Method and apparatus for predicting physical injury risk based on user posture recognition

Abstract

Provided are a method and an apparatus for predicting a risk of physical injury based on a user posture recognition infrastructure. **Solution**: According to the present invention, in a method for predicting a risk of physical injury based on a user posture recognition infrastructure, the method includes: providing a video related to body posture measurement; obtaining image data related to the user's body posture using at least one image sensor; extracting body joint coordinates through body part-by-part segmentation based on the image data; recognizing the user's body posture based on the extracted body joint coordinates; measuring a physical injury risk factor for at least one joint based on the recognized body posture; and predicting the risk level of physical injury for at least one body part based on the injury risk factor. A method for predicting a risk of physical injury is provided.

Description

Technical Field

[0001] The present invention relates to a method for predicting physical injury risk based on user posture recognition infrastructure and an apparatus using the same. More specifically, the present invention relates to a method, apparatus or system capable of predicting and managing physical injury risk with improved reliability through measurement of postures and behaviors in both conscious and unconscious states of a user.

Background Art

[0002] In the case of conventional body measurement or posture measurement devices, in most cases, the person to be measured takes a certain action and the measurement is performed by an operator through operation. At this time, the measurement is performed using an image sensor such as a camera. After capturing an image, the body size or body posture is measured through a certain calculation, and after the measurement, the measurement result is transmitted to the person to be measured through a measuring instrument, an application or a printed matter.

[0003] However, such conventional body measurement or posture measurement technology is a method of measurement when the person to be measured takes a posture according to the guidance of the administrator. Therefore, it is not the natural posture of the person to be measured, but is likely to be a posture deformed in a conscious state by consciously imitating the measurement guidance. In such a conscious state, there is a problem that it is difficult to accurately obtain normal physical vulnerability points or make an accurate diagnosis.

[0004] Therefore, there is a need for a method and an apparatus for more accurately predicting physical vulnerability points and injury risks caused by them by separately measuring the unconscious state and the conscious state during body and posture measurement of the person to be measured and finally making a comprehensive judgment.

Summary of the Invention

Problems to be Solved by the Invention

[0005] The object of the present invention is to provide a method and an apparatus that can predict and manage the risk of physical injury with improved reliability by recognizing both the conscious state and the unconscious state of a user and performing posture measurement in predicting the risk of physical injury based on user posture recognition.

[0006] Furthermore, the object of the present invention is to provide a method and an apparatus that can more accurately identify physical vulnerability points shown in ordinary natural movements in measuring physical injury risk factors through physical posture recognition in the unconscious state by classifying the state of the user into a conscious state and an unconscious state.

[0007] Moreover, the object of the present invention is to provide a method and an apparatus that can more accurately predict the risk of physical injury without increasing the inconvenience to the user by recognizing the natural postures before and after posture measurement without requiring an additional separate operation during user posture measurement.

[0008] In addition, the object of the present invention is to provide an injury management solution for minimizing the risk of injury to the relevant part of the user based on the prediction result of the risk of physical injury, and to provide a method and an apparatus that can provide feedback related to body movement by re-measuring the posture of the user through the injury management solution.

[0009] The problems to be solved by the present invention are not limited to the content mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the following description.

Means for Solving the Problems

[0010] According to an embodiment of the present invention, in a method for predicting physical injury risk of a user posture recognition base, a step of providing an image related to body posture measurement; a step of acquiring image data related to the user's body posture using at least one image sensor; a step of extracting body joint coordinates through body part-by-part segmentation based on the image data; a step of recognizing the user's body posture based on the extracted body joint coordinates; a step of measuring a physical injury risk factor for at least one joint based on the recognized body posture; and a step of predicting the risk degree of physical injury for at least one body part based on the physical injury risk factor; a physical injury risk prediction method including these steps can be provided.

[0011] Also, the step of measuring the physical injury risk factor is also to measure a joint-by-joint injury risk factor based on the inclination of each joint related to the physical injury and the position deviation from a reference point.

[0012] Also, a step of providing an image related to a body posture for injury management based on the predicted risk degree of the physical injury; a step of acquiring image data related to the body posture for the user's injury management using at least one image sensor; and a step of re-measuring a physical injury risk factor for at least one joint based on the recognized body posture based on the image data related to the body posture for the user's injury management; may further be included.

[0013] Also, after the step of recognizing the user's body posture, a step of determining whether the user's body posture is a conscious body posture or an unconscious body posture is further included, and the step of measuring the physical injury risk factor can measure a conscious state joint-by-joint injury risk factor from the conscious state body posture and an unconscious state joint-by-joint injury risk factor from the unconscious state body posture.

[0014] Further, in the step of predicting the risk associated with the bodily injury, the risk of bodily injury to the body part can be calculated based on the injury risk factors for each joint in the conscious state and the injury risk factors for each joint in the unconscious state.

[0015] Also, different weighting values are applied to the injury risk factors for each joint in the conscious state and the injury risk factors for each joint in the unconscious state respectively, to calculate the risk associated with the bodily injury. The weighting value can be variably determined based on at least one of the user's age, gender, and body shape.

[0016] Also, in the step of determining whether the user's body posture is a conscious body posture or an unconscious body posture, the user's conscious state can be determined based on whether the user's position exists within a predetermined body posture measurement position range.

[0017] Also, in the step of determining whether the user's body posture is a conscious body posture or an unconscious body posture, the user's conscious state can be determined based on whether the acquisition time point of the image data related to the user's body posture exists within the time range when the video related to the body posture is provided.

[0018] Also, in the step of determining whether the user's body posture is a conscious body posture or an unconscious body posture, the user's conscious state can be determined based on whether the recognized body posture of the user is similar to the body posture in the video related to the body posture.

[0019] According to another embodiment of the present invention, in an apparatus for predicting physical injury risk of a user posture recognition infrastructure, a display unit configured to provide an image related to body posture measurement; an image data acquisition unit configured to acquire image data related to the user's body posture using at least one image sensor; a joint extraction unit configured to extract body joint coordinates through body part-by-part segmentation based on the image data; a body posture recognition unit configured to recognize the user's body posture based on the extracted body joint coordinates; a physical injury risk factor measurement unit configured to measure a physical injury risk factor for at least one joint based on the recognized body posture; and a physical injury risk prediction unit configured to predict the risk of physical injury for at least one body part based on the physical injury risk factor. A physical injury risk factor measurement device can be provided that includes these components.

[0020] Further, the physical injury risk factor measurement unit may be configured to measure a joint-specific injury risk factor based on the inclination of each joint related to the physical injury and the position deviation from a reference point.

[0021] Further, the display unit is configured to provide an image related to a body posture for injury management based on the predicted risk of the physical injury, the image data acquisition unit is configured to acquire image data related to a body posture for the user's injury management using at least one image sensor, and the physical injury risk factor measurement unit may be configured to re-measure a physical injury risk factor for at least one joint based on the recognized body posture based on the image data related to the body posture for the user's injury management.

[0022] In addition, after the body posture recognition unit recognizes the user's body posture, it further includes a conscious state determination unit configured to determine whether the user's body posture is a conscious body posture or an unconscious body posture. The body injury risk factor measurement unit can be configured to measure the joint-specific injury risk factors in the conscious state from the conscious body posture and the joint-specific injury risk factors in the unconscious state from the unconscious body posture.

[0023] In addition, the body injury risk prediction unit can be configured to calculate the risk degree of body injury to the body part based on the joint-specific injury risk factors in the conscious state and the joint-specific injury risk factors in the unconscious state.

[0024] In addition, the body injury risk prediction unit is configured to calculate the risk degree related to the body injury by applying different weighting values to the joint-specific injury risk factors in the conscious state and the joint-specific injury risk factors in the unconscious state respectively. The weighting value can be variably determined based on at least one of the user's age, gender, and body shape.

[0025] In addition, the conscious state determination unit can determine the user's conscious state based on whether the user's position exists within a predetermined body posture measurement position range.

[0026] In addition, the conscious state determination unit can determine the user's conscious state based on whether the acquisition time point of the image data related to the user's body posture exists within the time range when the video related to the body posture is provided.

[0027] In addition, the conscious state determination unit can determine the user's conscious state based on whether the recognized body posture of the user is similar to the body posture in the video related to the body posture.

Advantages of the Invention

[0028] According to the present invention, in the prediction of physical injury risk of a user posture recognition base, by recognizing both the conscious state and the unconscious state of the user and performing posture measurement, a method and an apparatus capable of predicting and managing a physical injury risk with improved reliability can be provided.

[0029] Also, according to the present invention, a method and an apparatus capable of more accurately identifying physical vulnerability points shown in normal natural movements in the measurement of physical injury risk factors through the recognition of the physical posture in the unconscious state by classifying the state of the user into a conscious state and an unconscious state can be provided.

[0030] Also, according to the present invention, a method and an apparatus capable of more accurately predicting the physical injury risk without increasing the inconvenience of the user by recognizing the natural postures before and after the posture measurement without requiring an additional separate operation during the user posture measurement can be provided.

[0031] Also, according to the present invention, a method and an apparatus capable of providing an injury management solution for minimizing the injury risk of the corresponding part of the user based on the prediction result of the physical injury risk and providing feedback related to the body movement by re-measuring the posture of the user through the injury management solution can be provided.

[0032] The effects of the present invention are not limited to the contents mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

Brief Description of Drawings

[0033]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0034] Hereinafter, based on the accompanying drawings, the present invention will be described in detail so that those having ordinary knowledge in the technical field to which the present invention belongs can easily implement it. However, the present invention can be embodied in various mutually different forms and is not limited to the embodiments described herein.

[0035] The terms used in this specification are for explaining the embodiments and are not intended to limit the present invention. In this specification, the singular expression includes the plural expression unless otherwise specifically mentioned.

[0036] As used in this specification, "comprises" and "comprising" do not exclude the presence or addition of one or more other components, steps, operations, and / or elements.

[0037] In addition, in the description of the present invention, when a specific description of related known art is determined to obscure the gist of the present invention, the detailed description thereof will be omitted.

[0038] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. The configuration of the present invention and the effects thereof will be clearly understood through the following detailed description.

[0039] FIG. 1 is a block diagram for explaining the configuration of a body injury risk prediction device of a user posture recognition base according to an embodiment of the present invention.

[0040] Referring to FIG. 1, the body injury risk prediction device of the user posture recognition base includes a display unit 100, an image sensor 200, a body injury risk prediction processing unit 300, and an input unit 400, and is configured in various forms including a display device and a processing device. For example, it may be configured in any one form or a combination of two or more of a smart mirror, a kiosk, a smart phone, a tablet computer, a desktop computer, a laptop computer, a notebook personal computer, a workstation, a portable computer, a wireless phone, a mobile phone, a digital camera, a television, a wearable device, and is not limited thereto.

[0041] The display unit 100 is configured to provide a guide video, a selection menu, etc. related to body posture measurement to a user such as the person to be measured. For example, it includes, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, an organic light emitting diode (OLED) display, a quantum dot display, a micro LED, a micro electro mechanical systems (MEMS) display, etc. Also, a part of such a display unit 100 may be combined with the input unit 400 and embodied in the form of a touch screen.

[0042] The image sensor 200 is configured to acquire image data related to the user's body posture, and may be composed of a plurality of them so as to be able to acquire image data at one or more positions as necessary. The image sensor 200 is composed of at least one or more 2D or 3D cameras that can photograph and capture the user's outer shape, body posture or movement actions. For example, it is configured in the form of a depth camera capable of distance measurement or a general color camera, and may include one or more of a depth sensor or an infrared sensor.

[0043] In the present invention, the user's body posture has a meaning that includes not only static body postures but also connection information between body postures by dynamic continuous action behaviors, such as action attributes such as the speed and softness of posture movements. For example, when raising the arm in a cautious posture, the case where the postures are smoothly connected, the case of cutting in time, or the case of raising it gradually may be judged as different body postures from each other.

[0044] The body injury risk prediction processing unit 300 extracts body joint coordinates through body part-by-part segmentation based on the image data acquired via the image sensor 200, recognizes the user's body posture based on the extracted body joint coordinates, measures a body injury risk factor for at least one joint based on the recognized body posture, and may be configured to predict the risk degree of body injury for at least one body part based on the body injury risk factor. Further, after recognizing the user's body posture (including actions), the body injury risk prediction processing unit 300 may be configured to determine whether the user's body posture is a conscious body posture or an unconscious body posture, measure the injury risk factor for each joint in the conscious state from the conscious body posture, and measure the injury risk factor for each joint in the unconscious state from the unconscious body posture.

[0045] The body injury risk prediction processing unit 300 performs various control and processing operations related to the user's body posture, motion recognition, analysis, user-provided screens, and video generation, and performs calculations and data processing related to the control and communication of a plurality of internal components. For example, the body injury risk prediction processing unit 300 includes a central processing unit (CPU), an application processor (AP), etc., and includes internally a memory that can store instructions or data related to at least one other component, or can communicate with an external memory to access necessary information when necessary. It also includes programs or program modules that can be executed by one or more processors, can receive user input in conjunction with the user input unit 400, and can output the processed results in conjunction with the display unit 100. In particular, the programs or program modules included in the body injury risk prediction processing unit 300 are configured in the form of an operating system, an application program, or a program, etc., and can be physically stored on various types of widely used storage devices. Such programs or program modules include various forms for performing one or more routines, subroutines, programs, objects, components, instructions, data structures, and specific tasks, and are not limited to those forms. The description related to the specific components of the body injury risk prediction processing unit 300 will be described in detail later with reference to FIG. 2.

[0046] The input unit 400 includes at least one of, but is not limited to, a touch pad, a touch panel, a key pad, a dome switch, a physical button, a jog shuttle, and a microphone for receiving the user's voice, and various other types of sensors in order to receive various inputs for user operations and selections. Further, the input unit 400 may have a form of a touch-input-enabled display integrated with the display unit 100.

[0047] Also, although not shown in FIG. 1, the body injury risk prediction device of the user posture recognition infrastructure further includes a communication unit, and may be configured to transmit and receive various data and the like via a wired or wireless communication network or the like to / from an external user terminal or an external server by the user's input / output operations.

[0048] FIG. 2 is a block diagram for explaining the configuration of a body injury risk prediction processing unit for predicting body injury risk in a user posture recognition infrastructure according to an embodiment of the present invention.

[0049] Referring to FIG. 2, the body injury risk prediction processing unit 300 includes an image data acquisition unit 310, a joint extraction unit 320, a body posture recognition unit 330, a subjective state determination unit 340, a body injury risk factor measurement unit 350, a body injury risk prediction unit 360, an injury management solution providing unit 370, and a feedback providing unit 380, and some of the components may be configured in an integrated form.

[0050] First, the image data acquisition unit 310 may be configured to acquire and store image data related to the user's body posture using at least one image sensor 200.

[0051] The joint extraction unit 320 may be configured to extract body joint coordinates through body part-by-part segmentation based on the image data received from the image data acquisition unit 310. For example, the body image of the user, which is the object of interest in the image including the user's pose, may be extracted first, and the joint coordinates of each body may be extracted through the body part-by-part segmentation operation on the extracted body image.

[0052] The body posture recognition unit 330 may be configured to recognize the user's body posture based on the body joint coordinates extracted from the joint extraction unit 320.

[0053] After the body posture recognition unit 330 recognizes the user's body posture (including actions), the self-awareness state determination unit 340 is configured to determine whether the user's body posture is a self-aware state body posture that is aware of posture measurement, or otherwise, an unconscious state body posture that is not aware of posture measurement. The self-awareness state determination unit 340 is configured to classify the user's state into a self-aware state and an unconscious state based on the user's posture and actions for body posture measurement. For example, as a first condition, it is determined how aware the user is based on whether the user's position exists within a predetermined body posture measurement position range. As a second condition, it is determined how aware the user is based on whether the acquisition time point of the image data related to the user's body posture exists within the time range when the guide video related to the body posture measurement that the user must take is provided or played. As a third condition, it may be configured to determine how aware the user is based on whether the recognized body posture of the user is similar to the body posture of the guide video related to the body posture. For example, such three conditions may all be used as AND conditions to discriminate the user's self-awareness state. In that case, if any one of the above three conditions is not satisfied, the user may be discriminated as being in an unconscious state. Also, if necessary, only two conditions may be used as AND conditions, or all three conditions may be used as OR conditions to discriminate the user's self-awareness state. When the user's state is not a self-aware state, it may all be determined as an unconscious state.

[0054] The physical injury risk factor measurement unit 350 may be configured to measure a physical injury risk factor for at least one joint based on the body posture recognized via the body posture recognition unit 330. The physical injury risk factor measurement unit 350 may be configured to measure a joint-specific injury risk factor based on the inclination of each joint related to physical injury, such as the neck joint, spinal joint, shoulder joint, and pelvis, and the position deviation from the reference point of the joint. A guidance video related to the measurement of the body posture that the user must take is provided via the display unit 100. When the user takes the posture, the physical injury risk factor measurement unit 350 measures the physical injury risk factor in real time.

[0055] Also, the physical injury risk factor measurement unit 350 may be configured to measure the joint-specific injury risk factor in the conscious state from the body posture in the conscious state and individually measure the joint-specific injury risk factor in the unconscious state from the body posture in the unconscious state by distinguishing and determining the conscious state and the unconscious state with the conscious state determination unit 340 before measuring the physical injury risk factor.

[0056] The physical injury risk prediction unit 360 is configured to predict the risk level of physical injury for at least one body part based on the physical injury risk factor measured by the physical injury risk factor measurement unit 350, and is also configured to calculate the risk level of physical injury for the body part based on the joint-specific injury risk factor in the conscious state and the joint-specific injury risk factor in the unconscious state, and may be configured to calculate the comprehensive injury risk level together based on the injury risk level for each body part.

[0057] The physical injury risk prediction unit 360 is also configured to calculate the risk level related to physical injury by applying different weighting values to the joint-specific injury risk factor in the conscious state and the joint-specific injury risk factor in the unconscious state respectively. Here, the weighting value may be variably determined based on at least one of various factors that affect the physical state of the user, such as age, gender, body shape, occupation, lifestyle pattern, exercise history, injury history, and weight. A more specific explanation related to the application of the weighting value will be described later with reference to FIG. 3.

[0058] The injury management solution providing unit 370 generates an injury management solution based on the injury risk prediction result in the physical injury risk prediction unit 360, and can provide, for example, an injury management solution-related video to the user via the display unit 100. Here, the injury management solution is also, for example, a guidance video in the form of stretching or exercise content that can minimize the physical injury risk of the user according to the injury risk prediction result.

[0059] Based on the guidance video provided by the injury management solution, if the user performs the stretching or exercise actions shown in the video, the user can perform body posture recognition in the conscious and unconscious states through body posture image data acquisition, body joint coordinate extraction, body posture recognition, conscious / unconscious state judgment, etc. as performed in the primary body posture measurement stage, and individually analyze the body postures in the conscious and unconscious states to re-measure the physical injury risk factors for the joint. At this time, it is possible to compare the physical injury risk factors re-measured for the joint with the previously measured physical injury risk factors to determine whether the injury risk degree for the body part of the joint has decreased.

[0060] The feedback providing unit 380 can predict the physical injury risk degree of the body part based on the re-measured physical injury risk factors and provide feedback related to body movement at the same time (S770). That is, provide feedback related to the stretching and exercise execution of the user carried out in relation to the injury management solution. Here, the feedback can include content related to precautions and recommendations for exercise behaviors related to injury risk. Additionally, the re-measured physical injury risk factors and the previously measured physical injury risk factors can be compared and the result of the decrease in the physical injury risk degree can be provided together.

[0061] FIG. 3 is an exemplary diagram for explaining a method of predicting physical injury risk through the judgment of the user's conscious and unconscious body postures according to an embodiment of the present invention.

[0062] The real-time determination is performed via the awareness state determination unit 340 as to whether the user's body posture is the body posture in the awareness state or the body posture in the unconscious state, and the posture-specific injury risk factor measurement unit 350 can perform the posture-specific injury risk factor measurement by individually using the body posture data in the awareness state and the body posture data in the unconscious state. At this time, the posture-specific injury risk factor is measured based on the inclination angles of joints, for example, the measured values for each part of the body posture such as the angle of the head, the spinal angle, the shoulder angle, the pelvic angle, etc., and the individually measured body injury risk factors in the awareness and unconscious states are quantitatively listed and prioritized in the order of urgency and importance, and respective weighted values can be assigned individually.

[0063] In the body injury risk prediction unit 360, by applying different weighted values w1 and w2 to the measured values (evaluation scores) of the body injury risk factors in the awareness state and the measured values (evaluation scores) of the body injury risk factors in the unconscious state respectively, the body injury risk factors individually scored can finally perform the body injury risk degree prediction through comprehensive evaluation. The final comprehensive evaluation calculates the body injury risk factors measured in the awareness state and the unconscious state as quantified scores with weighted values, and the comprehensive evaluation items, for example, score the injury risk degrees of each body part such as the neck, shoulders, knees, pelvis (hip joint), waist, etc. respectively, and in order to make it easier for the user to intuitively grasp the comprehensive injury risk degree score based on the injury risk degree scores of each body part together, it can be provided as a graphic such as a figure.

[0064] Thus, the injury risk factor of the body part = (the evaluation score of the body injury risk factor in the awareness state * the weighted value (w1)) + (the evaluation score of the body injury risk factor in the unconscious state * the weighted value (w2)) is calculated. For example, among the body parts, the predicted score of the neck's body injury risk = 79 (the awareness state turtle neck risk evaluation score) * 0.44 (w1) + 85 (the unconscious state turtle neck risk evaluation score) * 0.54 (w2) = 80.7 points can be calculated.

[0065] Figure 4 is an exemplary diagram for explaining a method of determining the user's body postures in the awareness state and the unconscious state according to an embodiment of the present invention.

[0066] The self-awareness state determination unit 340 is configured to classify the user's state into a self-aware state and an unconscious state based on the user's posture and actions for body posture measurement, and can perform body posture recognition via the body posture recognition unit 330 previously (S410).

[0067] For example, based on the body posture recognition result, first, whether the user's position exists within a predetermined body posture measurement position range as a first condition can be used to determine the user's self-awareness state (S420). For example, the user's position can be obtained using the joint coordinates (x, z) extracted from the image data shown by the user.

[0068] Next, as a second condition, whether the acquisition time point of the image data related to the user's body posture exists within the time range when the guide video related to the body posture measurement that the user must take is provided or played can be used to determine the user's self-awareness state (S430). Since the user consciously imitates the body posture that must be taken through the video provided on the display unit 100, within the time range when the guide video is played on the display unit 100, the user's body posture can be judged in a self-aware state, and within the time range when the guide video is not played on the display unit 100, the user's body posture can be judged in an unconscious state where the user is not aware of the posture measurement.

[0069] Next, as a third condition, it can be configured to determine the user's self-awareness state based on whether the recognized body posture of the user is similar to the body posture of the guide video related to the body posture (S440). For example, when the similarity of the body postures is above a predetermined standard based on the image analysis of the user's body posture and the body posture of the guide video, it is determined that the two postures are similar. More specifically, the similarity of the two body postures can be judged based on the joint angles of the three points of interest and the relative positions between the joints set according to the body posture.

[0070] For example, when all three such conditions are AND conditions and all of the first condition, the second condition, and the third condition are satisfied, it is determined as the conscious state of the user, and if any one of the above three conditions is not satisfied, it can be determined as the unconscious state of the user.

[0071] Also, for example, the first condition of S420 is an AND condition, the second condition of S430 and the third condition of S440 are OR conditions, and it is also possible to determine the conscious state of the user through a conditional expression of the first condition AND (the second condition OR the third condition), and the conscious state and unconscious state of the user can be determined by logical combinations such as AND and OR of the first condition, the second condition, and the third condition.

[0072] Through such a method for determining the conscious state of the user, without requiring the user to perform an additional separate operation during the measurement of the user's body posture, by recognizing the natural postures before and after the measurement, it is possible to perform the posture recognition in the unconscious state without increasing the inconvenience of use, while ensuring the accurate body vulnerability points of the user and improving the reliability of the posture measurement.

[0073] FIG. 5 is an exemplary diagram for explaining a method of generating weighted values applied to the injury risk factors in the conscious state and the injury risk factors in the unconscious state, respectively, according to an embodiment of the present invention.

[0074] When calculating the body injury risk factors measured separately by distinguishing between the conscious state and the unconscious state with quantified scores to which weighted values are assigned, instead of assigning consistent weighted values to all users, variable weighted values can be assigned according to user characteristics.

[0075] Referring to FIG. 5, the variable weighted value (w1) in the conscious state is determined by various factor elements that can affect the physical state such as the user's age, gender, body type (body shape), occupation, lifestyle pattern, exercise history, injury history, and weight. Each element can apply different specific gravity parameter values of a, b, and c according to the type of the body injury risk factor.

[0076] In addition, the variable weighted value (w2) in the unconscious state, similar to the variable weighted value (w1) in the conscious state, is determined by various factors that can affect the user's physical condition, such as age, gender, body type (body shape), occupation, lifestyle, exercise history, injury history, and weight. Depending on the type of physical injury risk factors, different weights of a', b', and c' can be applied to each other. For example, body type is a type classification based on body size such as the ratio of the upper and lower body, back, and shoulder width, and can be scored according to established criteria for each body type.

[0077] For example, when a 21-year-old thin female measures her body posture, the following formula can be applied to perform the application of the variable weighted value when evaluating the risk factor of the turtle neck injury of the neck, a body part.

[0078] w1 (conscious state weighted value) = 21 (years old) * 0.01 (a) + 1 (female score) * 0.05 (b) + 6 (body shape score) * 0.03 (c) = 0.44

[0079] w2 (unconscious state weighted value) = 21 (years old) * 0.01 (a') + 1 (female score) * 0.09 (b') + 6 (body shape score) * 0.04 (c') = 0.54

[0080] FIG. 6 is a flowchart for explaining a method for predicting the physical injury risk of a user posture recognition base according to an embodiment of the present invention.

[0081] Referring to FIG. 6, for the method for predicting the physical injury risk of the user posture recognition base, a guide video related to the body posture measurement can be provided to the user via the display unit 100 (S610).

[0082] Image data related to the user's body posture can be acquired using at least one image sensor 200 via the image data acquisition unit 310 (S620).

[0083] Next, body joint coordinates can be extracted through body part-by-part segmentation based on the image data via the joint extraction unit 320 (S630).

[0084] Next, the user's body posture can be recognized based on the body joint coordinates extracted via the body posture recognition unit 330 (S640).

[0085] Next, it is possible to determine whether the user's body posture is a conscious state body posture or an unconscious state body posture via the conscious state determination unit 340 (S650). The determination of the user's conscious state and unconscious state can be performed by combining each condition based on, for example, whether the user's position exists within a predetermined body posture measurement position range, whether the acquisition time point of the image data related to the user's body posture exists within the time range when the video related to the body posture is provided, and whether the recognized body posture of the user is similar to the body posture of the video related to the body posture.

[0086] Next, at least one body injury risk factor can be measured for at least one joint based on the recognized body posture and the determination of the conscious state via the body injury risk factor measurement unit 350 (S660). At this time, the joint-specific injury risk factor can be measured based on the inclination of each joint related to the body injury and the position deviation from the reference point. Also, in the stage of measuring the body injury risk factor, the conscious state joint-specific injury risk factor can be measured from the conscious state body posture, and the unconscious state joint-specific injury risk factor can be individually measured from the unconscious state body posture.

[0087] Next, it is possible to predict the risk of physical injury to at least one body part based on the physical injury risk factors measured via the physical injury risk prediction unit 360 (S670). At this time, in the stage of predicting the risk of physical injury, it is possible to calculate the risk of physical injury to the body part based on the injury risk factors for each joint in the conscious state and the injury risk factors for each joint in the unconscious state. Further, by applying different weighting values to each of the injury risk factors for each joint in the conscious state and the injury risk factors for each joint in the unconscious state, the risk of physical injury is calculated. Here, the weighting value can be variably determined based on at least one of various factors that can affect the physical state of the user, such as age, gender, body shape, occupation, lifestyle pattern, exercise history, injury history, and weight.

[0088] Finally, based on the injury risk prediction result, it is possible to generate an injury management solution via the injury management solution providing unit 370 and provide an injury management solution-related video via the display unit 100 (S680). The injury management solution is also in the form of stretching or exercise content that can minimize the physical injury risk of the user according to the injury risk prediction result, and the user can be provided with such an injury management solution via various methods such as a personal mobile terminal, a kiosk, and e-mail.

[0089] FIG. 7 is a flowchart for explaining a method of predicting the physical injury risk of a user posture recognition base and providing feedback according to another embodiment of the present invention.

[0090] Similar to the last stage of FIG. 6, it is possible to provide an injury management solution-related video to the user via the display unit 100 (S710). The injury management solution-related video is also related to the body part with the highest injury risk according to the initial injury risk prediction result, and injury management solutions for a plurality of body parts can be provided.

[0091] If the user is provided with an injury management solution and performs the stretching or exercise movements shown in the video, they can perform the recognition of the conscious and unconscious body postures through the acquisition of the user's body posture image data (S720), the extraction of body joint coordinates (S730), the recognition of body postures (S740), the judgment of conscious / unconscious states (S750), etc. executed in the body posture measurement stage as shown in FIG. 6.

[0092] Also, similar to FIG. 6, the body postures in the conscious and unconscious states can be individually analyzed to measure the body injury risk factors respectively (S760).

[0093] Based on the measured body injury risk factors, the body injury risk levels of the respective body parts can be predicted, and at the same time, feedback related to the user's body movements can be provided (S770). Based on the injury management solution provided by the initial injury risk prediction result, the body injury risk level of the respective body parts re-measured can be predicted, and feedback related to the stretching and exercise execution of the user carried out in relation to the injury management solution can be provided. Here, the feedback may include content regarding precautions and recommendations for exercise behaviors related to injury risks.

[0094] The embodiments disclosed in the specification of the present invention are merely examples, and the present invention is not limited thereto. The scope of the present invention should be interpreted by the claims, and all technologies within the equivalent scope thereof should also be interpreted as being included in the scope of the present invention.

Explanation of Reference Numerals

[0095] 300 Body Injury Risk Prediction Processing Unit 310 Image Data Acquisition Unit 320 Joint Extraction Unit 330 Body Posture Recognition Unit 340 Conscious State Judgment Unit 350 Body Injury Risk Factor Measurement Unit 360 Body Injury Risk Prediction Unit 370 Injury Management Solution Providing Unit 380 Feedback Providing Unit

Claims

1. In a method for predicting physical injury risk of a user posture recognition infrastructure, providing a video related to body posture measurement; acquiring image data related to the user's body posture using at least one image sensor; extracting body joint coordinates through segmentation by body part based on the image data; recognizing the user's body posture based on the extracted body joint coordinates; measuring a physical injury risk factor for at least one joint based on the recognized body posture; predicting the risk of physical injury for at least one body part based on the physical injury risk factor, a method for predicting physical injury risk.

2. The step of measuring the physical injury risk factor measures a joint-specific injury risk factor based on the inclination of each joint related to the physical injury and the position deviation from the reference point. The method for predicting physical injury risk according to Claim 1.

3. providing a video related to the body posture for injury management based on the predicted risk of the physical injury; acquiring image data related to the body posture for the user's injury management using at least one image sensor; further comprising re-measuring a physical injury risk factor for at least one joint based on the recognized body posture based on the image data related to the body posture for the user's injury management. The method for predicting physical injury risk according to Claim 1.

4. further comprising, after the step of recognizing the user's body posture, determining whether the user's body posture is a conscious body posture or an unconscious body posture; The step of measuring the physical injury risk factor measures a conscious joint-specific injury risk factor from the conscious body posture and an unconscious joint-specific injury risk factor from the unconscious body posture. The method for predicting physical injury risk according to Claim 1.

5. The step of predicting the risk of the physical injury calculates the risk of physical injury for the body part based on the conscious joint-specific injury risk factor and the unconscious joint-specific injury risk factor. The method for predicting physical injury risk according to Claim 4.

6. The risk degree related to the physical injury is calculated by applying different weighting values to the injury risk factors for each joint in the conscious state and the injury risk factors for each joint in the unconscious state, and the weighting value is variably determined based on at least one of the age, gender, and body shape of the user. The method for predicting physical injury risk according to claim 5.

7. The step of determining whether the user's body posture is a conscious body posture or an unconscious body posture determines the user's conscious state based on whether the user's position exists within a predetermined body posture measurement position range. The method for predicting physical injury risk according to claim 4.

8. The step of determining whether the user's body posture is a conscious body posture or an unconscious body posture determines the user's conscious state based on whether the acquisition time point of the image data related to the user's body posture exists within the time range when the video related to the body posture is provided. The method for predicting physical injury risk according to claim 4.

9. The step of determining whether the user's body posture is a conscious body posture or an unconscious body posture determines the user's conscious state based on whether the recognized body posture of the user is similar to the body posture in the video related to the body posture. The method for predicting physical injury risk according to claim 4.

10. In an apparatus for predicting physical injury risk based on user posture recognition A display unit configured to provide a video related to body posture measurement; An image data acquisition unit configured to acquire image data related to the user's body posture using at least one image sensor; A joint extraction unit configured to extract body joint coordinates through body part-by-part segmentation based on the image data; A body posture recognition unit configured to recognize the user's body posture based on the extracted body joint coordinates; A physical injury risk factor measurement unit configured to measure a physical injury risk factor for at least one joint based on the recognized body posture; A physical injury risk prediction unit configured to predict the risk degree of physical injury for at least one body part based on the physical injury risk factor, including a physical injury risk factor measurement device.

Images