Attitude determination device and attitude determination method

Abstract

An attitude determination device and an attitude determination method are provided that can reduce the processing load associated with attitude determination. [Solution] The posture determination device 1 is roughly configured to include an acquisition unit 10 that acquires from the detection device a point cloud 43, which is a collection of points 44 in which position information and the intensity P of the reflected wave 41 are associated based on the reflection of the electromagnetic wave 40 output toward the detection area 42; an extraction unit 14 that extracts the maximum value of the intensity P of the point cloud 43 for each of multiple divided areas obtained by dividing the detection area 42 in a direction intersecting the height direction; and a determination unit 26 that determines the posture of the person to be detected 9 present in the detection area 42 based on the maximum value of the intensity P for each divided area.

Description

[Technical Field]

[0001] The present invention relates to an attitude determination device and an attitude determination method. [Background technology]

[0002] As a conventional technique, a human body posture detection device is known that includes a first determination unit that determines the location of a human body based on fluctuations in radar angle FFT (Fast Fourier Transform) data, a statistical unit that performs statistics on height information of a human body at the location of the human body based on differences in radar angle FFT data, and a detection unit that detects the posture of the human body based on the height information of the human body (see, for example, Patent Document 1).

[0003] This human body posture detection device does not use at least a point cloud, but performs posture detection using signal fluctuations caused by human body movement. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2023-026123 Summary of the Invention [Problem to be solved by the invention]

[0005] Conventional human body posture detection devices must calculate the position of the human body based on fluctuations in angle FFT data, which imposes a higher processing load than when using a point cloud containing position information.

[0006] SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide an attitude determination device and an attitude determination method that can reduce the processing load associated with attitude determination. [Means for solving the problem]

[0007] One aspect of the present invention provides a posture determination device that includes an acquisition unit that acquires a point cloud from a detection device, which is a collection of points in which position information and the intensity of the reflected wave are associated based on the reflection of electromagnetic waves output toward the detection area; an extraction unit that extracts the maximum value of the intensity of the point cloud for each of multiple divided areas obtained by dividing the detection area in a direction intersecting the height direction; and a determination unit that determines the posture of a person to be detected who is present in the detection area based on the maximum value of the intensity for each divided area.

[0008] Another aspect of the present invention is a method for detecting a point cloud, which is a set of points in which position information and the intensity of the reflected wave are associated based on the reflection of an electromagnetic wave output toward a detection area, from a detection device; extracting a maximum value of the intensity of the point cloud for each of a plurality of divided regions obtained by dividing the detection region in a direction intersecting with the height direction; The posture determination method determines the posture of a person to be detected who is present in the detection area based on the maximum value of the intensity for each of the divided areas. [Effects of the Invention]

[0009] According to the present invention, it is possible to reduce the processing load associated with posture determination. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is an example of a block diagram of a posture determination device according to an embodiment. [Figure 2] FIG. 2(a) is a diagram showing an example of the detection area of ​​the posture determination device according to the embodiment as viewed from above, and FIG. 2(b) is a diagram showing an example of the detection area as viewed from the side. [Figure 3] FIG. 3 is a diagram illustrating an example of a point cloud according to the embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of the maximum value of the intensity for each divided region of the posture determination device according to the embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of intensity for each divided region in a standing position according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of the intensity for each divided region in a sitting position according to the embodiment. [Figure 7] FIG. 7 is a diagram illustrating an example of the intensity for each divided region in the supine position according to the embodiment. [Figure 8] FIG. 8 is a flowchart illustrating an example of the operation of the posture determining device according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] (Summary of the embodiment) The posture determination device according to the embodiment is generally configured to include an acquisition unit that acquires a point cloud from the detection device, which is a collection of points in which position information and the intensity of the reflected wave are associated based on the reflection of the electromagnetic wave output toward the detection area; an extraction unit that extracts the maximum value of the intensity of the point cloud for each of multiple divided areas obtained by dividing the detection area in a direction intersecting with the height direction; and a determination unit that determines the posture of the person to be detected who is present in the detection area based on the maximum value of the intensity for each divided area.

[0012] Another embodiment of a posture determination method includes acquiring from a detection device a point cloud, which is a collection of points in which position information and the intensity of the reflected wave are associated based on the reflection of electromagnetic waves output toward a detection area, dividing the detection area in a direction intersecting the height direction to obtain a plurality of divided areas, extracting the maximum value of the intensity of the point cloud for each of the divided areas, and determining the posture of the person to be detected who is present in the detection area based on the maximum value of the intensity for each divided area.

[0013] This posture determination device and posture determination method determine the posture of the person to be detected using a point cloud containing position information, and therefore can reduce the processing load associated with posture determination compared to when this configuration is not adopted.

[0014] [Embodiment Mode] (Outline of posture determination device 1) Fig. 1 is an example of a block diagram of a posture determination device according to an embodiment. Fig. 2(a) is a diagram showing an example of a detection region of the posture determination device according to an embodiment viewed from above, and Fig. 2(b) is a diagram showing an example of the detection region viewed from the side. Fig. 3 is a diagram showing an example of a point cloud according to an embodiment. Fig. 4 is a diagram showing an example of the maximum intensity value for each divided region of the posture determination device according to an embodiment.

[0015] In the drawings relating to the embodiments described below, the ratios and shapes of figures may differ from the actual ratios and shapes. Also, in Fig. 1, arrows indicate the flow of main information. Below, we first provide an overview of the multiple posture determination devices 1.

[0016] As shown in Figures 1 to 2(b), the posture determination device 1 is generally configured to include an acquisition unit 10 that acquires a point cloud 43, which is a collection of points 44 in which position information and the intensity P of the reflected wave 41 are associated based on the reflection of the electromagnetic wave 40 output toward the detection area 42, from a 3D sensor 4 as a detection device; an extraction unit 14 that extracts the maximum value of the intensity P of the point cloud 43 for each of multiple divided areas obtained by dividing the detection area 42 in a direction intersecting the height direction; and a determination unit 26 that determines the posture of the person to be detected 9 present in the detection area 42 based on the maximum value of the intensity P for each divided area.

[0017] The 3D sensor 4 is, for example, a millimeter wave radar or a LiDAR (Light Detection and Ranging). The 3D sensor 4 of the present embodiment is, for example, a 79 GHz band MIMO radar, but is not limited to this. The 3D sensor 4 is, for example, placed at a height of 1.8 m from the floor, but is not limited to this.

[0018] 1 and 3, the 3D sensor 4 generates and outputs point cloud information S1, which is information about one period of point cloud 43 in the detection area 42. This point cloud information S1 is information in which coordinate values ​​indicating position information in an XYZ coordinate system are associated with intensities P at those coordinate values. Note that the 3D sensor 4 outputs the point cloud information S1 at intervals of, for example, 0.5 to 1.0 seconds.

[0019] The height direction is the positive direction of the Z axis. The directions intersecting the height direction are the XY axis directions intersecting the Z axis. Therefore, the detection area 42 is divided by the XY plane, but is not limited to this.

[0020] 1, the posture determination device 1 includes a center of gravity calculation unit 12 that calculates a center of gravity position 120 of the point cloud 43. An extraction unit 14 determines the posture of the detection target person 9 based on the center of gravity position 120.

[0021] 2(b), the divided regions in this embodiment are, by way of example only, the first divided region 14a to the nineteenth divided region 14s. The first divided region 14a to the nineteenth divided region 14s are regions obtained by dividing the detection region 42 at intervals of 10 cm.

[0022] Specifically, as shown in Fig. 3, the first divided region 14a is a region with a height of 0 to 10 cm. The second divided region 14b is a region with a height of 10 to 20 cm. The third divided region 14c to the eighteenth divided region 14r are regions with a height of 20 to 180 cm. The nineteenth divided region 14s is a region with a height of 180 to 190 cm.

[0023] 1 and 4, the posture determination device 1 includes a peak position extraction unit 18 that extracts a peak position 180 that has the greatest intensity and is located at the highest position among the maximum values ​​of the intensity P for each divided region. A determination unit 26 determines the posture of the detection target person 9 based on the center of gravity position 120 and the peak position 180.

[0024] 1 and 4, the posture determination device 1 includes an intensity extraction unit 20 that acquires the maximum value of intensity P of a divided region at a predetermined height. A determination unit 26 determines the posture of the detection target person 9 based on the center of gravity position 120, the peak position 180, and the maximum value of intensity of the divided region at the predetermined height.

[0025] 1 and 4, the posture determination device 1 has a threshold value 240 and includes a maximum height extraction unit 24 that extracts a high-position divided region 241 that is located at the highest position among the maximum values ​​of intensity P that are equal to or greater than the threshold value 240. A determination unit 26 determines whether the posture of the detection target person 9 is standing, sitting, or lying down based on the center of gravity position 120, the peak position 180, the maximum intensity value of divided regions at predetermined heights, and the high-position divided region 241.

[0026] The posture determination device 1 includes a threshold value correction unit 22 that varies the threshold value 240 depending on the distance from the 3D sensor 4, as shown in FIGS.

[0027] The posture determination device 1 further includes an averaging unit 16 that averages the maximum value of the intensity P of the point cloud 43 for each divided region acquired from the point cloud 43, and a communication unit 28 that communicates with connected electronic devices.

[0028] (Configuration of Acquisition Unit 10) The acquisition unit 10 is connected to the 3D sensor 4 and periodically acquires point cloud information S1, which is information on the point cloud 43. The acquisition unit 10 outputs the acquired point cloud information S1 to the center of gravity calculation unit 12 and the extraction unit 14.

[0029] (Configuration of the center of gravity calculation unit 12) The center of gravity calculation unit 12 calculates the center of gravity position 120 of the point cloud 43 based on the point cloud information S1 output from the acquisition unit 10. The center of gravity calculation unit 12 generates center of gravity information S2 as information on the calculated center of gravity position 120 and outputs it to the extraction unit 14 and the determination unit 26. The center of gravity calculation unit 12 calculates the center of gravity position 120 from multiple periods of point cloud information S1. In FIG. 4, as an example, the center of gravity position 120 is located in an eleventh divided region 14k.

[0030] (Configuration of extraction unit 14) The extraction unit 14 divides the detection area 42 into the first divided area 14a to the nineteenth divided area 14s based on the point cloud information S1, and extracts the maximum value of the intensity P for each divided area. As described above, the first divided area 14a to the nineteenth divided area 14s are divided every 10 cm. Therefore, the 3D sensor 4 detects the intensity P at heights from 0 to 190 cm.

[0031] The extraction unit 14 generates divided region intensity information S3, which is information on the maximum value of intensity P for each of the first divided region 14a to the nineteenth divided region 14s, and outputs the information to the averaging unit 16. The extraction unit 14 calculates the maximum value of intensity P for each of the acquired point cloud information S1 and outputs the information to the averaging unit 16.

[0032] (Configuration of averaging section 16) The averaging unit 16 averages the maximum values ​​of intensity P for the first divided region 14a to the nineteenth divided region 14s by using a moving average or the like based on the divided region intensity information S3 acquired from the extraction unit 14. The averaging unit 16 generates averaged information S4, which is information on the averaged maximum values ​​of intensity P for the first divided region 14a to the nineteenth divided region 14s, and outputs this information to the peak position extraction unit 18, the intensity extraction unit 20, and the threshold correction unit 22.

[0033] (Configuration of peak position extraction unit 18) The peak position extraction unit 18 extracts the divided region that has the highest position and has the maximum intensity P from the first divided region 14a to the nineteenth divided region 14s. In FIG. 4, the seventh divided region 14g to the thirteenth divided region 14m have the maximum intensity P, so the peak position 180 is the thirteenth divided region 14m, that is, 130 cm.

[0034] The peak position extraction unit 18 outputs peak position information S5, which is information on the divided region of the peak position 180, to the determination unit 26. In FIG.

[0035] (Configuration of the intensity extraction unit 20) The intensity extraction unit 20 acquires the intensity P of a divided region of a predetermined height. In the present embodiment, as an example, the intensity P of the 15th divided region 14o (150 cm) is acquired as the predetermined height. In FIG. 4, the intensity P of the 15th divided region 14o is illustrated as a designated high intensity 200. The intensity extraction unit 20 generates intensity information S6, which is information on the intensity P of the 15th divided region 14o (designated high intensity 200), and outputs it to the determination unit 26.

[0036] (Configuration of threshold correction unit 22) 4, the threshold value corrector 22 corrects the threshold value 240 in accordance with the distance from the 3D sensor 4. The threshold value corrector 22 generates threshold value information S7, which is information on the corrected threshold value 240, and outputs it to the maximum height extractor 24. Note that the threshold value corrector 22 may output the corrected threshold value 240, or may output the amount of correction from the threshold value 240.

[0037] (Configuration of maximum height extraction unit 24) The maximum height extraction unit 24 extracts a high position divided region 241 that is located at the highest position among the maximum values ​​of the intensity P for each divided region that are equal to or greater than the threshold value 240. In FIG. 4, the divided region that is located at the highest position that is equal to or greater than the threshold value 240 is the 17th divided region 14q, and this is illustrated as the high position divided region 241. The maximum height extraction unit 24 generates high position divided region information S8 that is information on the high position divided region 241, and outputs the information to the determination unit 26.

[0038] (Configuration of the determination unit 26) The determination unit 26 is a microcomputer including, for example, a CPU (Central Processing Unit) that performs calculations and processing on acquired data in accordance with a stored program 260, and semiconductor memories such as a RAM (Random Access Memory) and a ROM (Read Only Memory). The ROM stores, for example, the program 260 that causes the determination unit 26 to operate. The RAM is used, for example, as a storage area for temporarily storing calculation results. The determination unit 26 also has a means for generating a clock signal therein and operates based on this clock signal.

[0039] The determination unit 26 comprehensively controls the acquisition unit 10, the centroid calculation unit 12, the extraction unit 14, the averaging unit 16, the peak position extraction unit 18, the intensity extraction unit 20, the threshold correction unit 22, the maximum height extraction unit 24, and the communication unit. Note that at least a part of the centroid calculation unit 12, the extraction unit 14, the averaging unit 16, the peak position extraction unit 18, the intensity extraction unit 20, the threshold correction unit 22, and the maximum height extraction unit 24 may be realized by a program 260 executed by the determination unit 26, or may be realized by hardware such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array), or may be realized by a combination of both.

[0040] The determination unit 26 has a determination threshold value 261. This determination threshold value 261 is a threshold value that is compared with a predetermined height designated height intensity 200. When determining a standing position, the determination unit 26 sets a condition that at least the designated height intensity 200 is equal to or greater than the determination threshold value 261. When determining a sitting position or a lying position, the determination unit 26 sets a condition that at least the designated height intensity 200 is smaller than the determination threshold value 261.

[0041] The determination unit 26 determines whether the detection target person 9 is in a standing position, a sitting position, or a lying position according to the center of gravity position 120 based on the acquired center of gravity information S2, the peak position 180 based on the peak position information S5, the designated high intensity 200 based on the intensity information S6, and the high position divided area 241 based on the high position divided area information S8. The determination unit 26 generates determination information S9 indicating the determination result and outputs it to the electronic device 8 via the communication unit 28.

[0042] (About the judgment) FIG. 5 is a diagram showing an example of the intensity for each divided region in a standing position according to the embodiment. FIG. 6 is a diagram showing an example of the intensity for each divided region in a sitting position according to the embodiment. FIG. 7 is a diagram showing an example of the intensity for each divided region in a lying position according to the embodiment. In FIG. 5, as an example, since the position of the chest is close to the center of gravity, the center of gravity position 120 is located in the twelfth divided region 14l. In FIG. 6, as an example, the center of gravity position 120 is located in the tenth divided region 14j. In FIG. 7, the center of gravity position 120 is located in the second divided region 14b.

[0043] 5 to 7, for example, the lying position can be determined using at least one of the center of gravity position 120 and the peak position 180 because the center of gravity position 120 and the peak position 180 are located in a lower divided region compared to the others. However, compared to the lying position, the standing and sitting positions involve differences in the head position of the detection target person 9, but the intensity P may not be related to the detection target person 9, and the determination accuracy is low using the center of gravity position 120 and the peak position 180.

[0044] The designated high intensity 200 is extracted from the same divided region regardless of the posture of the detection target person 9. The other center of gravity position 120, peak position 180, and high position divided region 241 vary depending on the posture and physical characteristics of the detection target person 9, such as height. As shown in FIGS. 5 to 7, in a standing position, the peak position 180 and high position divided region 241 are located at positions higher than the designated high intensity 200, and in a sitting position and a lying position, the peak position 180 and high position divided region 241 are located at positions lower than the designated high intensity 200.

[0045] In both the sitting and lying positions, the peak position 180 and the high position divided region 241 are located at a position lower than the designated high intensity 200, but the size of the designated high intensity 200 is different. Therefore, the determination unit 26 compares the determination threshold value 261 with the designated high intensity 200 to determine whether the position is sitting or lying.

[0046] 6 and 7, the center of gravity position 120 is significantly different between the sitting position and the lying position. Therefore, the determination unit 26 further determines whether the position is sitting or lying based on the center of gravity position 120.

[0047] As described above, the determination unit 26 determines the posture based on the center of gravity position 120, the peak position 180, the designated high intensity 200, and the high position divided region 241.

[0048] An example of the operation of the posture determining device 1 will be described below with reference to the flowchart of FIG.

[0049] (operation) The acquisition unit 10 of the posture determination device 1 acquires point cloud information S1 from the 3D sensor 4 (Step 1).

[0050] The center of gravity calculation unit 12 calculates the center of gravity position 120 based on the acquired point cloud information S1 (Step 2), and outputs it to the determination unit 26 and the extraction unit 14.

[0051] The extraction unit 14 divides the detection area 42 into divided areas based on the point cloud information S1, and extracts the maximum value of the intensity P for each divided area (Step 3). The extraction unit 14 outputs divided area intensity information S3 to the averaging unit 16.

[0052] The averaging unit 16 performs averaging processing based on the divided region intensity information S3 (Step 4), and outputs averaging information S4 to the peak position extraction unit 18, the intensity extraction unit 20, and the threshold correction unit 22. The threshold correction unit 22 outputs threshold information S7 to the maximum height extraction unit 24, which corrects the threshold 240.

[0053] The peak position extraction unit 18, the intensity extraction unit 20, and the maximum height extraction unit 24 output peak position information S5 based on the peak position 180, intensity information S6 based on the specified high intensity 200, and high position division area information S8 based on the high position division area 241 to the judgment unit 26 (Step 5).

[0054] The determination unit 26 determines the posture of the person to be detected 9 based on the center of gravity position 120, peak position 180, specified high intensity 200 and high position division area 241, and outputs the determination information S9 to the connected electronic device 8 via the communication unit 28 (Step 6), thereby completing the determination process.

[0055] (Effects of the embodiment) The posture determination device 1 according to this embodiment determines the posture of the detection target person 9 based on point cloud information S1 including position information, and therefore can reduce the processing load compared to when this configuration is not adopted.

[0056] The posture determination device 1 divides the detection area 42 into multiple divided areas, extracts the maximum value of intensity P for each divided area, and determines the posture of the person to be detected 9 from the distribution of intensity P according to height, so that the posture can be more easily determined from intensity P for each height than when no division is made.

[0057] The posture determination device 1 determines the posture of the person to be detected 9 based on the center of gravity position 120, peak position 180, specified high intensity 200, and high-position division area 241, and therefore can accurately determine the posture of the person to be detected 9 as standing, sitting, or lying down, compared to when this configuration is not adopted.

[0058] In still another embodiment, the method may be provided as a program for executing a posture determination method, which includes acquiring from the detection device a point cloud 43, which is a collection of points 44 in which position information and intensity P of reflected wave 41 are associated based on the reflection of electromagnetic wave 40 output toward detection area 42, extracting the maximum value of intensity P of point cloud 43 for each of multiple divided areas obtained by dividing detection area 42 in a direction intersecting the height direction, and determining the posture of target person 9 present in detection area 42 based on the maximum value of intensity P for each divided area, or as a computer-readable recording medium having this program recorded thereon.

[0059] Although the embodiment of the present invention has been described above, this embodiment is merely an example and does not limit the scope of the invention as claimed. This novel embodiment can be embodied in various other forms, and various omissions, substitutions, modifications, etc. can be made without departing from the spirit of the present invention. Furthermore, not all combinations of features described in this embodiment are necessarily essential to the means for solving the problems of the invention. Furthermore, this embodiment is included within the scope and spirit of the invention, and is included in the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0060] 1... Posture determination device, 4... 3D sensor, 8... Electronic device, 9... Detection target, 10... Acquisition unit, 12... Center of gravity calculation unit, 14... Extraction unit, 14a to 14s... 1st to 19th divided areas, 16... Averaging unit, 18... Peak position extraction unit, 20... Intensity extraction unit, 22... Threshold correction unit, 24... Maximum height extraction unit, 26... Judgment unit, 28... Communication unit, 40... Electromagnetic wave, 41... Reflected wave, 42... Detection area, 43... Point cloud, 44... Point, 120... Center of gravity position, 180... Peak position, 200... Specified high intensity, 240... Threshold, 241... High position divided area, 260... Program, 261... Judgment threshold

Claims

1. an acquisition unit that acquires, from the detection device, a point cloud that is a set of points in which position information and the intensity of the reflected wave are associated based on the reflection of the electromagnetic wave output toward the detection area; an extraction unit that extracts a maximum value of the intensity of the point cloud for each of a plurality of divided regions obtained by dividing the detection region in a direction intersecting with a height direction; a determination unit that determines a posture of a person to be detected present in the detection area based on the maximum value of the intensity for each of the divided areas; An attitude determination device comprising:

2. a center of gravity calculation unit that calculates the center of gravity position of the point cloud; the determination unit determines the posture of the detection target person based on the center of gravity position. The posture determination device according to claim 1 .

3. a peak position extraction unit that extracts a peak position having the maximum intensity and at the highest position from the maximum values ​​of the intensity for each of the divided regions, the determination unit determines the posture of the detection target person based on the center of gravity position and the peak position. The posture determination device according to claim 2 .

4. an intensity extraction unit that extracts a maximum value of the intensity in the divided region at a predetermined height; the determination unit determines the posture of the detection target person based on the center of gravity position, the peak position, and the maximum value of the intensity of the divided region at the predetermined height. The posture determination device according to claim 3 .

5. a maximum height extraction unit that has a threshold value and extracts a high-position divided region that is at the highest position among the maximum values ​​of the intensity that are equal to or greater than the threshold value; the determination unit determines whether the posture of the detection target person is a standing posture, a sitting posture, or a lying posture based on the center of gravity position, the peak position, the intensity of the divided region at the predetermined height, and the high-position divided region. The posture determination device according to claim 4 .

6. a threshold correction unit that varies the threshold value according to the distance from the detection device; The posture determination device according to claim 5 .

7. A point cloud is obtained from the detection device, which is a set of points in which position information and the intensity of the reflected wave are associated based on the reflection of the electromagnetic wave output toward the detection area; extracting a maximum value of the intensity of the point cloud for each of a plurality of divided regions obtained by dividing the detection region in a direction intersecting with the height direction; A posture determination method for determining a posture of a detection target person present in the detection area based on the maximum value of the intensity for each of the divided areas.

Images