JP6750193B2 - Walking cycle detection method and detection apparatus - Google Patents
Walking cycle detection method and detection apparatus Download PDFInfo
- Publication number
- JP6750193B2 JP6750193B2 JP2015120165A JP2015120165A JP6750193B2 JP 6750193 B2 JP6750193 B2 JP 6750193B2 JP 2015120165 A JP2015120165 A JP 2015120165A JP 2015120165 A JP2015120165 A JP 2015120165A JP 6750193 B2 JP6750193 B2 JP 6750193B2
- Authority
- JP
- Japan
- Prior art keywords
- subject
- walking cycle
- walking
- camera
- height
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Description
本発明は、歩行周期の検出方法及び装置に関する。 The present invention relates to a walking cycle detection method and apparatus.
歩行周期は、歩行動作の詳細な解析を行う上で広く使用されている。 The gait cycle is widely used in detailed analysis of gait motion.
歩行周期の計測方法としては、被験者の画像を多方向から撮影し、３次元人物モデルの両脚のくるぶし間距離を算出し、その両脚のくるぶし間距離の最小値から歩行周期を算出する方法（特許文献１）、シート式圧力センサを使用して歩行周期を測定する方法（特許文献２）などが知られている。 As a method of measuring the walking cycle, a method of capturing images of a subject from multiple directions, calculating a distance between the ankles of both legs of a three-dimensional human model, and calculating a walking cycle from the minimum value of the distance between the ankles of both legs (Patent Document 1), a method of measuring a walking cycle using a sheet pressure sensor (Patent Document 2), and the like are known.
しかしながら、特許文献１に記載のように、足の位置情報を利用して歩行周期を検出する場合、足に左右差がある被験者では両脚のくるぶし間距離から歩行周期を正確に検出することができない。また、複数台の撮像装置を用いて多方向から撮影することが必要とされるので、大掛かりな計測装置を設置できる専用の大きなスペースを必要とし、検出システムが複雑である。 However, as described in Patent Document 1, when the walking cycle is detected by using the position information of the foot, a subject having a left/right difference in the foot cannot accurately detect the walking cycle from the distance between the ankles of both legs. .. Further, since it is necessary to take images from multiple directions using a plurality of imaging devices, a large dedicated space for installing a large-scale measurement device is required, and the detection system is complicated.
特許文献２に記載のようにシート式圧力センサを使用すると、歩行特徴の検査会場などで被験者にシート式圧力センサ上を歩行してもらったときの歩行周期は正確に得られるが、そうして得られた歩行周期が、日常生活における歩行周期と一致するとは限らない。 When the seat type pressure sensor is used as described in Patent Document 2, the walking cycle when the subject walks on the seat type pressure sensor at an inspection site of walking characteristics or the like can be accurately obtained. The obtained gait cycle does not always match the gait cycle in daily life.
これに対し、本発明は、日常生活の歩行周期を正確に省スペースで簡便なシステムで検出することを可能とする新たな技術の提供を課題とする。 On the other hand, an object of the present invention is to provide a new technique that enables the walking cycle of daily life to be accurately detected with a space-saving and simple system.
本発明者は、歩行中の被験者の頭部の高さの時間変化の計測から、簡便にかつ正確に歩行周期を検出できることを見出し、本発明を想到した。 The present inventor has found that the walking cycle can be easily and accurately detected from the measurement of the temporal change in the height of the head of the subject during walking, and has conceived the present invention.
即ち、本発明は、歩行中の被験者の頭部の高さの時間変化を計測し、その高さの時間変化の波形に基づいて歩行周期を検出する歩行周期の検出方法を提供する。 That is, the present invention provides a method for detecting a walking cycle in which the time change of the height of the head of a subject during walking is measured and the walking cycle is detected based on the waveform of the time change of the height.
また、歩行中の被験者の頭部の高さの時間変化を計測する計測手段、及びその高さの時間変化の波形に基づいて歩行周期を検出する演算手段を備えた歩行周期の検出装置を提供する。 Further, there is provided a walking cycle detection device including a measuring unit that measures a temporal change in the height of the head of a subject while walking and a calculating unit that detects a walking cycle based on a waveform of the temporal change in the height. To do.
本発明によれば、例えば、歩行中の被験者の頭部の高さの時間変化を、被験者の正面に向いたカメラで、該カメラに近づいてくる被験者の画像を撮ることにより計測し、その頭部の高さの時間変化から歩行周期を正確に検出することができる。よって、被験者の居住空間にある廊下などで、日常の歩行に対して簡便に歩行周期を検出することが可能となる。 According to the present invention, for example, a temporal change in the height of the head of a walking subject is measured by taking an image of the subject approaching the camera with a camera facing the front of the subject, and measuring the head. The walking cycle can be accurately detected from the temporal change of the height of the part. Therefore, it becomes possible to easily detect the walking cycle in daily walks in a corridor in the living space of the subject.
以下、図面を参照しつつ本発明を詳細に説明する。なお、各図中、同一符号は同一又は同等の構成要素を表している。 Hereinafter, the present invention will be described in detail with reference to the drawings. In each drawing, the same reference numerals represent the same or equivalent constituent elements.
図１は、本発明の一実施例の歩行周期の検出システム１０の模式図である。
この歩行周期の検出システム１０は、歩行中の被験者１の頭部２の高さの時間変化を計測する計測手段として、動画を撮るカメラ１１を有し、また、カメラ１１で計測した頭２の高さの時間変化の波形に基づいて歩行周期を算出する演算手段としてパーソナルコンピュータ１２を備えている。ここで、カメラ１１は、被験者１の歩行の進行方向にあり、被験者１の正面に向けて設置されている。
FIG. 1 is a schematic diagram of a walking cycle detection system 10 according to an embodiment of the present invention.
The walking cycle detection system 10 has a camera 11 that takes a moving image as a measuring unit that measures the temporal change in the height of the head 2 of the subject 1 during walking, and the head 2 of the head 2 measured by the camera 11 is measured. The personal computer 12 is provided as a calculation means for calculating the walking cycle based on the waveform of the height change over time. Here, the camera 11 is in the traveling direction of the subject 1, and is installed toward the front of the subject 1.
このカメラ１１としては、通常の動画を撮ると共に、カメラ１１から被験者１までの深度情報をリアルタイムに計測するデプスカメラを使用することができる。より具体的には、カメラ１１として、例えば、ＲＧＢカラー映像用カメラ１１ａと、奥行き測定用に赤外線カメラ１１ｂと赤外線発光部１１ｃを備えたものを使用することができる。このようにカメラ１１としてデプスカメラを使用することにより、被験者１のＲＧＢ画像を撮ると同時に、デプスカメラ１１から被験者１までの深度情報をリアルタタイムに計測することができ、それにより、被験者１の関節点３の位置情報も自動的に抽出し、被験者１の歩行画像に重ねて表示することが可能となる。このようなデプスカメラ１１としては、例えば、マイクロソフト社のＫｉｎｅｃｔを使用することができる。 As the camera 11, it is possible to use a depth camera that takes normal moving images and measures depth information from the camera 11 to the subject 1 in real time. More specifically, as the camera 11, for example, a camera including an RGB color image camera 11a, an infrared camera 11b for depth measurement, and an infrared light emitting unit 11c can be used. By using the depth camera as the camera 11 in this way, it is possible to take an RGB image of the subject 1 and at the same time measure the depth information from the depth camera 11 to the subject 1 in real time. It is possible to automatically extract the position information of the joint point 3 and display the position information on the walking image of the subject 1 in an overlapping manner. As such a depth camera 11, for example, Kinect of Microsoft Corporation can be used.
なお、本発明において、歩行周期の検出のために、被験者１の頭部２の高さの時間変化を計測する手段としては、カメラ１１から被験者１までの深度情報を計測できることは必ずしも必要ではなく、カメラ１１としては一般的なビデオカメラを使用することができる。ただし、検出した歩行周期をさらなる歩行解析で使用できるようにするため、デプスカメラを使用することが好ましい。また、カメラ１１としてデプスカメラを使用せず、一般的なビデオカメラを使用して被験者１の頭部２の高さの時間変化から歩行周期を検出する場合に、被験者１の頭部２の高さ自体を算出する必要はない。もっとも、歩行周期の検出精度を高めるためには、カメラ１１の垂直方向における設置位置は、被験者の直立静止状態の頭の高さに、概ね一致させておくことが好ましい。 In the present invention, it is not always necessary to be able to measure the depth information from the camera 11 to the subject 1 as means for measuring the temporal change in the height of the head 2 of the subject 1 for detecting the walking cycle. As the camera 11, a general video camera can be used. However, it is preferable to use a depth camera so that the detected gait cycle can be used for further gait analysis. When the walking cycle is detected from the time change of the height of the head 2 of the subject 1 without using the depth camera as the camera 11, the height of the head 2 of the subject 1 is detected. There is no need to calculate the size itself. However, in order to improve the detection accuracy of the walking cycle, it is preferable that the installation position of the camera 11 in the vertical direction substantially coincides with the height of the head of the subject in the upright still state.
図２（ａ）は、ＲＧＢカラー映像用カメラ１１ａで捉えた頭部２の高さの時間変化Ｙ(t)の波形２０である。なお、同図には、カメラ１１で計測された生データの波形と、それをスムージング処理した波形を示した。本発明では、歩行周期を検出するために、この波形２０の極大値２１及び極小値２２を算出する。 FIG. 2A shows a waveform 20 of the temporal change Y(t) of the height of the head 2 captured by the RGB color video camera 11a. In the figure, the waveform of raw data measured by the camera 11 and the waveform obtained by smoothing the raw data are shown. In the present invention, the maximum value 21 and the minimum value 22 of this waveform 20 are calculated in order to detect the walking cycle.
より具体的には、カメラ１１としてデプスカメラを使用する場合、その深度情報を計測し得る奥行き座標に対応する時間（ｔ）の波形２０を処理対象とする。 More specifically, when a depth camera is used as the camera 11, the waveform 20 at the time (t) corresponding to the depth coordinates at which the depth information can be measured is the processing target.
そして、波形２０からノイズを除去するため、歩行周期の検出に影響を与えない高周波成分を除去するローパスフィルターで平滑化する。例えば、歩行周期は通常０．５〜２Ｈｚの範囲であるため、カメラ１１で撮った画像から頭部の高さを１００Ｈｚで抽出して波形を得た場合に、１００Ｈｚの約６％である６Ｈｚよりも低い周波数成分を透過させることにより平滑化した波形を得る。 Then, in order to remove noise from the waveform 20, it is smoothed by a low-pass filter that removes high-frequency components that do not affect the detection of the walking cycle. For example, since the walking cycle is usually in the range of 0.5 to 2 Hz, when a waveform is obtained by extracting the height of the head at 100 Hz from the image taken by the camera 11, 6 Hz, which is about 6% of 100 Hz, is obtained. A smoothed waveform is obtained by transmitting lower frequency components.
次に、図２（ｂ）に示すように、頭部２の高さの座標Ｙ(ｔ)を微分してＹ'(ｔ)を得る。微分は、次式の中央差分により行うことができる。
Ｙ'(ｔ)＝（Ｙ(i＋1)−Ｙ(i−1)）／（ｔ(i＋1)−ｔ(i−1)）
Next, as shown in FIG. 2B, the coordinate Y(t) of the height of the head 2 is differentiated to obtain Y′(t). Differentiation can be performed by the central difference of the following equation.
Y'(t)=(Y(i+1)-Y(i-1))/(t(i+1)-t(i-1))
次にＹ'(ｔ)の正負反転を検出し、極大又は極小を判定する。
即ち、Ｙ'(i＋1)*Ｙ'(i)＜０となるiを検出し、Ｙ'(i＋1)の絶対値とＹ'(i)の絶対値で０に近い側のｔを極値を与える時刻とする。このとき、Ｙ'(i)＞０ならば極大値を与える時刻であり、Ｙ'(i)＜０ならば極小値を与える時刻である。
Next, the positive/negative inversion of Y'(t) is detected to determine the maximum or minimum.
That is, i that satisfies Y'(i+1)*Y'(i)<0 is detected, and the absolute value of Y'(i+1) and the absolute value of Y'(i) are set to the extreme value of t on the side close to 0. It is time to give. At this time, if Y'(i)>0, it is the time to give the maximum value, and if Y'(i)<0, it is the time to give the minimum value.
そして、図２（ｂ）に示すように、極大値と極小値からなる極値４区間（極値５点）に対応する時間を１歩行周期とすることができる。 Then, as shown in FIG. 2B, a time period corresponding to four extreme value zones (maximum value points: 5 points) having a maximum value and a minimum value can be set as one walking cycle.
このように、頭部の高さの時間変化から歩行周期を検出すると、頭部２を覆う服装は少ないため、服装によるアーティファクトを受けにくく、関節点３を使用して歩行周期を検出する場合に比して正確な歩行周期を得ることができる。 In this way, when the walking cycle is detected from the temporal change of the height of the head, the clothing that covers the head 2 is small, so that it is difficult to receive the artifact due to the clothing, and when the walking cycle is detected using the joint point 3. In comparison, an accurate walking cycle can be obtained.
また、歩行中の被験者の頭部の高さの時間変化を計測する計測手段としては、被験者の正面又は背面から被験者の頭の動きを撮るカメラが１台あればよく、歩行する被験者に対して複数台のカメラを向けられない幅狭い場所でも歩行周期を検出することが可能となる。したがって、例えば、床面積が比較的狭い店頭や、被験者の居住地の廊下などで歩行周期を計測することが可能となる。 Further, as the measuring means for measuring the temporal change in the height of the head of the walking subject, one camera for taking the movement of the subject's head from the front or back of the subject may be used. The walking cycle can be detected even in a narrow place where a plurality of cameras cannot be aimed. Therefore, for example, it becomes possible to measure the walking cycle at a storefront with a relatively small floor area, a corridor in the subject's residence, or the like.
なお、カメラ１１で関節点３も観察する場合には、当該カメラ１１の特性にもよるが、歩行により被験者１がカメラに近づいてくる向きで被験者の頭の画像を撮ることが好ましい。 When the joint point 3 is also observed by the camera 11, it is preferable to take an image of the subject's head in a direction in which the subject 1 approaches the camera while walking, depending on the characteristics of the camera 11.
また、歩行中の被験者１の頭部の高さの時間変化を検出する間に、被験者１の側方からｗｅｂカメラなどで被験者１の矢状面撮影を行い、歩行周期以外の歩行パラメータ（例えば、立脚期割合、遊脚期割合、歩幅等）を検出できるようにしてもよい。 Further, while detecting the temporal change in the height of the head of the subject 1 while walking, the subject 1 is photographed from the side with a sagittal plane with a web camera or the like, and a walking parameter other than the walking cycle (for example, , Stance period ratio, swing period ratio, stride length, etc.) may be detected.
こうして得られた歩行周期その他の歩行パラメータは、歩行年齢、健康リスク、歩行時の上体姿勢などを評価し、被験者の歩行特徴に対してアドバイスする場合に有用となる。 The gait cycle and other gait parameters obtained in this way are useful for evaluating gait age, health risk, upper body posture during gait, etc., and giving advice to the gait characteristics of the subject.
以下、実施例により本発明を具体的に説明する。
図１に示した歩行周期の検出システムにおいて、カメラ１１として、Ｋｉｎｅｃｔ（マイクロソフト社）を使用し、７０〜９０代の女性４９５名の歩行周期を検出した。
Hereinafter, the present invention will be specifically described with reference to examples.
In the gait cycle detection system shown in FIG. 1, Kinect (Microsoft Corporation) was used as the camera 11, and the gait cycle of 495 women in their 70s to 90s was detected.
一方、同じ被験者にシート式圧力センサ（ウォークＷａｙ（アニマ株式会社））上を歩行してもらい、歩行周期を測定した。そして図３に示すように、本発明の方法により得た歩行周期を、シート式圧力センサから得た歩行周期に対してプロットし、本発明の方法により得られた歩行周期の妥当性を調べた。 On the other hand, the same subject was made to walk on a sheet type pressure sensor (Walk Way (Anima Co., Ltd.)) and the walking cycle was measured. Then, as shown in FIG. 3, the walking cycle obtained by the method of the present invention was plotted against the walking cycle obtained by the sheet pressure sensor, and the validity of the walking cycle obtained by the method of the present invention was examined. ..
その結果、図３に示すように、本発明の方法により得た歩行周期は、シート式圧力センサから得た歩行周期と高い相関（相関係数Ｒ＝０．７５６）を示した。 As a result, as shown in FIG. 3, the walking cycle obtained by the method of the present invention showed a high correlation (correlation coefficient R=0.756) with the walking cycle obtained from the sheet pressure sensor.
１ 被験者
２ 頭部
３ 関節点
１０ 歩行周期の検出システム
１１ カメラ
１１ａ ＲＧＢカラー映像用カメラ
１１ｂ 赤外線カメラ
１１ｃ 赤外線発光部
１２ パーソナルコンピュータ（演算手段）
２０ 波形
２１ 極大値
２２ 極小値
1 Subject 2 Head 3 Joint Point 10 Walking Cycle Detection System 11 Camera 11a RGB Color Image Camera 11b Infrared Camera 11c Infrared Light Emitting Unit 12 Personal Computer (Calculation Means)
20 Waveform 21 Maximum value 22 Minimum value
Claims (4)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2015120165A JP6750193B2 (en) | 2015-06-15 | 2015-06-15 | Walking cycle detection method and detection apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2015120165A JP6750193B2 (en) | 2015-06-15 | 2015-06-15 | Walking cycle detection method and detection apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JP2017000601A JP2017000601A (en) | 2017-01-05 |
| JP6750193B2 true JP6750193B2 (en) | 2020-09-02 |
Family
ID=57751006
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP2015120165A Active JP6750193B2 (en) | 2015-06-15 | 2015-06-15 | Walking cycle detection method and detection apparatus |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JP6750193B2 (en) |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3655618B2 (en) * | 2003-03-28 | 2005-06-02 | 株式会社エヌ・ティ・ティ・データ | Pedestrian age determination device, walking state / pedestrian age determination method and program |
| JP2004344418A (en) * | 2003-05-22 | 2004-12-09 | Anima Kk | Three-dimensional motion analyzing device |
| JPWO2005096939A1 (en) * | 2004-04-07 | 2008-02-28 | 財団法人新産業創造研究機構 | Rehabilitation support measurement system and rehabilitation support measurement method |
| JP2009189671A (en) * | 2008-02-18 | 2009-08-27 | Nippon Telegr & Teleph Corp <Ntt> | Walking cycle detection method, walking cycle detection apparatus, and walking cycle detection program |
| JP2013101526A (en) * | 2011-11-09 | 2013-05-23 | Sony Corp | Information processing apparatus, display control method, and program |
| JP2014217692A (en) * | 2013-05-10 | 2014-11-20 | オムロンヘルスケア株式会社 | Walking posture meter and program |
-
2015
- 2015-06-15 JP JP2015120165A patent/JP6750193B2/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| JP2017000601A (en) | 2017-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101118654B1 (en) | rehabilitation device using motion analysis based on motion capture and method thereof | |
| CN107080540B (en) | System and method for analyzing gait and postural balance of a person | |
| Prakash et al. | Passive marker based optical system for gait kinematics for lower extremity | |
| JP6706809B2 (en) | Cognitive function evaluation device, cognitive function evaluation system, cognitive function evaluation method, and program | |
| JP2011027724A5 (en) | ||
| Zhu et al. | A computer vision-based system for stride length estimation using a mobile phone camera | |
| WO2016031313A1 (en) | Physical condition-detecting apparatus, physical condition-detecting method, and physical condition-detecting program | |
| JP6829988B2 (en) | Momentum estimation device, momentum estimation program, and momentum estimation system | |
| Yagi et al. | Gait measurement at home using a single RGB camera | |
| KR101725654B1 (en) | Respiratory monitoring system and method thereof | |
| CN112438724A (en) | Fall risk assessment method, fall risk assessment device, and recording medium | |
| Morita et al. | Comparative analysis of gait speed estimation using wideband and narrowband radars, thermal camera, and motion tracking suit technologies | |
| Wang et al. | An estimation of pedestrian action on footbridges using computer vision approaches | |
| CN112438723A (en) | Cognitive function evaluation method, cognitive function evaluation device, and storage medium | |
| JP6750193B2 (en) | Walking cycle detection method and detection apparatus | |
| EP2136548A3 (en) | Image processing apparatus, image processing method, and program | |
| Ruiz-Ruiz et al. | Evaluation of gait parameter estimation accuracy: a comparison between commercial IMU and optical capture motion system | |
| Pavone et al. | A Kinect-Based Portable Automatic Gait Analysis System-An Experimental Validation | |
| TWI736148B (en) | Posture detecting system and method thereof | |
| JP6578874B2 (en) | Walking cycle detection method and detection apparatus | |
| JP6482102B2 (en) | Measuring apparatus and measuring method | |
| KR20140142452A (en) | Scale that measures the area of human | |
| JP2015112215A (en) | Office space monitoring system | |
| EP3438601A1 (en) | Measurement device, measurement method, and computer readable recording medium | |
| JP6869205B2 (en) | Biological information detection device, program and method for detecting the beating of a living body |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A621 | Written request for application examination |
Free format text: JAPANESE INTERMEDIATE CODE: A621 Effective date: 20180302 |
|
| A977 | Report on retrieval |
Free format text: JAPANESE INTERMEDIATE CODE: A971007 Effective date: 20190125 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20190205 |
|
| A521 | Request for written amendment filed |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20190403 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20191001 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20200421 |
|
| A521 | Request for written amendment filed |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20200616 |
|
| TRDD | Decision of grant or rejection written | ||
| A01 | Written decision to grant a patent or to grant a registration (utility model) |
Free format text: JAPANESE INTERMEDIATE CODE: A01 Effective date: 20200714 |
|
| A61 | First payment of annual fees (during grant procedure) |
Free format text: JAPANESE INTERMEDIATE CODE: A61 Effective date: 20200727 |
|
| R151 | Written notification of patent or utility model registration |
Ref document number: 6750193 Country of ref document: JP Free format text: JAPANESE INTERMEDIATE CODE: R151 |
|
| R250 | Receipt of annual fees |
Free format text: JAPANESE INTERMEDIATE CODE: R250 |