阅读说明：本技术 一种实时的基于足底压力的步态分析方法 (Real-time gait analysis method based on plantar pressure ) 是由 杨翠微 何凯悦 刘森 于 2019-08-28 设计创作，主要内容包括：本发明涉及一种实时的基于足底压力的步态分析方法,具体步骤为：运用可穿戴鞋垫获得人体活动状态下的足底压力数据；基于动态足底压力信号的幅值阈值与时间阈值的综合比较,实现对各独立活动状态的提取、识别,同时计算步态的运动学参量,对活动状态进行评估分析；并最终对连续的活动状态做统计整理。本发明可实现基于可穿戴设备对人体长程活动信息的逐点实时提取、分析,可应用于步态识别、日常健康检测以及相关疾病的评估等领域。(the invention relates to a real-time gait analysis method based on plantar pressure, which comprises the following specific steps: obtaining plantar pressure data of a human body in an active state by using the wearable insole; based on the comprehensive comparison of the amplitude threshold value and the time threshold value of the dynamic plantar pressure signal, the extraction and the identification of each independent activity state are realized, meanwhile, the kinematic parameters of gait are calculated, and the activity states are evaluated and analyzed; and finally, performing statistical arrangement on the continuous activity states. The wearable device-based long-range activity information extraction and analysis method can realize point-by-point real-time extraction and analysis of the long-range activity information of the human body based on the wearable device, and can be applied to the fields of gait recognition, daily health detection, evaluation of related diseases and the like.)

1. A real-time gait analysis method based on plantar pressure signals is characterized by comprising the following specific steps:

(1) Initializing parameters:

Two thresholds for the plantar pressure signal of each subject were set based on subject-to-subject variability: a sole amplitude threshold th1 and a heel amplitude threshold th 2; setting the judgment time threshold of the sitting state as t _ sit and the corresponding count parameter count _ sit as 0; setting a judgment time threshold value of the station state as t _ std, and setting a corresponding counting parameter count _ std as 0; setting a judgment time threshold value of a standing state and a sitting state as t _ st and a judgment time threshold value of a walking state as t _ wk during state arrangement;

(2) Reading pressure signals of double feet or single foot of a human body under various moving states point by point:

The pressure signal is obtained by using a pressure sensor arranged on the sole of a foot in the insole at a certain sampling frequencyf Collected pressure data; the pressure sensors of the sole are generally positioned at the sole and the heel, and a plurality of pressure sensors can be respectively arranged at the sole and the heel so as to obtain multi-path pressure data;The number of the pressure sensors arranged at the sole of the foot is not less than 2, and the number of the pressure sensors arranged at the heel of the foot is not less than 1;

(3) Performing independent activity state identification and state evaluation on the pressure data acquired in the step (2):

Detecting whether at least one path of pressure data has a rising edge, namely judging whether the pressure amplitude at the sole is greater than an amplitude threshold th1 at the sole or whether the pressure amplitude at the heel is greater than an amplitude threshold th2 at the heel; if the rising edge is not detected in each path of pressure data, the judgment process of the sole powerless state (sitting) is carried out; if detecting that at least one path has a rising edge, adaptively selecting a window length value wlen of a sliding window, executing a judgment process of a sole forceful state (standing, walking, running and going up (down) stairs) and extracting corresponding characteristic parameters;

The window length value wlen of the sliding window needs to be adjusted according to the pressure condition at the heel: when the pressure peak value at the heel is larger than the amplitude threshold value th2 at the heel, setting the length of the observation window to be wlen 1; otherwise, considering that the high-frequency motion state, such as running, needs a smaller analysis window length, then selecting the observation window length as wlen2, where: wlen2< wlen 1;

wherein: the judgment of the state of sole weakness (sitting) comprises the following steps:

Counting data points before the rising edge detected by each path of pressure data, namely adding 1 to count _ sit until the rising edge is detected;

② judging stepwhether count _ sit after the end of counting is larger thanf * t _ sit; if the time interval is larger than the set time threshold t _ sit, the time interval is marked as a sitting state, the starting point and the ending point of the sitting state are recorded, and the counting parameter count _ sit is cleared;

The judgment of the sole forceful state comprises the following contents:

firstly, setting the initial length of a data segment to be analyzed through a self-adaptive sliding window length value wlen, namely setting the initial length of the data segment to be analyzed to be wlen1 when a pressure peak value at a heel is larger than an amplitude threshold value th2 at the heel; otherwise wlen 2;

respectively extracting characteristic parameters including step frequency, peak value, landing time delay, effective pressure duration and corresponding start and stop time from each path of pressure data;

judging whether the falling edge of the pressure data curve is detected in the current observation window, namely the pressure value at the sole is smaller than the amplitude threshold th1 at the sole or the pressure value at the heel is smaller than the amplitude threshold th2 at the heel; if so, identifying and evaluating each dynamic motion state (walking, running and going up (down) stairs); if not, judging and marking the station state;

(4) determining whether a state classification and evaluation process based on the pressure data continues; if yes, returning to the step (2); if not, the independent activity states are subjected to statistical arrangement, and corresponding transition states are marked.

2. the method of claim 1, wherein: in the judgment of the sole forceful state, the step frequency in the second step is the reciprocal of the time difference between two adjacent landings, wherein the landing time is defined as the minimum value of the peak time of each path of pressure data; the peak value is the maximum value of the pressure relative amplitude in each path of pressure data curve under the current sliding window; the landing time delay is the time difference between the pressure peak moment at the heel and the pressure peak moment at the sole; the effective pressure duration and the start-stop time are the time period and the start-stop time of each pressure curve in the range of the respectively set amplitude threshold value under the current observation window.

3. The method of claim 2, wherein: in the judgment of the sole forceful state, the identification and evaluation method for each dynamic motion state (walking, running and going up (down) stairs) in the step (c) respectively comprises the following steps: if the landing time delay is positive and small, when the step frequency is large, marking as Running (RUN); when the step frequency is small, marking as an upper (lower) Stair (Stair); if the landing time delay is positive and larger, marking as normal Walking (WALK); if the time delay of landing is negative, marking that the tiptoe lands before the abnormal walking (Abnorm) of the heel; if the pressure value at the sole or heel is lower under the limit condition of the given amplitude threshold (th 1 and th 2), marking the pressure value as a sole force application abnormal state; if the absolute value of the difference of the peak values of the two paths of pressure in the sole is larger than a set value, marking the balance condition of the sole application force as unbalance (balance: No), otherwise, marking the balance condition as balance (balance: Yes);

the station state judging and marking method comprises the following steps: will valuef assigning a counting parameter count _ std of a station state to the wlan, and calculating the duration of the pressure greater than a threshold value, namely, reading the pressure sensor data once, and if the falling edge is not detected, adding 1 to the count _ std; when count _ std is greater thanfwhen t _ std, marking the station state and clearing the counter count _ std; otherwise, once the falling edge is detected, the counter count _ std is cleared, the step (2) is updated, and the analysis of subsequent data is carried out.

4. the method of claim 1, wherein: the selection standard of the pressure peak moment at the sole in the step (3) is as follows: because the number of the sensors arranged at the sole is not less than 2, when the pressure peak value of each path of pressure sensor at the sole exists, the minimum value of each path of corresponding time is taken; when the pressure peak value of only one path of pressure sensor exists, selecting the corresponding moment as the moment of the pressure peak value at the sole; and if the peak value of each pressure sensor does not exist, taking the terminal point time of the current observation window as the pressure peak value time of the sole.

5. the method of claim 1, wherein: and (4) the abnormal sole force application state in the step (3) comprises two conditions of sole force application deficiency (toe-less) and heel force application deficiency (heel-less).

6. When the maximum value of each signal pressure peak value at the sole is less than th1 which is 2 times, the mark is toe-less; when the maximum value of the signal pressure peak at the heel is less than 1.5 times th2, it is marked as heel-less.

7. The method of claim 1, wherein: the statistical arrangement of the states in the step (4) refers to the statistical marking of each independent activity state, and the specific method is that the sitting (Sit) state and the standing (stand) state with the duration time longer than t _ st are respectively marked as Sitting (SIT) and Standing (STD); uniformly marking the walk, run and upstairs (downstairs) activities with the state duration greater than or equal to t wk as Walk (WLK); and according to the following steps: the transition sections among the corresponding states are marked by the definitions of WLK-TURN-WLK, WLK-WLKSIT-SIT, WLK-WALKSTD-STD, SIT-SITWLK-WLK, SIT-SITSTD-STD, STD-STDWLK-WLK and STD-STDSIT-SIT; wherein WLK-TURN-WLK represents that when the two sections of continuous states are marked as WLK (go), the transition state in the period is defined as TURN (TURN); WLK-WLKSIT-SIT represents that when the states of the front section and the rear section are WLK (walking) and SIT (sitting), respectively, the transition state in the period is defined as WLKSIT (walking to sitting); and so on; in addition, the unrecognizable state is UnKnown.

Technical Field

The invention relates to a real-time gait analysis method based on plantar pressure, in particular to a gait analysis and evaluation method based on wearable equipment.

Background

the dynamic information of the sole has high correlation with the physiological structure and the walking mode of the human body, is the intrinsic mechanical reflection of the movement, and has uniqueness and relative stability. The acquisition and analysis of the gait pressure information are beneficial to finding the detailed essence of human body movement, and have important significance for the evaluation of relevant illness states, such as small steps of Parkinson patients, frozen gait, hemiplegic gait of stroke patients and the like; meanwhile, the method can also be applied to the fields of gait recognition (identity recognition) and the like.

At present, wearable devices such as multi-view camera image sequence acquisition, nuclear magnetic resonance, three-dimensional force measuring platforms, pressure footpaths and intelligent insoles are mainly used as auxiliary means for gait analysis. The method has the defects of high manufacturing cost, complex test process or high operation requirement by means of a camera, a force measuring platform, nuclear magnetic resonance and the like, so that the popular test evaluation cannot be carried out. In contrast, monitoring the daily activity state of the patient and providing quantitative analysis results based on the wearable device can provide more objective basis for diagnosis of doctors. Meanwhile, scientific and effective guidance can be provided for the rehabilitation of the patient in real time.

the intelligent insole adopting the pressure sensor is more suitable for gait analysis in the daily activity state of the human body. However, the existing gait analysis based on the pressure sensor has the defects of insufficient utilization of pressure information; the state classification process is relatively complicated; the problems of interpretation and analysis and the like cannot be performed on other detailed information of the activity state, so that effective assessment of the activity state is difficult to achieve and the requirement of real-time performance is difficult to meet.

Disclosure of Invention

In view of the above problems, the present invention is directed to a real-time gait analysis method based on plantar pressure. The method can realize the classification and evaluation of different activity states of the human body by analyzing the dynamic change of the plantar pressure in real time.

The invention provides a real-time gait analysis method based on plantar pressure signals, which comprises the following specific steps:

(1) initializing parameters:

Two thresholds for the plantar pressure signal of each subject were set based on subject-to-subject variability: a sole amplitude threshold th1 and a heel amplitude threshold th 2; setting the judgment time threshold of the sitting state as t _ sit and the corresponding count parameter count _ sit as 0; setting a judgment time threshold value of the station state as t _ std, and setting a corresponding counting parameter count _ std as 0; setting a judgment time threshold value of a standing state and a sitting state as t _ st and a judgment time threshold value of a walking state as t _ wk during state arrangement;

(2) reading pressure signals of double feet or single foot of a human body under various moving states point by point:

the pressure signal is obtained by using a pressure sensor arranged on the sole of a foot in the insole at a certain sampling frequencyf collected pressure data; the pressure sensors of the sole are generally positioned at the sole and the heel, and a plurality of pressure sensors can be respectively arranged at the sole and the heel so as to obtain multi-path pressure data; the number of the pressure sensors arranged at the sole of the foot is not less than 2, and the number of the pressure sensors arranged at the heel of the foot is not less than 1;

(3) performing independent activity state identification and state evaluation on the pressure data acquired in the step (2):

detecting whether at least one path of pressure data has a rising edge, namely judging whether the pressure amplitude at the sole is greater than an amplitude threshold th1 at the sole or whether the pressure amplitude at the heel is greater than an amplitude threshold th2 at the heel; if the rising edge is not detected in each path of pressure data, the judgment process of the sole powerless state (sitting) is carried out; if detecting that at least one path has a rising edge, adaptively selecting a window length value wlen of a sliding window, executing a judgment process of a sole forceful state (standing, walking, running and going up (down) stairs) and extracting corresponding characteristic parameters;

The window length value wlen of the sliding window needs to be adjusted according to the pressure condition at the heel: when the pressure peak value at the heel is larger than the amplitude threshold value th2 at the heel, setting the length of the observation window to be wlen 1; otherwise, considering that the high-frequency motion state, such as running, needs a smaller analysis window length, then selecting the observation window length as wlen2, where: wlen2< wlen 1;

Wherein: the judgment of the state of sole weakness (sitting) comprises the following steps:

counting data points before the rising edge detected by each path of pressure data, namely adding 1 to count _ sit until the rising edge is detected;

② judging stepwhether count _ sit after the end of counting is larger thanf * t _ sit; if the time interval is larger than the set time threshold t _ sit, the time interval is marked as a sitting state, the starting point and the ending point of the sitting state are recorded, and the counting parameter count _ sit is cleared;

The judgment of the sole forceful state comprises the following contents:

firstly, setting the initial length of a data segment to be analyzed through a self-adaptive sliding window length value wlen, namely setting the initial length of the data segment to be analyzed to be wlen1 when a pressure peak value at a heel is larger than an amplitude threshold value th2 at the heel; otherwise wlen 2;

respectively extracting characteristic parameters including step frequency, peak value, landing time delay, effective pressure duration and corresponding start and stop time from each path of pressure data;

Judging whether the falling edge of the pressure data curve is detected in the current observation window, namely the pressure value at the sole is smaller than the amplitude threshold th1 at the sole or the pressure value at the heel is smaller than the amplitude threshold th2 at the heel; if so, identifying and evaluating each dynamic motion state (walking, running and going up (down) stairs); if not, judging and marking the station state;

(4) determining whether a state classification and evaluation process based on the pressure data continues; if yes, returning to the step (2); if not, the independent activity states are subjected to statistical arrangement, and corresponding transition states are marked.

In the invention, in the judgment of the sole forceful state, the step frequency in the step II is the reciprocal of the time difference between two adjacent landings, wherein the landing time is defined as the minimum value of the peak time of each path of pressure data; the peak value is the maximum value of the pressure relative amplitude in each path of pressure data curve under the current sliding window; the landing time delay is the time difference between the pressure peak moment at the heel and the pressure peak moment at the sole; the effective pressure duration and the start-stop time are the time period and the start-stop time of each pressure curve in the range of the respectively set amplitude threshold value under the current observation window.

in the determination of the sole forceful state, the method for identifying and evaluating each dynamic motion state (walking, running and going up (down) stairs) in the third step respectively comprises the following steps: if the landing time delay is positive and small, when the step frequency is large, marking as Running (RUN); when the step frequency is small, marking as an upper (lower) Stair (Stair); if the landing time delay is positive and larger, marking as normal Walking (WALK); if the time delay of landing is negative, marking that the tiptoe lands before the abnormal walking (Abnorm) of the heel; if the pressure value at the sole or heel is lower under the limit condition of the given amplitude threshold (th 1 and th 2), marking the pressure value as a sole force application abnormal state; if the absolute value of the difference of the peak values of the two paths of pressure in the sole is larger than a set value, marking the balance condition of the sole application force as unbalance (balance: No), otherwise, marking the balance condition as balance (balance: Yes);

the station state judging and marking method comprises the following steps: will valuef Assigning a counting parameter count _ std of a station state to the wlan, and calculating the duration of the pressure greater than a threshold value, namely, reading the pressure sensor data once, and if the falling edge is not detected, adding 1 to the count _ std; when count _ std is greater thanfwhen t _ std, marking the station state and clearing the counter count _ std; otherwise, once the falling edge is detected, the counter count _ std is cleared, the step (2) is updated, and the analysis of subsequent data is carried out.

In the invention, the selection standard of the pressure peak moment at the sole in the step (3) is as follows: because the number of the sensors arranged at the sole is not less than 2, when the pressure peak value of each path of pressure sensor at the sole exists, the minimum value of each path of corresponding time is taken; when the pressure peak value of only one path of pressure sensor exists, selecting the corresponding moment as the moment of the pressure peak value at the sole; and if the peak value of each pressure sensor does not exist, taking the terminal point time of the current observation window as the pressure peak value time of the sole.

in the invention, the abnormal sole force application state in the step (3) comprises two conditions of sole force application insufficiency (toe-less) and heel force application insufficiency (heel-less). When the maximum value of each signal pressure peak value at the sole is less than th1 which is 2 times, the mark is toe-less; when the maximum value of the signal pressure peak at the heel is less than 1.5 times th2, it is marked as heel-less.

in the invention, the statistical arrangement of the states in the step (4) refers to the statistical marking of each independent activity state, and the specific method is that the sitting (Sit) and standing (stand) states with the duration time longer than t _ st are respectively marked as Sitting (SIT) and Standing (STD); uniformly marking the walk, run and upstairs (downstairs) activities with the state duration greater than or equal to t wk as Walk (WLK); and according to the following steps: the transition sections among the corresponding states are marked by the definitions of WLK-TURN-WLK, WLK-WLKSIT-SIT, WLK-WALKSTD-STD, SIT-SITWLK-WLK, SIT-SITSTD-STD, STD-STDWLK-WLK and STD-STDSIT-SIT; wherein WLK-TURN-WLK represents that when the two sections of continuous states are marked as WLK (go), the transition state in the period is defined as TURN (TURN); WLK-WLKSIT-SIT represents that when the states of the front section and the rear section are WLK (walking) and SIT (sitting), respectively, the transition state in the period is defined as WLKSIT (walking to sitting); and so on; in addition, the unrecognizable state is UnKnown.

The invention has the following beneficial effects:

1. the processing algorithm is simple and convenient, can realize real-time data analysis, and can evaluate the activity state of the subject in time;

2. The wearable device can be used for extracting and analyzing the long-range activity information of the human body, and is expected to be applied to monitoring the daily activity state of the human body;

3. The characteristic parameters extracted by the method can be effectively used for evaluating the activity state, are suitable for the fields of gait analysis, identity recognition and the like, and can also assist the diagnosis, treatment and rehabilitation of related diseases.

Drawings

FIG. 1 is a general flow diagram of the present invention.

Fig. 2 is a schematic flow chart of "activity state classification and evaluation" in fig. 1.

FIG. 3 is a schematic diagram of the method of the present invention for extracting characteristic parameters of a single activity. The abscissa is time and the ordinate is the relative amplitude of plantar pressure. The three curves in the window are respectively the pressure curves corresponding to two points of the sole and one point of the heel. (t, V) represents the coordinates of the peak point of each curve, t being the time and V being the pressure amplitude. th1 and th2 are ball pressure amplitude thresholds and heel pressure amplitude thresholds, respectively. Hw1, Hw2, and Hw3 represent the effective pressure duration for two ball and one heel pressure data, respectively.

FIG. 4 is a diagram illustrating the evaluation result of the independent walking state based on the adaptive sliding window according to the present invention. The abscissa is time and the ordinate is the relative amplitude of plantar pressure. Each dark rectangular area in the figure corresponds to an observation window, three curves in the window are dynamic pressure change curves of two sole parts and one heel part respectively, wherein the curve with a higher peak value reflects the dynamic pressure change at the heel part, and the other two curves correspond to the two pressure change at the sole part respectively. In the figure, WALK is the record of the independent walking state; the balance is used for indicating the balance condition of the current walking state; delay is the time delay between the sole and the heel of the foot; toe-less refers to the abnormal sole force application state of which the current walking state is insufficient sole force application; freqz refers to the step frequency.

Fig. 5 is a schematic diagram of the state arrangement result of the walking-to-sitting exercise process of the present invention. The abscissa is time and the ordinate is the relative amplitude of plantar pressure. Three curves in each period are pressure dynamic change curves of two sole parts and one heel part respectively, wherein a light-color curve with a relatively high peak value represents the pressure change at the heel part, and the other two curves are the pressure changes of two signals at the sole parts respectively; stand, Sit represent independent standing and sitting states, respectively, and the remaining symbols are labeled as in fig. 4. In fig. 5, the color segments with different gray levels correspond to different active state sorting results, and the relationship between gray level and state is shown in fig. 6.

Fig. 6 is a schematic view of gray marks corresponding to the statistical arrangement of the active states in embodiment 1 of the present invention, where 0 to 10 are numbers corresponding to the states.

Detailed Description

the method and the application of the invention will be described in further detail with reference to the accompanying drawings and examples. The examples are intended to be illustrative of the invention and are not intended to be limiting. On the basis of the technical scheme of the invention, various modifications or amendments to the above embodiments according to the principle of the invention should not be excluded from the protection scope of the invention.