KR102501842B1 - Health care system using motion recognition algorithm - Google Patents

Abstract

Disclosed is a healthcare system using a motion recognition algorithm. The present invention comprises: a camera device that generates image data comprising a target; a computing device that receives the image data and runs a customized exercise program; and a display device that displays a screen according to the customized exercise program. The computing device recognizes the target from the image data and extracts skeleton data, and the skeleton data includes feature points for a specific body area of the target. The computing device generates exercise information of the target based on the feature points, and the image data may be generated based on a scenario provided by the customized exercise program. The present invention provides a personalized exercise program, detects the user's exercise information according to the customized exercise program, and takes care of the user's health based on the detected exercise information.

Description

Health care system using motion recognition algorithm {HEALTH CARE SYSTEM USING MOTION RECOGNITION ALGORITHM}

The present invention relates to a health care system using an exercise recognition algorithm. More specifically, the present invention relates to a health care system that generates skeleton data through vision recognition and recognizes exercise information by recognizing a specific body region.

Recently, as interest in health has increased, many technologies for more efficient health management have been studied. Technologies such as health devices capable of checking individual health, such as blood pressure monitors, blood glucose meters, and body fat measuring devices, and IT-integrated fall detection sensors, medication sensors, and electrocardiogram measurement sensors, are continuously developing.

Despite the individual development of these health devices, when a problem occurs in a part of the body, it can be fatal to health, so a technology for integrally managing information on various health devices is required.

Also, in general, a state in which humans are physically, mentally, and socially healthy is referred to as wellness, and such wellness can be realized through appropriate exercise.

Therefore, individuals who want to realize wellness need to accurately grasp their current health status, need to continue appropriate exercise to maintain and improve their body and health status, and need to accurately diagnose the occurrence of diseases. there is

However, since it is difficult for individuals to continuously and accurately identify their current health condition, continue appropriate exercise to maintain and improve their body and health condition, and accurately diagnose the occurrence of disease, it is difficult to realize such wellness. Various methods for this have been proposed.

Republic of Korea Patent Publication No. 10-2018-0088073 (Title of Invention: Healthcare Program Management Method and Electronic Device Therefor)

An object of the present invention is to provide a system capable of providing a personalized exercise program, detecting user's exercise information according to the customized exercise program, and taking care of the user's health based on the detected exercise information.

In order to solve the above problems, the present invention provides a camera device for generating image data including a target, a computing device for receiving the image data and driving a customized exercise program, and a display displaying a screen according to the customized exercise program. An apparatus, wherein the computing device recognizes the target from the image data and extracts skeleton data, the skeleton data includes feature points of a specific body region of the target, and moves the target based on the feature points. Information may be generated, and the image data may be generated based on a scenario provided by the tailored exercise program.

In addition, the computing device generates a first three-dimensional area and a second three-dimensional area based on the feature points, the first three-dimensional area moves along with the movement of a specific body region of the target, and the second three-dimensional area is the target It may be a reference area for calculating the movement of a specific body area of .

In addition, the computing device generates a first point where the first stereoscopic area overlaps with the second stereoscopic area as a starting point according to a definition input in advance, and the time at which the first stereoscopic area returns to the first point Information is acquired, and exercise information of the target may be generated based on the time information.

Also, the computing device may move the second stereoscopic area when the first stereoscopic area does not return to the first point for a predetermined time.

In addition, the computing device moves the second stereoscopic area based on trajectory information of the first stereoscopic area, and according to a pre-input definition, a second stereoscopic area overlapping the first stereoscopic area and the moved second stereoscopic area. Points can be recreated as starting points.

The specific body region may be a foot region of the target, the target's exercise information may be pedal exercise information of the target, and the pedal exercise information may include pedal exercise speed information and pedal exercise time information of the target. .

In addition, the health care system further includes a server that obtains body information or health information of the target based on exercise information of the target, and the server transmits diagnosis information according to the body information or health information of the target to the computer. can be sent to the device.

According to the present invention, a target in an image may be vision-recognized to extract skeleton data, and motion information of the target may be obtained based on the extracted skeleton data.

In addition, the present invention transmits the acquired exercise information to the outside and

Effects obtainable according to the present invention are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description below. will be.

1 shows a health care system using an exercise recognition algorithm according to the present invention.
2 schematically shows a camera device according to the present invention.
3 is a schematic representation of a computing device according to the present invention.
4 shows an example of skeleton data according to the present invention.
5 is an example explaining the trigger of skeleton data recognition according to the present invention.
6 and 7 are examples of three-dimensional regions according to the present invention.
8 shows an example of implementing a health care system according to the present invention.
9 to 12 show examples of scenarios provided by a customized exercise program according to the present invention.
13 illustrates an example of receiving diagnostic data based on exercise information according to the present invention.
FIG. 14 illustrates a motion recognition algorithm for the hwatu game of FIG. 11 .
15 is a diagram showing an example of an image correction filter according to the present invention.
16 shows a HSV graph according to the present invention.
The accompanying drawings, which are included as part of the detailed description to aid understanding of the present specification, provide examples of the present specification and describe technical features of the present specification together with the detailed description.

Hereinafter, the embodiments disclosed in this specification will be described in detail with reference to the accompanying drawings, but the same or similar elements are given the same reference numerals regardless of reference numerals, and redundant description thereof will be omitted.

In addition, in describing the embodiments disclosed in this specification, if it is determined that a detailed description of a related known technology may obscure the gist of the embodiment disclosed in this specification, the detailed description thereof will be omitted.

In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in this specification, the technical idea disclosed in this specification is not limited by the accompanying drawings, and all changes included in the spirit and technical scope of the present invention , it should be understood to include equivalents or substitutes.

Terms including ordinal numbers, such as first and second, may be used to describe various components, but the components are not limited by the terms. These terms are only used for the purpose of distinguishing one component from another.

Singular expressions include plural expressions unless the context clearly dictates otherwise.

In this application, terms such as "comprise" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Hereinafter, a health care system using an exercise recognition algorithm according to a preferred embodiment of the present specification will be described in detail based on the above information.

1 shows a health care system using a motion recognition algorithm according to the present invention, FIG. 2 schematically shows a camera device according to the present invention, and FIG. 3 schematically shows a computing device according to the present invention.

According to FIG. 1 , in a health care system using a motion recognition algorithm according to the present invention, the motion recognition system may include a camera device 100 , a computing device 200 , and a display device 300 . In addition, the health care system using the motion recognition algorithm according to the present invention may further include an RFID module 400 and a server 500.

According to FIGS. 1 and 2 , the camera device 100 according to the present invention may generate image data including a target. The camera device 100 is a component that generates image data from an optical image, and may be driven in units of camera modules.

According to FIG. 2 , the camera device 100 according to the present invention may include a housing including a through hole on a side wall, a lens 1321 installed in the through hole, and a driving unit 1323 that drives the lens 1321 . The through hole may have a size corresponding to the diameter of the lens 1321 . The lens 1321 may be inserted into the through hole.

The driving unit 1323 may be configured to control the lens 1321 to move forward or backward. The lens 1321 and the driving unit 1323 are connected in a conventionally known manner, and the lens 1321 may be controlled by the driving unit 1323 in a conventionally known manner. In order to obtain various image data, the lens 1321 needs to be exposed to the outside of the camera device 100 or the housing.

According to FIG. 1, the computing device 200 according to the present invention is a configuration for receiving image data and driving a customized exercise program, and includes a calculation function as a personal terminal such as a personal computer, server terminal, smart phone, mobile phone, laptop, etc. It can mean any configuration that does. The computing device 200 according to the present invention may transmit screen data to the display device 300 . The computing device 200 according to the present invention recognizes the target in image data and extracts skeleton data, the skeleton data includes feature points for a specific body region of the target, and generates exercise information of the target based on the feature points. can do.

In this case, the image data may be data generated based on a scenario provided by a customized exercise program, and various scenarios provided by a customized exercise program will be described later.

According to FIG. 1, the computing device 200 according to the present invention generates a first stereoscopic area and a second stereoscopic area based on feature points, the first stereoscopic area moves along with the movement of the specific body area of the target, and 2 The three-dimensional region may be a reference region for calculating motion of a specific body region of the target.

In this case, the computing device 200 according to the present invention may move the second stereoscopic area when the first stereoscopic area does not return to the first point for a predetermined time.

In addition, the computing device 200 according to the present invention moves the second stereoscopic area based on the trajectory information of the first stereoscopic area, and according to a pre-input definition, the first stereoscopic area overlaps the moved second stereoscopic area. The second point can be recreated as the starting point.

In this case, the specific body region may be the target's foot region, the target's exercise information may be pedal exercise information of the target, and the pedal exercise information may include pedal exercise speed information and pedal exercise time information of the target.

According to FIG. 1 , a display device 300 according to the present invention may refer to a device that receives screen data from a processor and displays it so that a user can check it through senses. The display device 300 may include a self-emissive display panel or a non-emissive display panel. For example, an OLED panel that does not require a backlight may be exemplified as a self-emissive display panel, and an LCD panel that requires a backlight may be exemplified as a non-emissive display panel, but is not limited thereto.

According to FIG. 1 , the health care system using the motion recognition algorithm according to the present invention may further include an RFID module 400 . The RFID module 400 may include user's unique ID information. The RFID module 400 may be provided in the form of a user's wrist strap or driven by utilizing an NFC module included in a user terminal.

According to FIG. 3 , a computing device 200 according to the present invention may include a processor 210 , a memory 220 and a communication module 230 .

The processor 210 may execute commands stored in the memory 220 to control other components. The processor 210 may execute instructions stored in the memory 220 .

The processor 210 is a component capable of performing calculations and controlling other devices. Mainly, it may mean a central processing unit (CPU), an application processor (AP), a graphics processing unit (GPU), and the like. Also, the CPU, AP, or GPU may include one or more cores therein, and the CPU, AP, or GPU may operate using an operating voltage and a clock signal. However, while a CPU or AP consists of a few cores optimized for serial processing, a GPU may consist of thousands of smaller and more efficient cores designed for parallel processing.

The processor 210 may provide or process appropriate information or functions to a user by processing signals, data, information, etc. input or output through the components described above or by running an application program stored in the memory 220.

The memory 220 stores data supporting various functions of the computing device 200 . The memory 220 may store a plurality of application programs (or applications) running in the computing device 200 , data for operating the computing device 200 , and instructions. At least some of these application programs may be downloaded from an external server through wireless communication. Also, the application program may be stored in the memory 220, installed in the computing device 200, and driven by the processor 210 to perform an operation (or function) of the computing device 200.

The memory 220 may be a flash memory type, a hard disk type, a solid state disk type, a silicon disk drive type, or a multimedia card micro type. ), card-type memory (eg SD or XD memory, etc.), RAM (random access memory; RAM), SRAM (static random access memory), ROM (read-only memory; ROM), EEPROM (electrically erasable programmable read -only memory), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, and an optical disk. Also, the memory 220 may include a web storage performing a storage function on the Internet.

The communication module 230 transmits and receives information with a base station or a camera having a communication function through an antenna. The communication module 230 may include a modulation unit, a demodulation unit, a signal processing unit, and the like. Also, the communication module 230 may perform a wired communication function.

Wireless communication may refer to communication using a wireless communication network using a communication facility previously installed by telecommunication companies and a frequency of the communication facility. At this time, the communication module 230 is CDMA (code division multiple access), FDMA (frequency division multiple access), TDMA (time division multiple access), OFDMA (orthogonal frequency division multiple access), SC-FDMA (single carrier frequency division multiple access), and the like, as well as the communication module 230 can also be used for 3rd generation partnership project (3GPP) long term evolution (LTE). In addition, not only 5G communication, which is currently being commercialized, but also 6G, which is scheduled to be commercialized later, can be used. However, the present specification may utilize a pre-installed communication network without being bound by such a wireless communication method.

In addition, as a short range communication technology, Bluetooth, BLE (Bluetooth Low Energy), Beacon, RFID (Radio Frequency Identification), NFC (Near Field Communication), infrared communication (Infrared Data Association; IrDA) ), UWB (Ultra Wideband), ZigBee, etc. may be used.

The server 500 according to the present invention is a component for acquiring body information or health information of the target based on the target's exercise information, and may include a communication module. The server 500 may communicate with a user terminal for medical staff and receive diagnosis information based on body information or health information of a target from the user terminal for medical staff. In addition, the server 500 includes a learning module using big data therein to generate automatic diagnostic data for the target's exercise information based on previously input basic information (age, gender, body size, etc.) of the target. can For example, based on the target's exercise information, the server 500 may derive information about the target's body age.

4 shows an example of skeleton data according to the present invention.

According to FIG. 4 , the skeleton data according to the present invention may include at least one feature point formed around the joint of the target. Skeleton data may be data for recognizing a person through vision recognition in order to roughly recognize motion information of a target.

In addition, the computing device 200 may remove z-axis coordinate values from coordinate information of feature points included in the skeleton data in order to generate the skeleton data. When the z-axis coordinate value is removed, the skeleton data or feature points included in the skeleton data can be treated as data in a two-dimensional space. In this case, the z-axis coordinate value may be a coordinate value corresponding to a depth value in 3D coordinates.

5 is an example explaining the trigger of skeleton data recognition according to the present invention.

According to FIG. 5 , the computing device 200 according to the present invention detects whether a target is located in a predetermined area, recognizes the skeleton data of the target as a trigger when the target is located in the predetermined area, and displays the display device 300 ) can be printed.

The computing device 200 according to the present invention may use a coordinate conversion algorithm to detect whether a target is located in a predetermined area. The camera device 100 is installed to face the target in front, but may be displayed looking down from above on the display screen of FIG. 5 . Through this, the present invention can induce the target to be accurately located in a predetermined area.

In this way, the reason why the target is positioned in the predetermined area is that a feature point corresponding to the foot of the target must be obtained when a motion of a bicycle is detected, and for this purpose, the angle of view of the camera must include the foot of the target.

6 and 7 are examples of three-dimensional regions according to the present invention.

According to FIG. 6 , a stereoscopic area according to the present invention may include a first stereoscopic area and a second stereoscopic area. The first stereoscopic region is generated based on feature points corresponding to a specific body region of the target, and the second stereoscopic region may be a criterion for determining motion of the first stereoscopic region. The first three-dimensional region may correspond to the target's foot, and may be composed of two to correspond to the left foot and the right foot. Since the second stereoscopic regions are generated corresponding to the two first stereoscopic regions, there may be two.

According to FIG. 7 , the first three-dimensional region may perform a trajectory movement according to a pedal movement. At this time, since the foot moves according to the trajectory of the pedal, the first three-dimensional region may move in the same way as the foot stepping on the pedal.

The second solid region may be a reference for trajectory motion of the first solid region. When the first three-dimensional region moves along a trajectory different from the pedal movement, a circle is drawn, and while one circle is drawn, the first three-dimensional region comes into contact with the second three-dimensional region once. In this case, the pedal movement may be measured based on a case in which contact between the first and second stereoscopic regions occurs. At this time, the two first three-dimensional regions and the two second three-dimensional regions can sense a contact only with respect to a pair corresponding to each other.

Also, when generating the first three-dimensional region, the computing device 200 may stop generating the first three-dimensional region according to a height value (y value) between the position of the foot of the target and the ground or the reference plane. That is, the computing device 200 may create the first three-dimensional region only when the difference between the height values (y values) of both feet of the target is greater than or equal to a predetermined size (ex-20 cm). Conversely, when the difference between the height values (y values) of both feet of the target is smaller than a predetermined size (ex-20 cm), the computing device 200 may fail without generating the first three-dimensional region. In this case, in order to compare the size of the y value, 2-dimensional skeleton data from which the z value has been previously removed may be used.

8 shows an example of implementing a health care system according to the present invention.

According to FIG. 8, the healthcare system according to the present invention may provide the RFID module 400 in the form of an RFID tag wristband, and may provide a display, a computing device 200, and a camera device 100 integrally. there is. Accordingly, the user (target) transmits his/her unique ID information to the computing device 200 by tagging the RFID included in the wristband, and the computing device 200 transmits various information according to user-customized data stored in advance based on the unique ID. Exercise scenarios can be provided.

9 to 12 show examples of scenarios provided by a customized exercise program according to the present invention.

9 illustrates a surfing program provided by a health care system using an exercise recognition algorithm according to the present invention. At this time, the computing device 200 recognizes both feet of the target with the skeleton data, creates first three-dimensional areas corresponding to both feet, and recognizes the surfing state of the target based on the change in position of the first three-dimensional areas. can Based on the recognition result, the computing device 200 may detect whether the first stereoscopic area accurately performs the mission operation of the given program.

10 illustrates an upper body exercise program provided by a health care system using an exercise recognition algorithm according to the present invention. At this time, the computing device 200 recognizes both hands of the target using the skeleton data, creates first three-dimensional regions corresponding to the two hands, and recognizes the motion state of the target's arms based on the change in position of the first three-dimensional regions. can Based on the recognition result, the computing device 200 may detect whether the first stereoscopic area accurately performs the mission operation of the given program.

11 illustrates an entertainment program, the Hwatu game, provided by the health care system using the exercise recognition algorithm according to the present invention. The entertainment program according to the present invention can induce a large motion of the target. At this time, the computing device 200 may recognize both hands of the target using the skeleton data, create first three-dimensional regions corresponding to both hands, and recognize the motion of the target based on the change in position of the first three-dimensional regions. . Based on the recognition result, the computing device 200 may detect whether the first stereoscopic area accurately performs the mission operation of the given program.

12 illustrates a bicycle exercise program provided by a health care system using an exercise recognition algorithm according to the present invention. The bicycle exercise program according to the present invention can recognize the pedal exercise of the target. In this case, the computing device 200 may recognize both feet of the target using the skeleton data, create first three-dimensional regions corresponding to both hands, and recognize the motion of the target based on the change in position of the first three-dimensional regions. . Based on the recognition result, the computing device 200 may detect whether the first stereoscopic area accurately performs the mission operation of the given program.

13 illustrates an example of receiving diagnostic data based on exercise information according to the present invention.

According to FIG. 13 , the computing device 200 according to the present invention may receive image data of medical staff from the server 500 . The health care system using the motion recognition algorithm according to the present invention transmits image data captured by the target to a terminal of a medical staff when the target operates according to the program, and can help the medical staff diagnose the target through the image data in real time. there is.

FIG. 14 illustrates a motion recognition algorithm for the hwatu game of FIG. 11 .

According to FIG. 14 , the motion recognition algorithm for the Hwatu game according to the present invention may be a motion recognition algorithm using a plurality of virtual triggers. The computing device 200 according to the present invention may extract skeleton data of a target based on image data generated by a camera device and recognize both hands of the target from the extracted skeleton data.

The computing device 200 may create two three-dimensional regions respectively corresponding to both hands of the target. The two three-dimensional areas can move along with the movement of both hands of the target. The two stereoscopic regions may be a third stereoscopic region and a fourth stereoscopic region.

The computing device 200 may define that a specific operation is performed when the third or fourth stereoscopic area collides with the virtual trigger. In this case, the first trigger box among the plurality of virtual triggers may be located in an area above the feature point corresponding to the head among the skeleton data. Also, a second trigger box among a plurality of virtual triggers may be located in a region to the left of a feature point corresponding to a bright spot among skeleton data. In addition, a third trigger box among a plurality of virtual triggers may be located in a right region of a feature point corresponding to a pituitary gland among skeleton data.

The computing device 200 may define that the left card (Hwatu hand) is selected when the third stereoscopic area and the second trigger box collide. In addition, the computing device 200 may define that the right card (Hwatu hand) is selected when the fourth stereoscopic area and the third trigger box collide. In addition, the computing device 200 may define that a card (hwatu hand) is submitted when the third or fourth stereoscopic area collides with the first trigger box.

15 is a diagram showing an example of an image correction filter according to the present invention, and FIG. 16 shows a HSV graph according to the present invention.

A vision recognition algorithm for obtaining skeleton data according to the present invention may include a CNN algorithm. A CNN algorithm may be a learning algorithm using a plurality of layers. The algorithm for vision recognition according to the present invention may be performed by a learning module pre-learned through big data, and a certain type of filter may be applied to image data for vision recognition.

According to FIG. 15, the types and functions of filters are shown. In other words, the CNN algorithm can automatically learn a filter that maximizes image classification accuracy, and by adding a new layer called a convolutional layer and a polling layer before the fully connected layer, the filtered image after applying the filtering technique to the original image. A classification operation can be performed on . The CNN algorithm is configured to apply a filtering technique to the original image by adding a new layer, called a convolutional layer and a pooling layer, before the fully-connected layer, and then perform a classification operation on the filtered image. can

At this time, the operation expression for the image filter using the CNN algorithm is as shown in Equation 1 below.

[Equation 1]

(step,

: 행렬로 표현된 필터링된 이미지의 i번째 행, j번째 열의 픽셀,

: pixels in the i-th row and j-th column of the filtered image expressed as a matrix,

: 필터,

: filter,

: 이미지,

: image,

: 필터의 높이 (행의 수),

: height of the filter (number of rows),

: 필터의 너비 (열의 수)이다. )

: The width of the filter (number of columns). )

Preferably, the operation expression for the image filter using the CNN algorithm is as shown in Equation 2 below.

[Equation 2]

(step,

: 행렬로 표현된 필터링된 이미지의 i번째 행, j번째 열의 픽셀,

: pixels in the i-th row and j-th column of the filtered image expressed as a matrix,

: 응용 필터

: Application filter

: 이미지,

: image,

: 응용 필터의 높이 (행의 수),

: height of application filter (number of rows),

: 응용 필터의 너비 (열의 수)이다.)

: The width (number of columns) of the application filter.)

는 응용 필터로서 타겟(사람)의 스켈레톤 데이터를 정확히 생성하기 위하여 상기 이미지 데이터에 적용되는 필터일 수 있다. 특히, 타겟(사람) 인식의 경우 형태 및 색감의 차이를 기초로 분류될 수 있으므로, 형태 및 색감 등을 효과적으로 인지하기 위한 응용 필터가 필요할 수 있다. 이러한 필요성을 충족하기 위하여 응용 필터

는 아래의 수학식 3에 의하여 연산될 수 있다. Preferably,

is an application filter, and may be a filter applied to the image data in order to accurately generate skeleton data of a target (human). In particular, since target (person) recognition can be classified based on differences in shape and color, an application filter for effectively recognizing shape and color may be required. Application filters to meet these needs

Can be calculated by Equation 3 below.

[Equation 3]

: 필터,

: 계수,

: 응용 필터)(step,

: filter,

: Coefficient,

: application filter)

에 따른 필터는 도 15에 따른 엣지 인식 필터(Edge detection), 샤픈 필터(sharpen) 및 박스 블러 필터(Box blur) 중 어느 하나의 행렬일 수 있다. At this time, each

A filter according to may be a matrix of any one of an edge detection filter according to FIG. 15, a sharpen filter, and a box blur filter.

를 구하는 연산식은 아래의 수학식 4와 같다. 이때,

는 필터의 효율을 높이기 위하여 사용되는 하나의 변수로서 해석될 수 있으며, 그 단위는 무시될 수 있다. Preferably,

The operation expression for obtaining is as shown in Equation 4 below. At this time,

can be interpreted as a variable used to increase the efficiency of the filter, and its unit can be ignored.

[Equation 4]

However, the diameter (diameter) of the lens of the camera device used to take the image is in mm, the f-value of the lens is the aperture value (F number) of the lens, and the HSV average value is the size value through the HSV graph of the color coordinates according to the image It may mean a value obtained by averaging the values converted to . Details of the HSV value are as follows.

According to FIG. 16, the HSV graph according to the present invention may mean a color space in which perceptual characteristics are reflected. H (Hue, 0 to 360°) may mean hue, S (Saturation, 0 to 100%) may mean saturation, and V (Value, 0 to 100%) may mean lightness. . Hue, saturation, and lightness values can be referred to as HSV values, which can be easily extracted using graphic tools such as Adobe illustrator cc 2019.

The average HSV value according to the present invention can be obtained through the 3-dimensional HSV coordinates of FIG. 16 . That is, the average HSV value according to the present invention may be calculated based on the HSV value obtained through a graphic tool. The distance value of the HSV coordinates measured with the origin coordinates on the HSV 3-dimensional coordinates as a reference point may constitute the above-described HSV average value. That is, the image according to the present invention includes an image of a target (human), and the HSV color coordinates of the image may be distributed in a certain area among HSV three-dimensional coordinates. Therefore, the HSV average value according to the present invention may mean a distance value from the origin coordinates calculated based on average coordinates using the average of HSV coordinates for colors of a certain region.

[Experimental example]

Regarding the image data according to the present invention, when the application filter F' of the present invention is applied, the accuracy of skeleton data for the target (human) is as follows. The table below shows the accuracy of the recognition result as a numerical value, depending on whether a filter is applied or not, commissioned by 10 experts in the relevant technical field.

no filter applied

적용filter

apply filter

apply Average Accuracy (Score) 75 90 97

Table 1 shows the scores for accuracy evaluated by experts for each case. In this experimental example, the accuracy of the learning module pre-learned through big data was investigated according to whether or not a filter was applied.

In addition, this experimental example was experimented with image data of 20 different people, and 10 experts surveyed the accuracy of the skeleton data generated in the image data with respect to the recognition result value.

As can be seen in Table 1, when the filter is not applied, there is a probability of error in image recognition, and the evaluation is relatively low accuracy. In contrast, the accuracy was slightly higher when the general CNN filter F was applied, but it was confirmed that the accuracy was remarkably improved when the applied filter F' was applied.

The above-described present invention can be modeled as a computer readable code on a medium on which a program is recorded. The computer-readable medium includes all types of recording devices in which data readable by a computer system is stored. Examples of computer-readable media include Hard Disk Drive (HDD), Solid State Disk (SSD), Silicon Disk Drive (SDD), ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. , and also includes those modeled in the form of carrier waves (eg, transmission over the Internet). Accordingly, the above detailed description should not be construed as limiting in all respects and should be considered illustrative. The scope of this specification should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of this specification are included in the scope of this specification.

Any or other embodiments of the present invention described above are not mutually exclusive or distinct. Certain or other embodiments of the present invention described above may be used in combination or combination of respective configurations or functions.

The above detailed description should not be construed as limiting in all respects and should be considered illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention.

10: health care system
100: camera
200: computing device
300: display device
400: RFID module
500: server

Claims (7)

a camera device generating image data including a target;
a computing device that receives the image data and drives a customized exercise program; and
A display device displaying a screen according to the customized exercise program; including,
The computing device,
Recognizing the target from the image data and extracting skeleton data, the skeleton data including feature points for a specific body region of the target, and generating exercise information of the target based on the feature points;
The video data is
It is created based on the scenario provided by the customized exercise program,
The computing device,
generating a first solid region and a second solid region based on the feature points;
The first three-dimensional region moves along with the movement of the specific body region of the target, and the second three-dimensional region is a reference region for calculating the movement of the specific body region of the target;
According to a definition input in advance, a first point where the first solid region and the second solid region overlap is created as a starting point;
obtaining information on a time when the first stereoscopic region returns to the first point, and generating motion information of the target based on the time information;
The image data is to which an image correction filter based on the equation 2 below is applied.
A health care system using an exercise recognition algorithm.
[Equation 2]

: pixels in the i-th row and j-th column of the filtered image expressed as a matrix,

: Equation 3 below is applied as an application filter.

: image,

: height of application filter (number of rows),

: The width (number of columns) of the application filter.)
[Equation 3]

(step,

: filter,

: Coefficient,

: application filter)
delete delete According to claim 1,
The computing device,
moving the second stereoscopic area when the first stereoscopic area does not return to the first point for a predetermined time period;
A health care system using an exercise recognition algorithm.
According to claim 4,
The computing device,
The second solid region is moved based on the trajectory information of the first solid region, and the second point where the first solid region overlaps the moved second solid region is regenerated according to a previously input definition as a starting point. to do,
A health care system using an exercise recognition algorithm.
According to claim 1,
The specific body region is a foot region of the target,
The movement information of the target is pedal movement information of the target, and the pedal movement information includes pedal movement speed information and pedal movement time information of the target.
A health care system using an exercise recognition algorithm.
According to claim 1,
The health care system,
Further comprising a server that obtains body information or health information of the target based on the exercise information of the target;
The server,
Transmitting diagnostic information according to the target's body information or health information to the computing device;
A health care system using an exercise recognition algorithm.

Images