CN106175778B - A kind of method that establishing gait data collection and gait analysis method - Google Patents
A kind of method that establishing gait data collection and gait analysis method Download PDFInfo
- Publication number
- CN106175778B CN106175778B CN201610517381.XA CN201610517381A CN106175778B CN 106175778 B CN106175778 B CN 106175778B CN 201610517381 A CN201610517381 A CN 201610517381A CN 106175778 B CN106175778 B CN 106175778B
- Authority
- CN
- China
- Prior art keywords
- gait
- data
- establishing
- parameters
- data set
- 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
- 230000005021 gait Effects 0.000 title claims abstract description 206
- 238000000034 method Methods 0.000 title claims abstract description 68
- 238000004458 analytical method Methods 0.000 title claims abstract description 37
- 238000013480 data collection Methods 0.000 title abstract description 24
- 238000012360 testing method Methods 0.000 claims abstract description 37
- 230000008569 process Effects 0.000 claims abstract description 33
- 238000013507 mapping Methods 0.000 claims description 14
- 241000282414 Homo sapiens Species 0.000 claims description 8
- 229920000742 Cotton Polymers 0.000 claims description 4
- 206010017577 Gait disturbance Diseases 0.000 claims description 4
- 230000006399 behavior Effects 0.000 claims description 4
- 239000010985 leather Substances 0.000 claims description 3
- 239000000843 powder Substances 0.000 claims description 3
- 239000000049 pigment Substances 0.000 claims description 2
- 239000004575 stone Substances 0.000 claims 1
- 210000002268 wool Anatomy 0.000 claims 1
- 238000011160 research Methods 0.000 abstract description 9
- 210000001699 lower leg Anatomy 0.000 abstract description 2
- 210000002683 foot Anatomy 0.000 description 28
- 210000003423 ankle Anatomy 0.000 description 11
- 238000010586 diagram Methods 0.000 description 9
- 238000012549 training Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 6
- 238000005259 measurement Methods 0.000 description 4
- 201000010099 disease Diseases 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 238000009434 installation Methods 0.000 description 3
- 210000002414 leg Anatomy 0.000 description 3
- 239000011435 rock Substances 0.000 description 3
- 239000002689 soil Substances 0.000 description 3
- 239000002023 wood Substances 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 230000005236 sound signal Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 206010008190 Cerebrovascular accident Diseases 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 235000019738 Limestone Nutrition 0.000 description 1
- 206010049214 Monopodia Diseases 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 208000006011 Stroke Diseases 0.000 description 1
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000037396 body weight Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 235000013312 flour Nutrition 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000002513 implantation Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000006028 limestone Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000001272 neurogenic effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Measuring devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb
- A61B5/112—Gait analysis
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
- A61B5/6804—Garments; Clothes
- A61B5/6807—Footwear
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6813—Specially adapted to be attached to a specific body part
- A61B5/6829—Foot or ankle
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physiology (AREA)
- Dentistry (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present invention provides a kind of method for establishing gait data collection, include: that tester's straight line in test zone 1) is allowed to be walked, the wearable sensors for being fixed on tester's left foot and right crus of diaphragm is used to acquire the sensing data of current walking process sample as the gait data of the sample；The wearable sensors include inertial sensor and sonic transducer；2) gait parameter of current walking process sample is obtained using the footprint left in video camera and/or the test zone preset in the test zone；3) gait data that corresponding sample is marked with the gait parameter that step 2) is obtained, thus the gait data collection after being marked.The present invention also provides the gait analysis methods accordingly based on labeled data collection.The present invention helps to improve the accuracy that accurate gait analysis is realized based on wearable sensors；Multi-modal data is merged, data set data volume obtained is big, classification is clear, facilitates analysis and research.
Description
Technical field
The present invention relates to gait research and analysis technical fields, specifically, the present invention relates to gait data acquisitions and survey
Measure technical field.
Background technique
The analysis and research of gait are the comprehensive studies to human motion function, including to human motion feature measurement,
The assessment of description and quantity.By the analysis and research to gait, can identify gait cycle, calculate gait kinematics and
Kinetic parameter etc..In recent years, the research of gait training, medical diagnosis on disease, rehabilitation medical, in terms of all rise
Very important effect and application are arrived.For example, in some trainings sportsman can be analyzed using gait analysis
Then some problems occurred in the training process help them to deduct a percentage achievement；In medical diagnosis on disease, sentence using gait analysis
Break some orthopaedics or neurogenic disease, such as apoplexy；In rehabilitation medical, controlling for patient can be guarded using gait analysis
More process；In biologic medical engineering, gait analysis has become the basic householder method of one kind to identify the motion feature of people；
In identification, different people is used as biometric identifier in the minor change of gait style to identify the people of individual.
In conclusion researcher will analyze and study since gait analysis and research have a wide range of applications,
With regard to needing a large amount of gait data.Most of the gait feature of disclosed available gait data collection is all based on both at home and abroad at present
Image.However, that the image shot in dynamic environment is illuminated by the light variation, the shadow of moving target, clothes block etc. is various
The influence of factor can bring larger difficulty to the Method of Gait Feature Extraction based on image.
On the other hand, M.Hofmann et al. was published in the name of J.Vis.Commun.Imag e.Represent in 2014
For " The TUM gait from audio, image and depth (GAID) database:Multimodal
In the article of recognition of subjects and traits ", discloses through Microsoft's Kinect tool, acquire people
Video, depth image and the footsteps of body gait are as feature, and the method for establishing data set.This scheme can pass through sound
Signal assists the Method of Gait Feature Extraction based on image, however, it still can not get rid of illumination variation, the shadow of moving target, clothing
Clothes such as block at the limitation of the image to dynamic environment shooting.Also, sound signal collecting is also required to preparatory installation sound in this scheme
Sound pick device simultaneously debugs voice pickup environment.Therefore, the obtained gait data collection of this method is difficult to apply to people's
In gait data acquisition and analysis in daily life.
M.U.B.Altaf et al. was published in the entitled " Acoustic of IEEE Trans.Biomed.Eng in 2015
In the article of Gaits:Gait Analysis with Footstep Sounds ", disclose by preset 16 in the room
Microphone array acquisition footsteps is acquired gait data as feature.The process employs sound as gait feature,
The image data of acquisition gait is not needed, but to realize this gait data acquisition based on sound, needs preparatory installation sound
Sound pick device simultaneously debugs voice pickup environment.However, environment locating for the daily behavior of people is difficult to carry out such peace in advance
Dress and debugging, this means that the obtained gait data collection of this method is difficult to apply to the gait data to people's daily life
In acquisition and analysis.
In conclusion can currently support the gait data acquisition and analysis in the daily life to people there is an urgent need to one kind
Gait data collection construct solution.
Summary of the invention
Therefore, it is an object of the invention to overcome the defect of the above-mentioned prior art, a kind of day that can be supported to people is provided
Often the gait data acquisition in life and the gait data collection of analysis construct solution.
The present invention provides a kind of methods for establishing gait data collection, including the following steps:
1) tester's straight line in test zone is allowed to be walked, with the wearable sensors for being fixed on tester's left foot and right crus of diaphragm
Acquire gait data of the sensing data of current walking process sample as the sample；The wearable sensors include that inertia passes
Sensor and sonic transducer；
2) it is obtained and is worked as using the footprint left in video camera and/or the test zone preset in the test zone
The gait parameter of preceding walking process sample；The gait parameter includes: gait distance parameter and time parameter, the gait distance
Parameter include at least step width and step-length, the time parameter include: the walking time, gait cycle, the support phase, shaking peroid, cadence,
Leg speed and step number；
3) gait data that corresponding sample is marked with the gait parameter that step 2) is obtained, thus the gait after being marked
Data set.
Wherein, in the step 1), the inertial sensor includes accelerograph, gyroscope, one in geomagnetic sensor
Item is multinomial；The sonic transducer includes microphone or ultrasonic sensor.
Wherein, in the step 1), the wearable sensors are fixed on the outside of the upper of a shoe of tested person, front side or rear side
Or the sole of tested person.
Wherein, the gait parameter further include: step pitch, stride, single step time, support initial stage, support mid-term and support end
Phase.
Wherein, described to obtain gait parameter using video camera preset in the test zone in the step 2)
Method includes: that the view of tested person's walking process can be tracked using the video camera shooting being previously placed in the test zone
Frequently, then by analyzing every frame image, determine the initial time of various gait behaviors, so obtain in the gait parameter when
Between parameter, the time parameter include: the single step time, gait cycle, support initial stage, support mid-term, support latter stage, in shaking peroid
It is one or more.
Wherein, described to obtain gait parameter using video camera preset in the test zone in the step 2)
Method further include: the rectangular region on the ground being located on track route lays readily identified coordinate (such as net in advance
The readily identified coordinate of trellis), in tester's walking process, is shot and walked simultaneously with horizontal and vertical video camera
Journey, the readily identified coordinate being then based in captured picture identify each step in walking process on the ground
Coordinate position, and then obtain the distance parameter in gait parameter.
Wherein, in the step 2), electronics footprint that the footprint is measured including the use of pressure test plate.
Wherein, in the step 2), the method for obtaining gait parameter using the footprint left in the test zone
It include: to spread one layer of thin powder on test zone ground in advance, or smear in advance in tested person's sole and easily leave obvious footprint
Pigment so that tested person walking after leave footprints on test zone ground, then again by measure footprint obtain step pitch, walk
Width, step-length, stride and step number.
Wherein, the step 1) further include: the affiliated classification information of record tested person's walking process, the classification information
Including the classification to acquisition main body: name, gender, height, weight, age, normal person, abnormal gait people；To acquisition time
Classification: season, date, time；Classification to collecting location: indoor and outdoor；The classification of shoes is worn to acquisition main body: soft bottom movement
Shoes, hard bottom sport footwear, leather shoes；Classification to acquisition ground: wood floor, rock land face, hair/cotton carpet ground, soil face.
The present invention provides a kind of gait analysis methods based on labeled data collection, comprising:
Step 100: obtaining the labeled data collection of body gait data, wherein the gait data is to be fixed on tested person
Sensing data in the wearable sensors of foot walking process collected, it includes multiple walked that the labeled data, which is concentrated,
The gait data of journey sample and gait parameter corresponding to each walking process sample, the wearable sensors include that inertia passes
Sensor and sonic transducer；
Step 200: establishing the mapping model from gait data to gait parameter, this is reflected with the labeled data collection training
Penetrate model；
Step 300: acquiring tested person in real time using the inertial sensor and sonic transducer that are fixed on current tested person foot
Gait data, be then based on surveyed gait data, the current step of current tested person obtained based on the mapping model after training
State parameter.
Wherein, in the step 100, body gait data are obtained using the previously described method for establishing gait data collection
Labeled data collection.
Compared with the prior art, the advantages of the present invention are as follows:
1, the present invention can carry out dataset acquisition by wearable equipment, not need to pacify in test environment in advance
Dress acquisition equipment；
2, the present invention can accurately mark wearable gait data using a variety of mask methods, to obtain
The data set precisely marked is obtained, this helps to improve the accuracy that accurate gait analysis is realized based on wearable sensors；
3, the present invention acquires gait data using voice signal and inertial sensor, has merged multi-modal data, has been obtained
Data set data volume is big, classification is clear, facilitate analysis and research.
Detailed description of the invention
Embodiments of the present invention is further illustrated referring to the drawings, in which:
Fig. 1 is the schematic diagram according to an embodiment of the invention for establishing gait data set method；
Fig. 2 is that gait distance parameter according to an embodiment of the invention illustrates schematic diagram；
Fig. 3 is according to an embodiment of the invention for carrying out bowing for the high-definition camera photographed scene of time-labeling
Depending on schematic diagram；
Fig. 4 is the schematic diagram according to an embodiment of the invention being mounted in gait acquisition device in advance on shoes；With left foot
For shoes, from left to right, the position of gait acquisition device be located at the outside of shoes, front side, rear side, bottom before, in bottom, bottom
Behind portion；
Fig. 5 is the schematic diagram according to an embodiment of the invention being worn on gait acquisition device at ankle；With the right side
For foot, from left to right, gait acquisition device is worn on outside, rear side, the front side of ankle respectively；
Fig. 6 shows the schematic diagram of a scenario measured to gait parameter in one embodiment of the invention.
Specific embodiment
As it was noted above, acquisition equipment is generally required in preset region (usually room in existing gait analysis technology
It is interior) it installs and debugs, this causes the acquisition of gait data that can only also complete in this specific region, and therefore, it is difficult to normal to the day for human beings
Gait in life is analyzed.Inventor overcomes drawbacks described above, by wearable inertial sensor and sonic transducer (such as wheat
Gram wind or ultrasonic sensor) combination be introduced into gait analysis technology, to realize the analysis to the gait in people's daily life.
It elaborates with reference to the accompanying drawings and detailed description to the present invention.
According to one embodiment of present invention, a kind of step based on wearable sonic transducer and inertial sensor is provided
The construction method of state data set.Data set constructed by this method has merged multi-modal data, the number in data set obtained
According to measuring, big, classification is clear, facilitates analysis and research.At the same time, sonic transducer, inertial sensor etc. wearable is also overcomed
Some inherent shortcomings of body, to make it possible to carry out the gait in people's daily life analysis.
Currently, inertial sensor has been widely used in step counting technology, however, the gait ginseng that gait analysis needs to obtain
Number requires higher precision, and since the intrinsic cumulative errors problem of inertial sensor can not be eliminated inherently, it is existing
In technology also accurate gait parameter directly can not be calculated with inertial sensor data.In the present embodiment, introduces microphone and carry out
The acquisition of voice data, microphone is small and exquisite, cheap, is very suitable to that wearable related smart machine is cooperated to use.Exist in both feet
When walking, footsteps is than more visible and reliable, by the detection to footsteps, some times that can be walked with accurate judgement gait
Parameter (such as single step period, swings initial stage, zero-speed detection at walking period).In this way, the step sound number by combining microphone
According to analysis, can largely eliminate and improve conventional inertia sensor carry out gait analysis when introducing cumulative errors.
Fig. 1 shows the schematic diagram of the construction method of the gait data collection of the present embodiment, the building side of the gait data collection
Method includes the following steps:
Step 101: allowing tester's straight line in test zone to walk, acquired using wearable sensors and traditional gait
Device, synchronous acquisition wearable sensors data and traditional gait information.In this step, wearable sensors include microphone and
Inertial sensor, wherein inertial sensor includes accelerograph, gyroscope, geomagnetic sensor etc..
In one embodiment, it can be classified according to specific acquisition situation to sample collected.For example, being led to acquisition
Body classification: name, gender, height, weight, age, normal person, abnormal gait people；To acquisition time classify: season, the date, when
Between；Classify to collecting location: indoor and outdoor；Shoes classification: sport footwear (soft bottom), sport footwear (hard bottom), skin is worn to acquisition main body
Shoes；To the classification of acquisition ground: wood floor, rock land face, hair/cotton carpet ground, soil face.It can incite somebody to action on all or part of
Classification information label is stated on the sample of wearable sensors gait data collected, to work for subsequent gait analysis
It provides and preferably supports.
In one example, it is included at least for acquiring the acquisition device of gait data: microphone and inertial sensor (example
Such as, accelerograph, gyroscope, geomagnetic sensor etc.).Wherein, acquisition device wearing mode can be worn simultaneously for left and right both feet.
By using two gait data acquisition device nodes simultaneously on biped, the data of left and right foot are subjected to analysis fusion, it can be with
Obtain information more more accurate than monopodia measurement method.Specifically, can be by different location of the acquisition device in shoes (with reference to figure
Shown in 4, it is generally mounted in shoe lining in advance when producing shoes) or be worn at double-legged ankle (with reference to shown in Fig. 5).In the present embodiment,
Acquisition device is worn in the symmetric position of left and right foot.
As shown in Figure 4 and Figure 5, when carrying out gait data acquisition, acquisition node (device) can be worn on two respectively
On foot or left and right foot is worn by the shoes of pre- implantation acquisition node respectively.
In Fig. 4 and Fig. 5, a indicates that gait acquisition device, b indicate elastic bandage, for fixing acquisition device, and meanwhile it is advantageous
In the comfort level of user's wearing.For wearing mode as shown in Figure 4, after both feet are worn by, it can according to need and wear shoes
Tightly, move shoes not on foot.It, may be to normal walking when due to being worn on gait acquisition device on the inside of ankle
It affects, and then influences gait and collected gait parameter, therefore gait acquisition device can be worn on the outside of ankle,
Rear side and front side.
For wearing mode as shown in Figure 5, after both feet are worn by, adjustment elastic bandage can according to need, keep it tight
It fastens in foot, does not move.When gait acquisition device is mounted in shoes in advance, before acquisition device can be located at upper of a shoe
Side, outside, rear side and sole.
Before the experiment of acquisition body gait data starts, following preparation can be selectively performed:
S1-0a: recording the personal information of subject, can be with for example, name, gender, height, weight, the age, whether
Abnormal gait people is diagnosed as by regular medical institutions.
S1-0b: record collecting location: indoor and outdoor.Since the present invention is using the multisensor of fusion voice signal
The method of acquisition gait data, should be as far as possible when choosing testing location so in order to reduce the interference to data collected
Avoid the noisy place of environment.
S1-0c: record acquisition time: season, date, time (specific to when).It should be understood that in different seasons
Section more or less will affect body gait since human body dress is different, dresss up the factors such as different, body weight bearing is different.Other one
Different period in it, food and drink, daily life etc. are also possible to influence body gait.
S1-0e: record acquisition main body wears shoes: sport footwear (soft bottom), sport footwear (hard bottom), leather shoes.
S1-0f: record acquisition ground: wood floor, rock land face, hair/cotton carpet ground, soil face.
Step 102: tested person's row can be tracked using the high-definition camera shooting being previously placed in the test zone
The video for walking process perhaps walks left footprint by manually distinguishing or passing through Computer Automatic Recognition using tested person
Technology obtains the gait parameter of tested person's walking process sample from the video and footprint.In the present embodiment, gait parameter can be with
Including with one of Types Below or a variety of: when step pitch, step width, step-length, stride, step number, cadence, leg speed, walking distance, walking
Between, the single step time, gait cycle, support initial stage (the support initial stage including left and right foot), support the mid-term (branch including left and right foot
Support mid-term), support latter stage (the support latter stage including left and right foot), shaking peroid (shaking peroid including left and right foot).Fig. 2 shows
The schematic diagram of gait distance parameter in the present embodiment.Wherein, gait distance parameter includes: step pitch, step width, step-length, stride etc..
For distance parameter and step number, the left footprint that can be walked by tested person is measured.It can be in advance on test zone ground
On spread one layer thin powder (such as flour perhaps pulverized limestone) or smear the face for easily leaving obvious footprint in advance in tested person's sole
Then material obtains step pitch, step width, step-length, stride and step by measuring footprint again so that leaving footprints on test zone ground
Number.Certainly, the method for above-mentioned measurement distance parameter is not unique, in another embodiment, can also be using professional pressure
Test board measures distance parameter.And for time parameter, for example, cadence, leg speed, walking distance, the walking time, the single step time,
Gait cycle, support initial stage (the support initial stage including left and right foot), support mid-term (the support mid-term including left and right foot), support
Latter stage (the support latter stage including left and right foot), shaking peroid (shaking peroid including left and right foot) etc., high-definition camera can be used
Camera shooting by analyzing every frame image, and then determines the initial time of gait behavior, and then obtain above-mentioned time parameter.Another
In a embodiment, support initial stage, support mid-term and the support latter stage in above-mentioned time parameter can be replaced with the support phase.Support
Phase is the integrated support period that support finish time in latter stage is carved at the beginning of support initial stage.
In the test zone, it can arrange that horizontal and vertical high-definition camera (sometimes referred to simply as video camera) arrives in advance
Predetermined position enables and takes in gait processes complete both feet ankle and ankle or less all in high-definition camera.Fig. 3
Show the schematic top plan view of the high-definition camera photographed scene for carrying out time-labeling in the present embodiment.As shown in figure 3,
Wherein: v indicates vertical range of longitudinal video camera apart from walking walking line (i.e. Walking Route)；When l indicates acquisition gait parameter
Walking distance.The distance of v and l should make that step can be taken in the case where high-definition camera is without movement and rotation
Part below capable double-legged ankle complete in the process and ankle.Lateral camera is the static camera shooting shot along walking direction
Machine, as shown in Figure 3.
May be noted that if tested person's walking process exceeds the range of l shown in Fig. 3, longitudinal video camera may be difficult
With the walking information of accurate recording people.Therefore, in order to increase the range shot of longitudinal video camera, in another embodiment
In, longitudinal video camera can be mounted on the track parallel with walking direction, enable longitudinal video camera with tested person
Walking speed same speed follows the tested person mobile, so that the longitudinal direction video camera can take walking always
Part below complete double-legged ankle and ankle in the process.Also, machine identifies for ease of calculation, involved in walking route
Horizontal and vertical line (or coordinate) is drawn on the ground in region in advance, when drawing horizontal and vertical line (or coordinate), adjacent two
The distance between bar line should not be too large, in order to avoid later period mark is caused to introduce more multiple error.In this way, in tested person's gait processes,
Image is acquired by horizontal and vertical high-definition camera, the moment corresponding to each frame image is then marked and identifies the frame
Position in image in the coordinate-system of left and right foot on the ground, and then obtain accurate gait parameter.Fig. 6 is shown based on the reality
The schematic diagram of a scenario that gait parameter is measured for applying example, with reference to Fig. 6, specific gait parameter measurement process in the embodiment
As follows: drawing grid in the rectangular region of ground one being located in track route, (in Fig. 6, rectangular region long 5.0m is wide
0.8 meter.Coordinate system has been demarcated on the ground, has been shot with horizontal and vertical high-definition camera, the two high-definition cameras can be used for marking
Infuse gait time parameter and gait distance parameter), it is many that this Rectangular grid, which is by the straight cuts in length and breadth of different colours,
A small square (such as square of 5cm × 5cm) is equivalent to one very big readily identified to having drawn on the ground in this way
Coordinate, so that the worst error of measured estimation gait distance parameter be made to control within 2-3cm.People is in this rectangle net
It on lattice in the walking process of normal walking, is shot simultaneously using the high-definition camera of vertical and horizontal, the grid can be relied on
Auxiliary, the distance parameter as labeled data is easily obtained from captured image.Specifically, one can be determined on the ground
Coordinate origin can be clearly seen that each step of walking on the ground from the image (one by one) that high-definition camera is shot
Coordinate position, and then distance parameter can be marked out.
Further, table 1 gives the definition and some methods for obtaining gait parameter of various gait parameters.
Table 1
In conclusion in above-described embodiment, that arranges by the video capture device arranged in advance and in advance is easy to leave
The acquisition measure of footprint, the footprint for obtaining the video that can track tested person's walking process and being left after test, by artificial
Measurement or Computer Automatic Recognition technology, obtain the corresponding gait parameter of gait data sample collected.It is easy to leave foot
The acquisition measure of print can be interpreted broadly, such as the electronics in professional pressure test plate acquisition tested person's walking process can be used
Then footprint information obtains corresponding gait parameter based on electronic foot official seal breath again.
In step 102, the gait parameter obtained can be considered measured directly as a result, accuracy is high, therefore can be used for marking
The gait data (the resulting sensing data based on wearable sensors of step 101) is infused, therefore these gait parameters can also
Referred to as labeled data.
Step 103: after obtaining the corresponding gait parameter of sample data collected, with the gait parameter mark pair obtained
The wearable sensors gait data answered, thus the sample data (being known as storing data in Fig. 1) after being marked.
Above-mentioned steps 101~103 are repeated, can be obtained the labeled data of the sample based on wearable sensors data
Collection, including wearable sensors acquisition sample data and to the mark of the sample data, which includes a series of tables
Levy the gait parameter of gait information.In one example, these sample datas and the gait feature vector being made of gait parameter
(gait feature vector that can be described as sample data) corresponds, to facilitate inquiry.
In above-described embodiment, labeled data collection is acquired gait data using voice signal and inertial sensor, merges
Multi-modal data, data volume in data set obtained is big, classification is clear, facilitates analysis and research；Meanwhile use is a variety of
Mask method accurately marks gait data, can accurately be marked for data set.
Further, according to another embodiment of the present invention, a kind of gait based on above-mentioned labeled data collection is additionally provided
Analysis method, comprising:
Step 100: establishing the labeled data collection of the body gait data acquired based on a variety of wearable sensors, the data
Collection includes at least the sample data of wearable sensors acquisition and the mark to the sample data, which includes a series of characterizations
The gait parameter of gait information.In the present embodiment, a variety of wearable sensors include inertial sensor and sonic transducer,
When acquiring gait data, they are all deployed in the foot of tested person.
Step 200: establishing the mapping model from gait data to gait parameter, this is reflected with the labeled data collection training
Penetrate model.Mapping model in this step is a kind of plan based on multisensor (including inertial sensor and sonic transducer) fusion
Mapping model slightly.It specifically, as the gait data of the mapping model input data include exactly inertial sensor gait
Data harmony sensing data.Inventor is the study found that if detect gait parameter by simple inertial sensor data
When, more missing inspection can be led to due to walking habits such as the weight speeds of step of wearer；And simple dependence sonic transducer
Data are come when detecting gait parameter, meeting lead to more false retrieval due to the weight dressed, speed etc. of walking.Therefore the present embodiment
In, when establishing the mapping model from labeled data to gait parameter, in this model of training, uses and merge both sensings
The strategy of device data, such inertial sensor gait data harmony sensing data can supply a gap mutually.
Mapping model in this step can be BP neural network model or SVM supporting vector machine model.BP nerve net
Network can refer to document: Learning internal representations by back-propagating errors, DE
Rumelhart, GE Hinton, RJ Williams- " Nature " -1986；SVM support vector machines can refer to document:
P.H.Chen,C.J.Lin,and B.A tutorial onν-support vector machines,
Appl.Stoch.Models.Bus.Ind.2005,21,111-136.。
Step 300: acquiring the step of tested person in real time using the inertial sensor and sonic transducer that are deployed in tested person foot
State data are then based on surveyed gait data, obtain the current gait parameter of tested person based on the mapping model after training.
Above-mentioned gait analysis scheme can carry out gait data acquisition by wearable equipment, not need surveying in advance
Installation acquisition equipment in test ring border makes to carry out the gait in people's daily life to widen the application field of gait analysis
Acquisition and analysis are possibly realized.Also, preliminary test show the mapping model based on convergence strategy compare single type sensing
The mapping model of device has higher accurate rate and recall rate.Wherein, accurate rate can be improved about 10%, and recall rate can be improved about
10%.
It should be noted last that the above examples are only used to illustrate the technical scheme of the present invention and are not limiting.On although
The invention is described in detail with reference to an embodiment for text, those skilled in the art should understand that, to skill of the invention
Art scheme is modified or replaced equivalently, and without departure from the spirit and scope of technical solution of the present invention, should all be covered at this
In the scope of the claims of invention.
Claims (9)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610517381.XA CN106175778B (en) | 2016-07-04 | 2016-07-04 | A kind of method that establishing gait data collection and gait analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610517381.XA CN106175778B (en) | 2016-07-04 | 2016-07-04 | A kind of method that establishing gait data collection and gait analysis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106175778A CN106175778A (en) | 2016-12-07 |
| CN106175778B true CN106175778B (en) | 2019-02-01 |
Family
ID=57465776
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610517381.XA Active CN106175778B (en) | 2016-07-04 | 2016-07-04 | A kind of method that establishing gait data collection and gait analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106175778B (en) |
Families Citing this family (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107016346A (en) * | 2017-03-09 | 2017-08-04 | 中国科学院计算技术研究所 | gait identification method and system |
| CN109222329B (en) * | 2017-04-12 | 2021-08-03 | 纵联汽车工业工程研究(天津)有限公司 | Walking length calculating method and intelligent insole configured with same |
| CN107170466B (en) * | 2017-04-14 | 2020-12-29 | 中国科学院计算技术研究所 | Audio-based mopping sound detection method |
| CN107403143A (en) * | 2017-07-06 | 2017-11-28 | 广东小天才科技有限公司 | Gait recognition method and electronic equipment |
| CN107696921B (en) * | 2017-08-30 | 2020-06-12 | 重庆延锋安道拓汽车部件系统有限公司 | Automobile seat adjusting system and control method thereof |
| CN108030497B (en) * | 2018-01-16 | 2023-12-19 | 大连乾函科技有限公司 | Gait analysis device and method based on IMU inertial sensor |
| CN109009147A (en) * | 2018-08-10 | 2018-12-18 | 柳州高华机械有限公司 | The terraced detection system of intelligent rehabilitation training and detection method |
| CN109009148A (en) * | 2018-08-24 | 2018-12-18 | 广东工业大学 | A kind of gait function appraisal procedure |
| WO2020077482A1 (en) * | 2018-10-15 | 2020-04-23 | 王长贵 | Composite ceramic tile and gait detection system having same |
| CN109528212B (en) * | 2018-12-29 | 2023-09-19 | 大连乾函科技有限公司 | Abnormal gait recognition equipment and method |
| CN110123330A (en) * | 2019-05-20 | 2019-08-16 | 浙江大学 | Foot type parameter measurement and pressure cloud atlas generation method based on plantar pressure data |
| CN110236550B (en) * | 2019-05-30 | 2020-07-10 | 清华大学 | Human gait prediction device based on multi-mode deep learning |
| CN110210392B (en) * | 2019-05-31 | 2022-12-09 | 吉林化工学院 | A Gait Recognition Device Based on Probability Theory |
| CN111028374B (en) * | 2019-10-30 | 2021-09-21 | 中科南京人工智能创新研究院 | Attendance machine and attendance system based on gait recognition |
| CN110916970B (en) * | 2019-11-18 | 2021-09-21 | 南京伟思医疗科技股份有限公司 | Device and method for realizing cooperative motion of weight-reducing vehicle and lower limb robot through communication |
| CN111079651A (en) * | 2019-12-17 | 2020-04-28 | 河南水滴智能技术有限公司 | Re-optimization of gait classification method |
| CN111178338A (en) * | 2020-03-18 | 2020-05-19 | 福建中医药大学 | A method for establishing a database and a standardized model in a gait analysis system |
| TWI737237B (en) * | 2020-03-25 | 2021-08-21 | 國泰醫療財團法人國泰綜合醫院 | Measuring system for measuring foot's inertial information |
| CN112472531A (en) * | 2020-12-17 | 2021-03-12 | 大连理工大学 | Gait smoothing algorithm of lower limb exoskeleton robot for medical rehabilitation and assisted walking |
| CN114052724B (en) * | 2022-01-13 | 2022-09-09 | 西安交通大学医学院第一附属医院 | Orthopedics traction abnormity detection system based on artificial intelligence |
| CN114267088B (en) * | 2022-03-02 | 2022-06-07 | 北京中科睿医信息科技有限公司 | Gait information processing method and device and electronic equipment |
| CN118470790A (en) * | 2024-04-30 | 2024-08-09 | 南京特殊教育师范学院 | Intelligent gait analysis and posture correction optimization method and system |
| CN118370534B (en) * | 2024-06-21 | 2024-09-17 | 浙江大学 | Gait analysis method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6836744B1 (en) * | 2000-08-18 | 2004-12-28 | Fareid A. Asphahani | Portable system for analyzing human gait |
| CN102243687A (en) * | 2011-04-22 | 2011-11-16 | 安徽寰智信息科技股份有限公司 | Physical education teaching auxiliary system based on motion identification technology and implementation method of physical education teaching auxiliary system |
| CN104729507A (en) * | 2015-04-13 | 2015-06-24 | 大连理工大学 | Gait recognition method based on inertial sensor |
| US20150257679A1 (en) * | 2011-03-24 | 2015-09-17 | MedHab, LLC | System and method for monitoring a runner's gait |
-
2016
- 2016-07-04 CN CN201610517381.XA patent/CN106175778B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6836744B1 (en) * | 2000-08-18 | 2004-12-28 | Fareid A. Asphahani | Portable system for analyzing human gait |
| US20150257679A1 (en) * | 2011-03-24 | 2015-09-17 | MedHab, LLC | System and method for monitoring a runner's gait |
| CN102243687A (en) * | 2011-04-22 | 2011-11-16 | 安徽寰智信息科技股份有限公司 | Physical education teaching auxiliary system based on motion identification technology and implementation method of physical education teaching auxiliary system |
| CN104729507A (en) * | 2015-04-13 | 2015-06-24 | 大连理工大学 | Gait recognition method based on inertial sensor |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106175778A (en) | 2016-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106175778B (en) | A kind of method that establishing gait data collection and gait analysis method | |
| US11033205B2 (en) | System and method for analyzing gait and postural balance of a person | |
| US20180350084A1 (en) | Techniques for object tracking | |
| Ghasemzadeh et al. | Coordination analysis of human movements with body sensor networks: A signal processing model to evaluate baseball swings | |
| CN103099602B (en) | Based on the physical examinations method and system of optical identification | |
| CN107578019B (en) | A gait recognition system and recognition method based on visual and tactile fusion | |
| US20170243057A1 (en) | System and method for automatic gait cycle segmentation | |
| Barth et al. | Subsequence dynamic time warping as a method for robust step segmentation using gyroscope signals of daily life activities | |
| JP6444813B2 (en) | Analysis system and analysis method | |
| US20180325467A1 (en) | Gait motion display system and program | |
| Ingwersen et al. | Sportspose-a dynamic 3d sports pose dataset | |
| Anwary et al. | Validity and consistency of concurrent extraction of gait features using inertial measurement units and motion capture system | |
| CN106846372B (en) | Human motion quality visual analysis and evaluation system and method thereof | |
| CN112438723B (en) | Cognitive function evaluation method, cognitive function evaluation device and storage medium | |
| Labuguen et al. | Performance evaluation of markerless 3D skeleton pose estimates with pop dance motion sequence | |
| TW202402231A (en) | Intelligent gait analyzer | |
| CN117529278A (en) | Gait monitoring method and robot | |
| CN113768471A (en) | An auxiliary diagnosis system for Parkinson's disease based on gait analysis | |
| Cai et al. | Single-camera-based method for step length symmetry measurement in unconstrained elderly home monitoring | |
| CN104688237B (en) | The time study method and system of physical examinations | |
| Tarek et al. | Yoga Trainer for Beginners Via Machine Learning | |
| WO2019025729A1 (en) | Analysis of a movement and/or of a posture of at least a portion of the body of a person | |
| Wang et al. | Gait analysis and validation using voxel data | |
| Nouredanesh et al. | Chasing feet in the wild: a proposed egocentric motion-aware gait assessment tool | |
| CN104331705B (en) | Automatic detection method for gait cycle through fusion of spatiotemporal information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20161207 Assignee: Beijing Zhongke Huicheng Technology Co., Ltd. Assignor: Institute of Computing Technology, Chinese Academy of Sciences Contract record no.: 2018110000005 Denomination of invention: Method for establishing gait data set and gait analysis method License type: Common License Record date: 20180222 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: Beijing Zhongke Huicheng Technology Co., Ltd. Assignor: Institute of Computing Technology, Chinese Academy of Sciences Contract record no.: 2018110000005 Date of cancellation: 20180309 |
|
| EC01 | Cancellation of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20161207 Assignee: Luoyang Zhongke Huicheng Technology Co., Ltd. Assignor: Institute of Computing Technology, Chinese Academy of Sciences Contract record no.: 2018110000009 Denomination of invention: Method for establishing gait data set and gait analysis method License type: Common License Record date: 20180319 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |