CN112906457A - Walking gait signal preprocessing method based on mobile phone acceleration sensor - Google Patents
Walking gait signal preprocessing method based on mobile phone acceleration sensor Download PDFInfo
- Publication number
- CN112906457A CN112906457A CN202110012482.2A CN202110012482A CN112906457A CN 112906457 A CN112906457 A CN 112906457A CN 202110012482 A CN202110012482 A CN 202110012482A CN 112906457 A CN112906457 A CN 112906457A
- Authority
- CN
- China
- Prior art keywords
- smot
- mobile phone
- points
- walking gait
- data
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/03—Arrangements for converting the position or the displacement of a member into a coded form
- G06F3/033—Pointing devices displaced or positioned by the user, e.g. mice, trackballs, pens or joysticks; Accessories therefor
- G06F3/0346—Pointing devices displaced or positioned by the user, e.g. mice, trackballs, pens or joysticks; Accessories therefor with detection of the device orientation or free movement in a 3D space, e.g. 3D mice, 6-DOF [six degrees of freedom] pointers using gyroscopes, accelerometers or tilt-sensors
Abstract
The invention discloses a walking gait signal preprocessing method based on a mobile phone acceleration sensor, which processes abnormal signals and burrs in a Gaussian smooth iteration mode, extracts the change trend and the change rule of acceleration sensor signals in the swing period and the support period of walking gait by a trend traversal method, performs signal preprocessing such as walking gait cycle division, normalization and the like, can accurately identify a single stride cycle of continuous walking gait data, interpolates the continuous walking gait signal data into uniform length data in a segmented manner, and packages and stores the uniform length data. The method solves the problems that the walking gait signal is processed by an amplitude threshold value method and a window extreme value method in the prior art, the variation difference of parameters such as amplitude, period and the like is large, the selection of the threshold value and the window width is difficult, the selected unreasonable threshold value and the window width influence the gait signal period segmentation accuracy, and the data mining effect is poor.
Description
Technical Field
The invention belongs to the technical field of walking gait signal preprocessing, and particularly relates to a walking gait signal preprocessing method based on a mobile phone acceleration sensor.
Background
The gait recognition is to achieve the recognition of human identity through the analysis of human or animal gait, and becomes a new biological feature recognition technology at present with the advantages of non-contact remote distance, difficult camouflage, no invasion and the like.
Gait refers to a periodic form and appearance of the progression of a human or animal through limb movement, and is a very complex behavioral characteristic. At present, it is an important detection method to obtain gait characteristics through an acceleration sensor, and portable devices such as mobile phones and hand rings generally have built-in acceleration sensors, and are concerned about the advantages of high popularity, high frequency of use, no additional burden and the like. Before data mining such as gait recognition is carried out on signals of the mobile phone acceleration sensor, preprocessing needs to be carried out on original data so as to obtain a better processing effect.
The walking gait signal has periodicity, the waveform data in the swing phase and the support phase have amplitude difference, and when the gait signal is collected, instability of the mobile phone in the direction and abnormal fluctuation in the amplitude are easily caused due to factors such as shaking and bumping. An amplitude threshold value method and a window extreme value method are adopted for walking gait cycle segmentation based on acceleration sensor signals in the prior art, but an ideal assumption is made, in an actual environment, parameters such as amplitude and cycle of walking gait signals have large variation difference, proper threshold values and window widths are difficult to select, and unreasonable threshold values and window widths directly influence cycle segmentation accuracy and hinder data mining effect.
Disclosure of Invention
The invention aims to solve the defects and defects of difficult selection of threshold values or window widths in the prior art, and provides a walking gait signal preprocessing method which does not need to set threshold values, window widths and the like and needs preset values.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a walking gait signal preprocessing method based on a mobile phone acceleration sensor comprises the following steps:
the method comprises the following steps: installing an acceleration data acquisition APP on the mobile phone, and acquiring a mobile phone acceleration sensor signal through the APP; placing the mobile phone in a front thigh pocket on the right side of a human body, wherein the mobile phone is placed vertically upwards along the positive direction of the y axis, the positive direction of the x axis is on the right side of the transverse direction of the mobile phone, the screen of the mobile phone is far away from the human body, and the positive direction of the z axis is on the direction far away from the human body;
step two: calculating acceleration synthetic sequence of x, y, z three axes
Wherein a isx(t)、ay(t)、az(t) is the acceleration value of x, y and z axes at the time t;
step three: normalizing the acceleration synthesis sequence S (t) obtained by the calculation in the step two,
wherein, min [ S (t)]Is the minimum value in the sequence S (t), max [ S (t)]Is the maximum value in the sequence S (t);
step four: and (3) carrying out iteration processing on the normalized data Sstd (t) by using Gaussian smoothing, wherein the iteration frequency is 2 times, the smoothing coefficient is 0.9, and the processing result is recorded as Ssmot(t)；
Step five: traverse Ssmot(t) retaining turning points, eliminating points with the same trend, and recording the processing result as Ssmot' (t); let Ssmot(t) any adjacent 3 numerical points are Ssmot(i)、Ssmot(i-1)、Ssmot(i-2) if satisfied
[Ssmot(i)-Ssmot(i-1)]*[Ssmot(i-1)-Ssmot(i-2)]<0, then Ssmot(i-1) is the turning point, and the result is reserved and recorded as Ssmot' (t); if not, continuing traversing along the time sequence;
step six: traverse Ssmot' (t), keeping minimum pointsEliminating the maximum value point and recording the processing result as Sfit(t); let Ssmot' (t) any of the immediately adjacent 2 numerical points is Ssmot‘(t)、Ssmot' (t-1) if S is satisfiedsmot‘(t)-Ssmot‘(t-1)<0, then Ssmot' (t) is a minimum point, and the retention and record result is denoted as Sfit(t); if not, continuing traversing along the time sequence;
step seven: traverse Sfit(t), retaining the minimum value points and removing the maximum value points; sfit(t) the minimum value point is a single-cycle division point of the continuous walking gait, and the position serial number of the division point is recorded;
step eight: with Sfit(t) mapping the position sequence number of the storage period segmentation point to an original sequence S (t), cutting and segmenting original data S (t) by taking the segmentation point as a starting point, and storing and recording the original data S' (t);
step nine: calculating the data length of each segment of S' (t), and calculating the least common multiple of the lengths of all the data segments;
step ten: linearly interpolating each segmented data of S' (t) according to the least common multiple to obtain a walking gait monocycle sequence with uniform length to form a final data set,
wherein s isiAnd (n) represents the ith gait cycle sequence, and 1, 2.
The invention has the beneficial effects that:
according to the method, abnormal signals and burrs are processed in a Gaussian smooth iteration mode, the change trends and the change rules of acceleration sensor signals in the swing phase and the support phase of walking gait are extracted by a trend traversal method, signal preprocessing such as walking gait cycle segmentation and normalization is performed, the single stride cycle of continuous walking gait data can be accurately identified, the continuous walking gait signal data are segmented and interpolated into uniform length data, and the uniform length data are packaged and stored; the method solves the problems that the amplitude threshold value method and the window extreme value method in the prior art have large variation difference of parameters such as amplitude, period and the like, are difficult to select proper threshold values and window widths, and unreasonable threshold values and window widths influence the period segmentation accuracy and hinder the data mining effect.
Drawings
FIG. 1 is a schematic flow chart of a walking gait signal preprocessing method based on a mobile phone acceleration sensor according to the present invention;
FIG. 2 is a schematic view of the three-axis direction of a three-axis acceleration sensor in the mobile phone according to the method of the present invention;
FIG. 3 is a schematic diagram of the synthetic data of the acceleration synthetic sequence S (t) according to the present invention;
FIG. 4 is a comparison graph of the effect before and after Gaussian smoothing iteration processing (in the graph, a is an original data graph before the processing, b is a data graph after the Gaussian smoothing processing, c is a data graph after the first Gaussian smoothing iteration processing, and d is a data graph after the second Gaussian smoothing iteration processing);
FIG. 5 is a graph of all data points after Gaussian smoothing iteration processing according to the method of the present invention;
FIG. 6 is a traversal S of the method of the present inventionsmot(t) curves connected by turning points retained after the step (a);
FIG. 7 is a traversal S of the method of the present inventionsmotA graph formed by connecting minimum points retained after' (t);
FIG. 8 is a block diagram of the method of the present invention utilizing traversal Sfit(t) a fitted function plot of the minimum points retained after (t);
FIG. 9 is a schematic view of a walking gait cycle division point according to the method of the invention;
FIG. 10 is a periodic partition point map of the original data S (t) according to the present invention.
Detailed Description
In order to better explain the present invention, the detailed description of the present invention is made below with reference to the accompanying drawings and examples.
Example (b): see fig. 1-10.
As shown in fig. 1, a walking gait signal preprocessing method based on a mobile phone acceleration sensor includes the following steps:
the method comprises the following steps: installing an acceleration data acquisition APP on the mobile phone, and acquiring a mobile phone acceleration sensor signal through the APP; as shown in fig. 2, the mobile phone is placed in the front thigh pocket on the right side of the human body, the mobile phone is placed vertically upwards along the y-axis forward direction, the right side of the mobile phone in the transverse direction is the x-axis forward direction, the screen of the mobile phone is far away from the human body, and the direction far away from the human body is the z-axis forward direction;
step two: calculating acceleration synthetic sequence of x, y, z three axes
The resulting acceleration synthesis sequence S (t) is shown in FIG. 3, where ax(t)、ay(t)、az(t) is the acceleration value of x, y and z axes at the time t;
step three: normalizing the acceleration synthesis sequence S (t) obtained by the calculation in the step two,
wherein, min [ S (t)]Is the minimum value in the sequence S (t), max [ S (t)]Is the maximum value in the sequence S (t);
step four: as shown in fig. 4-5, the normalized data sstd (t) is processed by gaussian smoothing iteration, the number of iterations is 2, the smoothing coefficient is 0.9, and the processing result is recorded as Ssmot(t)；
Step five: traverse Ssmot(t) retaining turning points, eliminating points with the same trend, and recording the processing result as Ssmot' (t); let Ssmot(t) any adjacent 3 numerical points are Ssmot(i)、Ssmot(i-1)、Ssmot(i-2) if [ S ] is satisfiedsmot(i)-Ssmot(i-1)]*[Ssmot(i-1)-Ssmot(i-2)]<0, then Ssmot(i-1) is the turning point, and the result is reserved and recorded as Ssmot' (t); if not, continuing traversing along the time sequence, wherein the traversing result is shown in FIG. 6;
step six: traverse Ssmot' (t), retaining minimum value point, eliminating maximum value point, and recording the processing result as Sfit(t); let Ssmot' (t) any of the immediately adjacent 2 numerical points is Ssmot‘(t)、Ssmot' (t-1) if S is satisfiedsmot‘(t)-Ssmot‘(t-1)<0, then Ssmot' (t) is a minimum point, and the retention and record result is denoted as Sfit(t); if not, continuing traversing along the time sequence, wherein the traversing result is shown in FIG. 7; a fitting function curve graph of the minimum value points obtained by fitting the traversed data is shown in fig. 8;
step seven: traverse Sfit(t), retaining the minimum value points and removing the maximum value points; sfit(t) the minimum value point is a single-cycle division point of the continuous walking gait, and the serial number of the division point position is recorded, and the processing result is shown in figure 9;
step eight: with Sfit(t) mapping the original sequence S (t) by storing the position sequence numbers of the periodic segmentation points, cutting and segmenting the original data S (t) by taking the segmentation points as starting points, and storing and marking the segmented original data S (t) as S' (t), wherein the mapping result is shown in figure 10;
step nine: calculating the data length of each segment of S' (t), and calculating the least common multiple of the lengths of all the data segments;
step ten: linearly interpolating each segmented data of S' (t) according to the least common multiple to obtain a walking gait monocycle sequence with uniform length to form a final walking gait signal data set,
wherein s isiAnd (n) represents the ith gait cycle sequence, and 1, 2.
The above description is only for the preferred embodiment of the present invention and is not intended to limit the scope of the present invention, and all equivalent structures or equivalent transformations made by the present specification and the attached drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (4)
1. A walking gait signal preprocessing method based on a mobile phone acceleration sensor is characterized by comprising the following steps:
the method comprises the following steps: installing an acceleration data acquisition APP on the mobile phone, and acquiring a mobile phone acceleration sensor signal through the APP; placing the mobile phone in a front thigh pocket on the right side of a human body, wherein the mobile phone is placed vertically upwards along the positive direction of the y axis, the positive direction of the x axis is on the right side of the transverse direction of the mobile phone, the screen of the mobile phone is far away from the human body, and the positive direction of the z axis is on the direction far away from the human body;
step two: calculating acceleration synthetic sequence of x, y, z three axes
Wherein a isx(t)、ay(t)、az(t) is the acceleration value of x, y and z axes at the time t;
step three: normalizing the acceleration synthesis sequence S (t) obtained by the calculation in the step two,
wherein, min [ S (t)]Is the minimum value in the sequence S (t), max [ S (t)]Is the maximum value in the sequence S (t);
step four: and (3) carrying out iteration processing on the normalized data Sstd (t) by using Gaussian smoothing, wherein the iteration frequency is 2 times, the smoothing coefficient is 0.9, and the processing result is recorded as Ssmot(t)；
Step five: traverse Ssmot(t) retaining turning points, eliminating points with the same trend, and recording the processing result as Ssmot‘(t)；
Step six: traverse Ssmot' (t), retaining minimum value point, eliminating maximum value point, and recording the processing result as Sfit(t)；
Step seven: traverse Sfit(t), retaining the minimum value points, eliminating the maximum value points, and recording the position serial numbers of the minimum value points;
step eight: with the result of Sfit(t) mapping the position sequence number of the minimum value point to the original sequence S (t) by Sfit(t) taking the minimum value point as a starting point, cutting and segmenting the original data S (t), and storing and recording as S' (t);
step nine: calculating the data length of each segment of S' (t), and calculating the least common multiple of the data length of the obtained segments;
step ten: linearly interpolating the data of each segment S' (t) according to the obtained least common multiple to obtain a walking gait monocycle sequence with uniform length so as to form a rowThe final data set for the walking gait,
wherein s isiAnd (n) represents the ith gait cycle sequence, and 1, 2.
2. The walking gait signal preprocessing method based on the acceleration sensor of mobile phone according to claim 1, characterized in that in step five, the traversal Ssmot(t), the turning points are reserved, and the processing method for eliminating the same trend points comprises the following steps:
let Ssmot(t) any adjacent 3 numerical points are Ssmot(i)、Ssmot(i-1)、Ssmot(i-2) if [ S ] is satisfiedsmot(i)-Ssmot(i-1)]*[Ssmot(i-1)-Ssmot(i-2)]<0, then Ssmot(i-1) is the turning point, and the result is reserved and recorded as Ssmot' (t); if not, then the traversal continues in temporal order.
3. The walking gait signal preprocessing method based on the acceleration sensor of mobile phone according to claim 1, characterized in that the step six is traversed by Ssmot' (t), the minimum value point is reserved, and the processing method for eliminating the maximum value point comprises the following steps:
let Ssmot' (t) any of the immediately adjacent 2 numerical points is Ssmot‘(t)、Ssmot' (t-1) if S is satisfiedsmot‘(t)-Ssmot‘(t-1)<0, then Ssmot' (t) is a minimum point, and the retention and record result is denoted as Sfit(t); if not, then the traversal continues in temporal order.
4. The walking gait signal preprocessing method based on the acceleration sensor of mobile phone according to claim 1, characterized in that the step seven is traversed by Sfit(t), retaining the minimum value points, eliminating the maximum value points, and recording the position serial numbers of the minimum value points; said Sfit(t) minimum points areAnd the position serial number of the minimum value point is the position serial number of the single period division point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110012482.2A CN112906457B (en) | 2021-01-06 | 2021-01-06 | Walking gait signal preprocessing method based on mobile phone acceleration sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110012482.2A CN112906457B (en) | 2021-01-06 | 2021-01-06 | Walking gait signal preprocessing method based on mobile phone acceleration sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112906457A true CN112906457A (en) | 2021-06-04 |
| CN112906457B CN112906457B (en) | 2022-08-09 |
Family
ID=76112255
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110012482.2A Active CN112906457B (en) | 2021-01-06 | 2021-01-06 | Walking gait signal preprocessing method based on mobile phone acceleration sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112906457B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114287921A (en) * | 2021-12-23 | 2022-04-08 | 常州信息职业技术学院 | Gait bilateral similarity analysis method, device and system |
| CN116548928A (en) * | 2023-07-11 | 2023-08-08 | 西安浩阳志德医疗科技有限公司 | Nursing service system based on internet |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140156215A1 (en) * | 2012-12-04 | 2014-06-05 | Mapmyfitness, Inc. | Gait analysis system and method |
| CN107103297A (en) * | 2017-04-20 | 2017-08-29 | 武汉理工大学 | Gait identification method and system based on mobile phone acceleration sensor |
| CN107403143A (en) * | 2017-07-06 | 2017-11-28 | 广东小天才科技有限公司 | A kind of gait recognition method and electronic equipment |
| CN110537922A (en) * | 2019-09-09 | 2019-12-06 | 北京航空航天大学 | Human body walking process lower limb movement identification method and system based on deep learning |
| CN111568436A (en) * | 2020-05-29 | 2020-08-25 | 常州大学 | Gait symmetry evaluation method based on regression rotation angle |
| CN112107397A (en) * | 2020-10-19 | 2020-12-22 | 中国科学技术大学 | Myoelectric signal driven lower limb artificial limb continuous control system |
-
2021
- 2021-01-06 CN CN202110012482.2A patent/CN112906457B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140156215A1 (en) * | 2012-12-04 | 2014-06-05 | Mapmyfitness, Inc. | Gait analysis system and method |
| CN107103297A (en) * | 2017-04-20 | 2017-08-29 | 武汉理工大学 | Gait identification method and system based on mobile phone acceleration sensor |
| CN107403143A (en) * | 2017-07-06 | 2017-11-28 | 广东小天才科技有限公司 | A kind of gait recognition method and electronic equipment |
| CN110537922A (en) * | 2019-09-09 | 2019-12-06 | 北京航空航天大学 | Human body walking process lower limb movement identification method and system based on deep learning |
| CN111568436A (en) * | 2020-05-29 | 2020-08-25 | 常州大学 | Gait symmetry evaluation method based on regression rotation angle |
| CN112107397A (en) * | 2020-10-19 | 2020-12-22 | 中国科学技术大学 | Myoelectric signal driven lower limb artificial limb continuous control system |
Non-Patent Citations (1)
| Title |
|---|
| 杨文辉: "《基于加速度传感器的步态分析与识别》", 《中国优秀硕士学位论文全文数据库 (信息科技辑)》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114287921A (en) * | 2021-12-23 | 2022-04-08 | 常州信息职业技术学院 | Gait bilateral similarity analysis method, device and system |
| CN116548928A (en) * | 2023-07-11 | 2023-08-08 | 西安浩阳志德医疗科技有限公司 | Nursing service system based on internet |
| CN116548928B (en) * | 2023-07-11 | 2023-09-08 | 西安浩阳志德医疗科技有限公司 | Nursing service system based on internet |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112906457B (en) | 2022-08-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112906457B (en) | Walking gait signal preprocessing method based on mobile phone acceleration sensor | |
| CN103471589A (en) | Method for identifying walking mode and tracing track of pedestrian in room | |
| CN110602647A (en) | Indoor fusion positioning method based on extended Kalman filtering and particle filtering | |
| Lu et al. | QRS detection based on improved adaptive threshold | |
| CN108510011B (en) | User travel mode analysis method based on mobile phone multi-sensor | |
| CN111277946A (en) | Fingerprint database self-adaptive updating method in Bluetooth indoor positioning system | |
| CN107274376A (en) | A kind of workpiece three dimensional point cloud smooth filtering method | |
| CN109540143B (en) | Pedestrian unconventional action direction identification method based on multi-sensing-source dynamic peak fusion | |
| CN114897025A (en) | Human body posture recognition model establishing method and human body posture recognition method | |
| CN106772572A (en) | A kind of pick-up method of micro-seismic monitoring first arrival | |
| CN108462939B (en) | Indoor positioning method for geomagnetic time sequence analysis | |
| CN111938627B (en) | Method for filtering magnetocardiogram signal baseline drift | |
| CN109766798A (en) | Gesture data processing method, server and awareness apparatus based on experience small echo | |
| CN105512480B (en) | The data-optimized processing method of wearable device based on editing distance | |
| US11373367B2 (en) | Method for characterization of respiratory characteristics based on voxel model | |
| CN110954109A (en) | Indoor space positioning equipment and positioning method based on fingerprint positioning and inertial navigation system | |
| WO2022111203A1 (en) | Heart rate detection method and device | |
| CN110231035A (en) | Climb mobile robot path guide method | |
| CN115439934A (en) | Self-adaptive step frequency detection method based on CNN-LSTM motion mode identification | |
| CN116051573A (en) | Model vibration measurement drawing vectorization method based on deep element learning | |
| CN112861098B (en) | Mobile phone identity verification method based on DTW algorithm and walking gait data | |
| CN115200575A (en) | Indoor positioning method for assisting pedestrian dead reckoning based on human activity recognition | |
| Chandel et al. | Airite: Towards accurate & infrastructure-free 3-d tracking of smart devices | |
| US20220218230A1 (en) | System and method of detecting walking activity using waist-worn inertial sensors | |
| CN107462247A (en) | A kind of indoor orientation method, device and computer-readable recording medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |