cm, 155-170 cm, 170-180 cm, and above 180 cm, a total of 7 intervals. The BMI value of the user is divided into below 20, 20-25, 25-30, and above 30, a total of 4 intervals. Finally, in combination with the height and weight data, 28 different user body types can be obtained.

[0102] ​​In this embodiment, different height intervals are coded in ascending order, denoted as 1-7, and different BMI indexes are coded in ascending order, denoted as 1-4. At this time, the label HxBy is used to represent the body type category of the user, where x represents the group corresponding to the height of the user, and y represents the group corresponding to the BMI index of the user.

[0103] For example, when the height of a certain user is 175 cm and the weight is 70 kg, the BMI value is calculated as 22.9. At this time, the height of the user belongs to the 6th group, and the BMI index is the 2nd group, so the body type category of the corresponding user is recorded as H6B2.

[0104] (b) Test the maximum motion range of the pen tip on the desktop of each user in the normal writing state in turn.

[0105] (c) Generate an inscribed rectangle corresponding to the maximum motion range of the pen tip, so that each side of the inscribed rectangle is parallel to each side of the rectangular desktop; and record the body position of the current user.

[0106] (d) The area of the above inscribed rectangle on one side of the body position is designated as the optimal writing area of the current user.

[0107] (e) A best writing area distribution database is constructed, and the mapping relationship between each user body type and the respective best writing area distribution is established in the best writing area distribution database.

[0108] II. Current user matching

[0109] (e) Obtain the height and weight of the current user, and calculate the BMI index of the user.

[0110] (f) Determine the body type of the current user in combination with the height and BMI index.

[0111] (g) According to the body type of the current user, query the best writing area distribution database to determine the optimal writing area of the current user.

[0112] S5: Real-time collection of the pen tip position when the user writes, and making the following judgments and decisions according to the pen tip position of the user:

[0113] (1) As shown in Figure 5 , when the pen tip position is within the optimal writing area, the next step is entered;

[0114] (2) As shown in Figure 6 , when the pen tip position is outside the optimal writing area, an early warning information indicating an abnormal pen holding posture is issued;

[0115] S6: Generate a virtual writing frame within the optimal writing area in combination with the pen tip position and the pose information of the user in the sitting state. Specifically, as shown inFigure 7 and Figure 8 As shown, the method for generating the writing box in this embodiment is as follows:

[0116] S61: Establish a planar coordinate system with the direction of the line connecting the two eyes as the horizontal axis and the direction perpendicular from the tip of the pen to the horizontal axis as the vertical axis.

[0117] S62: Draw a grid of preset size in a plane coordinate system with the pen tip position as the center P.

[0118] S63: Rotate the grid around the center P until each side of the grid is parallel to the horizontal or vertical axis of the plane coordinate system; the grid at this point is the required writing frame.

[0119] S7: Based on the reasonable offset range of the human body during the writing process, calculate the limit of the projection range of the pen in the writing frame, and generate a corresponding reasonable projection area for writing.

[0120] S8: Real-time acquisition of the position of any sampling point on the pen barrel near the pen cap when the user writes, and a virtual pen barrel model is built in the spatial coordinate system;

[0121] S9: Using the eye position as the light source position and the rectangular tabletop as the projection surface; calculate the projection position of the pen model on the projection surface under the illumination of the light source based on spatial projection geometry, and make the following decisions based on the projection position:

[0122] (1) As Figure 9 As shown, when the projection position is within the reasonable projection area for writing, it is confirmed that the user's writing posture is normal.

[0123] (2) Figure 10 As shown, when the projection position exceeds the reasonable projection area for writing, it is confirmed that the user's writing posture is abnormal and an early warning is issued.

[0124] In the solution provided in this embodiment, the reasonable projection area for writing reflects the spatial projection relationship between the eyes and the pen barrel for all users in a normal writing posture. This projection relationship has little correlation with physiological differences between individual users, and therefore can be used to characterize the user's current pen-holding posture and writing posture. This embodiment uses the spatial projection relationship between the eyes and the pen barrel as a criterion for evaluating the user's writing posture.

[0125] When using the projection relationship between the eyes and the pen as a criterion for evaluating a user's writing posture, it is also necessary to determine the typical distribution range of the spatial projection between the eyes and the pen when the user is writing normally. This embodiment defines the concept of a reasonable writing projection area to refer to this range, and uses a large number of representative sample data to finally determine a confidence interval for a reasonable writing projection area. Specifically, in step S7 of this embodiment, as follows... Figure 11As shown, the determination method of the reasonable writing projection area is as follows:

[0126] S01: Select people of different ages, body types and genders as test subjects.

[0127] S02: Obtain the relative positions of the eyes and pen barrels of each test subject when writing normally.

[0128] S03: Record multiple sets of sample data generated when each test subject adjusts the body and pen tip positions reasonably to form a sample data set.

[0129] S04: Determine the writing frame corresponding to each set of sample data in the sample data set, and calculate the projection of the pen barrel in the writing frame in each set of sample data.

[0130] S05: Define the distribution area of the projection corresponding to all sample data in the test process as the reasonable writing projection area.

[0131] In the actual application process of the embodiment, the judgment process of the writing posture in step S9 can be analyzed by using vector operation, in combination with Figure 12 , the specific operation process of the embodiment is as follows:

[0132] I. Assuming that the pen tip position collected is P and any point on the pen barrel is Q', the vector of the pen barrel is Further, the vector corresponding to the projection of the pen barrel in the writing plane can be calculated as

[0133] II. According to the pre-generated reasonable writing projection area, the boundary vectors thereof are determined as and

[0134] III. Define the direction vector of the Z axis as e z , and make the following judgments:

[0135] (1) If the vector mixed product and the vector mixed product are satisfied, it is determined that the pen barrel projection PQ is distributed in the reasonable writing projection area, and the current writing posture of the user is normal.

[0136] (2) If the vector mixed product and the vector mixed product are not satisfied, it is determined that the pen barrel projection PQ exceeds the reasonable writing projection area, and the current writing posture of the user is abnormal.

[0137] In the actual application process, the spatial positions of the user's eyes and the pen can be determined by various means. In this embodiment, image recognition algorithms and image-based spatial position detection algorithms are mainly used to realize the position detection of the target objects. In steps S3, S5, and S8, sample images of the corresponding regions of the user's face and hand are collected by the camera; then the regions of the eyes, the pen tip, and the pen shaft contained in the sample images are recognized by the image recognition algorithm; and finally, the spatial positions of the corresponding targets in the sample images are determined by using an image-based target spatial position calculation method. In order to obtain higher-precision position information, in steps S3, S5, and S8, a multi-camera or a depth camera is selected to obtain the sample images of the targets. There are mature related solutions for target recognition and positioning based on images in the prior art. For the technical content of this part, this embodiment will not be described in detail.

[0138] Embodiment 2

[0139] On the basis of the writing posture supervision method provided in Embodiment 1, this embodiment further provides an intelligent learning desk with a posture correction function. As shown in Figure 13 , it includes a table and chair body, lamps and lanterns 3, at least two groups of cameras 4, an alarm 5, and a device manager.

[0140] The table and chair body includes a table 1 with a horizontal desktop and a chair 2. The lamps and lanterns 3 include a lamp holder 31, an illuminating light source 32, and a projection lamp 33. The projection lamp 33 is used to generate guiding information for the installation and adjustment of the lamp holder 31 on the desktop. The lamp holder 31 includes multiple movable joints and has lifting, rotating, and pitching adjustment functions. The cameras 4 are respectively installed on the supports of the lamps and lanterns 3; the two groups of cameras 4 are respectively used to obtain user face images and user hand images. The alarm 5 is used to issue a posture correction prompt message when any kind of abnormal information is received.

[0141] In the intelligent learning desk with a posture correction function provided in this embodiment, as shown in Figure 14 , the lamp holder 31 includes a vertical holder 311, a horizontal holder 313, and a connecting holder 312. The bottom of the vertical holder 311 is fixedly connected with the table 1, and the top is movably connected with one end of the connecting holder 312 through a hinge one. One end of the horizontal holder 313 is movably connected with the other end of the connecting holder 312 through a hinge two. The hinge one and the hinge two enable the horizontal holder 313 to be pitch-adjusted relative to the vertical holder 311. The vertical holder 311 adopts a two-section structure, and at least one section adopts a telescopic holder. The two sections of the vertical holder 311 are connected by a joint one and can be coaxially rotated relative to each other. The horizontal holder 313 also adopts a two-section structure, and the two sections of the horizontal holder 313 are connected by a joint two and can be coaxially rotated relative to each other.

[0142] In the embodiment, the illumination light source 32, the projection lamp 33 and the camera 4 are respectively installed at different positions of the lamp stand 31. Specifically, the illumination light source 32 is installed at the bottom of the horizontal frame 313. The camera 4 includes a first camera 4 located at the side of the lamp 3 close to the user and a second camera 4 located at the ground of the lamp stand 31 close to the desktop, the first camera 4 collects the facial image of the user, and the second camera 4 collects the hand image of the user. The first camera 4 and the second camera 4 in the embodiment are both selected to be a depth camera or a multi-view camera. The projection lamp 33 is installed at the same side of the illumination light source 32 and is used for projecting a pattern on the desktop.

[0143] The device manager includes a control module and a data processing module. As shown in the figure, the control module is electrically connected with the projection lamp 33, the illumination light source 32 and the alarm 5; the data processing module is electrically connected with the camera 4, and the data processing module includes an information acquisition unit, a storage unit, a first judgment unit, a second judgment unit and a third judgment unit. The first judgment unit, the second judgment unit and the third judgment unit are used for judging whether the sitting posture, the pen holding posture and the writing posture of the user are normal respectively, and the three units send the abnormal results to the control module when an abnormality is monitored. Figure 15

[0144] The device manager of the embodiment further includes a communication unit, which is used for communication connection with other devices to realize data transmission between the other devices and the device manager; the other devices send the physiological characteristic data input by the user to the device manager.

[0145] In the data processing module of the device manager, the information acquisition unit is used for acquiring the physiological characteristic data of the user. The information acquisition unit requires the input physiological characteristic data of the user to at least include: age, gender, height, weight and dominant hand. The information of the dominant hand is mainly used for the second judgment unit to determine the distribution direction of the optimal writing area relative to the human eye: when the current user is left-handed, the optimal writing area is located in the specified area relative to the left side of the human eye; when the current user is right-handed, the optimal writing area is located in the specified area relative to the right side of the human eye.

[0146] The storage unit stores a best writing area distribution mapping table and a parameter set of a reasonable writing projection area. The best writing area is used to represent the allowed range of the pen tip landing point when the current user writes on the desktop with a normal posture; wherein the body type of the user is represented by combining height and BMI index; the corresponding best writing areas of two users with the same body type and different dominant hands have symmetry in the distribution position on the desktop.

[0147] The best writing area distribution mapping table establishes the distribution range of the best writing area corresponding to different body types of users. The best writing area is rectangular. In combination with​Figure 12 The data format for defining the optimal writing area distribution mapping table is: {type, a, L, H}; wherein, type represents the body type category of the current user; a represents the horizontal angle of the optimal writing area relative to the position of the eyes; L represents the length of the optimal writing area along the length direction of the table top; and H represents the length of the optimal writing area along the width direction of the table top.

[0148] The first judging unit runs a trained face recognition model. When the first judging unit acquires the user face image collected by the camera 4, the face recognition model can recognize and locate the eyes from the collected user face image, and then calculate the real space positions of the eyes in combination with the camera parameters and the pixel positions of the eyes in the user face image. Finally, whether the relative positions of the eyes of the user and the table top conform to the standard is determined to judge whether the sitting posture of the user is abnormal. If the relative positions of the eyes and the table top conform to the standard, it is indicated that the sitting posture of the user is normal, otherwise, it is determined that the sitting posture of the user is abnormal.

[0149] In the embodiment, the criterion for the first judging unit to judge that the sitting posture of the user is normal is that the following two conditions are met simultaneously:

[0150] (1) the average height value of the eyes of the user from the plane of the rectangular table top is within a preset height range;

[0151] (2) the angle between the space line connecting the eyes of the user and the long side of the rectangular table top is within a preset angle range.

[0152] In a more optimized scheme of other embodiments, the chair 2 and the positions of the seat cushion and the backrest are also provided with pressure sensors; the pressure sensors are in communication connection with the data processing module. The detection results of the pressure sensors are introduced into the judgment of the sitting posture of the user. At this time, the criterion for the first judging unit to judge that the sitting posture of the user is normal is that the following four conditions are met simultaneously:

[0153] (1) the average height value of the eyes of the user from the plane of the rectangular table top is within a preset height range;

[0154] (2) the angle between the space line connecting the eyes of the user and the long side of the rectangular table top is within a preset angle range;

[0155] (3) the fluctuation of the pressure signal at the seat cushion is within a preset amplitude;

[0156] (4) the fluctuation of the pressure signal at the backrest is within a preset amplitude.

[0157] The second judging unit runs an image recognition model of a hand or a pen, which can recognize and calculate the landing position of the pen tip held by the user on the desktop according to the user's hand image collected by the camera 4, and then query the best writing area distribution mapping table to obtain the best writing area of the current user, and finally judge whether the pen holding posture of the user is abnormal according to whether the real landing position is in the best writing area. When the pen tip held by the user lands in the best writing area, it is determined that the current pen holding posture of the user is normal, otherwise, it is determined that the current pen holding posture of the user is abnormal.

[0158] The third judging unit is used to obtain the user's hand image collected by the camera 4, recognize and calculate the spatial position of the pen barrel held by the user. Then calculate the projection position of the pen barrel on the desktop when the human eye is the light source; Finally, it is judged whether the projection position of the pen barrel is in the reasonable writing projection area, and then whether the writing posture of the user is abnormal.

[0159] The judgment process of the third judging unit is as follows:

[0160] (1) Calculate the center position E0 of the human eye, the pen tip position P, and the position Q' of any point on the pen barrel.

[0161] (2) Determine the vector of the pen barrel as and the projection vector of the pen barrel in the desktop when the center position E0 of the human eye is the light source

[0162] (3) Query the parameter set of the reasonable writing projection area to determine the boundary vectors of the reasonable writing projection area as and

[0163] (4) Define the Z-axis direction vector of the space above the desktop as e z , calculate the vector mixed product and the value of the vector mixed product , and make the following judgments and decisions:

[0164] When both are greater than 0, it is determined that the pen barrel projection PQ is distributed in the reasonable writing projection area, and the current writing posture of the user is normal; Otherwise, it is determined that the pen barrel projection PQ exceeds the reasonable writing projection area, and the current writing posture of the user is abnormal.

[0165] In order to make the advantages of the intelligent learning desk with posture correction function provided by the embodiment more obvious, the following further describes the intelligent school desk in combination with the installation test and use process of the intelligent school desk:

[0166] The smart study desk provided in this embodiment uses the same table and chair bodies as conventional tables and chairs, and can be assembled using a similar method. Specifically, for some upgraded models with pressure sensors installed in the seat cushion and backrest, the pressure sensors in chair 2 need to be paired with the device manager built into table 1 via a wireless communication module. After pairing, the pressure sensors send detection data to the device manager at a preset sampling frequency. To ensure product flexibility, the device manager, lamp 3, alarm 5, and camera 4 in table 1 are powered by 220V AC mains power. The pressure sensors and wireless communication module in chair 2 are powered by removable batteries.

[0167] Combination Figure 13 As can be seen, in this embodiment, the lamp 3 is installed on one side of the table 1, and the illumination angle of the light source 32 is adjusted by the lamp holder 31. Considering that the lighting effect of the light source can affect the user's experience and even affect the user's health, this embodiment designs the bracket as follows: Figure 14 The lamp holder 31 is a multi-jointed adjustable type. The vertical frame 311 of the telescopic stand can be used to adjust the height of the light source above the desktop. The rotatable design at both ends of the vertical frame 311 allows users to rotate the horizontal frame 313 in or out from above the desktop. The connecting frame 312 is hinged to both ends of the vertical frame 311 and the horizontal frame 313, allowing the three parts of the lamp holder 31 to perform arm-like bending movements, thereby achieving tilt adjustment of the lighting source 32. Users can change the coverage of the light source and the light intensity above the desktop through the above-mentioned various adjustment methods to meet their lighting needs during writing. In this embodiment, the horizontal frame 313 of the lamp holder 31 also adopts a two-section rotatable structure, which allows users to flexibly adjust the illumination angle or framing direction of the lighting source 32, projection lamp 33, and camera 4 mounted on it.

[0168] After the lamp 3 is assembled, the lighting area needs to be adjusted. In this embodiment, the projection lamp 33 under the crossbeam 313 is used to assist in adjusting the lighting area of ​​the lamp 3. Figure 16 As shown, the specific operation process is as follows:

[0169] 1. Turn on the projector lamp 33. When the projector lamp 33 is turned on, a pattern is generated below to represent the area range that meets the lighting requirements.

[0170] 2. Adjust the connection status of the horizontal frame 313, vertical frame 311 and connecting frame 312 in the lamp holder 31 so that the pattern generated by the projection lamp 33 coincides with the designated area on the table below, where lines will be added to the designated area on the table as guide lines.

[0171] III. After the above adjustment is completed, the states of the hinge one, the hinge two, the joint one and the joint two in the lamp holder 31 are fixed. At this time, the specified area in the lower desktop meets the lighting requirements after the lighting light source 32 is turned on.

[0172] After the smart learning desk is installed, the user should interact with the device manager, bind the user, and input user information to the device manager. This part can refer to the very mature interaction logic of smart home appliances. For example, the user binds the smart learning desk with the electronic device (such as a mobile phone, bracelet, watch, tablet computer, etc.) running a specific application (APP) in the user's hand through the way of scanning the code. Then upload the user data to the device manager in the application. The uploaded user data at least includes: age, gender, height, weight, dominant hand, etc. In addition, for products with pressure sensors installed in the chair 2, the user should also try the product in advance, and the device manager records the initial pressure signals detected by each pressure sensor when the user sits on the chair 2 in the normal use state. In addition, the physiological characteristic data of the user is a key parameter related to the normal operation of the smart learning desk, and also affects the correction effect of the smart learning desk; therefore, these data should be updated regularly. In this embodiment, for users under the age of 18, the initial pressure signal and other physiological characteristic data are re-collected and updated monthly, while for adults, the data update cycle can be appropriately extended.

[0173] After the above initialization is completed, the user can use the smart learning desk with posture correction function to supervise and correct his own writing posture. The use process of the smart learning desk is as shown in Figure 17 , including the following processes:

[0174] In the initial state, the lighting light source 32 is turned off, and the device manager remains in standby state. When the lighting light source 32 is turned on, the smart device manager runs, the first camera 4 collects the face image of the user at the position of the opposite chair 2. When repeated collection of the face image fails multiple times, it is determined that the light source is turned on by mistake, the device manager turns off the lighting light source 32 and returns to standby state.

[0175] When the face image collection is successful, the device manager enters the supervision stage, and the device manager repeatedly performs the following tasks in the supervision stage:

[0176] The first judging unit first analyzes whether the current sitting posture of the user is normal. If yes, the next step is entered. If not, the alarm 5 is driven to issue a reminder that the sitting posture of the user is abnormal, until the user adjusts the sitting posture to be normal. Next, the second judging unit obtains the user's hand image collected by the second camera 4, identifies and calculates the pen tip position of the pen held by the user, and then judges whether the pen tip position is located in the specified optimal writing area. If yes, the next step is entered. If not, the alarm 5 is driven to issue a reminder that the pen holding posture of the user is abnormal, until the user adjusts the pen holding posture to be normal. Finally, the third judging unit identifies the spatial positions of any two points of the pen held by the user, and calculates the spatial position of the pen barrel held by the user. Then, the projection position of the pen barrel on the desktop is calculated when the human eye is the light source. Finally, it is judged whether the projection position of the pen barrel is located in the reasonable projection area for writing. If yes, it is determined that the writing posture of the user is normal. If not, the alarm 5 is driven to issue a reminder that the writing posture of the user is abnormal, until the user adjusts the writing posture to be normal.

[0177] The above merely describes the preferred embodiments of the present application and is not used to limit the present application. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A writing posture supervising method characterized by comprising: It is used for supervising the sitting posture, pen holding posture and writing posture of a writing user, and the writing posture supervision method comprises the following steps: S1: taking the long side and the short side of a horizontal rectangular desktop as X and Y axes respectively, and taking the vertical direction as Z axis, a space coordinate system is established; S2: the lighting effect of the lighting fixture at each place on the rectangular desktop is determined, and the area in the rectangular desktop that meets the lighting and illumination hygiene standards is defined as a writing plane; S3: the positions of the eyes of the user in the space coordinate system are collected in real time, and whether the sitting posture of the user is correct is judged according to the relative positions of the eyes and the rectangular desktop: (1) when the relative positions of the eyes and the rectangular desktop meet the requirements, the next step is entered; (2) when the relative positions of the eyes and the rectangular desktop do not meet the requirements, a warning information representing the abnormal sitting posture is sent out; S4: the body size data of the user is obtained, and the body position of the user is determined according to the positions of the eyes of the user; then a best writing area meeting the ergonomics is specified in the writing plane by combining the body size data and the body position; S5: the position of the pen nib when the user writes is collected in real time, and the following judgment and decision are made according to the position of the pen nib of the user: (1) when the position of the pen nib is in the best writing area, the next step is entered; (2) when the position of the pen nib is out of the best writing area, a warning information representing the abnormal pen holding posture is sent out; S6: a virtual writing frame is generated in the best writing area by combining the position of the pen nib and the position and posture information of the user in the sitting state; S7: the projection range limit of the pen barrel in the writing frame is calculated by combining the reasonable offset interval of the body of the human in the writing process, and a corresponding writing reasonable projection area is generated; S8: the position of any sampling point on the side close to the pen cap of the pen barrel when the user writes is collected in real time, and a virtual pen barrel model is established in the space coordinate system; S9: taking the position of the eyes as the light source position and the rectangular desktop as the projection surface; the projection position of the pen barrel model on the projection surface under the illumination condition of the light source is calculated according to the space projection geometry, and the following decision is made according to the projection position: (1) when the projection position is in the writing reasonable projection area, it is confirmed that the writing posture of the user is normal; (2) when the projection position is out of the writing reasonable projection area, it is confirmed that the writing posture of the user is abnormal, and a warning is sent out.

2. The writing pose supervision method of claim 1, wherein: In step S2, under the illumination state of the specific lighting fixture, the demarcation of the writing plane in the rectangular desktop is determined by the test personnel after testing according to the relevant mandatory standards; the test indexes include: average illuminance value, surface uniformity, color temperature and color rendering index, anti-glare, anti-blue light and stroboscopic performance.

3. The writing pose supervision method of claim 1, wherein: In step S3, the indexes for judging whether the sitting posture of the user is correct include: (1) the average height value of the eyes of the user from the plane of the rectangular desktop is in a preset height range; (2) the included angle between the space connecting line of the eyes of the user and the long side of the rectangular desktop is in a preset angle range.

4. The writing pose supervision method of claim 1, wherein, The demarcation method of the best writing area comprises the following processes: I. Prior data generation (a) combining the height and BMI index, the user group is divided into multiple different body types; (b) the maximum movement range of the pen nib on the desktop of the user of each body type in the normal writing state is tested in turn; (c) generating a circumscribed rectangle corresponding to the maximum movement range of the pen tip, so that the sides of the circumscribed rectangle are parallel to the sides of the rectangular tabletop; and recording the current user's body position; (d) designating the area of the above circumscribed rectangle on one side of the body position as the optimal writing area for the current user, (e) constructing a best writing area distribution mapping table, wherein the mapping relationship between each type of user body shape and the respective optimal writing area distribution is established; II. Current user matching (e) obtaining the height and weight of the current user, and calculating the user's BMI index; (f) determining the user's body shape in combination with the height and BMI index; (g) determining the optimal writing area of the current user according to the user's body shape by querying the best writing area distribution mapping table.

5. The writing pose supervision method of claim 1, wherein: In step S6, the writing frame is generated as follows: S61: establishing a plane coordinate system with the direction of the line connecting the two eyes as the horizontal axis and the direction of the perpendicular foot of the pen tip to the horizontal axis as the vertical axis; S62: drawing a predetermined size of a cross-shaped grid in the plane coordinate system with the pen tip position as the center P, S63: rotating the cross-shaped grid around the center P until the sides of the cross-shaped grid are parallel to the horizontal axis or the vertical axis of the plane coordinate system; at this time, the cross-shaped grid is the required writing frame.

6. The writing pose supervision method of claim 1, wherein: In step S7, the determination method of the reasonable writing projection area is as follows: S01: selecting people of different ages, body shapes, and genders as test subjects; S02: recording the relative positions of the eyes and pen shaft during normal writing for each test subject; S03: recording multiple sets of sample data generated by each test subject when the body and pen tip positions are reasonably adjusted, and constructing a sample data set; S04: determining the writing frame corresponding to each set of sample data in the sample data set, and calculating the projection of the pen shaft in the writing frame in each set of sample data; S05: defining the distribution area of all sample data projections during the test as the reasonable writing projection area.

7. The writing pose supervision method of claim 1, wherein: In step S9, the judgment method of the writing posture is as follows: I. Assuming the collected pen tip position is P, and any point on the pen barrel is Q', then the vector of the pen barrel is The projection of the pen barrel in the writing plane can be calculated as II. According to the pre-generated writing reasonable projection area, determine its boundary vector is and III. Define the Z-axis direction vector as e z and make the following determination: (1) If the following is satisfied: vector mixture product And vector mixture product Then it is determined that the pen barrel projection PQ is distributed in the reasonable writing projection area, and the user's current writing posture is normal; (2) If not satisfied: vector mixed product and vector mixed product then determine that the pen barrel projection PQ exceeds the reasonable projection area of writing, and the current writing posture of the user is abnormal.

8. The writing pose supervision method of claim 1, wherein: A projection spotlight is provided on one side of the lamp corresponding to the lighting source, which is used to project a pattern on the lower rectangular tabletop surface that matches the area generated by the lighting source that meets the lighting and illumination hygiene standards, to prompt the user to write on the position of the plane.

9. The writing pose supervision method of claim 1, wherein: In steps S3, S5, and S8, the images of the corresponding regions of the user's face and hand are captured by the camera; then the regions containing the eyes, pen tip, and pen shaft are identified by image recognition algorithm; finally, the spatial positions of the corresponding targets are determined by image-based target spatial position calculation method.

10. The writing pose supervision method of claim 9, wherein: In steps S3, S5, and S8, a multi-camera or depth camera is used to obtain the image of the target, and the spatial position of the target object is calculated according to the image. In steps S3, S5, and S8, a multi-camera or depth camera is used to obtain the image of the target, and the spatial position of the target object is calculated according to the image.

Images