Pose estimation system, pose estimation method, and program

Abstract

The present application provides a posture estimation system capable of improving the estimation accuracy of a human posture using air pressure. The present application provides a posture estimation system including: an air pressure measurement unit that measures air pressure using at least one wearable sensor that measures air pressure mounted on a human body; and an estimation unit that estimates the posture of the human body using the measured air pressure. The posture is a standing posture, a sitting posture, or a lying posture. The posture estimation system estimates the posture of the human body by previously storing the air pressure difference at each posture.

Description

Technical Field

[0001] This invention relates to a pose estimation system, pose estimation method, and program. Background Technology

[0002] Patent document 1 describes an action status monitoring system that can manage measurement results in a common manner based on the installation orientation of the sensors.

[0003] Patent Document 1: Japanese Patent Application Publication No. 2022-34450 Summary of the Invention However, Patent Document 1 does not disclose the sensor's measurement of air pressure. Therefore, the object of the present invention is to provide a posture estimation system that improves the accuracy of human posture estimation by measuring air pressure.

[0004] The pose estimation system of the present invention is a pose estimation system comprising: A barometric pressure measuring unit that uses at least one wearable sensor, installed on the human body, to measure the barometric pressure and triaxial acceleration; and The estimation unit uses the measured air pressure to estimate the posture of the human body.

[0005] Based on the above structure, a posture estimation system is provided that can use air pressure to improve the estimation accuracy of human posture.

[0006] The pose estimation system of the present invention, wherein, The postures referred to are a person's standing, sitting, and lying postures.

[0007] The above structure is an example of the posture being estimated.

[0008] The pose estimation system of the present invention, wherein, The posture of the human body is estimated by pre-storing the air pressure difference at various postures.

[0009] The above structure is an example of a method for estimating human posture.

[0010] The pose estimation system of the present invention, wherein, The measured air pressure is classified into patterns, and the posture of the human body is estimated.

[0011] The above structure is an example of a method for estimating human posture.

[0012] The pattern classification in this invention is performed using machine learning.

[0013] The above structure indicates that the pose estimation system of the present invention uses artificial intelligence (AI).

[0014] The pattern classification in this invention is performed using an algorithm.

[0015] The above structure indicates that the pose estimation system of the present invention can also perform algorithm analysis.

[0016] The pose estimation system of the present invention, wherein, There are multiple wearable sensors. The posture of the human body is estimated by combining the triaxial acceleration, triaxial angular velocity, and the angle between the wearable sensors.

[0017] The above structure can improve the accuracy of pose estimation.

[0018] The pose estimation system of the present invention, wherein, The posture of the human body is estimated based on the measured pressure difference.

[0019] The above structure is an example of a method for estimating human posture.

[0020] The pose estimation method of the present invention is as follows, wherein, The air pressure is measured using at least one wearable sensor installed on the human body. The measured air pressure is used to estimate the posture of the human body.

[0021] Based on the above structure, a posture estimation method is provided that can use air pressure to improve the estimation accuracy of human posture.

[0022] The program of the present invention is a program that causes an information processing device to perform the following processes: The air pressure is measured using at least one wearable sensor for measuring air pressure, which is installed on the human body; and The measured air pressure is used to estimate the posture of the human body.

[0023] Based on the above structure, a program is provided that can use air pressure to improve the accuracy of human posture estimation.

[0024] Invention Effects According to the present invention, a posture estimation system is provided that can use air pressure to improve the estimation accuracy of human posture. Attached Figure Description

[0025] Figure 1 This is a block diagram illustrating the structure of the pose estimation system involved in the implementation.

[0026] Figure 2 This is a flowchart of the pose estimation method involved in the implementation.

[0027] Figure 3This is a graph showing the results of measurements taken using the sensors of the posture estimation system described in the embodiment.

[0028] Figure 4 This is a diagram showing an example of sensor installation in the posture estimation system according to the embodiment.

[0029] Figure 5 This is a diagram illustrating a first wearing example where the pressure difference is measured by the sensor of the posture estimation system involved in the embodiment.

[0030] Figure 6 This is a diagram illustrating a second wearing example where the pressure difference is measured by the sensor of the posture estimation system involved in the implementation.

[0031] Figure 7 This is a diagram illustrating a third wearing example where the pressure difference is measured by the sensor of the posture estimation system involved in the implementation.

[0032] Figure 8 This is a diagram illustrating a fourth wearing example where the pressure difference is measured by the sensor of the posture estimation system involved in the implementation.

[0033] Figure 9 This is a diagram showing an example of the results obtained by measuring the air pressure difference in the attitude estimation system involved in the implementation.

[0034] Figure 10 This is a diagram illustrating an example of the estimated posture involved in the implementation method.

[0035] Figure 11 This is a diagram showing the structure of the information processing device involved in the implementation method. Detailed Implementation

[0036] Implementation Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, the invention as described in the claims is not limited to the following embodiments. Furthermore, not all structures described in the embodiments are necessarily necessary means to solve the problem. For clarity of description, the following descriptions and drawings have been appropriately omitted and simplified. In the drawings, the same elements are labeled with the same symbols, and repeated descriptions are omitted as necessary.

[0037] (Description of the posture estimation system involved in the implementation) Figure 1 This is a block diagram illustrating the structure of the pose estimation system involved in the implementation. Figure 3 This is a graph showing the results of measurements taken using the sensors of the posture estimation system described in the embodiment. Figure 4 This is a diagram showing an example of sensor installation in the posture estimation system according to the embodiment. Figure 5 This is a diagram illustrating a first wearing example where the pressure difference is measured by the sensor of the posture estimation system involved in the embodiment. Figure 6 This is a diagram illustrating a second wearing example where the pressure difference is measured by the sensor of the posture estimation system involved in the implementation. Figure 7 This is a diagram illustrating a third wearing example where the pressure difference is measured by the sensor of the posture estimation system involved in the implementation. Figure 8 This is a diagram illustrating a fourth wearing example where the pressure difference is measured by the sensor of the posture estimation system involved in the implementation. Figure 9 This is a diagram showing an example of the results of measuring the air pressure difference in the attitude estimation system involved in the implementation.

[0038] Figure 10 This is a diagram illustrating an example of the estimated posture involved in the implementation method. (See reference) Figure 1 and Figures 3 to 10 The posture estimation system involved in the implementation method will be described.

[0039] like Figure 1 As shown, the posture estimation system 100 according to the embodiment includes a pressure measuring unit 101 and an estimation unit 102.

[0040] The barometric pressure measuring unit 101 uses at least one wearable sensor, installed on the human body, to measure triaxial acceleration, triaxial angular velocity, and barometric pressure. The three axes are the X, Y, and Z axes of a three-dimensional coordinate system. The wearable sensor is installed on the human body, particularly on the upper arm, forearm, thigh, calf, back, abdomen, or head, via a strap. When multiple wearable sensors are installed, their angles to each other can also be measured.

[0041] like Figure 3 As shown, wearable sensors can be used to observe a person's gait. Because the measurements can be performed with great precision, it can be used for rehabilitation of people with disabilities. If 24-hour activity levels can be monitored during rehabilitation, further rehabilitation can be achieved.

[0042] like Figure 4 As shown, wearable sensors do not need to be installed on all selected sites. Only a few wearable sensors need to be selected and installed. For example, for rehabilitation of the right foot, sensors can be installed on the right thigh and right calf for detailed measurements.

[0043] like Figure 5 As shown, the pressure difference can be the pressure difference between multiple wearable sensors. In this case, consider multiple wearable sensors installed at different heights to generate pressure differences, such as the lower leg and upper arm, the back and lower leg, the thigh and upper arm, the thigh and back, etc.

[0044] like Figure 6 As shown, the pressure difference can be the pressure difference between the mobile terminal 601, particularly a smartphone, and the wearable sensor. In this case, the wearable sensor can be a single device. Alternatively, an external terminal that has determined a reference position can replace the mobile terminal 601.

[0045] The estimation unit 102 uses the measured air pressure to estimate the posture of the human body.

[0046] like Figure 9 As shown, a pressure difference is generated between the ankles and back in a person's standing, sitting, and lying positions. Figure 9 It's a graph showing the measured air pressure difference. Areas with large air pressure differences are defined as standing, the middle areas as sitting, and the smallest areas as lying down. When walking, it can be identified as standing. Furthermore, this can be determined using accelerometers and angular velocity sensors.

[0047] Thus, by pre-storing the air pressure differences for each posture, it is possible to estimate the human body's posture. In particular, by pre-storing at least two air pressure differences, the remaining posture can also be determined.

[0048] Furthermore, it can perform pattern classification and pose determination based on various pieces of information. In pattern classification, supervised data is used to memorize the correct answer and make the determination. Alternatively, pattern classification can also be achieved through unsupervised learning in machine learning.

[0049] Alternatively, pattern classification can be performed using algorithms. The average value over a given period is divided into three segments for posture determination. In continuous patterns, moments of significant pressure change are considered posture transitions and are not included in the average calculation. In cases of slow changes, the average value is calculated when it crosses a certain threshold. In cases of slow changes where it's impossible to determine whether someone is standing or sitting, a moving average is used. In this case, the time period is shortened, and the posture is determined more precisely.

[0050] Furthermore, the initially measured pressure difference can be used as a calibration for a specific posture. The posture is divided into three stages; therefore, a high pressure difference can be detected as a standing posture, and a low pressure difference as a lying posture. In this way, posture estimation can be performed based on the magnitude of the measured pressure difference.

[0051] like Figure 7 and Figure 8 As shown, wearable sensors other than those for measuring air pressure differences can be installed. Figure 9 As shown, the air pressure difference changes significantly when postures are reversed. Therefore, false postures may be detected. Thus, by combining the triaxial acceleration, triaxial angular velocity, and angles between multiple wearable sensors to estimate human posture, the accuracy of posture estimation can be improved.

[0052] Based on the above structure, a posture estimation system can be provided that can use air pressure to improve the estimation accuracy of human posture.

[0053] (Explanation of the posture estimation method involved in the implementation) Figure 2 This is a flowchart of the pose estimation method involved in the implementation. Figure 11 This is a block diagram illustrating the structure of the information processing apparatus according to the embodiment. (Reference) Figure 2 and Figure 11 The posture estimation method involved in the implementation method will be explained.

[0054] First, the air pressure is measured (step S201). At least one wearable sensor, installed on the human body, measuring triaxial acceleration, triaxial angular velocity, and air pressure, is used to measure the air pressure difference. Multiple wearable sensors are used to measure the air pressure difference. Alternatively, wearable sensors and mobile terminal 601 are used to measure the air pressure. Next, the posture is estimated (step S202). The measured air pressure difference is used to estimate the human body's posture.

[0055] Based on the above structure, a posture estimation method can be provided that can use air pressure to improve the estimation accuracy of human posture.

[0056] The above method can be executed by processing information using an information processing device. For example... Figure 11 As shown, the information processing device 1100 includes a processor 1101 for executing programs and a memory 1102 for storing programs. The information processing device 1100 can be a single device or multiple devices. The information processing device 1100 can be a cloud server that provides part or all of the distributed processing functionality.

[0057] Part or all of the processing in the information processing apparatus 1100 can be implemented as a computer program. Such a program can be stored and supplied to the computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of physical recording media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., floppy disks, magnetic tapes, hard disks), optical-magnetic recording media (e.g., optical discs), compact disc read-only memory (CD-ROM), CD-R, CD-R / W, and semiconductor memory (e.g., mask ROM, programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), flash memory ROM, random access memory (RAM)). Furthermore, the program can be supplied to the computer via various types of temporary computer-readable media. Examples of temporary computer-readable media include electrical signals, optical signals, and electromagnetic waves. Temporary computer-readable media can provide the program to the computer via wired or wireless communication paths such as wires and optical fibers.

[0058] Furthermore, the present invention is not limited to the above-described embodiments and can be appropriately modified without departing from the spirit of the invention.

[0059] Symbol Explanation 100 - Attitude estimation system, 101 - Barometric pressure measurement unit, 102 - Estimation unit, 601 - Mobile terminal, 1100 - Information processing device, 1101 - Processor, 1102 - Memory.

Claims

1. A pose estimation system, characterized in that, have: A barometric pressure measuring unit that uses at least one wearable sensor for measuring barometric pressure, installed on the human body, to measure the barometric pressure; and The estimation unit uses the measured air pressure to estimate the posture of the human body.

2. The pose estimation system according to claim 1, characterized in that, The postures referred to are standing, sitting, and lying down.

3. The pose estimation system according to claim 1, characterized in that, The posture of the human body is estimated by pre-storing the air pressure difference at various postures.

4. The pose estimation system according to claim 1, characterized in that, The measured air pressure is classified into patterns, and the posture of the human body is estimated.

5. The pose estimation system according to claim 4, characterized in that, The pattern classification is performed using machine learning.

6. The pose estimation system according to claim 4, characterized in that, The pattern classification is performed using an algorithm.

7. The pose estimation system according to claim 1, characterized in that, There are multiple wearable sensors. The posture of the human body is estimated by combining the triaxial acceleration, triaxial angular velocity, and the angle between the wearable sensors.

8. The pose estimation system according to claim 1, characterized in that, The posture of the human body is estimated based on the measured pressure difference.

9. A pose estimation method, characterized in that, The air pressure is measured using at least one wearable sensor installed on the human body. The measured air pressure is used to estimate the posture of the human body.

10. A program, characterized in that, The information processing device shall perform the following processing: The air pressure is measured using at least one wearable sensor for measuring air pressure, which is installed on the human body; and The measured air pressure is used to estimate the posture of the human body.

Images