KR102202069B1 - 3D stereoscopic pillow manufacturing device - Google Patents

Abstract

The present invention relates to a device for manufacturing a 3D stereoscopic pillow, which is configured to 3D model a cervical vertebrae part shape of a user using a cervical vertebrae part image of the user, acquired by X-ray imaging, compare the 3D modeled shape with a standard cervical vertebrae shape to analyze conditions of the cervical vertebrae, derive a user-customized pillow shape suitable for the user based on an analysis result, and output the shape as a real object. According to the present invention, the device for manufacturing a 3D stereoscopic pillow comprises: a database in which the standard cervical vertebrae part shape and neck part shape are stored, and the cervical vertebrae part image of the user, acquired by X-ray imaging, is input and stored; a modeling unit which 3D models the cervical vertebrae part shape and the neck part shape of the user using the cervical vertebrae part image of the user, which is input, and 3D models a user-customized pillow shape derived based on the 3D modeling; an artificial intelligence unit which compares the cervical vertebrae shape and neck part shape of the user, which are 3D modeled, with the standard cervical vertebrae part shape and neck part shape, respectively, and analyzes conditions of the cervical vertebrae and neck of the user, and derives a user-customized pillow shape suitable for the user based on the analysis result; a generation unit which creates a 3D file for 3D output using the 3D modeled user-customized pillow shape; and an output unit which uses the generated 3D file to output a user-customized pillow as a real object. According to the present invention, a user can experience sound sleep.

Description

3D stereoscopic pillow manufacturing device}

The present invention relates to an apparatus for manufacturing a 3D three-dimensional pillow using an image of a user's cervical vertebrae, and more particularly, 3D modeling of the shape of a user's cervical vertebrae using an image of the user's cervical vertebrae acquired by X-ray photography. And, it is compared with the standard cervical shape to analyze the state of the cervical vertebrae, and based on the analysis result, it relates to a device for manufacturing a 3D three-dimensional pillow by deriving a customized pillow shape suitable for the user and outputting the shape as a real object.

In general, the top of the spine, which is composed of 7 spines and forms a C shape, is called the cervical vertebrae, and the shape of the cervical spine is gradually deformed according to the usual lifestyle, work environment, and the amount of muscle exercise in the cervical spine. , Turtle neck syndrome, and other symptoms may appear.

In the case of deformation of the cervical vertebrae in this way, or using a normal pillow that does not consider the shape of the cervical vertebrae that are different for each individual, the angle at which the pillow supports the neck during sleep is out of the recommended range, which may adversely affect the neck muscles and cervical spine. In addition, it may interfere with blood circulation, which soon leads to chronic pain and sleep disturbances, causing discomfort in daily life.

Therefore, in recent years, customized pillows of various shapes referred to as'cervical pillows' are in the spotlight, and as a conventional invention related to such a customized pillow, the “custom pillow selection system and its operation method” in Korean Patent Publication No. 10-0574783 ”, “Customized scapula fixing pillow using 3D printer and its manufacturing method” in Korean Patent Application Publication No. 10-2018-0052447 and “Customized pillow and manufacturing method thereof” in Korean Patent Publication No. 10-1979182 have been proposed. It has been released.

The “customized pillow selection system and its operation method” of the Korean Patent Publication No. 10-0574783 includes a head contour measurement unit, a data storage unit, a head type determination unit, and a pillow shape display unit. An invention related to a system and method for selecting the shape of the pillow more scientifically has been proposed, and “Personalized scapula fixing pillow using a 3D printer and its manufacturing method” in Korean Patent Publication No. 10-2018-0052447 Considering the shape of the user's scapula, it is configured to stably support the scapula by forming left and right armrests at the lower part of the neck support. This not only supports the cervical vertebrae stably, but also relieves the tension of the cervical vertebrae. Invention has been proposed for a pillow that can be prevented and corrected and a method for manufacturing the pillow.

In addition, in the “user-customized pillow and its manufacturing method” of Korean Patent Publication No. 10-1979182, a base including a seating portion, a pad supporting at least one of a user's cervical vertebrae or a head, and a pad surrounding the outer peripheral surface of the pad Invention related to a pillow that can conform to the shape of the head or cervical spine of different users, and a method for manufacturing the pillow, by being configured to include a fixed pad receiving unit and including a receiving unit selectively mounted on the seating unit Was suggested.

However, the prior inventions as described above are configured to measure the user's head contours using a 3D scanner or two or more CCD cameras, and to select a customized pillow based on the measured value, so the effectiveness may be questioned. , When configured to manufacture a customized pillow based on the measured value, there may be problems such as an increase in cost due to the provision of equipment, and an error in the measured value depending on the measurement method and skill level.

Therefore, by solving problems such as chronic pain and sleep disturbances due to the use of a general pillow where the cervical vertebrae have been deformed or the shape of the cervical vertebrae has not been taken into account, the user's cervical vertebrae shape is considered by solving the problems of the prior inventions. Invention of a manufacturing device capable of providing a customized pillow is required.

Korean Registered Patent Publication No. 10-0574783 (2006. 04. 21) Korean Patent Application Publication No. 10-2018-0052447 (2018. 05. 18) Korean Patent Publication No. 10-1979182 (2019. 05. 09)

The apparatus for manufacturing a 3D three-dimensional pillow according to the present invention relates to a device for manufacturing a 3D three-dimensional pillow using an image of a user's cervical spine,

There was a problem that the cervical vertebrae were deformed to a considerable extent depending on the usual lifestyle, work environment, and the amount of muscle exercise in the cervical vertebrae, or chronic pain and sleep disturbances occurred due to the use of a general pillow that did not consider the shape of the cervical vertebrae.

In the prior inventions as described above, in addition to the 3D printer required for the actual output of the customized pillow, the cost increases due to the provision of a 3D scanner or two or more CCD cameras, and a measurement method using a 3D scanner or two or more CCD cameras and Since there was a problem such as an error in the measured value depending on the skill level, the purpose of this is to provide a solution to help the user to sleep well.

3D three-dimensional pillow manufacturing apparatus according to the present invention to realize the above object,

A database in which a standard cervical vertebral region shape and a neck region shape are stored, and an image of a user's cervical vertebrae obtained through X-ray imaging is input and stored; A modeling unit for 3D modeling the shape of the user's cervical vertebrae and the shape of the neck, respectively, using the input image of the cervical vertebrae, and 3D modeling the shape of the user-customized pillow derived based on the shape of the user's cervical vertebrae; The user's cervical and neck conditions are analyzed by comparing the 3D modeled shape of the user's cervical and neck with the standard cervical and neck shapes, respectively, and based on the analysis results, a customized pillow shape suitable for the user is derived. An artificial intelligence unit; A generator for generating a 3D file for 3D output by using the 3D modeled pillow shape; And an output unit that outputs a user-customized three-dimensional pillow in real form by using the generated 3D file. It presents a 3D three-dimensional pillow manufacturing apparatus, characterized in that it is configured to include.

3D three-dimensional pillow manufacturing apparatus according to the present invention,

Since the user's cervical vertebra shape is 3D modeled using the user's cervical vertebra image acquired by X-ray, and the pillow shape derived based on this is configured to output the actual shape, By providing the considered customized pillow in a new method different from the conventional method, an effect that can help the user's sleep is generated,

Since it is configured to 3D model the shape of the user's cervical vertebrae and the shape of the neck using an image of the user's cervical vertebrae acquired by X-ray, instead of a measurement method using a 3D scanner, it is possible to prevent an increase in cost compared to the conventional invention. And minimize the occurrence of errors.

1 is a configuration diagram of a 3D three-dimensional pillow manufacturing apparatus according to the present invention.
Figure 2 is a reference diagram showing the function of a database that is a component of the 3D three-dimensional pillow manufacturing apparatus according to the present invention.
3(a) and 3(b) are exemplary views showing images of cervical vertebrae stored in a database.
Figure 4 is a reference diagram showing the function of the artificial intelligence that is a component of the 3D three-dimensional pillow manufacturing apparatus according to the present invention.
5(a) to 5(c) are exemplary diagrams showing reference materials stored in a database.
6 is a reference diagram showing in detail one of the functions of the artificial intelligence unit, which is a component of the 3D three-dimensional pillow manufacturing apparatus according to the present invention.

The present invention relates to an apparatus for manufacturing a 3D three-dimensional pillow using an image of a user's cervical spine,

A database 100 in which a standard cervical vertebral region shape and a neck region shape are stored, and an image of the user's cervical vertebrae obtained by X-ray imaging is input and stored; A modeling unit 110 for 3D modeling the shape of the user's cervical vertebrae and the shape of the neck, respectively, using the input image of the cervical vertebrae, and 3D modeling the shape of a user-customized pillow derived based on the shape of the user's cervical vertebrae; The user's cervical and neck conditions are analyzed by comparing the 3D modeled shape of the user's cervical and neck with the standard cervical and neck shapes, respectively, and based on the analysis results, a customized pillow shape suitable for the user is derived. An artificial intelligence unit 120; A generation unit 130 for generating a 3D file for 3D output by using the 3D modeled user-customized pillow shape; And an output unit 140 for outputting a user-customized pillow in real form using the generated 3D file. It relates to a 3D three-dimensional pillow manufacturing apparatus, characterized in that configured to include.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

First, the relationship between the database 100, the modeling unit 110, the artificial intelligence unit 120, the generation unit 130 and the output unit 140, which are major components constituting the 3D three-dimensional pillow manufacturing apparatus according to the present invention Is as shown in FIG. 1.

The database 100 is a storage device capable of inputting and outputting data, and as shown in FIG. 2, the shape of the standard cervical vertebrae and the shape of the neck are stored in the form of image files, respectively, and The image of the cervical spine is also input and saved in the form of an image file.

At this time, the standard cervical shape and neck shape among the data stored in the database 100 are comparison data for determining the state of the user's cervical vertebrae through comparison with the user's cervical vertebral region image, and the 2D image and the corresponding It is desirable to include a 3D shape, and the 2D image of the 2D image may be an X-ray photograph of a cervical spine region having a medically ideal or normal shape, or a 2D modeled medically ideal or normal shape, and the 3D shape is It may be 3D modeled using the 2D image or may be 3D modeled in a medically ideal or normal shape.

Therefore, among the data stored in the database 100, it is preferable that the standard cervical spine and neck shape are verified by a specialist related to musculoskeletal, and the data is verified by the service provider or administrator of the present invention. After being stored in the base 100, it may be provided to the user.

In addition, as shown in FIGS. 3(a) and 3(b), among the data stored in the database 100, the image of the user's cervical vertebrae is an X-ray photograph of the user's cervical vertebrae from one side and the back. It is a basic configuration, and additionally, an X-ray photograph taken from the other side and the front may be included, and a neck image taken of the user's neck in general may be included for clear identification of the shape.

That is, among the data stored in the database 100, the image of the user's cervical spine is provided to the user through X-ray photographing at a public medical institution or an officially certified medical institution, and the like. It can be transmitted and stored in a wired or wireless manner from the terminal.

The modeling unit 110 is a device capable of generating a 3D three-dimensional shape using a plurality of 2D images. A known image conversion software is installed, and the user's cervical vertebra image displayed in the user's cervical vertebra image input to the database 100 is installed. The shape of the cervical spine and the shape of the neck can be modeled in 3D, respectively.

Therefore, the user's cervical vertebra portion and the neck portion shown in the user's cervical vertebra portion image stored in the database 100 need to be clearly distinguished before the 3D modeling operation by the modeling unit 110, and such a user's cervical vertebra The artificial intelligence unit 120 is involved in the process of distinguishing the user's cervical vertebra and neck from the site image.

The artificial intelligence unit 120 is a device that implements artificial intelligence by applying Google's Tensorflow, and as shown in FIG. 4, the user's cervical spine region input to the database 100 It is characterized in that it is configured so that the user's cervical vertebra and neck can be identified and identified based on the image.

That is, in the X-ray photograph of the user's cervical vertebrae, not only the user's cervical vertebrae, but also the neck region can be displayed to a specific degree, so that the artificial intelligence unit 120 is a 2D image stored in the database 100. Based on the shape of the user's cervical vertebrae and the shape of the neck, machine learning is performed to identify and identify the user's cervical and neck regions from X-ray images of the cervical vertebrae.

Therefore, in the database 100, it is preferable that X-ray pictures of the cervical vertebrae having various forms required for machine learning of the artificial intelligence unit 120 are stored as reference data. -ray pictures are preferably configured to include all of the cervical vertebrae state within the normal range as shown in Fig. 5(a) and abnormal cervical vertebrae states as shown in Figs. 5(b) and 5(c) in addition to the medically ideal cervical spine state.

In addition, in the process of discriminating and specifying the cervical vertebrae and the neck by the artificial intelligence unit 120, an image of the user's neck, which can be additionally stored in the database 100, is used as a reference material for clear identification of the shape. Can be utilized.

On the other hand, generation of 3D stereoscopic shapes using 2D images generally requires a plurality of 2D images taken at various angles, but the user's cervical spine image is an X-ray photograph of the user's cervical spine from one side and the back. As a basic composition of photos, additionally, X-ray photographs taken from the other side and front may be included, but with only such limited angle images, the 3D modeled 3D shape is the shape of the actual user's cervical spine and neck. It may be difficult to implement the shape exactly as it is.

Therefore, the artificial intelligence unit 120 is stored in the database 100, and iteratively learns deep learning by using the standard cervical spine shape and neck shape including 2D images and 3D shapes. It is possible to accurately model a 3D three-dimensional shape just by taking an X-ray of the cervical spine from one side and the back.

In other words, the cervical vertebrae of a person consisting of 7 spines and forming a C-shape differs in size from person to person, but the basic shape is the same or similar, so if sufficient learning is preceded, 2D photographed at a limited angle such as the side and the back. It is possible to create a 3D three-dimensional shape relatively accurately with only an image.

Therefore, the artificial intelligence unit 120 is configured to regenerate the user's cervical vertebral region image at various angles based on the user's cervical vertebral region image, which is a 2D image, by deep learning learning using the standard cervical vertebral region shape and the neck region shape. Then, based on this, 3D modeling through the modeling unit 110 can be implemented.

In addition, in the database 100, for more effective deep learning learning, various shapes of cervical vertebrae and neck shapes composed of 2D images and corresponding 3D shapes may be stored as reference data.

In addition, the artificial intelligence unit 120 includes the shape of the user's cervical vertebrae and the shape of the neck 3D modeled by the modeling unit 110 to the standard cervical vertebral region shape and the neck region shape input to the database 100, respectively. It is characterized by analyzing the state of the cervical spine and the state of the neck of the user by comparison.

Specifically, the artificial intelligence unit 120 recognizes the position and shape of each of the cervical vertebrae 1 to 7 constituting the user's cervical vertebrae from the shape of the user's cervical vertebrae 3D modeled by the modeling unit 110, and the data After recognizing the position and shape of each of the 1 to 7 cervical vertebrae from the standard cervical vertebrae input to the base 100, it will be configured to analyze the position and shape of each of the 1 to 7 cervical vertebrae constituting the user's cervical vertebrae. I can.

That is, the cervical spine of a general person is composed of 7 spines to form a C-shape, but it can be gradually deformed according to the usual lifestyle or work environment, and the amount of muscle exercise in the cervical spine, so that the artificial intelligence unit 120 is transformed. By classifying the shape of the user's cervical vertebrae into any one of a plurality of preset states according to the degree of, it can be reflected in derivation of a pillow shape suitable for the user.

As an embodiment of the present invention, first, the artificial intelligence unit 120 refers to some of the X-ray pictures of various cervical vertebrae stored in the database 100 and machine learning the medically ideal cervical vertebrae state. You can learn in a way, and you can learn the state of the cervical spine, which is within the normal range in which problems such as chronic pain and sleep disorders do not occur in daily life, although some transformations have occurred, and the deformation occurs greatly and affects daily life. You can learn abnormal cervical spine conditions.

Therefore, each of the X-ray pictures of various cervical vertebrae stored in the database 100 is classified into any one of an ideal state, a normal state, or an abnormal state by consultation of a specialist, etc. It is preferable that machine learning learning by the artificial intelligence unit 120 is possible by being classified and stored by state, but the machine learning learning method is not necessarily limited to this example.

In addition, the artificial intelligence unit 120 uses other pictures of X-ray pictures of various cervical vertebrae stored in the database 100 except for pictures used for machine learning method learning. The shape of the cervical spine may be classified into one of an ideal state, a normal state, or an abnormal state, and deep learning learning may be performed using the classification result.

That is, the artificial intelligence unit 120 randomly extracts any one or more of a plurality of X-ray pictures stored in the database 100 in a state classified as an ideal state, and machine learning learns the ideal state of the cervical spine. The normal state of the cervical spine may be machine-learned by randomly extracting any one or more of a plurality of X-ray pictures stored in the database 100 in a state classified as a normal state, and One or more of a plurality of X-ray pictures stored in the database 100 in a classified state may be randomly extracted to perform machine learning learning of an abnormal state of the cervical spine.

In addition, the artificial intelligence unit 120 randomly extracts any one or more of a plurality of X-ray pictures stored in the database 100, and selects a cervical spine portion shown in each of the extracted one or more X-ray pictures. After classifying into either an ideal state, a normal state, or an abnormal state according to the shape, the classification result and the original classification state of each of the extracted one or more X-ray images can be compared, and deep learning can be learned based on the comparison result. have.

At this time, the original classification state of each of the one or more X-ray images means that each of the X-ray images of the cervical vertebra having various shapes is either an ideal state, a normal state, or an abnormal state by consulting a specialist about the musculoskeletal structure. It means a state classified as a state of.

In summary, the artificial intelligence unit 120 can basicly learn the state of the cervical vertebrae and the neck according to the shape of the cervical vertebrae by the machine learning method as described above, and the cervical vertebrae by the deep learning learning method as described above. It is possible to analyze the state of the cervical vertebrae and neck according to the shape of the part, and through the result, in-depth learning is possible, and through this repetitive learning, the accuracy of the analysis is improved.

On the other hand, as shown in Fig. 6, when the shape of the user's cervical vertebrae is classified as a normal state or an abnormal state, the artificial intelligence unit 120 determines each state of the cervical vertebrae 1 to 7 as an ideal state, a normal state, or an abnormal state. It can be classified into any one of the states, and the degree of deformation of each of the cervical vertebrae 1 to 7 which is a normal state or an abnormal state can be derived and expressed as a percentage.

That is, the artificial intelligence unit 120 is normal based on the position and shape of each of the 1st to 7th cervical vertebrae, which are ideal states, and the position and shape of each of the 1st to 7th cervical vertebrae, which are the most severe abnormal states. By displaying the degree of deformation of each of the cervical vertebrae 1 to 7 in a state or abnormal state as a percentage, each displayed percentage can be reflected in derivation of a pillow shape suitable for the user.

In this regard, as an embodiment of the present invention, the artificial intelligence unit 120 may classify the shape of the cervical vertebrae shown in the X-ray image into one of an ideal state, a normal state, or an abnormal state, excluding the ideal state. The degree of deformation of the cervical vertebrae can be expressed as a percentage by comparing the shape of the cervical vertebrae in a normal or abnormal state with the shape of the cervical vertebrae in an ideal state, and if the degree of deformation of the cervical vertebrae expressed as a percentage is 8% or more, the deformation It is possible to derive the shape of a pillow that can improve the degree of s to 2%.

In addition, when the degree of deformation of the cervical vertebrae expressed as a percentage is within the range of 4 to 8%, the artificial intelligence unit 120 can derive a pillow shape to improve the degree of deformation to 2% or less, When the degree of deformation of the cervical vertebrae is less than 4%, a pillow shape that can completely improve the degree of deformation can be derived.

At this time, the user-customized pillow shape derived by the artificial intelligence unit 120 considers not only the shape of the user's cervical vertebrae, but also the shape of the neck directly in contact with the pillow, and in addition, the length and thickness of the user's neck are More detailed dimensions can be considered.

Accordingly, various types of customized pillow shapes can be stored in the database 100, and one of the shapes is derived by the artificial intelligence unit 120, and a specific detailed dimension is changed according to the user, thereby optimizing for the user. It is a configuration capable of manufacturing customized pillows.

In addition, the generation unit 130 is a device that generates a 3D file required for 3D output, and outputs the user-customized pillow shape as an actual shape by using the user-customized pillow shape 3D modeled by the modeling unit 110. It generates a 3D file for, and the output unit 140 is a device that outputs a user-customized pillow in real form by the modeling unit 110, and may be configured with any of the 3D printers distributed in the market. Since it is a device, detailed descriptions of the configuration and usage method of the device will be omitted.

The embodiments introduced above are provided by way of example in order to sufficiently convey the technical idea of the present invention to those of ordinary skill in the technical field to which the present invention pertains, and the present invention is based on the above-described embodiments. It is not limited and may be embodied in other forms.

In order to clearly describe the present invention, parts irrelevant to the description are omitted from the drawings, and in the drawings, the width, length, thickness, etc. of the components may be exaggerated or reduced for convenience.

In addition, the same reference numbers throughout the specification indicate the same elements.

100: database
110: modeling unit
120: Artificial Intelligence Department
130: generation unit
140: output

Claims (6)

A database 100 in which a standard cervical vertebral region shape and a neck region shape are stored, and an image of a user's cervical vertebrae obtained by X-ray imaging is input and stored;
A modeling unit 110 for 3D modeling the shape of the user's cervical vertebrae and the shape of the neck, respectively, using the input image of the cervical vertebrae, and 3D modeling the shape of a user-customized pillow derived based on the shape of the user's cervical vertebrae;
Artificial intelligence that analyzes the user's cervical and neck conditions by comparing the 3D modeled shape of the user's cervical and neck with the standard cervical and neck shapes, and derives a customized pillow shape based on the analysis results. Part 120;
A generation unit 130 for generating a 3D file for 3D output by using the 3D modeled user-customized pillow shape; And,
An output unit 140 for outputting a user-customized pillow in real form using the generated 3D file; It is composed including,
Among the data stored in the database 100, the shape of the standard cervical vertebra and the shape of the neck,
Consisting of a 2D image and a corresponding 3D shape so that the user's cervical vertebrae can be determined through comparison with the image of the user's cervical vertebrae,
The artificial intelligence unit 120,
It is stored in the database 100, by performing machine learning learning based on the standard cervical shape and neck shape, which are configured including the 2D image and the 3D shape,
It is configured to distinguish and specify the user's cervical vertebrae and neck from the user's cervical vertebral region image input into the database 100,
The artificial intelligence unit 120,
By performing deep learning learning based on the shape of the standard cervical vertebra and the shape of the neck stored in the database 100,
It is configured to regenerate the image of the user's cervical vertebrae input to the database 100 at various angles,
In the database 100,
It is medically configured to include both an ideal cervical state, a cervical state in which the degree of deformation is within the normal range, and an abnormal cervical state,
X-ray pictures of the cervical vertebrae required for machine learning learning and deep learning learning of the artificial intelligence unit 120 are stored,
The artificial intelligence unit 120,
Compare the shape of the cervical vertebrae shown in the image of the user's cervical vertebrae with the shape of the standard cervical vertebrae and indicate the degree of deformation of the cervical vertebrae as a percentage,
If the degree of deformation is more than 8%, a customized pillow shape that can improve the degree of deformation to the level of 2% is derived,
When the degree of deformation is within the range of 4 to 8%, a customized pillow shape that can improve the degree of deformation to 2% or less is derived,
3D three-dimensional pillow manufacturing apparatus, characterized in that configured to derive a user-customized pillow shape capable of completely improving the degree of deformation when the degree of deformation of the cervical spine is less than 4%.
delete delete delete delete The method of claim 1,
In the artificial intelligence unit 120,
A 3D three-dimensional pillow manufacturing apparatus, characterized in that Google's Tensorflow is applied.

Images