KR20230099896A - Object information recognition and prediction system applying virtual reality technology to rehabilitation treatment - Google Patents

Abstract

The present invention relates to an object information recognition and prediction system to which virtual reality technology is applied to rehabilitation treatment, and more particularly, by performing numerical analysis based on a 3D model, feature points, and mark information loaded from a 3D model library, The position, direction, and type of the captured object can be grasped, and the 3D modeling information, current position, direction, and future predicted position and direction, which are loaded from the 3D model library, are input to the game logic, so that the virtual It can be reproduced with images and sounds, and by applying game elements using virtual reality (VR) technology to rehabilitation treatment, it is possible to improve the treatment effect by helping patients undergoing rehabilitation treatment overcome the tedious and difficult treatment process, By replacing the visual and auditory signals with the visual and auditory signals of virtual reality with game elements while leaving the tactile sense that the patient feels when interacting with objects without discomfort, the patient continues to feel a sense of achievement even before the therapeutic effect appears. It is possible to motivate by making it feel, and by matching the type, location, action/reaction of the object that the patient feels with touch, and the visual stimulus and auditory signal that can be predicted from it, visual and auditory signals are combined with virtual signals. When replacing it, there is an effect that the rehabilitation subject does not feel confused.

Description

Object information recognition and prediction system applying virtual reality technology to rehabilitation treatment}

The present invention performs numerical analysis based on 3D models, feature points, and mark information loaded from a 3D model library, so that the position, direction, and type of an object caught in a current image can be grasped, and the 3D modeling information loaded from the 3D model library As the current position, direction, and predicted position and direction of the future are input into the game logic, it can be reproduced as virtual images and sounds according to the conversion of the game logic, and the game-like By applying elements, it is possible to improve the treatment effect by helping patients undergoing rehabilitation treatment overcome the tedious and difficult treatment process, and visual and auditory signals can be displayed in a game-like way while leaving the tactile sense that patients feel when interacting with objects. By replacing the visual and auditory signals of virtual reality with elements without discomfort, it is possible to motivate the patient by continuously feeling a sense of achievement even before the treatment effect appears, and the type, location, and action of objects felt by the patient's tactile sense /Virtual reality technology is applied to rehabilitation treatment that can prevent the rehabilitation subject from feeling confused when visual and auditory signals are replaced with virtual signals by matching the reaction and the visual stimulus and auditory signal that can be predicted from it. It is a technology related to object information recognition and prediction system.

Due to the development of science and technology and medical technology, the aging of the population is rapidly progressing not only in Korea but also around the world, and industrial trends are changing to prepare for an aging society.

The field of greatest interest is healthcare. According to the National Health Insurance Corporation, as a result of analyzing health insurance treatment cost payment data from 2005 to 2010, the total medical cost due to geriatric diseases increased by 271% and the public corporation co-payment increased by 272.5%. It shows the importance of health care for the elderly.

One of the most representative of the geriatric diseases is degenerative joint disease, which recently occurs not only in the elderly and middle/elderly, but also in the young.

The main causes for this are incorrect posture habits, excessive intake of sodium and sugar, weight gain, heavy work and stress, lack of exercise or joint injuries due to strenuous exercise.

Rehabilitation exercise is very important because joints do not recover immediately even after surgery, but the range of joint motion and the speed of recovery vary according to subsequent recovery treatment.

However, since a lot of pain is accompanied during joint rehabilitation exercise, it is difficult for patients to exercise on their own, and eventually, limitation of motion (LOM) appears due to contracture and inflammation.

To prevent this, range of motion exercises are performed. The most common type of joint motion is CPM (Continuous Passive Motion), which is a motion performed by pure external force such as a therapist or mechanical device. It restores the function of the joint by continuously providing passive movement to patients who have come or undergo surgery.

Since CPM is a passive exercise by a mechanical device, not by the patient's own movement, it helps to recover the range of motion (ROM) without causing muscle fatigue and helps to reduce pain and prevent complications. As a result, the number of companies lending CPM machines so that joint disease patients can exercise at home has also increased significantly.

However, existing CPM devices have a problem in that a physical therapist or a patient himself arbitrarily sets an exercise angle, speed, number of times, etc., and it is practically difficult to change the initially set exercise setting in real time according to the patient's condition.

In addition, there is also a problem that side effects occur due to excessive exercise according to these problems, or rehabilitation exercise is not properly performed due to too weak exercise.

In addition, since the degree of recovery of the patient cannot be grasped with only the CPM device, there is a problem that customized rehabilitation treatment cannot be realized unless data such as pain during exercise for each patient and data such as ROM, number of times, and time are separately recorded and managed.

On the other hand, in general, since the treatment effect through rehabilitation appears gradually, it is difficult for patients to know their achievement and is highly likely to give up in the middle. The treatment effect can be improved by enabling the patient to overcome the tedious and difficult treatment process, and the visual and auditory signals of virtual reality with game elements can be provided while leaving the tactile sense that the patient feels when interacting with objects. Since it is replaced without discomfort, it is possible to motivate the patient by continuously making the patient feel a sense of achievement even before the treatment effect appears, and the type, location, action/reaction of the object felt by the patient's tactile sense, and the visual stimulus that can be predicted from it However, since the auditory signal is matched, there is still no system that can prevent the rehabilitation subject from feeling confused when visual and auditory signals are replaced with virtual signals.

Therefore, based on the 3D model, feature points, and mark information loaded from the 3D model library, the location, direction, and type of the object caught in the current image can be identified through numerical analysis, and the 3D modeling information loaded from the 3D model library and the current Position, direction, and future predicted position and direction are entered into game logic, so virtual images and sounds can be reproduced according to the conversion of game logic, and game elements using virtual reality (VR) technology are applied to rehabilitation treatment. It is possible to improve the treatment effect by enabling patients undergoing rehabilitation to overcome the tedious and difficult treatment process, and the virtual reality with game elements to transmit visual and auditory signals while leaving the tactile sense that patients feel when interacting with objects. Since the visual and auditory signals are replaced without discomfort, it is possible to motivate the patient by continuously making the patient feel a sense of achievement even before the therapeutic effect appears, and the type, location, action/reaction of the object felt by the patient's tactile sense and predictable Development of an object information recognition and prediction system applying virtual reality technology to rehabilitation treatment that can prevent rehabilitation recipients from feeling confused when visual and auditory signals are replaced with virtual signals by matching possible visual stimuli and auditory signals This is an urgent need.

Registration No. 10-1501838

Therefore, the present invention was conceived to solve the above problems, and by performing numerical analysis based on the 3D model, feature points, and mark information loaded from the 3D model library, the position, direction, and type of the object captured in the current image can be determined. The purpose is to provide an object information recognition and prediction system that applies virtual reality technology to rehabilitation treatment that can be grasped.

Another object of the present invention is to obtain 3D modeling information from a 3D model library, current position, direction, and future predicted position and direction by inputting them into game logic, which can be reproduced as virtual images and sounds according to conversion of game logic. It is to provide an object information recognition and prediction system applying virtual reality technology to rehabilitation treatment.

Another object of the present invention is to improve the treatment effect by applying a game element using virtual reality (VR) technology to rehabilitation treatment, enabling patients receiving rehabilitation treatment to overcome boring and difficult treatment processes. It is to provide an object information recognition and prediction system to which virtual reality technology is applied.

Another object of the present invention is to replace visual and auditory signals with visual and auditory signals of virtual reality with game elements without discomfort while leaving the patient's tactile sense when interacting with objects, even before the therapeutic effect appears. It is to provide an object information recognition and prediction system applying virtual reality technology to rehabilitation treatment that can continuously make patients feel a sense of achievement and motivate them.

Another object of the present invention is to match the type, position, action/reaction of an object felt by the patient with the visual stimulus and auditory signal that can be predicted therefrom, thereby replacing visual and auditory signals with virtual signals. It is to provide an object information recognition and prediction system applying virtual reality technology to rehabilitation treatment that can prevent rehabilitation subjects from feeling confused.

Object information recognition and prediction system applying virtual reality technology to rehabilitation treatment according to a preferred embodiment of the present invention for achieving the above object is to recognize an object and display it in virtual reality while a patient interacts with the object and photographs it. with a camera that does; a feature point extraction unit extracting feature points from the image input from the camera; a relationship derivation unit between feature points that grasps a relationship between a distance, a direction, and an invisible feature point between each feature point from the feature point extractor; a marker recognizing unit recognizing a marker from an image received from the camera; a relationship derivation unit between the marks recognizing a plurality of marks from the mark recognizing unit and determining a distance and relationship between the marks; a location, direction, and type of object that determines the location, direction, and type of an object captured in the current image through numerical analysis based on 3D model, feature points, and mark information retrieved from a 3D model library; If there is a delay in the recognition time from the position, direction, and type of the object, the sound must be reproduced within a time that a person cannot perceive, or if there is another need, the degree of change in position and direction is measured a future position and direction prediction unit that predicts a position or changing direction of an object in the future and predicts current position and direction information of the object in consideration of the patient's body movement; a game logic unit for inputting 3D modeling information retrieved from the 3D model library, a current position and direction, and a predicted future position and direction; a virtual image display unit for displaying a virtual image according to the conversion of the game logic in the game logic unit; a virtual sound reproducing unit that reproduces a virtual sound according to the conversion of the game logic in the game logic unit; It is characterized by including.

In the present invention, the 3D model library is characterized by including storing information on objects necessary for recognizing and expressing objects with which the patient currently interacts and data for reproducing through computer graphics. .

In the present invention, the configuration of the 3D model library reflects the dimensions of the real object and the 3D scanning solution that provides a method of configuring, modifying, and writing three-dimensional data of a real object, so that it is configured and used, or used in combination. A 3D modeling tool, a 3D model unit composed of the 3D scanning solution and the 3D modeling tool, and feature point information that can be used to determine the type, location, and direction of the 3D model from the 3D model unit. To derive feature point information A feature point information addition unit and a marker information addition unit that adds a process of registering a marker to assist feature point information because there is a limit to accurately grasping the type and direction of an object using only the feature points from the feature point information addition unit. And, a recognizable 3D model unit that creates a recognizable 3D model by combining the 3D modeling, feature point information, and marker information, and the information of the recognizable 3D model unit is stored so that it can be used for later recognition through a camera and reproduction in computer graphics, 3D model library; It is characterized by including.

In the present invention, the feature point information in the feature point information addition unit includes photographing a real object with a camera and statistically analyzing the relationship between each feature point and using it to estimate the type, location, direction, etc. of the current object. characterized by

In the present invention, the camera can use a plurality of cameras to reduce blind spots when a patient covers an object while interacting, and since a plurality of cameras are used, a photogrammetry technique is used to determine the relationship between feature points from the camera. It is characterized in that it includes being able to grasp the distance, direction, etc. of an object.

In the present invention, the marker in the mark recognizing unit is characterized in that it includes information on the type of object and can be used to identify the object in the 3D model library.

In the present invention, when the future position and direction prediction unit tries to reproduce the sound at the moment the chopsticks collide, it is expected that the chopsticks will collide after 0.1 second, but if it takes 0.2 seconds to recognize the position, the effect sound will be played in advance before image recognition It is characterized by including what can be.

As described above, the object information recognition and prediction system applying the virtual reality technology to the rehabilitation treatment of the present invention has the following effects.

First, the present invention can grasp the position, direction, and type of an object caught in a current image by performing numerical analysis based on a 3D model, feature points, and mark information loaded from a 3D model library.

Second, according to the present invention, 3D modeling information loaded from a 3D model library, current position and direction, and future predicted position and direction are input to game logic, so that virtual images and sounds can be reproduced according to conversion of game logic.

Third, the present invention can improve the treatment effect by applying a game element using virtual reality (VR) technology to rehabilitation treatment, so that the patient receiving rehabilitation treatment can overcome a tedious and difficult treatment process.

Fourth, the present invention replaces visual and auditory signals with virtual reality visual and auditory signals with game elements without discomfort while leaving the patient's tactile sense when interacting with objects, so that the treatment effect continues even before the appearance. This can motivate patients by making them feel a sense of accomplishment.

Fifth, the present invention matches the type, location, action/reaction of objects that the patient feels by touch, and visual stimulation and auditory signals that can be predicted therefrom, thereby providing rehabilitation when visual and auditory signals are replaced with virtual signals. It can make the subject not feel confused.

1 is a view to explain the configuration of an object information recognition and prediction system to which virtual reality technology is applied to rehabilitation treatment according to an embodiment of the present invention.
2 is a diagram showing a detailed configuration of a 3D model library among configurations of an object information recognition and prediction system to which virtual reality technology is applied to rehabilitation treatment according to an embodiment of the present invention.

Hereinafter, looking at preferred embodiments of the present invention together with the accompanying drawings, the detailed description of the known technology or configuration related to the present invention may unnecessarily obscure the subject matter of the present invention. Descriptions will be omitted, and the terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of the user or operator, so the definition is an object to which virtual reality technology is applied to rehabilitation treatment, which is the present invention. It should be made based on the contents throughout this specification describing the information recognition and prediction system.

Hereinafter, an object information recognition and prediction system to which virtual reality technology is applied to rehabilitation treatment according to a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram showing the configuration of an object information recognition and prediction system to which virtual reality technology is applied to rehabilitation treatment according to an embodiment of the present invention, and FIG. It is a diagram shown to explain the detailed configuration of the 3D model library among the configurations of the object information recognition and prediction system to which real technology is applied.

Object information recognition and prediction system applying virtual reality technology to rehabilitation treatment of the present inventors includes a 3D scanning solution (1), a 3D modeling tool (2), a 3D model unit (3), a feature point information addition unit (4), a mark information addition unit (5), 3D model unit (6), 3D model library (7), camera 10, feature point extraction unit 20, mark recognition unit 30, relationship derivation unit 40 between feature points, relationship derivation between marks unit 50, object location, direction, object type unit 60, future location, direction prediction unit 70, game logic unit 80, virtual image display unit 90, virtual sound reproduction unit (100) and the like.

As shown in FIGS. 1 and 2, the object information recognition and prediction system to which virtual reality technology is applied to the rehabilitation treatment of the present invention is a camera that photographs objects while a patient interacts with them in order to recognize objects and display them in virtual reality. (10) and; a feature point extraction unit 20 that extracts feature points from the image input from the camera 10; a relationship derivation unit 40 between feature points that grasps the relationship between the distance, direction, and invisible feature points between the feature points from the feature point extractor 20; a marker recognizing unit 30 recognizing a marker from an image received from the camera 10; a relation derivation unit 50 between the marks recognizing a plurality of marks from the mark recognizing unit 30 and determining a distance and relationship between the marks; Object location, direction, and object type unit 60 that identifies the location, direction, and type of the object caught in the current image through numerical analysis based on the 3D model, feature points, and mark information imported from the 3D model library (7) and; If there is a delay in the recognition time from the position, direction, and object type unit 60 of the object, the sound must be reproduced within a time that a person cannot perceive, or if there is another need, the degree of change in position and direction a future position and direction prediction unit 70 that measures and predicts the position or changing direction of an object in the near future and predicts the current position and direction information of the object in consideration of the patient's body movement; a game logic unit 80 for inputting 3D modeling information retrieved from the 3D model library 7, current position and direction, and future predicted position and direction; a virtual image display unit 90 that displays a virtual image according to the conversion of the game logic in the game logic unit 80; a virtual sound reproducing unit 100 that reproduces a virtual sound according to the conversion of the game logic in the game logic unit 80; to provide

The function of each technical means constituting the object information recognition and prediction system to which virtual reality technology is applied to the rehabilitation treatment of the present invention will be described as follows.

The camera 10 captures objects while the patient interacts with them in order to recognize objects and display them in virtual reality.

Here, the camera 10 uses a plurality of cameras 10 to reduce blind spots when an object is covered while the patient interacts, and since the plurality of cameras 10 are used, a photogrammetry technique is used between feature points. It is possible to grasp the distance, direction, etc. of an object from the camera 10 in relation to .

In addition, in general, there are a total of four HMD (Head Mounted Display) cameras, two each on the left and right, and it is a technology to match the 3D modeled object to the actual position after identifying the actual position of the object using this, and stereo camera through Asynchronous Spacewarp technology It is possible to accurately match by calculating the position of the real object using the captured image.

The feature point extraction unit 20 extracts feature points from an image input from the camera 10 .

The relationship derivation unit 40 between the feature points determines the relationship between the distance, direction, and invisible feature points between the feature points from the feature point extractor 20 .

The mark recognizing unit 30 recognizes a marker from an image received from the camera 10 .

Here, in the mark recognizing unit 30, a marker has type information of an object and can be used to identify an object in a 3D model library.

The relationship derivation unit 50 between the marks recognizes a plurality of marks from the mark recognizing unit 30 and determines the distance and relationship between the marks.

The object location, direction, and object type unit 60 determines the location, direction, and type of the object captured in the current image through numerical analysis based on the 3D model, feature points, and mark information retrieved from the 3D model library 7. is to figure out

Here, the 3D model library 7 stores information on objects necessary for recognizing and expressing objects with which the patient currently interacts and data for reproducing through computer graphics.

In addition, the 3D model library 7 includes a 3D scanning solution 1, a 3D modeling tool 2, a 3D model unit 3, a feature point information addition unit 4, and a mark information addition unit 5 and the configuration of the 3D model unit 6 and the 3D model library 7.

The 3D scanning solution 1 provides a method of constructing, modifying, and creating three-dimensional data of a real object.

The 3D modeling tool 2 is configured to reflect the dimensions of a real object, or used in combination.

The 3D model unit 3 is composed of the 3D scanning solution 1 and the 3D modeling tool 2.

The feature point information addition unit 4 derives feature point information that can be used to determine the type, location, and direction of the 3D model from the 3D model unit 3.

Here, the feature point information in the feature point information adding unit 4 is used to estimate the type, location, direction, etc. of the current object by statistically analyzing the relationship between each feature point by photographing a real object with a camera.

In addition, in the feature point information, for example, in the case of a cube, the vertices become feature points, and distance information between the respective vertices should be recorded.

The marker information adding unit 5 has limitations in accurately grasping the type and direction of an object with only the feature points from the feature point information adding unit 4, so the process of registering a marker to assist feature point information is performed. is to add

Here, the case where there is a limit in the feature point is, for example, a case of having a highly reflective surface or having a shape in which it is difficult to derive the feature point.

In addition, in the process of registering the markers of the marker information adding unit 5, a plurality of markers are attached to objects, and different colors are assigned to each marker to record object type information in the form of color codes. It can be used to identify the direction and position of an object by recording the relational information between the signs.

In addition, it is possible to record object type information on the marker, attach a plurality of markers to the object, and use distance information of the markers to determine the position and direction of the object, while the patient interacts with the object. The markers can be covered, and the effect of blind spots can be minimized by using a plurality of markers.

The recognizable 3D model unit 6 creates a recognizable 3D model by combining the 3D modeling, feature point information, and marker information.

The 3D model library 7 stores the information of the recognizable 3D model unit 6 so that it can be used for later recognition through a camera and reproduction in computer graphics.

Here, when the information of the recognizable 3D model unit 6 is reproduced in virtual reality, the color and surface texture of the stored 3D model can be changed, or a static object can be changed to a dynamically changing object. When it is expressed virtually, it can be expressed in red, and a pottery cup can be expressed as a metal cup, and it can be displayed as a continuously changing color cup.

The future position and direction prediction unit 70 must reproduce a sound within a time that a person cannot perceive when a delay occurs in the recognition time from the position, direction, and object type unit 60 of the object, or other If there is a need, the degree of change in position and direction is measured to predict the position or changing direction of the object in the near future, and the current position and direction information of the object is predicted in consideration of the patient's body movement.

Here, when the future position and direction prediction unit 70 tries to reproduce the sound at the moment the chopsticks collide, it is expected that the chopsticks will collide after 0.1 second, but if it takes 0.2 seconds to recognize the position, the effect sound can be played in advance before image recognition There is.

In addition, when predicting the current position and direction information of an object in consideration of the patient's body movement, for example, when the arm is rotated while holding the object by hand, the object held in the hand makes a circular motion, and the object Rather than predicting only the position or direction information of the hand, the prediction accuracy increases when the information held in the hand is also considered.

The game logic unit 80 inputs the 3D modeling information retrieved from the 3D model library 7, the current position and direction, and the predicted future position and direction.

The virtual image display unit 90 displays a virtual image according to conversion of the game logic in the game logic unit 80 .

The virtual sound reproducing unit 100 reproduces the virtual sound according to the conversion of the game logic in the game logic unit 80 .

As described above, the present invention can replace visual and auditory signals with virtual reality visual and auditory signals having game elements without discomfort while leaving the patient's tactile sense when interacting with objects.

For example, in the case of training to pick up and move a water cup, the tactile sense of holding the water cup is felt as it is, but the shape of the water cup is reproduced on a display for virtual reality as a virtual object implemented by computer graphics, It is possible to create and synthesize background music or sound effects that do not occur in reality.

In addition, since the present invention matches the type, location, action/reaction of an object felt by the patient with the visual stimulus and auditory signal that can be predicted therefrom, when visual and auditory signals are replaced with virtual signals, rehabilitation subjects can make you feel less confused.

For example, when training in the process of lifting chopsticks and picking up food, the shape, position, and direction of the chopsticks felt by the hand must be matched with virtual visual cues, and the virtual sound produced when the chopsticks collide can guide people. When playing a sound within an imperceptible time difference, it must coincide with the moment of collision.

In particular, to represent objects virtually without discomfort, 1) the patient must be able to recognize the type and number of objects that the patient is currently interacting with, 2) be able to calculate the location and direction of objects, and 3) use computer graphics. You need a 3D model to represent it.

And since some or all of the objects may be obscured while the patient is interacting, a technique to reduce blind spots is needed.

In addition, as described above, since the object information recognition and prediction system applying virtual reality technology to rehabilitation treatment can be applied to the medical field of rehabilitation treatment as well as the industrial field related to education and training, its use and application are wide.

The best embodiment has been disclosed in the drawings and specifications, and the terms used herein are only used for the purpose of describing the present invention, and are not used to limit the scope of the present invention described in the claims or limitation of meaning. Therefore, those skilled in the art will be able to make various modifications and equivalent other embodiments therefrom, and therefore, the true technical scope of protection of the present invention should be defined by the technical spirit of the appended claims.

1: 3D scanning solutions 2: 3D modeling tools
3: 3D model part 4: feature point information addition part
5: Mark information addition part 6: 3D model part
7: 3D Model Library 10: Camera
20: feature point extraction unit 30: mark recognition unit
40: relationship derivation unit between feature points 50: relationship derivation unit between marks
60: object location, direction, type of object
70: future position, direction prediction unit 80: game logic unit
90: virtual image display unit 100: virtual sound playback unit

Claims (7)

In the object information recognition and prediction system applying virtual reality technology to rehabilitation treatment,
a camera for capturing an object while the patient interacts with it in order to recognize the object and display the object in virtual reality;
a feature point extraction unit extracting feature points from the image input from the camera;
a relationship derivation unit between feature points that grasps a relationship between a distance, a direction, and an invisible feature point between each feature point from the feature point extractor;
a marker recognizing unit recognizing a marker from an image received from the camera;
a relationship derivation unit between the marks recognizing a plurality of marks from the mark recognizing unit and determining a distance and relationship between the marks;
a location, direction, and type of object that determines the location, direction, and type of an object captured in the current image through numerical analysis based on 3D model, feature points, and mark information retrieved from a 3D model library;
If there is a delay in the recognition time from the position, direction, and type of the object, the sound must be reproduced within a time that a person cannot perceive, or if there is another need, the degree of change in position and direction is measured a future position and direction prediction unit that predicts a position or changing direction of an object in the future and predicts current position and direction information of the object in consideration of the patient's body movement;
a game logic unit for inputting 3D modeling information retrieved from the 3D model library, a current position and direction, and a predicted future position and direction;
a virtual image display unit for displaying a virtual image according to the conversion of the game logic in the game logic unit;
a virtual sound reproducing unit that reproduces a virtual sound according to the conversion of the game logic in the game logic unit; Object information recognition and prediction system applying virtual reality technology to rehabilitation treatment, characterized in that it comprises a. According to claim 1,
The 3D model library stores information about objects necessary for recognizing and expressing objects with which the patient currently interacts and data for reproducing through computer graphics Virtual reality technology in rehabilitation treatment, characterized in that Object information recognition and prediction system applied. According to claim 1,
The configuration of the 3D model library includes a 3D scanning solution that provides a method of configuring, modifying, and creating three-dimensional data of a real object, a 3D modeling tool that reflects the dimensions of a real object and configures and uses it, or A 3D model unit composed of the 3D scanning solution and a 3D modeling tool, and a feature point information addition unit for deriving feature point information that can be used to determine the type, location, and direction of the 3D model from the 3D model unit; A marker information addition unit for adding a process of registering a marker to assist feature point information because there is a limit to accurately grasping the type and direction of an object only with the feature point from the feature point information addition unit, and the 3D modeling and A recognizable 3D model unit that creates a recognizable 3D model by combining feature point information and marker information, and a 3D model library that stores the information of the recognizable 3D model unit so that it can be used for later recognition through a camera and reproduction in computer graphics; Object information recognition and prediction system applying virtual reality technology to rehabilitation treatment, characterized in that it comprises a. According to claim 3,
Characterized in that the feature point information in the feature point information addition unit is used to estimate the type, location, direction, etc. of the current object by photographing a real object with a camera and statistically analyzing the relationship between each feature point. Object information recognition and prediction system applied with virtual reality technology. According to claim 1,
The camera uses a plurality of cameras to reduce blind spots when the patient covers an object while interacting with the patient. Since the camera uses a plurality of cameras, the distance and direction of the object from the camera are determined by the relationship between feature points using a photogrammetry technique. Object information recognition and prediction system applying virtual reality technology to rehabilitation treatment, characterized in that it includes being able to grasp the etc. According to claim 1,
The marker in the mark recognition unit has object type information and can be used to identify objects in a 3D model library. Object information recognition and prediction system applying virtual reality technology to rehabilitation treatment. According to claim 1,
When trying to play a sound at the moment the chopsticks collide in the future position and direction prediction unit, it is expected to collide after 0.1 seconds, but if it takes 0.2 seconds for position recognition, it includes being able to play sound effects in advance before image recognition Object information recognition and prediction system applying virtual reality technology to rehabilitation treatment characterized by.

Images