Posture analysis system and posture analysis method

Abstract

This invention provides a posture analysis system and method for analyzing the posture of a subject walking in an environment with ambient noise. [Solution] The posture analysis system is a posture analysis system that analyzes the posture of a subject walking in an environment with ambient sounds, and comprises a sound collection unit that collects sounds in the environment, and an analysis unit that extracts sounds having a specific periodicity from the collected sounds, and analyzes the posture of the subject that produced the extracted sounds based on the changes in the characteristics of the extracted sounds.

Description

Technical Field

[0001] The present invention relates to a posture analysis system and a posture analysis method.

Background Art

[0002] Since walking is a daily activity, even if one's state and posture during walking gradually change, it is difficult to notice such changes. Such changes in the state and posture during walking may affect walking. Meanwhile, techniques for analyzing the footsteps of a subject have been reported.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of Invention

Problems to be Solved by the Invention

[0004] However, when walking or the like is performed using a movable floor device inside a building, vibration and sound are generated. If there is environmental sound other than the subject's footsteps inside the building, it may become difficult to analyze the posture during walking using the subject's footsteps and the like.

[0005] The problem to be solved by the present invention is to provide a posture analysis system and a posture analysis method for analyzing the posture of a subject walking in an environment with environmental sound.

Means for Solving the Problems

[0006] (1) The present invention has been made to solve the above problems, and a posture analysis system according to a first aspect of the present invention is a posture analysis system for analyzing the posture of a subject walking in an environment with ambient sounds, comprising: a sound collection unit for collecting sounds in the environment; and an analysis unit for extracting sounds having a specific periodicity from the collected sounds, and analyzing the posture of the subject who generated the extracted sounds based on changes in the characteristics of the extracted sounds. (2) The posture analysis system described above includes a camera that photographs the subject in the environment, and a periodic behavior identification unit that detects the periodicity of the subject's behavior from the image of the subject and determines the specific period from the periodicity of the subject's behavior, and the analysis unit extracts the sound using the information of the specific period determined by the periodic behavior identification unit. (3) The posture analysis system described above comprises a movable floor device having a movable belt on which a subject walks, a vibration sensor that detects vibrations of the movable floor device, and a periodic behavior identification unit that detects a periodic signal from the detection results of the vibration sensor obtained during the subject's actions and determines the specific period from the periodic signal, and the analysis unit extracts the sound using the information of the specific period determined by the periodic behavior identification unit. (4) One aspect of the present invention is a computer-based posture analysis method for analyzing the posture of a subject walking in an environment with ambient sounds, the method comprising extracting sounds having a specific periodicity from the sounds in the environment collected by a sound collection unit, and analyzing the posture of the subject who produced the extracted sounds based on changes in the characteristics of the extracted sounds. [Effects of the Invention]

[0007] According to the present invention, it is possible to provide a posture analysis system and a posture analysis method for analyzing the posture of a subject walking in an environment with ambient noise. [Brief explanation of the drawing]

[0008] [Figure 1] This is a schematic diagram of the posture analysis system according to the embodiment. [Figure 2A] This is a diagram illustrating the configuration of the posture analysis system according to the embodiment. [Figure 2B] This is an elevation view illustrating the gait of a subject in the embodiment. [Figure 3] This is a plan view illustrating the gait of a subject in the embodiment. [Figure 4A] This is a diagram illustrating the detection of periodicity in the embodiment. [Figure 4B] This is a diagram illustrating the detection of periodicity in the embodiment. [Figure 5] This is a flowchart of the analysis process for the embodiment. [Figure 6] This is a diagram illustrating the determination of sound during walking in the embodiment. [Figure 7] This is a schematic diagram of the posture analysis system according to the embodiment. [Figure 8] This is a flowchart of the analysis process for the embodiment. [Modes for carrying out the invention]

[0009] The posture analysis system and posture analysis method of the embodiment will be described with reference to the figures.

[0010] (First Embodiment) First, an overview of the posture analysis system 1 of the first embodiment will be described. Figure 1 is a schematic diagram of the posture analysis system 1 of the embodiment. Figure 2A is a configuration diagram of the posture analysis system 1 of the embodiment. Figure 2B is an elevation view illustrating the gait of subject U in the posture analysis system 1 of the embodiment.

[0011] The posture analysis system 1 analyzes the posture of a subject (U) walking in an environment with ambient noise.

[0012] This posture analysis system 1 includes, for example, a movable floor device 2, an analysis device 3 (analysis unit 33), a sound collection unit 4, a camera 5, and a display terminal 6. The sound collection unit 4 includes a microphone and other components and collects sounds from the surrounding environment. The camera 5 is disposed at a fixed point, for example, and captures images of surrounding conditions. For example, the camera 5 outputs an image obtained by photographing the subject (U) in the environment. This camera 5 may also serve as an indoor surveillance camera.

[0013] The movable floor device 2 is an example of an exercise apparatus used for indoor walking. The movable floor device 2 may sometimes be referred to as a treadmill, a walking machine, or the like. The movable floor device 2 is configured to be capable of adjusting the exercise load for walking or the like of a user.

[0014] For example, the movable floor device 2 includes a frame 21 disposed substantially along the floor surface, a movable belt 22 (movable floor), a pair of rollers (not shown), an electric motor (not shown), and a controller unit 24.

[0015] The frame 21 of the movable floor device 2 also serves as the housing of the movable floor device 2, and supports each part of the movable floor device 2. For example, the frame 21 (movable floor device 2) is disposed along the floor surface.

[0016] The movable belt 22 of the movable floor device 2 is supported by a pair of rollers and the like provided at the front and rear of the frame 21 of the movable floor device 2, and rolls as the pair of rollers rotate. For example, the shaft of an electric motor (not shown) is connected to the shaft of one of the pair of rollers, and by driving the electric motor, the movable belt 22 of the movable floor device 2 can be caused to roll at a desired speed. By adjusting the speed of this electric motor, the speed at which the movable belt 22 of the movable floor device 2 rolls can be adjusted. The configuration for causing the movable belt 22 of the movable floor device 2 to roll is not limited to the above, and may be another known configuration.

[0017] The controller unit 24 performs processing for speed control of the electric motor described above. The controller unit 24 executes the above control in accordance with a command from the analysis device 3. This controller unit 24 is formed as a so-called embedded computer.

[0018] The analysis device 3 of this embodiment may be built into the movable floor device 2 described above, or it may be configured separately from the movable floor device 2.

[0019] The analysis device 3 includes, for example, a storage unit 31, an image extraction unit 32, an analysis unit 33, a communication processing unit 34, and a periodic behavior identification unit 35.

[0020] The memory unit 31 is composed of semiconductor memory and other components, and stores data and programs used for various processes in the analysis device 3.

[0021] The communication processing unit 34 communicates with, for example, the sound collection unit 4, the camera 5, and the terminal device 6 (not shown), which are located outside the analysis device 3. The communication processing unit 34 receives commands from the terminal device 6 and other devices regarding the control of the movable floor device 2, commands for the analysis device 3 to execute analysis processing, etc. The communication processing unit 34 outputs a display of the control status of the movable floor device 2 and a display of the results of the analysis processing. The terminal device 6 is, for example, a portable computer such as a smartphone.

[0022] The image extraction unit 32 acquires image data from the camera 5, for example, via the communication processing unit 34. The image extraction unit 32 detects an image of subject U from the image based on the image data from the camera 5.

[0023] The analysis unit 33 acquires the sound collected by the sound collection unit 4, for example, via the communication processing unit 34. The analysis unit 33 extracts sounds with a specific periodicity from the sounds collected by the sound collection unit 4.

[0024] The periodic behavior identification unit 35 detects the periodicity of subject U's behavior from the image of subject U acquired by the image extraction unit 32. The periodic behavior identification unit 35 determines a specific period from the periodicity of subject U's behavior. The periodic behavior identification unit 35 detects the actions of subject U from the image of subject U, including, for example, arm swings and leg movements associated with walking. These movements include the periodicity of subject U's actions. Based on the above, the periodic behavior identification unit 35 determines a specific period related to subject U's actions from the periodicity of subject U's actions.

[0025] The posture analysis system 1 of this embodiment determines the above-mentioned specific period using a camera 5 and a periodic behavior identification unit 35. Based on the changes in the characteristics of the sound having components of that period, the posture analysis system 1 analyzes the posture of the subject who generated the sound.

[0026] As shown in Figure 2A, the movable floor device 2 is positioned on the floor (floor panel).

[0027] The frame 21 of the movable floor device 2 is formed at a predetermined height. The movable belt 22, a pair of rollers, and an electric motor (not shown) are arranged within the cross-section of the frame 21.

[0028] Figure 2B shows an example of a view of subject U from the side while he is walking. Figure 2B(a) shows the situation of walking with short steps (stride length L1), and Figure 2B(b) shows the situation of walking with long steps (stride length L2). In each figure, the left foot of subject U is shown in black, and its right foot is shown in white.

[0029] For example, HC indicates the position when the left foot is stepped forward and the heel touches the walking surface (the surface of the movable floor formed by the movable belt 22), and HO indicates the position of the heel when the right foot is lifted and then swung forward. The distance from the HO corresponding to the right foot to the HC of the left foot represents the stride length of the left foot. Similarly, for the left foot, the distance from the HO corresponding to the left foot to the HC of the right foot represents the stride length of the right foot.

[0030] The stride is defined as the distance from the left foot's HO position to the left foot's HC position, after lifting the foot and swinging it forward. The same applies to the right foot.

[0031] As a result of subject U's walking, footsteps are generated periodically.

[0032] Figure 3 is a plan view illustrating the gait of subject U in the posture analysis system 1 of the embodiment. This figure shows an example of subject U as viewed from above during walking. Figures 3(a) and 3(b) correspond to the previously mentioned Figures 2B(a) and 2B(b), respectively.

[0033] Walking speed can be calculated by summing the stride length per unit time to obtain the distance traveled per unit time. Dividing this by the length of the unit time gives the walking speed. Instead of summing the stride length per unit time to obtain the distance traveled, the average stride length can be used. The product of the average stride length and the number of steps per unit time can be considered as the distance traveled. The frequency of footsteps is related to the time it takes for the left and right legs to move alternately.

[0034] The processes of the embodiment will be described in order with reference to Figures 4A, 4B, and 5. (Analysis process) First, please refer to Figure 5 to explain the overall flow of the analysis process. Figure 5 is a flowchart of the analysis process in the embodiment.

[0035] The walking speed during the gait evaluation is pre-set using information from the terminal device 6 (S11). For example, the analysis unit 33 sets the speed at which the movable belt 22 rotates (command speed) based on the walking speed during the gait evaluation identified by the subject U's actions. Subject U stands on the movable belt 22 and, when ready to walk, operates the controller of the movable floor device 2 to start its operation. Subject U begins walking on the movable belt 22 at an appropriate speed corresponding to the movement speed of the movable belt 22 and starts recording to evaluate their own walking state. Accordingly, the analysis unit 33 starts imaging with the camera 5 (S12) and continues recording for a predetermined period of time. The analysis unit 33 acquires image data of the recorded images for a predetermined period (S13).

[0036] The analysis unit 33 extracts periodic information based on the subject U's actions from the image data (S14). The analysis unit 33 uses the periodic information from the subject U's actions to extract characteristic features of subject U's walking (S15). The analysis unit 33 identifies whether or not there is an abnormality in the posture during walking using predetermined judgment conditions (S16). For example, the difference in stride length between the left and right sides, and the degree of variation in stride length changes may be used for the above identification. The analysis unit 33 displays the identification result on the terminal device 6 (S17).

[0037] Through this process, an analysis of subject U's walking state is performed using information from subject U's footsteps and images. Through the above process, the analysis unit 33 becomes able to easily identify the walking state of subject U.

[0038] Incidentally, assuming that subject U has no major impairment to walking, it is presumed that the stride length, walking posture (range of motion of the feet), walking rhythm, and characteristics of footsteps (period, sound components) are generally consistent. However, if there is a major impairment to walking, some phenomenon different from a healthy state may occur, such as inconsistencies in the stride length, walking posture (range of motion of the feet), walking rhythm, and characteristics of footsteps (period, sound components). In this embodiment, the presence or absence of impairment is identified based on the sound of walking (footsteps).

[0039] (Detection of sounds during walking) The detection of walking sounds in this embodiment will be described with reference to Figures 4A and 4B. Figures 4A and 4B are diagrams illustrating the detection of walking sounds in the embodiment. Figure 4A(a) shows a model of the detected sound waveform. Footsteps upon landing are detected periodically. Figure 4A(b) shows the absolute values ​​of the waveforms in Figure 4A(a). Figure 4A(c) shows a sine wave with a period of two steps. Figure 4A(d) shows the product of the instantaneous value of the waveform in Figure 4A(b) and the instantaneous value of the waveform in Figure 4A(c). Figure 4A(e) shows a comparison of the foot positions related to the generation of the footsteps described above.

[0040] Figure 4A shows an example where continuity is observed in the period of one step. The posture analysis system 1 uses the above-mentioned information on specific periods to detect sounds during walking. As shown in Figure 4A(a), the sounds during walking include footsteps and other sounds from subject U's walking behavior (see Figure 4A(e)).

[0041] For example, let's assume that one cycle is the time it takes for subject U to take two steps. The repeating signal described above is defined as a signal whose positive and negative values ​​change at a predetermined period, for example, a sine wave signal (Figure 4A(c)). For example, the 90-degree phase of the sine wave in Figure 4A(c) is aligned with the timing at which the waveform shown in Figure 4A(a) or Figure 4A(b) reaches its maximum sound pressure. In an ideal walking state, the 270-degree phase of the sine wave will coincide with the timing at which the maximum sound pressure of the next step is reached.

[0042] The instantaneous value of the waveform in Figure 4A(b) is multiplied by the instantaneous value of the waveform in Figure 4A(c), and the result is shown in Figure 4A(d). The waveform shown in Figure 4A(d) is divided into positive and negative for the left and right footsteps. Positive and negative peak values ​​appear in the result of this calculation. If loud sounds are detected at the 90-degree and 270-degree timings, the variation in the peak values ​​will be small. In an ideal walking state, footsteps occur roughly periodically. If the left and right sides are balanced, the time it takes to take each step will also be roughly the same on both sides. In this case, there is no significant difference in the peak values ​​of the waveforms in Figure 4A(b) and Figure 4A(d).

[0043] In contrast, Figure 4B shows an example where the period of one step becomes discontinuous. The waveforms shown in Figures 4B(a) to 4B(e) correspond to the waveforms shown in Figures 4A(a) to 4A(e) mentioned above, respectively. Figure 4B(e) shows the case where the landing of the second step with the left foot (L) is delayed. In this case, the waveforms shown in Figures 4B(a), 4B(b), and 4B(d) are all delayed by the amount of the delay in landing of the second step of the left foot (L). The sine wave shown in Figure 4B(c) is continuous. Therefore, if we use the phase of the sine wave shown in Figure 4B(c) as a reference, a portion of the waveforms shown in Figures 4B(a), 4B(b), and 4B(d) (the waveform of the second step of the left foot (L)) will be out of phase with the sine wave shown in Figure 4B(c).

[0044] Consequently, the timing of the peak value will be further away from the phases of the 90-degree and 270-degree sine waves. In this case, the larger the phase difference, the smaller the value of the sine function becomes. As a result, the product of these values ​​also becomes smaller.

[0045] Referring to Figure 6, we will explain how to use this relationship to determine sounds during walking. Figure 6 is a diagram illustrating the determination of sound during walking in the embodiment. In this graph, the horizontal axis is assigned to the time of one step, and the vertical axis is assigned to the loudness of the footsteps (landing sound). The negative direction of each axis is assigned to the value for the right foot (R), and the positive direction is assigned to the value for the left foot (L).

[0046] • Time from the sound of the left foot landing to the sound of the right foot landing: X1 • Time from the sound of the right foot landing to the sound of the left foot landing x 2 • Sound level of the right foot landing: Y1 • The loudness of the sound of the left foot landing (Y2)

[0047] Based on the above calculation results, the analysis unit 33 plots the left foot P1 (X1, Y1) and the right foot P2 (-X2, -Y2) on a two-dimensional XY coordinate system. In an ideal state where the person is perfectly balanced while walking, the line segment P1P2 passes through the origin O. Even if it's not a perfectly balanced, ideal state, if it's roughly balanced, it can be expected to pass near the origin O without deviating too far.

[0048] Therefore, the analysis unit 33 first determines the position of the left foot P1, draws a line segment from the left foot P1 through the origin O, determines the estimated position P2* which is the position of the right foot P2 corresponding to the left foot P1, and determines a predetermined range based on this. This predetermined range is shown as a rectangle.

[0049] The analysis unit 33 performs the same calculations as described above on the sound actually detected and plots the right foot P2 on the XY plane. The analysis unit 33 determines whether or not this right foot P2 falls within the aforementioned range.

[0050] The analysis unit 33 repeats the above determination, alternating between left and right, one step at a time, using the same method. For example, the sound of the left foot P3 was detected following the sound of the right foot P2. This left foot P3 corresponds to a case where the continuity of the period is not satisfied, as shown in Figure 4B above. The position of the left foot P3 deviates from the predetermined range of the rectangle.

[0051] The analysis unit 33 indicates that there was a discontinuous gait if the result of the determination shows that the left foot P3 deviates from the range. The analysis unit 33 may, in addition to or instead of indicating whether or not discontinuous walking occurs, classify and display the degree of discontinuity.

[0052] As described above, the analysis unit 33 extracts sounds to be used for analysis using information on specific periods determined by the periodic behavior identification unit 35. Based on the changes in the characteristics of the extracted sounds, the analysis unit 33 analyzes the posture of the subject U who produced the extracted sounds. The periodic behavior identification unit 35 detects the behavior of subject U from the image of subject U, and it is desirable that this behavior includes, for example, arm swings associated with walking and movements of the left and right legs.

[0053] The above determination method allows for real-time processing.

[0054] The posture analysis system 1 of this embodiment analyzes the posture of a subject U walking in an environment with ambient noise. The posture analysis system 1 comprises a sound collection unit 4 that collects sounds in the environment, and an analysis unit 33 that extracts sounds having a specific periodicity from the collected sounds and analyzes the posture of the subject U that generated the extracted sounds based on changes in the characteristics of the extracted sounds. This makes it possible to analyze the posture of a subject U walking in an environment with ambient noise.

[0055] In this embodiment, the periodic behavior identification unit 35 of the posture analysis system 1 uses a camera 4 that photographs the subject U in the environment to detect the periodicity of the subject U's behavior from the image of the subject U. Based on this detection result, the periodic behavior identification unit 35 may determine a specific period from the periodicity of the subject U's behavior. In this case, the analysis unit 33 may use the information of the specific period determined by the periodic behavior identification unit 35 to extract the footsteps of the subject U.

[0056] According to the above embodiment, it is possible to detect irregularities in the walking motion cycle based on the sounds (footsteps) detected during the walking motion of subject U. The sounds produced by subject U while walking are objectively evaluated, and the evaluation results are presented to the user. For example, by extracting differences in stride length and spacing between the left and right feet as sound features and applying the above evaluation method, the posture and condition of subject U can be evaluated. By performing this evaluation while adjusting the load of the movable floor device 2, it becomes possible to quantify the effect of changing the exercise load on subject U's motor ability and posture, thereby improving the reproducibility of the events during the evaluation.

[0057] It should be noted that there are individual differences among subjects U, and the length of each person's left and right legs and stride length are not necessarily the same. Even with individual differences among subjects U, it can be expected that the continuity of walking movement will be generally maintained. However, due to some factor during walking, it may be possible to not maintain the continuity of walking movement. By applying this embodiment and detecting walking states affected by some factor, it becomes possible to identify this factor. Eliminating this factor or avoiding a causal relationship with it would contribute to correcting the gait of non-test subject U.

[0058] (Second embodiment) The posture analysis system 1A of the second embodiment will be described with reference to Figures 7 and 8. Figure 7 is a schematic diagram of the posture analysis system 1A according to this embodiment. The posture analysis system 1 of the first embodiment described above identified periodic behavior using image information. In contrast, the posture analysis system 1A of this embodiment identifies periodic behavior using vibration information A. The differences will be explained below.

[0059] The posture analysis system 1A includes a periodic behavior identification unit 35A that replaces the periodic behavior identification unit 35. The movable floor device 2, which has a movable belt on which subject U1 or subject U2 walks, is equipped with a vibration sensor 25 that detects vibrations of the movable floor device 2. The periodic behavior identification unit 35A detects a periodic signal from the detection results of the vibration sensor 25 obtained during the behavior of subject U1 or subject U2, and determines a specific period from that periodic signal.

[0060] The analysis unit 33 extracts sounds to be used for analysis using information on specific periods determined by the periodic behavior identification unit 35A.

[0061] Referring to Figure 8, the analysis process of the embodiment will be described. Figure 8 is a flowchart of the analysis process in the embodiment.

[0062] The walking speed during the walking evaluation is pre-set using information from terminal device 6 (S11). Subject U stands on the movable belt 22 and, when ready to walk, operates the controller of the movable floor device 2 to start its operation. Subject U begins walking on the movable belt 22 at an appropriate speed corresponding to the movement speed of the movable belt 22 and starts taking images to evaluate their walking state. Accordingly, the analysis unit 33 starts detecting vibrations using the vibration sensor 25 (S12A). The analysis unit 33 acquires vibration data from the vibration sensor 25 (S13A).

[0063] The analysis unit 33 extracts periodic information from this vibration data due to the actions of subject U (S14A). The analysis unit 33 uses the periodic information from the subject U's actions to extract characteristic features of subject U's walking (S15). The analysis unit 33 identifies whether or not there is an abnormality in the posture during walking using predetermined judgment conditions (S16). The analysis unit 33 displays the identification result on the terminal device 6 (S17).

[0064] Through this process, information on subject U's footsteps and vibrations is used to analyze subject U's walking state. Through the above process, the analysis unit 33 becomes able to easily identify the walking state of subject U.

[0065] The posture analysis system 1A of the above embodiment comprises a movable floor device 2 having a movable belt on which a subject walks, and a periodic behavior identification unit 35A. The periodic behavior identification unit 35A detects vibrations of the movable floor device 2 from a vibration sensor 25, detects a periodic signal from the detection results of the vibration sensor 25 obtained during the subject U's actions, and determines the specific period from the periodic signal. The analysis unit 33 extracts the sound of the subject U using the information of the specific period determined by the periodic behavior identification unit 35A. As a result, posture analysis system 1A can analyze the posture of a subject walking in an environment with ambient noise, similar to posture analysis system 1.

[0066] (First modified example common to the embodiment) In such cases, the walking patterns desired by subject U1 and subject U2 generally do not coincide. Therefore, the sounds detected by the sound collection unit 4 include both the footsteps of subject U1 and subject U2. Furthermore, the periods of subject U1's footsteps and subject U2's footsteps do not coincide. Consequently, when analyzing the footsteps of one subject, the footsteps of the other subject become noise.

[0067] Furthermore, if the footsteps have different periods as described above, components with different periods can be suppressed using the same procedure as in the first embodiment.

[0068] (Second modified example common to the embodiment) In the above embodiment, an example was described in which image information or vibration information is used to estimate the period, and whether or not footsteps are included in the estimated period and the actually detected sound information. In this modified example, an example is shown in which sound information is used to estimate the period. In this case, a similar configuration can be achieved without using a device for obtaining image information or vibration information. When applying this modified example, it is desirable that the sound pressure level of ambient noise be relatively low. If this condition cannot be met, the method of the embodiment may be applied instead.

[0069] According to any of the above embodiments, the posture analysis system is a posture analysis system for analyzing the posture of a subject walking in an environment with ambient noise, comprising: a sound collection unit for collecting sounds in the environment; and an analysis unit for extracting sounds having a specific periodicity from the collected sounds, and analyzing the posture of the subject who produced the extracted sounds based on changes in the characteristics of the extracted sounds. This makes it possible to analyze the posture of a subject walking in an environment with ambient noise.

[0070] The embodiments of the present invention have been described above, but the posture analysis system 1 has a computer system inside. The series of processes related to the above-described process are stored in the form of a program on a computer-readable storage medium, and the above process is performed when the computer (CPU) reads and executes this program. Here, the computer-readable storage medium refers to a magnetic disk, magneto-optical disk, CD-ROM, DVD-ROM, semiconductor memory, etc. Alternatively, this computer program may be distributed to a computer via a communication line through a communication interface, and the computer that receives this distribution executes the program. Furthermore, the term "computer system" here includes an OS, etc. Image information from the camera 5 may also be collected via the communication interface.

[0071] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of symbols]

[0072] 1. 1A Posture Analysis System 2 Movable floor device 3 Analysis device 4 Sound collection section 5 Cameras 6 Terminal devices 25 Vibration Sensor 33 Analysis Department 35, 35A Periodic behavior identification unit

Claims

1. A posture analysis system for analyzing the posture of subjects in an environment where multiple subjects are walking, A sound collection unit for collecting sounds within the aforementioned environment, From the collected sounds, a sound having a specific periodicity is extracted. Based on the changes in the characteristics of the extracted sound, an analysis unit analyzes the posture of the subject who generated the extracted sound. A posture analysis system equipped with the following features.

2. A camera for photographing the subject in the aforementioned environment, A periodic behavior identification unit detects the periodicity of the subject's behavior from the subject's image and determines the specific period from the periodicity of the subject's behavior. Equipped with, The aforementioned analysis unit, The sound is extracted using the information of the specific period determined by the periodic behavior identification unit. The posture analysis system according to claim 1.

3. A movable floor device having a movable belt on which the subject walks, A vibration sensor for detecting vibrations of the movable floor device, A periodic behavior identification unit detects a periodic signal from the vibration sensor detection results obtained during the subject's behavior and determines the specific period from the periodic signal. Equipped with, The aforementioned analysis unit, The sound is extracted using the information of the specific period determined by the periodic behavior identification unit. The posture analysis system according to claim 1.

4. A computer-based posture analysis method for analyzing the posture of subjects in an environment where multiple subjects are walking, From the sounds in the environment collected by the sound collection unit, sounds having a specific periodicity are extracted. Based on the changes in the characteristics of the extracted sound, the posture of the subject who produced the extracted sound is analyzed. A posture analysis method that includes this.

Images