KR20230109213A - Exercise system through an image recognition-based experiential simulator robot - Google Patents

Abstract

A smart exercise system using an experiential simulator robot based on image recognition of the present invention includes a display that outputs an exercise image including an exercise motion, a camera that photographs a user located near the display and generates a photographed image, and the above A simulator having a robot arm driven according to movement; A database storing a plurality of motion images, a motion output module outputting one of the motion images through the display, and a robot driving module controlling driving of the robot arm according to motion motions of the motion images; An image input module that receives a captured image of a user, a motion analysis unit that analyzes the user's exercise motion based on the captured image, and the exercise motion of the exercise image and the user's motion included in the captured image and a main server including a guide providing module including a matching rate calculation unit that detects a matching rate between the two devices and outputs the matching rate to the display.

Description

Smart exercise system through an image recognition-based experiential simulator robot

The present invention relates to a smart exercise system through an experiential simulator robot based on image recognition. More specifically, an exercise image is output through a display of a simulator, and at the same time, the robot arm is moved to guide the exercise operation, and the user can It relates to a smart exercise system through an experiential simulator robot based on image recognition, which can provide guidelines for exercise by comparing and processing the exercise video and the user's motion after watching and following the exercise video. .

In recent years, modern people are suffering from various chronic diseases, and social demands are also increasing due to the demographic change trend following the entry into an aging society.

Therefore, modern people are reducing the importance of health management compared to before, and the paradigm for health management is shifting from treatment-oriented management through hospitals to consumer-level management that prevents diseases and maintains health. there is a trend

Accordingly, the healthcare market related to health management is also growing, and the healthcare market is growing to account for 18% of the world's GDP. In addition, the accumulated amount of healthcare data is abundant, and global IT companies are attracting attention for technological innovation related to the processing and provision of the healthcare data.

In addition, in relation to such healthcare, an equipment-based health management system using exercise equipment can also be said to be in charge of one side. Characterized in that the purpose is to improve exercise efficiency through.

As a prior art for this, Korea Patent Registration No. 10-2061232 discloses a 'user health management system based on exercise data of exercise equipment'.

The present invention relates to a user health management system based on motion data of an exercise device, comprising: a measurement unit configured to measure motion data of a user including a plurality of motion analysis elements for at least one exercise device; an input unit configured to receive an input about body data including a plurality of body analysis elements from a user; a processing unit generating correlation data by analyzing a correlation between the plurality of motion analysis elements and at least two or more elements among the plurality of body analysis elements, and outputting user-customized health information based on the correlation data; and a display unit displaying the user-customized health information.

However, in the case of such exercise data analysis, input for body data is only received, accurate exercise posture guidance and accuracy judgment of exercise posture rely on human subjective judgment, and therefore exercise system that can accurately perform exercise without a separate trainer is not provided.

Therefore, in order to solve the above-described problems, a simulator equipped with a robot arm guides accurate motion motion through an image and a robot arm, and analyzes the matching rate by photographing the user's motion to provide a more accurate motion method. There is a need to develop a smart exercise system that enables the provision of information.

The main object of the present invention is to increase exercise efficiency by guiding an exercise through an exercise image and a robot arm, and analyzing a degree of agreement with respect to an exercise motion followed by a user.

Another object of the present invention is to reduce diffuse reflection of a display provided in a simulator.

Another object of the present invention is to provide an additive capable of maximizing matting properties.

In order to achieve the above object, a smart exercise system through an experiential simulator robot based on image recognition of the present invention includes a display that outputs an exercise image including an exercise motion, and a user located near the display to photograph the captured image. A simulator having a camera and a robot arm that is driven according to the movement motion; A database storing a plurality of motion images, a motion output module outputting one of the motion images through the display, and a robot driving module controlling driving of the robot arm according to motion motions of the motion images; An image input module that receives a captured image of a user, a motion analysis unit that analyzes the user's exercise motion based on the captured image, and the exercise motion of the exercise image and the user's motion included in the captured image and a main server including a guide providing module including a matching rate calculation unit that detects a matching rate between the two devices and outputs the matching rate to the display.

Furthermore, the surface of the display is characterized in that a reflection softening functional agent containing titanium dioxide (Titanium Dioxide) is coated.

In addition, the antireflection function is prepared by mixing 5 to 15 parts by weight of titanium dioxide nano powder, 10 to 20 parts by weight of polymethyl acrylate, and 60 to 80 parts by weight of purified water to prepare a first solution. doing; preparing a second solution by mixing 90 to 98 parts by weight of the first solution, 0.5 to 5 parts by weight of gum Arabic, and 1 to 3 parts by weight of rosin; 85 to 95 parts by weight of the second solution and 1 to 10 parts by weight of a light-matting reinforcing additive including perfluorooctyltrethoyxysilane and porous silica are mixed to complete a reflection softening functional agent. It is characterized in that it is prepared through; the step of doing.

According to the smart exercise system through the experiential simulator robot based on image recognition of the present invention,

1) It guides the exercise motion through the exercise video, shows the user detailed exercise motion through the robot arm at the same time, and analyzes the filmed video of the user's exercise scene to determine the difference between the user's exercise motion and the exercise motion of the exercise video. By analyzing and providing the coincidence rate, the user can perform the exercise with more accurate motion,

2) Minimize interference by an external light source and prevent glare while lowering the refractive index and reflectance on the display surface and increasing the transmittance through a reflection softening functional agent containing organopolysilazane,

3) Through the matting property enhancing additive, the matting property is maximized based on the numerous pores of porous silica, and at the same time, the density of the particles is increased through silver nanoparticles to increase the film strength, and furthermore, perfluorooctyltrethoxysilane Through this, there is an effect of providing a dense film with high wear resistance.

1 is a conceptual diagram showing a schematic configuration of the system of the present invention.
2 is an exemplary diagram of a simulator of the present invention;
3 is a block diagram showing the configuration of a main server;
Figure 4 is a process diagram showing an example of posture imitation through the simulator of the present invention.
5 is a graph illustrating the operation of the robot arm.
Figure 6 is a photograph showing an example of exercise motion analysis through the motion analysis unit.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The accompanying drawings are not drawn to scale, and like reference numbers in each drawing indicate like elements.

1 is a conceptual diagram showing a schematic configuration of a system of the present invention, and FIG. 2 is an exemplary view of a simulator of the present invention.

Referring to FIGS. 1 and 2 , the smart exercise system through the experiential simulator robot based on image recognition of the present invention is characterized in that it includes a simulator 2 and a main server 1.

The simulator 2 is characterized by having a display 21, a camera 22, and a robot arm 23, through which a motion image is output and a robot arm 23 driven according to the motion of the motion image Through this, it guides the user more accurately to the exercise motion and performs a function of generating a photographed image of the user performing the exercise along with the exercise image.

Therefore, a display 21 capable of outputting an exercise image is provided on one side of the simulator 2, and a photographed image can be generated by photographing a user watching an exercise image and following an exercise motion on one side around the display 21. A camera 22 is provided so that

At this time, the output motion image may be an image including motion of the arm part and motion of the whole body, but most preferably, motion of moving the arm so that the robot arm 23 to be described later can be driven according to the motion. It is based on including action.

Here, the display 21 may be formed of any one of LCD, LED, and OLED, and the display 21 may provide only a simple exercise image output function, but may be implemented as a touch screen and have an input function.

In addition, the simulator 2 is provided with a robot arm 23 on one side, such a robot arm 23 may be provided with one as shown in the drawing, but the robot arm 23 on both sides of the simulator 2 ) is provided one by one, and it is also possible to look as if both arms are provided. At this time, in the case of the robot arm 23, it can be driven according to the motion of the motion image by interlocking with the main server 1 to be described later. The robot arm 23 is characterized in that it can move in various directions, including at least one link, a joint formed on the link, and a motor that rotates the joint.

Furthermore, the simulator 2 may be used standing on one side, but it may be additionally provided with wheels, so that it may be moved through the wheels, of course.

Therefore, the smart exercise system of the present invention allows the user to check the exercise image on the display 21 through the simulator 2 and at the same time provides a guide for a more detailed posture through the movement of the robot arm 23 , When an exercise is actually performed according to the motion image and the driving of the robot arm 23, the camera 22 photographs the user and outputs the match rate between the motion image and the actual user's motion on the display 21 so that the user The effectiveness of the exercise can be enhanced by allowing the person to perform the exercise in a more accurate posture.

In the smart exercise system described above, the main server 1 controls the storage and output of motion images and at the same time controls the driving of the robot arm 23 accordingly, and takes pictures through analysis of the captured images. It analyzes the user's exercise motion from the video and calculates the matching rate between the model exercise motion of the motion image and the motion motion of the photographed image.

That is, in the smart exercise system of the present invention, it can be said that it controls the driving of the simulator 2 and performs the function of analyzing and processing the input data. It can be said to be the subject of analysis and driving control.

Although this main server 1 may be provided separately from the simulator 2, it is also possible to be provided inside the main body of the simulator 2, and it is said that the main server 1 serves as a controller for the simulator 2. You can also.

Such a main server 1 is based on hardware having a central processing unit (CPU) and memory and storage means such as a hard disk, and a program that can be executed in the central processing unit, that is, software is installed and the software can be executed. A series of detailed configurations of the software will be described later as units such as 'modules', 'parts', and 'parts'.

Configurations such as 'modules' or 'parts' or 'interfaces' or 'parts' are software or hardware such as FPGAs or ASICs executed via CPU and memory in a state installed and stored in the storage means of the main server 1 means the work composition of

At this time, the configuration of 'module' or 'unit' or 'interface' is not limited to hardware, and may be configured to be in an addressable storage medium or configured to reproduce one or more processors.

As an example, 'module' or 'part' or 'interface' refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, and attributes. fields, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables.

Functions provided by these 'modules' or 'units' or 'interfaces' may be combined into a smaller number of components and 'units' or 'modules', or may be combined into additional components and 'units' or 'modules'. can be further separated.

In addition, the main server 1 means all types of hardware devices including at least one processor, and may be understood as encompassing software configurations operating in the corresponding hardware devices according to embodiments.

For example, a computing device as an example of a server may be understood as including all of smart phones, tablet PCs, desktops, laptops, and user clients and applications running on each device, but is not limited thereto.

In this way, based on each component constituting the system of the present invention, the configuration of the main server 1 will be described in more detail as follows.

3 is a block diagram showing the configuration of a main server, and FIG. 4 is a process diagram showing an example of posture imitation through a simulator of the present invention.

3 and 4, the main server 1 of the present invention includes a database 100, a motion output module 200, a robot driving module 300, an image input module 400, and a guide providing module. It is characterized by including (500).

The database 100 stores a plurality of exercise images including exercise motions. In this case, the exercise videos include expert exercise motions to enable teaching on posture, or animations for performing exercise motions in exemplary postures. may include That is, a photograph of an exercise expert performing an exercise motion or an animation performing an exercise motion may be stored as an exercise image.

The plurality of exercise images may be classified and stored according to motion names, or may be classified and stored according to body parts that can be strengthened through the corresponding exercise images, or may be classified and stored according to the storage date of the exercise images. It is based on not limiting the exercise motion classification storage method in (100).

The motion output module 200 performs a function of outputting any one of the exercise images stored in the database 100 through the display 21 provided in the simulator 2, and at this time, any one of the stored motion images is randomly selected. It may be extracted and output to the display 21, but more preferably, one of a plurality of exercise images may be selected from a user using the simulator 2, and the selected exercise image may be output through the display 21.

To this end, the display 21 of the simulator 2 may preferably be made of a touch screen, lists the titles or thumbnails of exercise images, outputs them to the display 21, and displays thumbnails of a plurality of exercise images output from the user After selecting one of the titles or titles, the selected exercise image can be output to the display 21 through a method of outputting the selected exercise image to the display 21 and provided to the user.

The robot driving module 300 performs a function of controlling the driving of the robot arm 23 according to the motion of the motion image. Preferably, the robot arm 23 can follow the human arm, When the motion motion included is a motion of extending the arm in a straight line, the robot arm 23 also takes an motion of extending the arm in a straight line, and when the motion motion included in the motion image is a motion of bending the arm at a specific angle, the robot arm 23 also The drive may be controlled to bend at a specific angle.

Therefore, preferably, the driving of the robot arm 23 is differentially controlled according to the motion included in the motion image stored in the database 100, and for this purpose, data for driving the robot arm 23 is stored in the database 100. It is also possible that is stored for each exercise image. In other words, in the case of a motion image including a motion motion in which the arm is repeatedly bent and extended at an angle of 30 °, the driving of the robot arm 23 is controlled to repeat bending and unfolding at an angle of 0 ° together with the motion image. . Furthermore, as described above, the data programmed to control the driving of the robot arm 23 is stored together in the database 100, so that the driving of the robot arm 23 can be controlled according to the data stored in the database 100. .

Here, the robot arm 23 may be most preferably a 3-axis chain type robot. This three-axis chain type robot arm is characterized in that it can be driven by imitating the motion included in the motion image as much as possible to fold or unfold through a joint and to rotate as much as possible.

At this time, a mathematical expression in Forward Kinetics, which is a mathematical expression method for driving the robot arm 23, is described as follows.

5 is a graph illustrating driving of a robot arm.

Based on the graph of FIG. 5, the equation for explaining the driving of the robot arm 23 is as follows.

First, when a series of joint angles are given through the regular kinematics method, the position and orientation of the end-effector on Cartesian coordinates can be obtained through the following method. In other words, a method for determining the position/posture of the distal end of the robot with respect to a given joint variable may be performed as follows.

Equation 1,

Equation 2,

(here, is the coordinate value (position) in the x direction, is the coordinate value (position) in the y direction, is the length of the first link, is the length of the second link, is the length of the third link, is the angle of the first joint, is the angle of the second joint, means the angle of the third joint.)

Next, when the position and attitude (Euler angle) of the distal end of the robot arm 23 are given through inverse kinematics, the process of determining the rotation angle of the corresponding joint may be performed as follows. Here, both theoretical and numerical methods may be applied to determine the rotation angle by the inverse kinematics method.

The closed form solution means obtaining the rotation angle of each driving joint directly at the position level for a given distal end position path in closed form, and is mainly used in industrial robot arms. It includes algebraic methods and geometric methods using homogeneous transformations.

The numerical solution means to obtain the velocity of each drive joint numerically by calculating the Jacobian inverse matrix given the velocity path of the distal end. Due to the iterative algorithm to reach the range of convergence, the time to find a solution is longer compared to the theoretical method. Also, there is no guarantee that all solutions will be found in the presence of multiple solutions.

Therefore, preferably, in the present invention, the rotation angle of the joint may be determined through a theoretical analysis method, and the following equation may be applied here.

Equation 3,

Equation 4,

Equation 5,

Equation 6,

Equation 7,

Equation 8,

Equation 9,

(here, is the coordinate value (position) in the x direction, is the coordinate value (position) in the y direction, is the length of the first link, is the length of the second link, is the length of the third link, is the angle of the first joint, is the angle of the second joint, means the angle of the third joint.)

Therefore, the rotation angle of the joint constituting the robot arm 23 or the position/posture of the robot end can be determined according to the motion through forward kinematics and inverse kinematics, and conventional robot arm control methods can be applied in addition to the methods mentioned above. Of course there is.

Continuing the description with reference to FIGS. 3 and 4 again, the image input module 400 views the exercise image output through the display 21 and the driven robot arm 23 and photographs a user following the exercise motion. As a function of receiving a captured image, this may be preferably implemented by receiving a captured image captured through the camera 22 of the simulator 2 .

Preferably, the motion image is output and the robot arm 23 is driven through the motion output module 200 and the robot drive module 300, and the camera 22 is output when the motion image is output and the robot arm 23 is driven. ) controls driving of the camera 22 to generate a captured image, thereby generating a captured image when the user executes an exercise motion, and transmitting the generated captured image to the main server 1.

6 is a photograph showing an example of exercise motion analysis through a motion analyzer.

Referring to FIG. 6 together with FIGS. 3 and 4, the guide providing module 500 analyzes the user's exercise motion based on the captured image, and identifies the matching rate between the motion motion of the motion image and the motion motion of the captured image. Then, it performs a function of outputting it to the display 21. To this end, the guide providing module 500 may include a motion analyzer 510 and a match rate calculator 520 .

The motion analysis unit 510 performs a function of analyzing the user's exercise motion by analyzing the input captured image. At this time, analyzing the user's motion is most preferably, as in the configuration of FIG. 6 described above, by analyzing the length of each joint and the angle of each joint constituting the user's arm, the rotation angle of each joint, and further, the direction of each joint. It can be grasped by calculating the direction and the coordinates of the tip of the hand, and in addition, the movement of the user's arm can be analyzed in various ways through a conventional software program. In other words, it can be said to analyze which direction the arm was stretched, how each joint was moving, and how much each joint was rotated in which direction.

To this end, an object, that is, a user, must be recognized in the photographed image, and among them, the user's arm and each node constituting the arm, which is the subject of analysis, must be recognized. Furthermore, it recognizes the joints that make up the nodes, and based on the direction the arm is facing and the coordinates of the fingertips, in which direction the arm is extended, how each node is moving, and how each joint is rotated in which direction. It can be said to figure out what happened. This may be done based on software loaded on the main server 1 .

The coincidence rate calculating unit 520 performs a function of detecting the coincidence rate between the motion motion of the exercise image and the motion motion of the photographed image and outputting it to the display 21 . In other words, the rate of agreement between the user's motion analyzed through the motion analysis unit 510 and the motion motion of the photographed image is determined.

At this time, it is possible to compare the motion motion of the expert or animation character included in the captured image with the motion motion of the user analyzed through the motion analysis unit 510, but more conveniently, the motion motion of the robot arm 23 and the motion motion of the user. Based on the comparison, the concordance rate can be determined.

In the case of the robot arm 23, since more detailed analysis can be performed based on the rotation angle of the joint or the coordinates of the end in driving, the result of analysis of the motion motion of the robot arm 23 and the motion motion of the user is compared and processed. It is possible to perform a simpler comparison by identifying the digitized concordance rate.

That is, the rotation angle of each joint/joint, furthermore, the coordinates of the extremity are compared and processed to compare the motion agreement rate. In this case, the agreement rate may be expressed as a digitized score, preferably with 100% being the perfect score. That is, the matching rate may be 80%, 90%, 100%, etc., and the motor motion of the robot arm 23, that is, the motor motion of the motion image and the user's motor motion perfectly match, the closer to 100% the coincidence rate can indicate

Therefore, the match rate calculated in this way is output to the display 21 to determine how accurately the user is performing the exercise when compared with the provided exercise image, and at the same time, the match rate is a preset reference value (for example, 85%). ), it is possible to provide a guide that allows the user to perform an exercise with an accurate motion by outputting another exercise image only when an abnormality is indicated.

Therefore, according to the smart exercise system through the experiential simulator robot based on image recognition of the present invention, the exercise motion is guided through the exercise image and at the same time, the detailed exercise motion is shown to the user through the robot arm 23, and the user can exercise. It provides an effect that allows the user to perform the exercise with more accurate motion by analyzing and providing the coincidence rate between the user's exercise motion and the exercise motion of the exercise video by analyzing the captured image of the scene.

Here, a captured image is received and stored whenever the user performs an exercise, and the captured image may have a large capacity according to accumulation, so that it occupies a large amount of storage space in a storage unit that may be separately provided in the system of the present invention.

In order to solve these disadvantages, the video input module 400 of the present invention generates a plurality of frames from a live video, extracts a plurality of frames, sets them as main frames, and edits the main frames time-sequentially. It may include an image compression processing unit 410 that compresses the capacity of the captured image and an image storage unit 420 that stores and processes the compressed captured image.

The image compression processing unit 410 generates a plurality of frames from a captured image input through the camera 22, extracts a plurality of frames to generate a main frame, and then sequentially sequentially converts the generated main frames into time series. It is possible to generate a captured image whose capacity is compressed by editing. To this end, the image compression processing unit 410 includes a reset execution determination part 411, a frame generation process part 412, a sector generation part 413, and a main frame extraction part 414.

The reset execution determination part 411 compares the capacity of the input captured image with a preset reference value to determine whether to reset the captured image capacity, and the capacity of the captured image exceeds the reference value preset by the system administrator of the present invention. , it is determined that the captured image capacity is reset, and if it is less than or equal to the reference value, it is determined that the reset is not performed. In addition, it is also possible to reset the capacity of the captured image when the cumulative capacity of the stored captured image exceeds a reference value.

At this time, if it is determined that the captured image capacity is to be reset, a process for resetting the captured image capacity is performed through a frame generation processing part 412, a sector generation part 413, and a main frame extraction part 414, which will be described later.

When it is determined that the capacity of the captured image is to be reset, the frame generation processing part 412 generates a plurality of frames by dividing the captured image at regular time intervals. Frames can be generated by dividing an image, and a time interval for generating frames can be arbitrarily set in the system of the present invention or set in advance by a system manager.

The sector creation part 413 creates a plurality of sectors by grouping a plurality of frames. For example, assuming that 1000 frames are generated from a captured image, the frames are grouped by 10 to create a total of 10 sectors. can

The main frame extraction part 414 extracts at least one frame for each sector and sets it as a main frame. Looking back at the above example, assuming that each sector includes 10 frames, at least one of the 10 frames is selected. One can be arbitrarily selected and extracted as the main frame. In this way, if one main frame is selected for each sector, 10 main frames can be extracted from a 3D image including a total of 10 sectors.

Accordingly, the image compression processing unit 410 may re-edit the main frame extracted in the main frame extraction part 414 to regenerate the captured image, thereby producing a captured image that is more compressed than the existing captured image and has a smaller capacity.

As such, the captured image having a compressed capacity is stored in the image storage unit 420, so that the storage space can be managed more efficiently.

In the simulator 2 installed outdoors or indoors, preferably indoors such as inside an exercise center, when light generated from a light source such as a separate light provided indoors is reflected on the surface of the display 21, it is caused by light reflection. Motion images are difficult to be easily identified by users and may cause users to feel glare.

To prevent this, a surface of the display 21 of the simulator 2 may be coated with an antireflection function agent. The antireflection functional agent performs a function of preventing light reflection of an external light source by including organopolysilazane having a function of preventing light reflection as an active ingredient.

Here, the organopolysilazane included in the antireflection functional agent has high transparency and at the same time performs a function of imparting hydrophobicity, water resistance, detergency, and antifouling properties to the coated surface. Furthermore, when curing by heat treatment, organopolysilazane is converted into silica to form a number of pores inside the silica film, and the pores formed inside the film lower the refractive index and reflectance and increase the transmittance to prevent reflection of light from an external light source. High transparency can be guaranteed.

Furthermore, the reflection moderating functional agent may further include an additional composition in addition to the organopolysilazane, and the reflection moderating functional agent is prepared by preparing a first solution, preparing a second solution, and completing the reflection moderating functional agent. It can be manufactured through the steps of

Preparing the first solution

First, a first solution is prepared by mixing 5 to 15 parts by weight of organo polysilazane, 10 to 20 parts by weight of polymethyl methacrylate, and 60 to 80 parts by weight of purified water.

As described above, organopolysilazane imparts hydrophobicity, water resistance, detergency, and antifouling properties to the coating surface, lowers refractive index and reflectance, and increases transmittance, and is dissolved in purified water as a solvent.

Furthermore, the added polymethyl methacrylate is an adhesive with excellent transparency, high refractive index, and strong chemical reaction, and has excellent corrosion resistance. As it is added to the antireflection additive, it increases adhesion to the surface of the display 21 and contributes to the creation of a highly transparent coating film.

Preparing the second solution

Subsequently, a second solution is prepared by mixing 90 to 98 parts by weight of the first solution, 0.5 to 5 parts by weight of gum Arabic, and 1 to 3 parts by weight of rosin.

Gum arabic is a rubber made by drying liquid secreted from trees of the genus Acacia, and as a substance easily soluble in purified water, it increases tackiness and viscosity and improves emulsion stability. Therefore, it is added to enhance the emulsion stability in mixing various components while assisting the adhesiveness of the antireflection functional agent. Such gum arabic exhibits a pale yellow color but has a transparent property, so it has a property that does not impair the transparency of the coating agent.

Rosin is a natural resin obtained by distilling rosin, which contains abietic acid as a main component and resin acids such as neoabietic acid, lepopimaric acid, hydroabietic acid, pimaric acid, and dextonic acid. It is pale yellow or brown, transparent, has a glassy luster, and exhibits a high degree of adhesion. When added to the antireflection functional agent, it increases the adhesive strength and imparts a glassy luster to the coating film formed at the same time.

Steps to complete the antireflection functional agent

Finally, 90 to 95 parts by weight of the second solution, 1 to 5 parts by weight of benzotriazol, perfluorooctyltrethoyxysilane and porous silica. 1 to 10 parts by weight of additives are mixed to complete the antireflection functional agent.

Benzotriazole is a colorless, particulate compound with a needle-like crystal structure, which can be added to function as a stabilizer for UV. Such benzotriazole can help to reduce and delay the penetration of ultraviolet light, thereby preventing the coating film coated on the surface of the display 21 from being oxidized due to exposure to ultraviolet light, thereby reducing its quality.

Finally, a matting-enhancing additive is added to suppress diffused reflection more efficiently. This matting-enhancing additive basically contains perfluorooctyltrethoyxysilane and porous silica. to be characterized

In the case of porous silica, it is a silica with pores inside, and since the size of the particles is different from that of silica produced by curing organopolysilazane, it has irregularity and has a function of enhancing the anti-reflective effect of the surface.

Perfluorooctyltroxysilane is a fluorosilane-based compound that improves abrasion resistance and enhances the dispersibility of porous silica, thereby assisting the matting effect.

As the surface of the display 21 provided in the simulator 2 is coated with such an antireflection functional agent, the light quenching property of the surface of the display 21 is increased to reduce diffuse reflection caused by external lighting or light sources, thereby improving the user's performance. At the same time as preventing glare, it is possible to provide a clearer motion image, and furthermore, the coating film has a high level of adhesion so that it is not easily separated and has antifouling, water repellency, and water resistance.

The present invention will be described in detail through Examples and Comparative Examples below. The following examples are only exemplified for the practice of the present invention, and the content of the present invention is not limited by the following examples.

The examples are composed of Examples 1 and 2, which are coating compositions containing the organopolysilazane of the present invention, and comparative examples are conventional coating compositions.

Table 1 is a table showing component ratios of Examples 1 and 2 and Comparative Examples. (Unit: % by weight) Example control example One 2 organopolysilazanes 8.9 8.9 - polymethyl methacrylate 13.4 13.4 13.4 Purified water 67.0 77.7 77.7 sword arabic 1.9 - - rosin 2.8 - - Benzotriazole 1.0 - - Perfluorooctyltrethoxysilane 2.5 - - porous silica 2.5 - - tetraethoxysilane - - 8.9

Each component was mixed according to the component ratio of Table 1 to prepare a coating composition, and each composition was coated on a glass substrate as follows.

A glass substrate having a width of 15 cm, a length of 15 cm, and a thickness of 1 cm was prepared, and the glass substrate was washed to remove dust, oil, organic compounds, and contaminants present on the substrate.

Coating compositions of Examples 1 and 2 and Control Example were applied to the glass substrate as a bar coating, respectively, to form a coating film.

The coating film was completed by a low-temperature curing treatment at 400 ° C. for 10 minutes.

The following experiments were conducted on the completed coating film, and the results of the experiments are as follows.

[Experiment 1] Reflectance test

Specular reflectance was measured at an incident angle of 5° and an exit angle of 5° in a wavelength range of 380 to 780 nm by attaching an adapter MPC 2200 to a spectrophotometer UV 2450, and an average reflectance of 450 to 650 nm was calculated.

[Experiment 2] Transmittance test

In accordance with ASTM D 1003, the total light transmittance was measured using a transmittance meter (HM-150).

[Experiment 3] Surface hardness test

Using a pencil hardness tester, a 500 g load was applied and the pencil hardness of the coating films prepared from Examples and Comparative Examples was measured. As for the pencil, a Mitsubishi product was used, and the test was performed 5 times per pencil hardness.

[Experiment 4] Water droplet contact angle experiment

In order to examine the degree of hydrophilicity of the formed coating film, distilled water was dropped on the surface of the porous silica film and the water droplet contact angle was measured.

Table 2 is a table showing the experimental results. Example 1 Example 2 comparative example reflectivity(%) 1.2 1.4 2.2 Transmittance (%) 95.4 95.8 90.4 surface hardness 7H 6H 4H Water droplet contact angle (°) 14 17 24

As shown in the above Table 2, the coating composition containing the antireflection functional agent of Examples 1 and 2 of the present invention has lower reflectance and higher transmittance than the coating composition of Comparative Example, and is excellent in surface hardness and water droplet contact angle. properties can be identified.

Furthermore, the above-described additive for enhancing matting properties may further include an additional composition in addition to perfluorooctyltrethoyxysilane and porous silica. This additive for enhancing matting properties may be prepared through the following process. .

Preparing a Silica Dispersion

First, a silica dispersion is first prepared by mixing 20 to 40 parts by weight of porous silica, 30 to 50 parts by weight of distilled water, and 20 to 40 parts by weight of silver nitrate.

Here, since the use of porous silica can maximize the effect of preventing irregular reflection through surface irregularities and enhancing the matting property based on pores, porous silica in the form of particles is used as a starting material for the matting property enhancing additive.

Furthermore, by dispersing the porous silica in distilled water and simultaneously providing silver nanoparticles based on silver nitrate, dense filmability based on the silver nanoparticles is provided. At this time, a heating and stirring process may be accompanied to dissolve the silver nitrate, and there is no particular limitation on this.

Preparing a Silver Reduction Solution

When the preparation of the silica dispersion is completed, 10 to 30 parts by weight of the prepared silica dispersion, 40 to 60 parts by weight of N,N-dimethylformamide, and 5 to 60 parts by weight of polyvinylpyrrolidone A silver reducing solution is prepared by mixing 15 parts by weight.

N,N-dimethylformamide is added as a polar organic solvent that will serve as a medium between the silica dispersion and the precursor aqueous solution to be described later, and acts as a reducing agent for silver ions contained in silver nitrate to reduce silver ions to silver nanoparticles perform the function Furthermore, it functions as a solvent capable of mixing both polyvinylpyrrolidone and silica dispersion.

In the case of polyvinylpyrrolidone, it is added to improve the adhesion of the coating film created through the reflection softening agent, and at the same time, it serves as a reducing agent for silver ions together with N,N-dimethylformamide to form silver nano It performs the function of obtaining particles.

Preparing an aqueous precursor solution

Next, an aqueous precursor solution is prepared by mixing 15 to 35 parts by weight of methyltrimethoxysilane, 20 to 40 parts by weight of distilled water, and 30 to 60 parts by weight of methanol.

Methyltrimethoxysilane is added to improve transparency by increasing light transmittance while enhancing the effect of forming a coating film through a matting enhancing additive. At this time, distilled water and methanol are used as solvents for methyltrimethoxysilane.

Preparing a Silica Sol Solution

When the preparation of the silver reducing solution and the aqueous precursor solution are all completed, 20 to 60 parts by weight of the silver reducing solution and 40 to 80 parts by weight of the aqueous precursor solution are stirred at 20 to 40 ° C. for 1 to 4 hours, and then heated at 15 to 30 ° C. to 200 hours to prepare a silica sol solution.

At this time, the prepared silica sol has a property in which the porous silica is mixed with particulate silver nanoparticles and the optical properties are improved through methyltrimethoxysilane. Furthermore, after the stirring treatment, a maintenance treatment is performed at 15 to 30° C. for 160 to 200 hours, through which the silica sol is stabilized to obtain a chemically stable silica sol solution.

Steps to complete the matting enhancement additive

Finally, 98 to 99.5 parts by weight of the silica sol solution and 0.5 to 2 parts by weight of perfluorooctyltrethoyxysilane are mixed and maintained at 20 to 35 ° C. for 1 to 3 hours to complete the matting enhancement additive. do.

Perfluorooctyltrethoxysilane is added last as a fluorosilane-based compound, and since it is well soluble in distilled water and methanol included in the silica sol solution, it can be uniformly mixed in the quenching additive. At this time, when perfluorooctyltroxysilane is added, the abrasion resistance is improved and the dispersibility of the silica sol is further increased so that the coating film in which the matting enhancing additive is mixed can form a more dense film.

In the case of the additive for enhancing matting property prepared in this way, the matting property is maximized based on the plurality of pores of porous silica, and at the same time, the density of the particles is increased through silver nanoparticles to increase the film strength, and furthermore, the perfluorooctyl Through retoxysilane, it provides an effect capable of providing a dense film with high abrasion resistance.

As described so far, the configuration and operation of the smart exercise system through the experiential simulator robot based on image recognition according to the present invention are expressed in the above description and drawings, but this is only an example and the spirit of the present invention is not reflected in the above description. And it is not limited to the drawings, and various changes and changes are possible without departing from the technical spirit of the present invention.

1: Main Server 2: Simulator
21: display 22: camera
23: robot arm 100: database
200: motion output module 300: robot drive module
400: video input module 410: video compression processing unit
411: reset execution determination part 412: frame generation processing part
413: sector creation part 414: main frame extraction part
420: image storage unit 500: guide providing module
510: motion analysis unit 520: coincidence rate calculation unit

Claims (7)

As a smart exercise system through an experiential simulator robot based on image recognition,
A simulator having a display that outputs an exercise image including an exercise motion, a camera that captures a user located near the display and generates a photographed image, and a robot arm driven according to the exercise motion included in the motion image;
A database storing a plurality of motion images, a motion output module outputting one of the motion images through the display, and a robot driving module controlling driving of the robot arm according to motion motions of the motion images; An image input module that receives a captured image of the user, a motion analyzer that analyzes the motion of the user based on the captured image, and a rate of agreement between the motion of the motion image and the motion of the user is determined. Characterized in that, the smart exercise system comprises a; main server including a guide providing module including a matching rate calculation unit for outputting to the display. According to claim 1,
The simulator,
Characterized in that it is movable through wheels, a smart exercise system. According to claim 1,
The video input module,
An image compression processing unit generating a plurality of frames from the captured image, extracting the plurality of frames, setting them as main frames, and compressing the capacity of the captured image by time-sequentially editing the main frames;
A smart exercise system comprising an image storage unit for storing the compressed image. According to claim 3,
The video compression processing unit,
a reset execution determining part for determining whether to reset the captured image capacity by comparing the capacity of the captured image with a preset reference value;
A frame generation processing part generating a plurality of frames by dividing the captured image at predetermined time intervals when it is determined that the capacity of the captured image is to be reset;
a sector creation part for generating a plurality of sectors by grouping a plurality of the frames;
And a main frame extraction part for extracting at least one frame for each sector and setting it as a plurality of main frames. According to claim 1,
On the surface of the display,
A smart exercise system, characterized in that a reflection softening functional agent containing organopolysilazane is coated. According to claim 5,
The antireflective function agent,
preparing a first solution by mixing 5 to 15 parts by weight of organo polysilazane, 10 to 20 parts by weight of polymethyl methacrylate, and 60 to 80 parts by weight of purified water;
preparing a second solution by mixing 90 to 98 parts by weight of the first solution, 0.5 to 5 parts by weight of gum Arabic, and 1 to 3 parts by weight of rosin;
90 to 95 parts by weight of the second solution, 1 to 5 parts by weight of benzotriazol, perfluorooctyltrethoyxysilane and porous silica. to 10 parts by weight to complete a reflection softening functional agent; characterized in that manufactured through, a smart exercise system. According to claim 6,
The matting property enhancing additive,
preparing a silica dispersion by mixing 20 to 40 parts by weight of porous silica, 30 to 50 parts by weight of distilled water, and 20 to 40 parts by weight of silver nitrate;
A silver reduction solution is prepared by mixing 10 to 30 parts by weight of the silica dispersion, 40 to 60 parts by weight of N,N-dimethyl formamideanhydrous, and 5 to 15 parts by weight of polyvinylpyrrolidone doing;
preparing an aqueous precursor solution by mixing 15 to 35 parts by weight of methyltrimethoxysilane, 20 to 40 parts by weight of distilled water, and 30 to 60 parts by weight of methanol;
Preparing a silica sol solution by stirring 20 to 60 parts by weight of the silver reducing solution and 40 to 80 parts by weight of the precursor aqueous solution at 20 to 40 ° C. for 1 to 4 hours and maintaining at 15 to 30 ° C. for 160 to 200 hours ;
98 to 99.5 parts by weight of the silica sol solution and 0.5 to 2 parts by weight of perfluorooctyltrethoyxysilane are mixed and maintained at 20 to 35 ° C. for 1 to 3 hours to complete a matting property enhancing additive; Characterized in that it is manufactured through, a smart exercise system.