Standing position determination device, standing position determination system, standing position determination method, and standing position determination program

Abstract

To provide a standing posture determination device, a standing posture determination system, a standing posture determination method, and a standing posture determination program capable of easily grasping a standing posture type by a graph.SOLUTION: Standing position determination device 13 includes captured image acquisition unit 21, point specification unit 22, standing position estimation unit 23, standing position reference storage unit 24, standing position type determination unit 25, and standing position type output unit 26. The captured image acquisition unit 21 acquires a captured image captured from the side of the whole body of the target person in the standing position. The point specification unit 22 specifies a plurality of points for obtaining the degree of curvature of the back and the forward inclination angle of the pelvis. The standing position and orientation estimator 23 estimates a standing position and orientation of the subject. The standing position standard storage unit 24 stores a standing position standard. Standing position type determination unit 25 determines a standing position of the target person. Standing posture type output unit 26 outputs the plurality of standing posture types and the first graph indicating the result determined by standing posture type determination unit 25.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a standing posture determination device, a standing posture determination system, a standing posture determination method, and a standing posture determination program. [Background technology]

[0002] There is a technology for evaluating the posture of a subject using an image output by capturing an image of the subject using an imaging device. For example, Patent Document 1 discloses a posture evaluation device including a memory unit, an acquisition unit, a posture estimation unit, and an evaluation unit. The memory unit pre-stores a plurality of posture references that serve as standards used in posture evaluation, and posture feature amounts corresponding to each of the plurality of posture references. The acquisition unit acquires information regarding the positions of the subject's body parts. The posture estimation unit calculates the posture feature amounts of the subject's posture based on the information. The evaluation unit evaluates the subject's posture by comparing the calculated posture feature amounts of the subject's posture with the posture feature amounts corresponding to each of the plurality of stored posture references, and outputs the evaluation result. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2023 / 223880 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the posture evaluation device described in Patent Document 1 does not allow the evaluation result of the evaluated posture to be easily grasped.

[0005] Therefore, an object of the present invention is to provide a standing posture determination device, a standing posture determination system, a standing posture determination method, and a standing posture determination program that allow a user to easily grasp a standing posture type using a graph. [Means for solving the problem]

[0006] The standing posture determination device of the present invention includes an image acquisition unit, a point identification unit, a standing posture estimation unit, a standing posture reference storage unit, a standing posture type determination unit, and a standing posture type output unit. The image acquisition unit acquires an image of a subject in a standing position, capturing the subject's entire body from the side. The point identification unit identifies a plurality of points on the subject's image in the captured image for determining the degree of back curvature and the forward tilt angle of the pelvis. The standing posture estimation unit estimates the subject's standing posture from the plurality of points identified by the point identification unit. The standing posture reference storage unit stores standing posture references, which are standards for each of a plurality of standing posture types obtained by classifying human standing postures based on combinations of the degree of back curvature and the forward tilt angle of the pelvis. The standing posture type determination unit determines which of the plurality of standing posture types the subject's standing posture is by comparing the standing posture of the subject estimated by the standing posture estimation unit with the plurality of standing posture references. The standing posture type output unit has a first axis representing the degree of curvature and a second axis representing the forward tilt angle, and outputs a first graph showing a plurality of standing posture types as well as the results determined by the standing posture type determination unit.

[0007] The point specifying unit may further specify a plurality of points on the subject image of the subject in the captured image to determine whether the subject has a sway-back posture or a normal posture. The standing posture reference storage unit may further store sway-back posture references, which are standards for a plurality of standing posture types classified as either a sway-back posture or a normal posture. The standing posture type determining unit may further determine whether the subject's standing posture is a sway-back posture or a normal posture by comparing the subject's standing posture estimated by the standing posture estimating unit with the sway-back posture reference. The standing posture type output unit may further output the result of determining whether the subject has a sway-back posture or a normal posture as a second graph.

[0008] The standing posture type output unit preferably outputs a first graph in which the standing posture types are classified into six types, and a second graph in which the standing posture types are classified into two types, swayback posture and normal posture.

[0009] The standing posture reference storage unit may further store references for muscle shortening and weakening for each human standing posture type. The standing posture type determination unit may further determine muscle shortening and weakening of the subject by comparing the subject's standing posture estimated by the standing posture estimation unit with the references for muscle shortening and weakening. It is preferable that the standing posture type output unit further outputs the results regarding muscle shortening and weakening.

[0010] A standing posture determination system of the present invention includes a terminal device including a camera and a transmission unit that transmits an image captured by the camera, and the above-described standing posture determination device.

[0011] The standing posture determination method of the present invention includes an image acquisition step, a point identification step, a standing posture estimation step, a standing posture reference storage step, a standing posture type determination step, and a standing posture type output step. The image acquisition step acquires an image of a subject in a standing position, capturing the subject's entire body from the side. The point identification step identifies a plurality of points for determining the degree of back curvature and the forward tilt angle of the pelvis in the subject image in the captured image. The standing posture estimation step estimates the subject's standing posture from the plurality of points identified by the point identification unit. The standing posture reference storage step stores standing posture references, which are standards for each of a plurality of standing posture types obtained by classifying human standing postures based on combinations of the degree of back curvature and the forward tilt angle of the pelvis. The standing posture type determination step determines which of a plurality of standing posture types the subject's standing posture is by comparing the subject's standing posture estimated by the standing posture estimation unit with the plurality of standing posture references. The standing posture type output step outputs a first graph having a first axis representing the degree of curvature and a second axis representing the forward tilt angle, and showing the results determined by the standing posture type determination unit along with a plurality of standing posture types.

[0012] The standing posture determination program of the present invention causes a computer to execute the above steps. [Effects of the Invention]

[0013] According to the present invention, the standing posture type can be easily understood from the graph. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is an explanatory diagram of a standing posture determination system according to an embodiment; [Figure 2] 1 is a configuration diagram of a standing posture determination device according to an embodiment; [Figure 3] FIG. 10 is an explanatory diagram of imaging from the side. [Figure 4] FIG. 10 is an explanatory diagram of an image displayed on the client terminal. [Figure 5] FIG. 10 is an explanatory diagram of an image displayed on the client terminal. [Figure 6] FIG. 10 is an explanatory diagram of pixel coordinates. [Figure 7] FIG. 1 is an explanatory diagram illustrating how to determine the kyphosis index. [Figure 8] 10A and 10B are explanatory diagrams illustrating point identification by a point identification unit and estimation of a standing posture by a standing posture estimation unit. [Figure 9] 10 is a processing flow of a standing posture determination system. DETAILED DESCRIPTION OF THE INVENTION

[0015] 1 is a standing posture determination system that determines the standing posture type of a person to be determined (hereinafter simply referred to as "subject") and displays the determination result on a display unit (display) of a terminal device. The standing posture determination system 10 includes a client terminal 11, which is an example of a terminal device, a standing posture determination device 13, and a management terminal 15. The client terminal 11, the management terminal 15, and the standing posture determination device 13 communicate with each other via a communication network CN, i.e., transmit and receive various information.

[0016] The client terminal 11 is a terminal used by clients such as physical therapists, chiropractors, and trainers who provide physical training to subjects. The client is not limited to these and may also be a subject to be assessed. The client terminal 11 includes a camera 11a (see FIG. 2) that captures an image of the subject. In this example, the client terminal 11 further includes a transmitter 11b that transmits the captured image to the standing posture assessment device 13 and a display 11c (see FIG. 2) that displays the assessment results. The camera 11a may be any device capable of capturing a still image of the subject. In this example, the display 11c is a touch panel display and also functions as an input unit for performing input operations. However, the input unit may be a known input device capable of various input operations, such as a mouse or keyboard. The client terminal 11 may be any device, such as a mobile terminal, smartphone, or personal computer, as long as it has a camera and a transmitter. Although only one client terminal 11 is illustrated, the number of client terminals 11 may be multiple, one for each client.

[0017] The standing posture determination device 13 performs predetermined processing in response to input of each piece of information from the client terminal 11, and transmits and displays information as a processing result to the client terminal 11. For example, in response to input of a captured image captured by the camera 11a of the client terminal 11, the standing posture determination device 13 generates various data for displaying on the client terminal 11 a determination result related to the standing posture shown by the input captured image, and displays it on the display unit 11a of the client terminal 11.

[0018] The management terminal 15 is a terminal that manages various settings and / or data of the standing posture determination device 13. Examples of various settings include registering reference values, which will be described later, when determining standing posture. Examples of data management include managing (storing) reference values, which will be described later. In this example, the management terminal 15 is configured as a computer equipped with an input unit and a display unit (display), but from the perspective of managing (storing) data, the input unit and display unit are not necessarily required. The management terminal 15 may be any of a mobile terminal, a smartphone, a personal computer, etc., and is not particularly limited.

[0019] The client terminal 11, the standing posture determination device 13, and the management terminal 15 are each composed of a computer. The client terminal 11 and the management terminal 15 may operate on a browser when a program is sent from the standing posture determination device 13, or may be equipped with predetermined application software and operate by executing the program. The standing posture determination device 13 is equipped with a predetermined program, and by executing this program, it functions as each unit described below and performs predetermined processing.

[0020] The program causes a computer to execute a captured image acquisition step, a point identification step, a standing posture estimation step, a standing posture reference storage step, a standing posture type determination step, and a standing posture type output step. The captured image acquisition step acquires a captured image of a subject in a standing position, capturing the subject's entire body from the side. The point identification step identifies a plurality of points for determining the degree of back curvature and the forward tilt angle of the pelvis for the subject image in the captured image. The standing posture estimation step estimates the subject's standing posture from the plurality of points identified by the point identification unit. The standing posture reference storage step stores standing posture references, which are standards for each of a plurality of standing posture types obtained by classifying human standing postures based on combinations of the degree of back curvature and the forward tilt angle of the pelvis. The standing posture type determination step determines which of a plurality of standing posture types the subject's standing posture is by comparing the subject's standing posture estimated by the standing posture estimation unit with the plurality of standing posture references. The standing posture type output step outputs a first graph having a first axis representing the degree of curvature and a second axis representing the angle of forward tilt, and showing the results determined by the standing posture type determination unit along with a plurality of standing posture types. When this program is used in the application software program and incorporated into the client terminal 11, the client terminal 11 can be operated and used as the standing posture determination device 13.

[0021] 2, standing posture determination device 13 includes an image acquisition unit 21, a point identification unit 22, a standing posture estimation unit 23, a standing posture reference storage unit 24, a standing posture type determination unit 25, and a standing posture type output unit 26. Standing posture estimation device 13 preferably further includes a controller 33, and this is also the case in this example. Standing posture reference storage unit 24 constitutes a database DB.

[0022] The controller 33 comprehensively controls each component constituting the standing posture determination device 13. The controller 33 also functions as a transmitting / receiving unit that transmits and receives data to and from the client terminal 11 and the management terminal 15, and as a display control unit that generates images to be displayed on the display unit 11c of the client terminal 11 and the display of the management terminal 15.

[0023] The captured image acquisition unit 21 acquires a captured image of the subject to be determined, the subject's entire body being captured from the side in a standing position, from the client terminal 11. The captured image acquisition unit 21 sends the acquired captured image to the point identification unit 22.

[0024] The point specifying unit 22 specifies a plurality of points for determining the degree of curvature of the back and the angle of forward tilt of the pelvis for the subject image of the subject in the captured image input from the captured image acquisition unit 21. The point specifying unit 22 sends the specified plurality of points to the standing posture estimation unit 23. It is preferable that the point specifying unit 22 further specifies a plurality of points for determining whether the posture is a swayback posture or a normal posture.

[0025] The point identification unit 22 is a posture estimation AI (Artificial Intelligence) having a trained model 22a described below, and uses this trained model 22a to estimate pixel coordinates of multiple points in the subject image of the captured image input from the captured image acquisition unit 21 and contour information indicating at least the dorsal contour line of the subject image. The point identification unit 22 sends the pixel coordinates and contour information of the estimated multiple points to the standing posture estimation unit 23. The pixel coordinates will be described later using another drawing. Standing posture references, which are standards for multiple standing posture types for the standing postures estimated from the multiple points, are stored in the standing posture reference storage unit 24.

[0026] The trained model 22a learns a dataset that uses a side image showing the side of a human, a human region in the side image, and pixel coordinates of multiple points identified in the side image as input parameters, and a pth (PyTorch) file as an estimation model file as output parameters. When a side image is input, the trained model 22a estimates a subject image of the subject P (see FIG. 3), estimates pixel coordinates of multiple points, generates a masked image in which the subject image region is masked, and estimates contour information of the subject image. Thus, when a captured image showing the side of the subject P is input, the point identification unit 22 estimates pixel coordinates of multiple points in the subject image in the captured image, generates a masked image, estimates contour information, and outputs the masked image to the standing posture estimation unit 23. The contour information is composed of pixel coordinates of multiple points that form the contour of the subject image. The point identification unit 22 can use a publicly available library (open source library) and is not particularly limited. The point identification unit 22 may be configured with one AI or two AIs. When configured with two AIs, one of them estimates the pixel coordinates of multiple points, while the other generates a masked image in which the area of ​​the subject image is masked to estimate contour information of the subject image. The other AI can use MediaPipe, a known open-source library. When configured with two AIs, one of them has a trained model trained to estimate a human subject image and output the pixel coordinates of multiple points when a side image is input, using a dataset including a side image showing the side of a human, a human area in the side image, and each pixel coordinate of multiple points identified in the side image. When configured with two AIs, one of them has a trained model trained to estimate a human subject image and output the pixel coordinates of multiple points when a side image is input, using a dataset including a side image showing the side of a human and a human area in the side image. When configured with two AIs, the other has a trained model trained to estimate a human subject image and output a masked image in which the area of ​​the subject image is masked and contour information of the subject image when a side image is input, using a dataset including a side image showing the side of a human and a human area in the side image.

[0027] The side image used for learning may be a captured image of a person taken from the side, or an image generated using image generation AI; in this example, the latter is used. As is well known, unlike AI that uses a structured trained model that clearly separates explanatory variables from target variables, generation AI uses the unstructured trained model to output (generate) a certain degree of creative processing results. When a generation request is made to such a trained model (a prompt indicating the request is sent), the image generation AI generates an image as a creative processing result. A publicly available image generation AI can be used, such as Stable-Diffusion, which generates an image based on input text or image prompts.

[0028] The identification of the multiple points in the side image may be performed in advance, for example, in the management terminal 15, or in a terminal device separate from the management terminal 15 and the client terminal 11; in this example, the latter is used. In this example, AI is used for the point identification unit 22, but the invention is not limited to AI. For example, markers may be attached to the body surface of the subject P at multiple points, and pixel coordinates of the marker images in the captured image may be identified. Examples of markers include markers used in motion capture, such as reflective markers that reflect light and emitting markers that emit light. Furthermore, when a point on the captured image is designated with a finger, the pixel coordinates of that point may be identified.

[0029] The standing posture estimation unit 23 estimates the standing posture of the subject by calculating the degree of curvature of the back and the forward tilt angle of the pelvis from the multiple identified points input from the point identification unit 22. The standing posture estimation unit 23 sends the estimated standing posture of the subject to the standing posture type determination unit 25.

[0030] The standing posture reference storage unit 24 stores standing posture references, which are standards for each of a plurality of standing posture types obtained by classifying a person's standing posture according to a combination of the degree of back curvature and the angle of forward tilt of the pelvis. The standing posture reference storage unit 24 sends the stored standing posture references to the standing posture type determination unit 25. The standing posture reference storage unit 24 may further store a swayback posture reference, which is a standard for each of a plurality of standing posture types obtained by classifying a person's standing posture as either a swayback posture or a normal posture. The standing posture reference storage unit 24 may further store standards related to muscle shortening and weakening for each of the person's standing posture types.

[0031] The reference values ​​set as standards for multiple standing posture types, which classify human standing postures according to the combination of the degree of back curvature and the angle of pelvic forward tilt, are preset by, for example, an administrator using the management terminal 15. When a reference value registration request for registering the reference values ​​is sent from the management terminal 15, the reference values ​​are stored in the standing posture reference storage unit 24 by the controller. Instead of storing the reference values ​​in the standing posture reference storage unit 24 in response to a reference value registration request, the reference values ​​may be stored in the standing posture reference storage unit 24 by being set in advance in a program. The reference values ​​are not particularly limited and can be set as appropriate, and may be set by referring to values ​​disclosed in papers, etc. In this example, the reference values ​​can be updated when a new reference value registration request is sent from the management terminal 15. When the reference values ​​are stored in the management terminal 15, the standing posture reference storage unit 24 may be provided in the management terminal 15 instead of the standing posture determination device 13. In this case, the above-mentioned reference values ​​may be read out from the standing posture reference memory unit 24 provided in the management terminal 15 via the controller 33 at an appropriate timing, and may be stored in the standing posture reference memory unit 24.

[0032] The standing posture type determination unit 25 determines which of a plurality of standing posture types the subject's standing posture is by comparing the estimated standing posture input from the standing posture estimation unit 23 with the standing posture reference stored in the standing posture reference storage unit 24. The standing posture type determination unit 25 sends the determined standing posture type to the standing posture type output unit 26. It is preferable that the standing posture type determination unit 25 further determines whether the subject has a sway-back posture or a normal posture by comparing the subject's standing posture estimated by the standing posture estimation unit 23 with the sway-back posture reference.

[0033] The standing posture type output unit 26 has a first axis A1 (see FIG. 4) representing the degree of curvature of the back and a second axis A2 (see FIG. 4) representing the angle of forward tilt of the pelvis, and outputs a first graph DG1 (see FIG. 4) showing the results of the determination by the standing posture type determination unit 25 along with a plurality of standing posture types to the client terminal 11. It is preferable that the standing posture type output unit 26 further outputs the determination result of whether the posture is swayback or normal posture as a second graph DG2 (see FIG. 4). It is also preferable that the standing posture type output unit 26 outputs a first graph classifying the standing posture types into six types and a second graph classifying the standing posture types into two types, swayback or normal posture. Furthermore, the standing posture type output unit 26 may output the determination results of muscle shortening and weakening.

[0034] In FIG. 3, the client terminal 11 is arranged so that the camera 11a mounted thereon faces the side of the subject P. This allows for a captured image of the subject from the side. This allows for a captured image of the subject P from the side in the vertical direction. It is preferable to arrange the camera 11a so that the entire body BW of the subject P from the top of the head to the soles of the feet is captured, as this will result in more accurate determination results. In this example, too, an image of the entire body of the subject from the side is obtained. It is preferable to arrange the camera 11a so that the imaging optical axis OA intersects with the sagittal plane SP of the body of the subject P, as this will result in more accurate determination results, and it is more preferable to arrange it so that it is perpendicular. In this example, the camera 11a is also arranged so that the imaging optical axis OA is as perpendicular as possible to the sagittal plane SP.

[0035] As shown in FIG. 4, the display unit 11c (see FIG. 2) of the client terminal 11 (see FIG. 2) displays a first image G1 having a first image portion G1a and a first graph image portion G1b showing the determination result. The first image portion G1a displays an image including a human body diagram showing the outer contours and bones of the whole body from the right side, imitating the subject's standing posture type. In this example, one of 12 patterns of images is displayed, which are combinations of six standing posture types (described below) and either a swayback posture or a normal posture. The first graph image portion G1b shows the determination result of the first graph DG1, etc. In this way, the display unit 11c simultaneously displays an image of the human body diagram and a graph. Since the first image portion G1a and the first graph image portion G1b showing the determination result are simultaneously displayed, the subject's posture and the determination result of the subject can be compared, making it easier for the client to understand the posture. Furthermore, since a graph is displayed as the result of the judgment, the judgment result can be understood as being based on a predetermined method, regardless of the client's level of experience, knowledge, and insight.

[0036] In the example shown in FIG. 4, the first image portion G1a and the first graph image portion G1b are arranged horizontally, with the first graph image portion G1b located to the right of the first image portion G1a. However, the horizontal positional relationship between the first image portion G1a and the first graph image portion G1b may be reversed. Furthermore, instead of being arranged horizontally, they may be arranged vertically. In the case of being arranged vertically, the vertical positional relationship between the first image portion G1a and the first graph image portion G1b is not particularly limited. Furthermore, the first image portion G1a and the first graph image portion G1b may be superimposed on a partial area of ​​the other.

[0037] The human body diagram in the image of the first image section G1a reflects the posture of the subject. In the example shown in Fig. 3, the degree of curvature of the back is "weak," the angle of forward tilt of the pelvis is "backward tilt," the standing posture type determination result is "unarmed," and the determination result of whether the posture is swayback or normal is "normal." Instead of the human body diagram, a captured image of the subject may be displayed as an image in the first image section G1a.

[0038] The first graph image section G1b has a first vertical axis A1 representing the degree of curvature of the back and a second horizontal axis A2 representing the angle of forward tilt of the pelvis. It is composed of a first graph DG1 showing multiple standing posture types and the results of the standing posture type determination, and a second graph DG2 showing the results of the determination of whether the posture is swayback or normal on the vertical axis. The first graph DG1 classifies the standing posture types into six types. The second graph DG2 classifies the standing posture types into two types: swayback or normal.

[0039] In the first graph DG1, the first axis (vertical axis) A1 indicates the result of the assessment of the degree of curvature of the back, and the second axis (horizontal axis) A2 indicates the result of the assessment of the angle of forward tilt of the pelvis, and is shown in six types of standing posture. In this example, the results of the assessment of the standing posture type are indicated by stars on the first graph DG1, but this is not limited to this and may be indicated by other symbols, etc.

[0040] The second graph DG2 shows two classifications, a swayback posture and a normal posture, on the vertical axis. In this example, the determination result of which posture falls is shown by a star on the second graph DG2, but this is not limited to this and other symbols may be used.

[0041] In this way, the client can easily understand the standing posture type of the subject from the first graph DG1 and the second graph DG2.

[0042] In the example shown in FIG. 5, the display unit 11c (see FIG. 2) of the client terminal 11 (see FIG. 2) further displays a second image G2 having a second image portion G2a and a list image portion G2b showing the determination results for muscle shortening and weakening. Muscle shortening and weakening are determined by combining the six standing posture type classifications of the first graph DG1 and two classifications of either a swayback posture or a normal posture of the second graph DG2. Here, muscle shortening is treated as the same as muscle hypertonia, and no distinction is made between the two. Muscle weakening is treated as the same as muscle elongation, and no distinction is made between the two. Muscle shortening and weakening are stored in the standing posture reference storage unit 24 for each standing posture type and the determination result of whether the posture is a swayback posture or a normal posture. The standing posture type determination unit 25 outputs the determination result to the standing posture type output unit 26, and the controller 33 sends the second image G2 to the display unit 11c, where the second image G2 is displayed. The second image area G2a displays a schematic representation of a human body diagram, showing shortened muscles and weakened muscles in different colors. For example, shortened muscles are colored red and weakened muscles are colored blue. In FIG. 5, the color coding of shortened muscles and weakened muscles is represented by different hatching patterns on the body diagram. The list image area G2b displays a list of shortened muscles and a list of weakened muscles arranged vertically. Alternatively, the list may be arranged horizontally. It is not necessary to display shortened muscles and weakened muscles separately. In this way, the determination results of muscle shortening and weakening can be more easily understood.

[0043] The standing posture estimation unit 23 estimates the degree of curvature of the back, the angle of forward tilt of the pelvis, and whether the posture is swayback or normal. First, the degree of curvature of the back will be described. In this example, the degree of curvature of the back is estimated using a kyphosis index, but this is not particularly limited and can be set appropriately. The standing posture estimation unit 23 for determining the degree of curvature of the back of the present invention includes a length calculation unit 23a (see FIG. 2) and a kyphosis index calculation unit 23b (see FIG. 2). The length calculation unit 23a calculates the maximum second length L2, which is the longest of the first length L1 of the line segment connecting the seventh cervical vertebra and the fourth lumbar vertebra identified based on an image captured from the side of the subject to be determined to have hunched back, and the second length L2 from the line segment to the dorsal contour line in the subject image in the image. max (See FIG. 7.) The kyphosis index calculation unit 23b calculates (L2 max The kyphosis index is calculated using the formula: kyphosis index (L1 / L1) × 100. The kyphosis index calculation unit 23b sends the calculated kyphosis index to the standing posture type determination unit 25. The standing posture type determination unit 25 reads a reference value from the standing posture reference storage unit 24, and determines that the degree of curvature of the back is "strong" if the kyphosis index is equal to or greater than a preset reference value, and determines that the degree of curvature of the back is "weak" if the kyphosis index is less than the reference value.

[0044] When the masking image, the pixel coordinates of the seventh cervical vertebra Pa1 and the fourth lumbar vertebra Pb1, and the contour information are input from the point specifying unit 22, the length calculation unit 23a (see FIG. 2) responds to this input by calculating the longest of the first length L1 of the line segment LP connecting the seventh cervical vertebra Pa1 and the fourth lumbar vertebra Pb1 in the masking image and the second length L2 from the line segment LP to the dorsal contour line. maxand is calculated (see FIG. 7). Since the masking image is obtained by masking the subject image based on the captured image, the dorsal contour line in the masking image can be considered to match the dorsal contour line in the subject image in the captured image. In this example, when imaging the subject P, the subject P is imaged from the right lateral side, and a captured image of the right side of the subject P is obtained, so in the captured image and the masking image, the contour line existing to the left of the seventh cervical vertebra Pa1 and the fourth lumbar vertebra Pb1 is specified as the dorsal contour line (see FIGS. 7 and 8). The length calculation unit 23a calculates the calculated first length L1 and maximum second length L2. max The first length L1, the second length L2, the maximum second length L2 are sent to the kyphosis index calculation unit 23b (see FIG. 2) together with the user ID and posture classification input from the length calculation unit 23a. max The method for determining this will be described in detail later with reference to another drawing.

[0045] The kyphosis index calculation unit 23b (see FIG. 2) calculates the first length L1 and the maximum second length L2 from the length calculation unit 23a (see FIG. 2). max When the input is given, the kyphosis index is calculated in response to this input. The kyphosis index is (L2 max / L1)×100. Kyphosis index calculation unit 23b sends the kyphosis index input from length calculation unit 23a to standing posture type determination unit 25, which reads out a reference value from standing posture reference storage unit 24. Standing posture type determination unit 25 compares the read-out reference value with the calculated kyphosis index, and determines that the degree of back curvature is "strong" if the calculated kyphosis index is equal to or greater than the reference value (17.0 or greater in this example), or determines that the degree of back curvature is "weak" if the calculated kyphosis index is less than the reference value.

[0046] The pixel coordinates of the seventh cervical vertebra Pa1 and the fourth lumbar vertebra Pb1 identified in the lateral image, and the pixel coordinates of the seventh cervical vertebra Pa1 and the fourth lumbar vertebra Pb1 estimated by the point identification unit 22 in the captured image, and the pixel coordinates constituting the contour information, are identified and estimated using the pixel coordinates of the upper left corner of the lateral image and the captured image as the origin D0 of (0,0), as shown in Fig. 6. Note that in the pixel coordinate system, the X coordinate increases toward the right in the horizontal direction (lateral direction), and the Y coordinate increases toward the bottom in the vertical direction (longitudinal direction).

[0047] First length L1, second length L2, maximum second length L2 max The method for calculating the kyphosis index will be described with reference to FIG. 7. The captured image and the masked image G3 generated from this captured image have the same size in the vertical and horizontal directions, and pixel coordinates in the captured image and pixel coordinates in the masked image G3 are the same. The length calculation unit 23a (see FIG. 2) calculates a line segment LP with the seventh cervical vertebra Pa1 and the fourth lumbar vertebra Pb1 as its endpoints based on the pixel coordinates of the seventh cervical vertebra Pa1 and the fourth lumbar vertebra Pb1 in the masking image G3, and calculates a first length L1. The coordinates of points on the line segment LP are defined in an XY coordinate system in which the upper left corner of the masked image G3 is set as the origin (0,0) and orthogonal XY axes are set, with the X coordinate increasing toward the right and the Y coordinate increasing toward the bottom, as shown in FIG. 7.

[0048] The length calculation unit 23a (see FIG. 2) then virtually scans the line segment LP, sequentially determining the distance from each of the plurality of points on the line segment LP to the dorsal contour line OL as a second length L2. The number of points selected on the line segment LP to determine the second length L2 is not particularly limited, and in this example, a plurality of points are selected with a difference in Y coordinate of 1. The length calculation unit 23a then determines the maximum value of the determined second lengths L2 as the maximum second length L2. max The kyphosis index calculation unit 23b specifies the kyphosis index as (L2 maxThe kyphosis index is calculated using the formula: / L1) × 100. Note that in FIG. 7, for convenience of explanation, outlines of bones are shown by two-dot chain lines (virtual lines) within the masking region, but bones are not depicted in the masking image G3. In the example shown in FIG. 7, the masking region is the entire area within the contour line of the subject image, but the masking region is not limited to this example. For example, a masking region of a predetermined constant width along the contour line may be formed inside the contour line of the subject image.

[0049] The standing posture estimation unit 23 for calculating the kyphosis index includes a length calculation unit 23a (see FIG. 2) and a kyphosis index calculation unit 23b (see FIG. 2). The length calculation unit 23a calculates the maximum second length L2, which is the longest of the first length L1 of the line segment LP connecting the seventh cervical vertebra Pa1 and the fourth lumbar vertebra Pb1 identified based on the captured image, and the second length L2 from the line segment LP to the dorsal contour line OL. max The kyphosis index calculation unit 23b calculates the kyphosis index based on these and sends the calculated kyphosis index to the standing posture type determination unit 25. The standing posture type determination unit 25 reads out the reference values ​​stored in the standing posture reference storage unit 25 and outputs the determination result as the determination reference. This allows the degree of back curvature to be determined easily and quickly. Also, since the reference values ​​are set and stored in advance, the determination by the standing posture type determination unit 25 is extremely quick. Furthermore, since the line segment LP is calculated based on the pixel coordinates of the seventh cervical vertebra Pa1 and the fourth lumbar vertebra Pb1 in the image, the time required for calculation is extremely short. In this example, the degree of back curvature corresponds to "weak."

[0050] When points Pc1 (right-asis (right upper anterior iliac spine)) and Pd1 (right-psis (right upper posterior iliac spine)) of the subject image shown in FIG. 8 are input from point identification unit 22, pelvic forward tilt angle calculation unit 23c (see FIG. 2) calculates the pelvic forward tilt angle in response to this input. The pelvic forward tilt angle is calculated as the angle between a line connecting two points Pc1 (right-asis (right upper anterior iliac spine)) and Pd1 (right-psis (right upper posterior iliac spine)) of the subject image shown in FIG. 8 and a horizontal line passing through Pd1 (right-psis (right upper posterior iliac spine)). Angle in the clockwise direction is positive, and angle in the counterclockwise direction is negative. Pelvic forward tilt angle calculation unit 23c calculates the pelvic forward tilt angle from the above-mentioned multiple points for calculating the pelvic forward tilt angle input from point identification unit 22, and sends the calculated angle to standing posture type determination unit 25. The standing posture type determination unit 25 reads a reference value from the standing posture reference storage unit 24. The standing posture type determination unit 25 compares the read reference value with the calculated angle. If the calculated angle is equal to or greater than the reference value, the determination is made as "forward tilt." If the calculated angle is equal to or greater than the reference value but less than the reference value, the determination is made as "normal." If the calculated angle is less than the reference value, the determination is made as "backward tilt." In this example, the second axis representing the pelvic forward tilt angle in the first graph DG1 is classified into three types from left to right: backward tilt, normal, and forward tilt. In this example, if the pelvic forward tilt angle is 26.0° or greater, the determination is forward tilt. If the pelvic forward tilt angle is 14.0° or greater but less than 26.0°, the determination is normal. If the pelvic forward tilt angle is less than 14.0°, the determination is backward tilt. This classification is merely an example, and the present invention is not limited to this and can be set as appropriate. In this example, the pelvic forward tilt angle corresponds to "backward tilt" (see FIG. 4).

[0051] In this example, standing postures are classified into six types depending on the combination of the degree of curvature of the back and the angle of forward tilt of the pelvis. When the degree of curvature is "weak" and the angle of forward tilt of the pelvis is "backward tilt," it is classified as "unshaven." When the degree of curvature is "strong" and the angle of forward tilt of the pelvis is "backward tilt," it is classified as "flat back." When the degree of curvature is "weak" and the angle of forward tilt of the pelvis is "normal," it is classified as "normal." When the degree of curvature is "strong" and the angle of forward tilt of the pelvis is "normal," it is classified as "hunchback." When the degree of curvature is "weak" and the angle of forward tilt of the pelvis is "forward tilt," it is classified as "arched back." When the degree of curvature is "strong" and the angle of forward tilt of the pelvis is "forward tilt," it is classified as "hunchback x arched back." Each standing posture is always classified into one of these types. This classification is an example and is not limited to this. In this example, the degree of curvature is "weak" and the angle of forward tilt of the pelvis is "normal," so the patient is "unarmed" (see FIG. 4).

[0052] To determine whether the posture is sway-back or normal, when three points Pe1 (right shoulder), Pf1 (right hip), and Pg1 (right ankle) of the subject image shown in FIG. 8 are input from point identification unit 22, the points input from point identification unit 22 are sent to standing posture estimation unit 23. Standing posture estimation unit 23 calculates an angle for determining whether the posture is sway-back or normal and sends the calculated angle to standing posture type determination unit 25, which then reads out a reference value from standing posture reference storage unit 24. At this time, the angle is calculated as the angle formed on the dorsal side between a line connecting Pe1 and Pf1 with Pf1 as the starting point and a line connecting Pf1 and Pg1. The standing posture type determination unit 25 compares the read reference value with the determined angle, and if the determined angle is equal to or greater than the reference value, determines that the posture is a "swayback posture," and if the determined angle is less than the reference value, determines that the posture is a "normal posture." In this example, if the angle is less than 176.4°, it is a swayback posture, and if it is 176.4° or greater, it is a normal posture. This classification is an example and is not limiting. In this example, it is a "normal posture" (see FIG. 4).

[0053] Muscle shortening and weakening are determined based on a combination of six types of standing posture types and whether the posture is swayback or normal. In this example, the posture types are "hunchback x arched lower back" and "swayback posture" (see FIG. 5). The standing posture type determination unit 25 reads out criteria for muscle shortening and weakening for each human standing posture type from the standing posture criteria storage unit 24, and determines muscle shortening and weakening of the subject by comparing the subject's standing posture estimated by the standing posture estimation unit 23 with the criteria for muscle shortening and weakening. In this case, muscle shortening and weakening are determined as follows: Pectoralis major shortening Pectoralis minor shortening Tensor femoris muscle shortening External oblique muscle weakness Upper internal oblique muscle shortening Weakened lower internal oblique muscle Upper rectus abdominis shortening Lower rectus abdominis muscle weakness Transverse abdominis weakness iliopsoas muscle weakness Rectus femoris weakness Weakened neck flexors Intercostal muscle shortening Serratus anterior muscle shortening Hamstring shortening Neck extensor muscle shortening Thoracic erector spinae muscle weakness Lumbar erector spinae muscle weakness Rhomboid muscle weakness Levator scapulae shortening Upper trapezius muscle shortening Middle trapezius muscle weakness Lower trapezius muscle weakness Gluteus maximus weakness

[0054] The results of the assessment of the degree of curvature of the back, the angle of forward tilt of the pelvis, the assessment of whether the posture is swayback or normal, and the corresponding assessment of muscle shortening and weakening are shown in the two tables below.

[0055] [Table 1]

[0056] [Table 2]

[0057] The operation of the above configuration will be described with reference to Fig. 9. When a client terminal 11 logs in, a captured image is sent from the transmitter 11b of the client terminal 11 (S1). When the standing posture determination device 13 receives the transmission request from the client terminal 11, it responds to the transmission request by transmitting the captured image to the client terminal 11 under the control of the controller 33, and the client terminal 11 acquires the captured image (S2).

[0058] When the standing posture determination device 13 acquires the captured image, the point identification unit 22 identifies a plurality of points for determining the degree of back curvature and the forward tilt angle of the pelvis for the subject image of the subject in the captured image (S3). In response to acquiring these identified points, the point identification unit 22 outputs these points to the standing posture estimation unit 23.

[0059] In response to the input of the identified point data, the standing posture estimation unit 23 estimates the standing posture (S4) and outputs the estimated standing posture to the standing posture type determination unit 25. In response to the input of the estimated standing posture, the standing posture type determination unit 25 compares the estimated standing posture with a standing posture reference to determine which of a plurality of standing posture types the estimated standing posture is (S5). The standing posture type determination unit 25 outputs the determination result to the standing posture type output unit 26.

[0060] In response to the input of the determination result, the standing posture type output unit 26 generates a first image G1 (S6a) and also generates a second image G2 (S6b). Either step S6a or step S6b may precede. The standing posture type output unit 26 sends the generated data as image data to the client terminal 11 (S7).

[0061] The client terminal 11 displays the image on the display unit 11c (S8). [Explanation of symbols]

[0062] 10 Standing Posture Judgment System 11 Client terminal 13 Standing posture determination device 21 Image acquisition unit 22 point specific part 23 Standing posture estimation unit 24 Standing posture reference memory unit 25 Standing posture type determination unit 26 Standing posture type output unit P Target Audience DG1 1st graph DG2 2nd graph

Claims

1. an image acquisition unit that acquires an image of the subject from the side of the subject's entire body in a standing position; a point specifying unit that specifies a plurality of points for determining the degree of back curvature and the forward tilt angle of the pelvis with respect to the subject image of the subject in the captured image; a standing posture estimation unit that estimates a standing posture of the subject from the plurality of points identified by the point identification unit; a standing posture standard storage unit that stores standing posture standards that are standards for a plurality of standing posture types obtained by classifying human standing postures according to a combination of the degree of curvature of the back and the angle of forward tilt of the pelvis; a standing posture type determination unit that determines which of the plurality of standing posture types the standing posture of the subject is, by comparing the standing posture of the subject estimated by the standing posture estimation unit with a plurality of the standing posture standards; a standing posture type output unit that has a first axis representing the degree of curvature and a second axis representing the forward tilt angle, and outputs a first graph showing the plurality of standing posture types and the results determined by the standing posture type determination unit; A standing posture determination device comprising:

2. the point specifying unit further specifies a plurality of points for determining whether the subject image of the subject in the captured image is in a swayback posture or a normal posture; The standing posture reference storage unit further stores a swayback posture reference, which is a reference for each of the plurality of standing posture types classified as either a swayback posture or a normal posture, the standing posture type determination unit further determines whether the standing posture of the subject is a sway-back posture or a normal posture by comparing the standing posture of the subject estimated by the standing posture estimation unit with the sway-back posture reference; 2. The standing posture determination device according to claim 1, wherein the standing posture type output unit further outputs the result of determining whether the posture is a swayback posture or a normal posture in the form of a second graph.

3. 3. The standing posture determination device according to claim 2, wherein the standing posture type output unit outputs the first graph in which the standing posture types are classified into six types and the second graph in which the standing posture types are classified into two types, a swayback posture and a normal posture.

4. The standing posture reference storage unit further stores a reference for muscle shortening and weakening for each standing posture type of a person, the standing posture type determination unit further determines muscle shortening and weakening of the subject by comparing the standing posture of the subject estimated by the standing posture estimation unit with a standard regarding muscle shortening and weakening; 4. The standing posture determination device according to claim 1, wherein the standing posture type output unit further outputs a determination result of muscle shortening and weakening.

5. a terminal device including a camera and a transmission unit that transmits an image captured by the camera; The standing posture determination device according to claim 1 or 2, A standing posture determination system comprising:

6. an image acquisition step of acquiring an image of the subject from the side of the subject's entire body in a standing position; a point specifying step of specifying a plurality of points for determining the degree of back curvature and the forward tilt angle of the pelvis on the subject image of the subject in the captured image; a standing posture estimating step of estimating a standing posture of the subject from the plurality of points identified by the point identifying unit; a standing posture standard storage unit that stores standing posture standards that are standards for a plurality of standing posture types obtained by classifying human standing postures according to a combination of the degree of curvature of the back and the angle of forward tilt of the pelvis; a standing posture type determination step of determining which of the plurality of standing posture types the standing posture of the subject is by comparing the standing posture of the subject estimated by the standing posture estimation unit with the plurality of standing posture standards; a standing posture type output step of outputting a first graph having a first axis representing the degree of curvature and a second axis representing the forward tilt angle, the first graph showing the plurality of standing posture types and the results determined by the standing posture type determination unit; A standing posture determination method comprising:

7. an image acquisition step of acquiring an image of the subject from the side of the subject's entire body in a standing position; a point specifying step of specifying a plurality of points for determining the degree of back curvature and the forward tilt angle of the pelvis on the subject image of the subject in the captured image; a standing posture estimating step of estimating a standing posture of the subject from the plurality of points identified by the point identifying unit; a standing posture standard storage unit that stores standing posture standards that are standards for a plurality of standing posture types obtained by classifying human standing postures according to a combination of the degree of curvature of the back and the angle of forward tilt of the pelvis; a standing posture type determination step of determining which of the plurality of standing posture types the standing posture of the subject is by comparing the standing posture of the subject estimated by the standing posture estimation unit with the plurality of standing posture standards; a standing posture type output step of outputting a first graph having a first axis representing the degree of curvature and a second axis representing the forward tilt angle, the first graph showing the plurality of standing posture types and the results determined by the standing posture type determination unit; A standing posture determination program that causes a computer to execute the above.

Images