Footprint analysis system

Abstract

To improve the analysis accuracy of footprints.SOLUTION: A footprint analysis system includes: a load distribution sensor in which multiple detection cells which detect loads and on which a user walks are arranged in a matrix in a row direction and a column direction; footprint generation means that generates time-series data of a load distribution during walking of the user based on load values detected by the respective detection cells of the load distribution sensor, and generates a footprint of the user by superimposing the generated time-series data of the load distribution; and footprint analysis means that analyzes the footprint of the user generated by the footprint generation means. The footprint generation means generates the footprint of the user by excluding, from the time-series data of the load distribution, load values of detection cells whose response times are equal to or less than a threshold.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a footprint analysis system for analyzing a user's gait. [Background technology]

[0002] A load distribution sensor is known that is made up of an arrangement of multiple detection cells that detect loads, and a user walks on the detection cells, and a method is known for estimating the footprints of a user walking based on the load detected by each detection cell of the load distribution sensor (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-103512 Summary of the Invention [Problem to be solved by the invention]

[0004] However, for example, if a user walks with a drag, the shape of the footprints may be unclear. If such unclear footprints are used to analyze the footprints, the accuracy of the footprint analysis may be reduced.

[0005] The present disclosure has been made in consideration of such problems, and has as its main object to provide a footprint analysis system that can improve the accuracy of footprint analysis. [Means for solving the problem]

[0006] In order to achieve the above object, one aspect of the present disclosure is to a load distribution sensor having a plurality of detection cells arranged in a matrix in row and column directions to detect a load, the detection cells being over which a user walks; a footprint generating means for generating time series data of load distribution when the user is walking based on the load values ​​detected by each detection cell of the load distribution sensor, and for generating footprints of the user by superimposing the generated time series data of load distribution; a footprint analysis means for analyzing the user's footprint generated by the footprint generation means; Equipped with the footprint generating means generates the footprint of the user by excluding, from the time-series data of the load distribution, load values ​​of the detection cells whose reaction times are equal to or less than a threshold value. Footprint Analysis System is. In this aspect, The footprint generating means may generate the footprints of the user by excluding, from the time-series data of the load distribution, load values ​​detected by detection cells of the load distribution sensor in a final double support period. In this aspect, The footprint generating means may generate footprints of the user by excluding load values ​​detected by the detection cells located at the ends of the load distribution sensor from the time-series data of the load distribution. In this aspect, The footprint generating means may approximate the generated footprints to an ellipse, calculate the major axis of the ellipse, and exclude from the generated footprints any footprints whose major axis length is equal to or less than a threshold value. In this aspect, The footprint analysis means may analyze the footprints by calculating walking distance factors including at least one of overlapping stride distance, stride length, stride width, and foot angle based on the footprints generated by the footprint generation means. [Effects of the Invention]

[0007] According to the present disclosure, it is possible to provide a footprint analysis system that can improve the accuracy of footprint analysis. [Brief explanation of the drawings]

[0008] [Figure 1]1 is a block diagram showing a schematic system configuration of a footprint analysis system according to an embodiment of the present invention. [Figure 2] FIG. 10 shows footprints excluding the foot dragging portion. [Figure 3] FIG. 10 is a diagram showing overlapping stride distance, stride length, stride width, and foot angle. [Figure 4] 1 is a flowchart showing the flow of a footprint analysis method according to the present embodiment. [Figure 5] FIG. 1 is a diagram showing a walking cycle including an initial double support period, a right single support period, a final double support period, and a left single support period. [Figure 6] 10 is a flowchart showing an example of the flow of a footprint analysis method by the footprint analysis system. DETAILED DESCRIPTION OF THE INVENTION

[0009] Embodiment 1 Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Fig. 1 is a block diagram showing a schematic system configuration of a footprint analysis system according to this embodiment. The footprint analysis system 1 according to this embodiment includes a load distribution sensor 2 and a processing unit 3.

[0010] The load distribution sensor 2 is a panel-shaped sensor in which a plurality of detection cells 21 for detecting loads are arranged in a matrix in the row and column directions. The detection cells 21 are composed of, for example, piezoelectric elements or pressure-sensitive elements.

[0011] A user walks on the load distribution sensor 2. Each detection cell 21 of the load distribution sensor 2 detects the load value received from the sole of the user's foot while the user is walking. At this time, the load distribution sensor 2 acquires the load value of each detection cell 21 corresponding to the coordinate value of each detection cell 21. Then, the load distribution sensor 2 acquires a two-dimensional load distribution caused by the user's walking based on the load value of each detection cell 21 corresponding to the coordinate value of each detection cell 21.

[0012] The arithmetic processing device 3 performs a calculation process, which will be described later, based on the two-dimensional load distribution obtained by the load distribution sensor 2 when the user is walking, thereby analyzing the footprints.

[0013] The arithmetic processing device 3 has a hardware configuration of a typical computer, including, for example, a processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), internal memory such as a RAM (Random Access Memory) or a ROM (Read Only Memory), storage devices such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive), an input / output I / F for connecting peripheral devices such as a display, and a communication I / F for communicating with devices outside the device.

[0014] The arithmetic processing device 3 according to this embodiment includes a footprint generating unit 31 and a footprint analyzing unit 32, as shown in FIG.

[0015] The footprint generating unit 31 generates time-series data of the load distribution when the user is walking, based on the load values ​​detected by each detection cell 21 of the load distribution sensor 2. The footprint generating unit 31 superimposes the generated time-series data of the load distribution to generate footprints of the user.

[0016] For example, the footprint generating unit 31 detects the two-dimensional load distribution continuously acquired by the load distribution sensor 2 for a fixed time at fixed time intervals and generates multiple time-series load distribution images (time-series data of load distribution). The footprint generating unit 31 generates footprints of the user by superimposing the generated time-series load distribution images in the time direction.

[0017] In this superimposition process, the maximum value of the load values ​​of each detection cell 21 may be used as the value of the superimposed image at the position corresponding to that detection cell 21. Also, the sum of the load values ​​of each detection cell 21 may be used as the value of the superimposed image at the position corresponding to that detection cell 21.

[0018] For example, if a user, such as a hemiplegic patient due to a stroke, walks with a dragging foot, the shape of the footprint may become unclear due to the dragging foot, as shown in Figure 2(a). If such unclear footprints are used for footprint analysis, the accuracy of the footprint analysis may decrease.

[0019] In contrast, the footprint generator 31 according to this embodiment can generate footprints of an appropriate shape that exclude the dragging portion of footprints when a user walks with a drag as described above, as shown in Fig. 2(b). By using footprints of this appropriate shape, the accuracy of footprint analysis can be improved.

[0020] If the user drags their foot, the reaction time of the detection cell 21 to the dragged area will be shorter. The reaction time of the detection cell 21 refers to, for example, the time it takes for the detection cell 21 to react to a load and detect the load.

[0021] Therefore, the footprint generating unit 31 performs the above-mentioned superimposition process to generate the user's footprints by excluding the load values ​​of the detection cells 21 whose reaction times are equal to or less than the threshold from the time-series data of the load distribution. This makes it possible to generate footprints of an appropriate shape that exclude the dragging portion of footprints when the user walks with a drag as described above. Note that an optimal value obtained experimentally may be set in the footprint generating unit 31 in advance as the threshold.

[0022] In addition, the footprint generation unit 31 may perform the above-mentioned superposition process to generate the user's footprint from the time series data of the load distribution, excluding the load values ​​of the detection cells 21 whose proportion of the reaction time of the detection cells 21 at the time of footprint generation is below a threshold value.

[0023] As described above, footprint analysis unit 32 analyzes the footprints based on the footprints of appropriate shapes generated by footprint generation unit 31. This makes it possible to improve the accuracy of footprint analysis.

[0024] The footprint analysis unit 32 may analyze the footprints by calculating a distance factor of walking based on the footprints generated by the footprint generation unit 31 as described above. The distance factor includes, for example, overlapping stride distance, stride length, stride width, foot angle, etc., as shown in FIG.

[0025] The footprint analysis unit 32 may calculate the stride length, step width, and foot angle, for example, as follows: The footprint analysis unit 32 approximates the footprints generated by the footprint generation unit 31 into an ellipse, and calculates the major axis of this ellipse (hereinafter referred to as the footprint approximation ellipse). The footprint analysis unit 32 calculates the heel position of the footprint based on the center point, major axis angle, and footprint shape of the footprint approximation ellipse. The footprint analysis unit 32 calculates the stride length and step width based on the heel positions of the left and right footprints. The footprint analysis unit 32 calculates the foot angle based on the angle of the major axis of the footprint approximation ellipse.

[0026] Next, a footprint analysis method using the footprint analysis system according to the present embodiment will be described. Fig. 4 is a flowchart showing the flow of the footprint analysis method according to the present embodiment.

[0027] A user walks on the load distribution sensor 2 (step S101). Each detection cell 21 of the load distribution sensor 2 detects the load value received from the sole of the user's foot while the user is walking (step S102).

[0028] The footprint generating unit 31 generates time-series data of the load distribution when the user is walking, based on the load values ​​detected by each detection cell 21 of the load distribution sensor 2 (step S103).

[0029] The footprint generating unit 31 excludes the load values ​​of the detection cells 21 whose reaction times are equal to or less than a threshold value from the time series data of the load distribution (step S104). The footprint generating unit 31 superimposes the time series data of the load distribution to generate footprints of the user (step S105).

[0030] Footprint analysis unit 32 analyzes the footprints based on the footprints of appropriate shapes generated by footprint generation unit 31 (step S106).

[0031] As described above, the footprint analysis method according to this embodiment can generate correct footprints that exclude the dragging portion of footprints made when a user walks with a drag. By analyzing footprints based on these correct footprints, the accuracy of footprint analysis can be improved.

[0032] Embodiment 2 The footprint generation unit 31 in this embodiment excludes the load values ​​detected by the detection cells 21 of the load distribution sensor 2 during the final double support phase described below from the time series data of the load distribution, and performs the above-mentioned superposition process to generate the user's footprints.

[0033] For example, the footprint generating unit 31 distinguishes between a left foot load that is a load applied to the left foot and a right foot load that is a load applied to the right foot, based on the load values ​​detected by each detection cell 21 of the load distribution sensor 2.

[0034] Here, the user's walking motion alternates between a stance state in which the user's feet are on the ground and a swing state in which the user's feet are off the ground. Therefore, the footprint generating unit 31 can determine the swing state and stance state of the left and right feet when the user is walking, based on the determined left foot load and right foot load.

[0035] The footprint generating unit 31 determines the final double support period (hereinafter referred to as the final double support period) based on the swing state and stance state of the left and right feet determined above.

[0036] Here, the single-leg support period refers to a period in which only one of the right and left feet is in a standing position and touches the ground, while the double-leg support period refers to a period in which both the right and left feet are in a standing position and touch the ground.

[0037] For example, as shown in FIG. 5, if the right foot is used as a reference, the period from the initial contact of the right foot with the ground to the lift-off of the left foot is the initial double support period (referred to as the initial double support period). Next, the right foot enters the stance period (stance state) and the left foot enters the swing period (swing state), resulting in a right single support period. After that, the period from the initial contact of the left foot with the ground to the lift-off of the right foot is the final double support period. In the final double support period, the right foot, which serves as the reference, is positioned behind the user. In this final double support period, footprint drag may occur when the right foot kicks off the ground. Then, the right foot enters the swing period and the left foot enters the stance period, resulting in a left single support period.

[0038] The user walks by repeating a series of walking cycles, such as the initial double support period, the right single support period, the final double support period, and the left single support period, as described above.

[0039] The footprint generating unit 31 can determine each phase of the walking cycle described above and determine the final double support phase based on the swing state and stance state of the left and right feet. The footprint generating unit 31 then excludes the load values ​​detected by the detection cells 21 of the load distribution sensor 2 in the final double support phase from the time-series data of the load distribution. The footprint generating unit 31 superimposes the excluded time-series data of the load distribution to generate the user's footprints.

[0040] In this way, by generating footprints while excluding the load values ​​of the detection cells 21 in the final double support phase, it is possible to eliminate the dragging portion of the footprints when the user's feet leave the ground, thereby improving the accuracy of footprint analysis.

[0041] The footprint generator 31 may generate time-series data of the load distribution while the user is walking, based on the load values ​​detected by each detection cell 21 of the load distribution sensor 2 from the single support phase to the initial double support phase. The footprint generator 31 generates the user's footprints by superimposing the time-series data of the load distribution. As a result, similar to the above, by generating footprints while excluding the load values ​​of the detection cells 21 in the final double support phase, it is possible to exclude the dragging portion of the footprints when the user's feet take off the ground.

[0042] Furthermore, the footprint generating unit 31 may generate footprints from the time-series data of load distribution, excluding the load values ​​detected by the detection cells 21 at the four corners of the load distribution sensor 2. This is because there is a possibility that the footprint may extend outside the range of the load distribution sensor 2, and such protruding parts of the footprint can be excluded. This can improve the accuracy of footprint analysis.

[0043] The footprint generation unit 31 may also exclude from the generated footprints any footprints whose major axis length of the footprint approximation ellipse is less than a threshold. When walking on tiptoes or heels, the major axis of the footprint approximation ellipse becomes shorter, so such footprints can be excluded. This improves the accuracy of footprint analysis.

[0044] The footprint generating unit 31 may generate the user's footprints by any combination of the above-mentioned excluding load values ​​of detection cells 21 whose reaction times are below a threshold, generating time series data of the load distribution based on the load values ​​detected by each detection cell 21 of the load distribution sensor 2 from the initial double support period to the single support period, excluding load values ​​detected by detection cells 21 at the four corners of the load distribution sensor 2, and excluding footprints whose length of the major axis of the footprint approximation ellipse is below a threshold.

[0045] FIG. 6 is a flowchart showing an example of the flow of the footprint analysis method by the footprint analysis system described above.

[0046] The footprint generating unit 31 according to this embodiment generates time-series data of the load distribution when the user is walking, based on the load values ​​detected by each detection cell 21 of the load distribution sensor 2 (step S201).

[0047] The footprint generating unit 31 excludes the load values ​​detected by the detection cells 21 of the load distribution sensor 2 in the final double support phase from the time-series data of the load distribution (step S202).

[0048] The footprint generating unit 31 excludes the load values ​​of the detection cells 21 whose response times are equal to or less than the threshold value from the time series data of the load distribution (step S203).

[0049] The footprint generating unit 31 excludes the load values ​​detected by the detection cells 21 at the four corners of the load distribution sensor 2 from the time-series data of the load distribution (step S204).

[0050] The footprint generation unit 31 superimposes the time-series data of the load distribution to generate footprints of the user (step S205). The footprint generation unit 31 calculates the major axis of the footprint approximate ellipse of the generated footprints (step S206). The footprint generation unit 31 excludes footprints whose major axis length of the footprint approximate ellipse is equal to or less than a threshold value from the generated footprints (step S207).

[0051] The footprint analysis unit 32 analyzes the footprints by calculating a walking distance factor based on the footprints generated by the footprint generation unit 31 (step S208).

[0052] Embodiment 3 In this embodiment, the load distribution sensor 2 is fixedly disposed under the belt of the treadmill. When the load distribution sensor 2 is disposed under the belt of the treadmill and footprints of a person walking on the belt of the treadmill are generated, the feet move as the belt moves.

[0053] Therefore, the footprint generator 31 according to this embodiment generates footprints by superimposing the time-series data of the load data detected by each detection cell 21 of the load distribution sensor 2 so as to cancel out the moving speed of the treadmill belt. For example, the footprint generator 31 generates the user's footprints by superimposing the generated time-series load distribution image in the time direction so as to cancel out the moving speed of the treadmill belt. This makes it possible to calculate the distance factor even for walking on a treadmill.

[0054] It should be noted that when the treadmill belt accelerates or decelerates, or during the first walking cycle, the user's gait may become unstable, making it difficult to obtain accurate footprints.

[0055] In contrast to this, the footprint generating unit 31 may perform the above-mentioned superimposition process to generate the user's footprints by excluding the load values ​​detected by the detection cells 21 of the load distribution sensor 2 when the treadmill belt is accelerating from the time-series data of the load distribution. This makes it possible to obtain accurate footprints and improve the accuracy of footprint analysis.

[0056] In addition, the footprint generating unit 31 may exclude the load value detected by the detection cell 21 of the load distribution sensor 2 when the treadmill belt is decelerating from the time series data of the load distribution, and perform the above-mentioned superimposition process to generate the user's footprints.

[0057] Furthermore, the footprint generating unit 31 may exclude the load values ​​detected by the detection cells 21 of the load distribution sensor 2 in the first predetermined number of walking cycles from the time-series data of load distribution, and perform the above-mentioned superimposition process to generate the user's footprints. The footprint generating unit 31 may exclude the load values ​​detected by the detection cells 21 of the load distribution sensor 2 in, for example, the first two walking cycles from the time-series data of load distribution, and perform the above-mentioned superimposition process to generate the user's footprints.

[0058] Although several embodiments of the present disclosure have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the scope of the invention and its equivalents as defined in the claims.

[0059] The present disclosure can also be implemented by causing a processor to execute a computer program to perform the processes shown in FIG. 4 or FIG. 6, for example.

[0060] The program can be stored and supplied to a computer using various types of non-transitory computer readable media. Non-transitory computer readable media include various types of tangible storage media. Examples of non-transitory computer readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)).

[0061] The program may be provided to the computer by various types of transitory computer-readable media. Examples of transitory computer-readable media include electrical signals, optical signals, and electromagnetic waves. The transitory computer-readable media can provide the program to the computer via a wired communication path such as an electrical wire or optical fiber, or via a wireless communication path.

[0062] Each component of the footprint analysis system 1 according to the above-described embodiment can be realized not only by a program, but also in part or in whole by dedicated hardware such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field-Programmable Gate Array). [Explanation of symbols]

[0063] 1. Footprint analysis system 2 Load distribution sensor 3. Processing Unit 31 Footprint generator 32 Footprint Analysis Department

Claims

1. a load distribution sensor having a plurality of detection cells arranged in a matrix in row and column directions to detect a load, the detection cells being over which a user walks; a footprint generating means for generating time series data of load distribution when the user is walking based on the load values ​​detected by each detection cell of the load distribution sensor, and for generating footprints of the user by superimposing the generated time series data of load distribution; a footprint analysis means for analyzing the user's footprint generated by the footprint generation means; Equipped with the footprint generating means generates the footprint of the user by excluding, from the time-series data of the load distribution, load values ​​of the detection cells whose reaction times are equal to or less than a threshold value. Footprint analysis system.

2. 2. The footprint analysis system of claim 1, the footprint generating means generates the footprints of the user by excluding the load values ​​detected by the detection cells of the load distribution sensor in the final double support period from the time-series data of the load distribution. Footprint analysis system.

3. 2. The footprint analysis system of claim 1, the footprint generating means generates footprints of the user by excluding load values ​​detected by the detection cells at the ends of the load distribution sensor from the time-series data of the load distribution. Footprint analysis system.

4. 2. The footprint analysis system of claim 1, the footprint generating means approximates the generated footprint to an ellipse, calculates the major axis of the ellipse, and excludes from the generated footprints any footprint whose major axis length is equal to or less than a threshold value; Footprint analysis system.

5. 2. The footprint analysis system of claim 1, the footprint analysis means calculates walking distance factors including at least one of overlapping stride distance, stride length, stride width, and foot angle based on the footprints generated by the footprint generation means, thereby analyzing the footprints; Footprint analysis system.

Images