CN108520242A - The method and apparatus that a kind of heel for Gait Recognition detects - Google Patents
The method and apparatus that a kind of heel for Gait Recognition detects Download PDFInfo
- Publication number
- CN108520242A CN108520242A CN201810322257.7A CN201810322257A CN108520242A CN 108520242 A CN108520242 A CN 108520242A CN 201810322257 A CN201810322257 A CN 201810322257A CN 108520242 A CN108520242 A CN 108520242A
- Authority
- CN
- China
- Prior art keywords
- heel
- rotation center
- collision
- radial acceleration
- acceleration
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
- G06V40/25—Recognition of walking or running movements, e.g. gait recognition
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The purpose of the application is to provide a kind of method and apparatus that the heel for Gait Recognition detects, and the application determines key frame and the position of the heel contact of the target user by the gait image sequence of acquisition target user in the gait image sequence；It is determined positioned at the target area of front foot according to the key frame, the radial acceleration of the target user is detected from the target area；The dense mapping of the corresponding rotation center in the position is determined according to the radial acceleration；Heel position of collision is determined according to the dense mapping of the rotation center and the radial acceleration；To accomplish real-time dynamic.
Description
Technical field
This application involves the method and apparatus that computer realm more particularly to a kind of heel for Gait Recognition detect.
Background technology
In some form of gait analysis, it is essential that the occurrence time of heel strike can be captured.Based on mark
The Gait Recognition of quasi- image has been widely studied.Currently, the target of most methods is to use gait as biological characteristic, thus
Identification individual.For example, background subtraction, feature extraction and classification.Method based on model has with human body and its movement
Close relationship, the method for model-free concentrate on the movement of the shape or entire gait process of body.Method based on model is
Visual angle and scale are changeless, but the cost is relatively high for calculating, and this method is very sensitive to picture quality.Model-free
Method is the alternative of the method based on model, but less sensitive for picture quality, and calculating cost is also higher, if visual angle
With the variation of scale, this method itself also and is not suitable for.The problem of heel movement is difficult to capture in gait analysis is still public
The problem opened.
Invention content
The purpose of the application is to provide a kind of method and apparatus that the heel for Gait Recognition detects, and solves current
The problem of being difficult to capture for heel movement in computation vision gait analysis.
According to the one side of the application, a kind of method that the heel for Gait Recognition detects, this method packet are provided
It includes：
The gait image sequence for obtaining target user determines the heel of the target user in the gait image sequence
The key frame to land and position；
It is determined positioned at the target area of front foot according to the key frame, the target user is detected from the target area
Radial acceleration；
The dense mapping of the corresponding rotation center in the position is determined according to the radial acceleration；
Heel position of collision is determined according to the dense mapping of the rotation center and the radial acceleration.
Further, the dense mapping of the corresponding rotation center in the position is determined according to the radial acceleration, including：
The video frame of the heel strikes of the target user is positioned according to the radial acceleration；
Using the point of the heel strikes as rotation center, determined in the rotation according to the video frame of the heel strikes
The dense mapping of the heart.
Further, heel position of collision is determined according to the dense mapping of the rotation center and the radial acceleration,
Including：
The density of the rotation center is determined according to the dense mapping；
The density of the rotation center and the radial acceleration are determined into the heel position of collision according to calculation formula,
Wherein, the calculation formula indicates as follows：
Wherein, (ih, jh) it is heel position of collision, I (i, j) is the density of the rotation center at position (i, j), and s is described
Quantity of the radial acceleration in target area.
Further, according to the dense mapping of the rotation center and the radial acceleration determine heel position of collision it
Afterwards, including：
By the data of the determining heel position of collision comprising multiple visual angles and walking direction in several benchmark databases
It is assessed, with the determining heel position of collision of detection.
Further, the method includes：
The key frame of heel contact point is detected according to the quantity of the radial acceleration in the target area；
According to the fixed heel position of collision of crucial filtering frames for the heel contact point that detection obtains.
According to the application another aspect, a kind of computer-readable medium is additionally provided, is stored thereon with computer-readable
Instruction, the computer-readable instruction can be executed by processor to realize the method as described in aforementioned.
According to the application another aspect, a kind of equipment that the heel for Gait Recognition detects is additionally provided, wherein institute
Stating equipment includes：
One or more processors；And
It is stored with the memory of computer-readable instruction, the computer-readable instruction makes the processor when executed
Execute the operation such as aforementioned the method.
Compared with prior art, the application by obtain target user gait image sequence, in the gait image sequence
Key frame and the position of the heel contact of the target user are determined in row；The target positioned at front foot is determined according to the key frame
The radial acceleration of the target user is detected in region from the target area；According to radial acceleration determination
The dense mapping of the corresponding rotation center in position；Foot is determined according to the dense mapping of the rotation center and the radial acceleration
With position of collision；Solve the problems, such as that heel movement is difficult to capture in vision gait analysis, compared with prior art, greatly improves
The accuracy of heel positioning, and reached approximate real time detection.
Description of the drawings
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon：
Fig. 1 shows the method flow detected according to a kind of heel for Gait Recognition that the one side of the application provides
Schematic diagram；
Fig. 2 shows the schematic diagrames of the key frame that heel contact is detected in image sequence of the embodiment in the application；
Fig. 3 shows the target area manikin ratio chart of the extraction front foot in the application；
Fig. 4 (a) and 4 (b) be shown respectively in the application for multiple test databases, more true heels of people's hand labeled
The error amount of position；
Fig. 5 shows the acceleration detection of the embodiment in the application to the knot for blocking immunity test in detection zone
Fruit schematic diagram；
Fig. 6 shows what the acceleration detection of the embodiment in the application tested the immunity of Gauss zero-mean white noise
Result schematic diagram.
Same or analogous reference numeral represents same or analogous component in attached drawing.
Specific implementation mode
The application is described in further detail below in conjunction with the accompanying drawings.
In a typical configuration of this application, terminal, the equipment of service network and trusted party include one or more
Processor (CPU), input/output interface, network interface and memory.
Memory may include computer-readable medium in volatile memory, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology realizes information storage.Information can be computer-readable instruction, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), fast flash memory bank or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storages, magnetic tape cassette, magnetic tape disk storage or other magnetic storage apparatus or
Any other non-transmission medium can be used for storage and can be accessed by a computing device information.As defined in this article, computer
Readable medium does not include non-temporary computer readable media (transitory media), such as data-signal and carrier wave of modulation.
Fig. 1 shows the method flow detected according to a kind of heel for Gait Recognition that the one side of the application provides
Schematic diagram, this method include：Step S11~step S14, wherein in step s 11, obtain the gait image sequence of target user
Row determine key frame and the position of the heel contact of the target user in the gait image sequence；In step s 12,
Determine that, positioned at the target area of front foot, the radial direction that the target user is detected from the target area adds according to the key frame
Speed；In step s 13, the dense mapping of the corresponding rotation center in the position is determined according to the radial acceleration；In step
In rapid S14, heel position of collision is determined according to the dense mapping of the rotation center and the radial acceleration.In the application
In, by the way that the key frame of heel contact is estimated in image sequence by heel contact detecting step, and can be according to walking body Model
Extraction is positioned at the target area (ROI) of front foot, and export the dense of rotation center reflects from the radial acceleration detected in ROI
It penetrates, to determine heel position of collision (ih, jh), by the acceleration detection of offer allow in gait analysis heel movement compared with
Good capture.The application is applicable to social safety management, is equally applicable to human bioequivalence and medical research etc..
Further, the diversity of different types of movement is considered in this application, some objects are moved with speed, are had
A little objects are with acceleration movement.In dynamics, instantaneous acceleration is derivative of the instantaneous velocity relative to the time, in order to estimate
The acceleration fields in image sequence are counted, the limited difference by considering the velocity field between frame and front and back frame is close in frame to formulate
Seemingly.A picture frame is provided in time, can estimate acceleration fields by the time-derivative of the light stream between successive frame：
A (t)=V (t+1)-(- V (t)), wherein V (t) and V (t+1) indicates the movement of frame t-1 to t and t to t+1 respectively
.
Radially and tangentially component is decomposed the acceleration into this application, to determine their influences to movement.It can manage
Solution, when object is moved along the path of bending, the size of speed and the direction of movement are all changing, tangential acceleration
Change the size of speed, direction increases or decreases speed on the tangent line of track.Radial acceleration (also referred to as centripetal acceleration)
Change mobile direction, and is directed toward the center of circular trace.In addition, if the time is short enough, the track of movement may be considered that
It is linear motion or circular motion.Therefore we assume that the object in image sequence in three continuous frames along identical
Path is moved, and this movement is linear either circular.Then, acceleration can be decomposed into tangential component and normal direction
(or radial) component, in order to determine the rotation center in circular motion, it is assumed that following the point of circular motion at any three
Along identical circle rotation in continuous frame.To radial component arWith tangential component atThey and resultant acceleration a can be passed through
Geometrical relationship estimate：
ar=sin θ * a
at=cos θ * a；
Wherein, θ indicates a and atBetween angle, value is equal to angle between resultant acceleration a and speed v, speed side
To along tangent line：
The rotation center of so radial acceleration is similar with circular motion：
ar=v2/r
R=v2/(sinθ*a)；
The position at center can subtract radius to export by the coordinate of currently processed pixel：
O=(x, y)-(rx, ry)
To provide data basis for detection acceleration, to be detected heel strikes with the detection method of acceleration, into
And solve the problems, such as that heel movement is difficult to capture in the computer vision gait analysis of present society safety management, and estimating
Need to only three frames be used by counting in the time masterplate of acceleration stream, accomplish approximate real time detection.
In step s 13, the video frame of the heel strikes of the target user is detected according to the radial acceleration；According to
The dense mapping of the rotation center of the key frame of the heel strikes.Further, the dense of the rotation center is reflected
It penetrates and the radial acceleration determines the heel position of collision according to calculation formula, wherein the calculation formula indicates as follows：
Wherein, (ih, jh) it is heel position of collision, I (i, j) is the density of the rotation center at position (i, j), and s is described
Quantity of the radial acceleration in target area.
Here, when heel will land, heel becomes to strike ground from swing forward, while foot is in heel and is
The circular motion at center.In the process, a large amount of radially accelerated stream is generated around foot.Cause by using by heel strike
Radial acceleration as prompt, key frame heelstrike and position can be estimated in image sequence.Heelstrike detect
The frame number of system is as shown in Figure 2, wherein abscissa is key frame, and ordinate is the quantity of the acceleration detected in the roi.
In heelstrike moment, foot is stepped on the ground, its speed is forced to stop in a short time.Therefore, because speed disappearance (quickly subtracts
Speed), the acceleration of front foot sharply increases.In addition, when heelstrike, which is approximate circular motion.Therefore, by foot
It is radial mostly with acceleration caused by shock.The video of heel strike can be positioned by the size of radial acceleration
Frame.When target user walks, the movement of body is similar to several pendulums, includes acceleration in the movement of pendulum, therefore foot
It may obscure with radial acceleration caused by other limbs with radial acceleration caused by shock.In order to reduce interference, according to row
Body Model extraction is walked positioned at the target area (ROI) of front foot.As shown in figure 3, the size of ROI is 0.133H × 0.177H, wherein H
Represent the height of objective body.When heel strike, front foot carries out circular motion centered on heel.If owned in ROI
Radial acceleration be all caused by heel strike, then the rotation center of all radial accelerations is all answered in the region
It should be located at heel.Heel-strike position (ih, jh) determined by following formula：
Wherein, (ih, jh) it is heel position of collision, I (i, j) is the density of the rotation center at position (i, j), and s is described
The value of radial acceleration.
Further, after step s 14, to being assessed comprising multiple visual angles and walking direction gait data library, with
Detect determining heel position of collision.Here, assessing above-mentioned steps S11~step S14 in several benchmark databases, wherein
Benchmark database can be CASIA, SOTON and OU-ISIR.The data used to acquire under different controlled environments,
Testing time can be 100 or so heel strikes in each scene, and test data include multiple visual angles and walking direction,
Test data is as shown in table 1：
Table 1
Wherein, 45 ° refer to target user's direction of travel with the face of camera shooting angle, 90 ° refer to target user's walking
The angle vertical in direction and the face of camera shooting.When being assessed using benchmark database, heelstrike error is in SOTON numbers
According to the variance for showing bigger on collection, because the picture size from the database is maximum, therefore larger detection faces manufacture
At higher error.
It is understood that acceleration decomposition algorithm based on the object moved perpendicular to background, therefore, divides in vision gait
In analysis, it is most effective perpendicular to the direction of camera, the gait data being imaged in multiple views is carried out in this application
The assessment of robustness.It specifically includes：
The key frame of heel contact point is detected according to the quantity of the radial acceleration in the target area；According to detecting
The fixed heel position of collision of crucial filtering frames of the heel contact point arrived.Here, according to the number of the radial acceleration in ROI
It measures and is used for detecting the key frame (instantaneous) heelstrike put, the histogram of radial acceleration can more obviously be shown in walking sequence
Show key frame, in the sequence, radial acceleration often occurs and notable, and can show the periodicity of gait.Institute in fig. 2
In the sequence shown, heel strike is happened between frame number 13,27,41 and frame number 54 and 55.Due to the people during gait cycle
The shape of body changes, therefore is not necessarily located in the sequence of front foot according to the ROI that gait ratio is extracted, in addition, in other bodies
Body region, such as shank, there is also radial accelerations, because the movement of limbs is the movement of the pendulum of several connections.These are wrong
The rotation center of radial acceleration accidentally can also form invalid followed by position of collision.In order to reduce the radially accelerated of these mistakes
The influence of degree, the key frame detected are used to the alternative heelstrike position of filtering.When heel is hit between two frame numbers
When, amount of acceleration is used as exporting the weighted factor of position.Fig. 4 (a) shows the couple candidate detection of heelstrike position, Fig. 4 (b)
It is the filtered result of key frame.
Further include performance detection, with actual value in addition to above-mentioned key frame detects when being assessed using benchmark database
It is compared, following methods is used due to not returning to key frame during Corner Detection, when so carrying out heel positioning：When described
When the numerical value of the heel actual value of target user is more than or equal to -30 pixel values and is less than or equal to 30 pixel values, corresponding foot
Corner location with place is the key frame.Corner location i.e. in actual value ± 30 pixel of heel contact is judged as
It is that key frame is successfully detected.For radial acceleration detector, the heel is judged according to the radial acceleration
Whether the frame where touchdown point is within 1, if so, the frame where heel contact point is to close with the frame difference apart from ground
Key frame.Whether the standard of real key frame is detection frame in ± 1 frame of actual value.
Further, in order to assess heel contact detection method to the immunity blocked, judge the gait in actual monitoring
Whether information can be become totally visible, and the target area of preset area is covered using random grain, with from the target area
Carry out blocking the filming apparatus when gait image sequence for obtaining target user in front of domain.Use random grain covering part target
Region is blocked with being realized from front, and acceleration as shown in Figure 5 is blocked to blocking immunity test schematic diagram in detection zone
Under performance decline steadily, when area of detection be more than entire area 30% when, judgement use heel contact described herein
The method of detection fails.
In one embodiment of the application, the performance of the acceleration detector to using is detected, including equal to Gauss zero
The detection that three factor immunitys are reduced with resolution ratio is blocked in value white noise, detection zone.As shown in fig. 6, acceleration is examined
The performance for surveying device slowly smoothly reduces with the increase of Gaussian noise variance, and increase of the Corner Detection to Gaussian noise variance
It is more sensitive.
The method that heel described herein for Gait Recognition detects, based on the thought for distinguishing type of sports, according to
Heelstrike detecting system estimates key frame heelstrike and position in image sequence, and according to running body model extraction
Positioned at the target area (ROI) of front foot, the dense mapping graph of rotation center is exported from the radial acceleration detected in ROI,
So that it is determined that heel position of collision.Solve the problems, such as that heel movement is difficult to capture in vision gait analysis.With conventional method phase
Than the detection method of the application has accomplished approximate real time dynamic.
According to the application another aspect, a kind of computer-readable medium is additionally provided, is stored thereon with computer-readable
Instruction, the computer-readable instruction can be executed by processor to realize the method as described in aforementioned.
According to the application another aspect, a kind of equipment that the heel for Gait Recognition detects is additionally provided, wherein institute
Stating equipment includes：
One or more processors；And
It is stored with the memory of computer-readable instruction, the computer-readable instruction makes the processor when executed
Execute the operation such as aforementioned the method；For example, the computer-readable instruction when executed execute the processor with
Lower step：
The gait image sequence for obtaining target user determines the heel of the target user in the gait image sequence
The key frame to land and position；
It is determined positioned at the target area of front foot according to the key frame, the target user is detected from the target area
Radial acceleration；
The dense mapping of the corresponding rotation center in the position is determined according to the radial acceleration；
Heel position of collision is determined according to the dense mapping of the rotation center and the radial acceleration.
Obviously, those skilled in the art can carry out the application essence of the various modification and variations without departing from the application
God and range.In this way, if these modifications and variations of the application belong to the range of the application claim and its equivalent technologies
Within, then the application is also intended to include these modifications and variations.
It should be noted that the application can be carried out in the assembly of software and/or software and hardware, for example, can adopt
With application-specific integrated circuit (ASIC), general purpose computer or any other realized similar to hardware device.In one embodiment
In, the software program of the application can be executed by processor to realize steps described above or function.Similarly, the application
Software program (including relevant data structure) can be stored in computer readable recording medium storing program for performing, for example, RAM memory,
Magnetic or optical driver or floppy disc and similar devices.In addition, hardware can be used to realize in some steps or function of the application, example
Such as, coordinate to execute the circuit of each step or function as with processor.
In addition, the part of the application can be applied to computer program product, such as computer program instructions, when its quilt
When computer executes, by the operation of the computer, it can call or provide according to the present processes and/or technical solution.
And the program instruction of the present processes is called, it is possibly stored in fixed or moveable recording medium, and/or pass through
Broadcast or the data flow in other signal loaded mediums and be transmitted, and/or be stored according to described program instruction operation
In the working storage of computer equipment.Here, including a device according to one embodiment of the application, which includes using
Memory in storage computer program instructions and processor for executing program instructions, wherein when the computer program refers to
When order is executed by the processor, method and/or skill of the device operation based on aforementioned multiple embodiments according to the application are triggered
Art scheme.
It is obvious to a person skilled in the art that the application is not limited to the details of above-mentioned exemplary embodiment, Er Qie
In the case of without departing substantially from spirit herein or essential characteristic, the application can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and scope of the present application is by appended power
Profit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent requirements of the claims
Variation is included in the application.Any reference signs in the claims should not be construed as limiting the involved claims.This
Outside, it is clear that one word of " comprising " is not excluded for other units or step, and odd number is not excluded for plural number.That is stated in device claim is multiple
Unit or device can also be realized by a unit or device by software or hardware.The first, the second equal words are used for table
Show title, and does not represent any particular order.
Claims (7)
1. a kind of method that heel for Gait Recognition detects, wherein the method includes：
The gait image sequence for obtaining target user determines the heel contact of the target user in the gait image sequence
Key frame and position；
It is determined positioned at the target area of front foot according to the key frame, the diameter of the target user is detected from the target area
To acceleration；
The dense mapping of the corresponding rotation center in the position is determined according to the radial acceleration；
Heel position of collision is determined according to the dense mapping of the rotation center and the radial acceleration.
2. according to the method described in claim 1, wherein, being determined in the corresponding rotation in the position according to the radial acceleration
The dense mapping of the heart, including：
The video frame of the heel strikes of the target user is positioned according to the radial acceleration；
Using the point of the heel strikes as rotation center, the rotation center is determined according to the video frame of the heel strikes
Dense mapping.
3. according to the method described in claim 2, wherein, according to the dense mapping of the rotation center and the radial acceleration
Determine heel position of collision, including：
The density of the rotation center and the radial acceleration are determined into the heel position of collision according to calculation formula,
In, the calculation formula indicates as follows：
Wherein, (ih, jh) it is heel position of collision, I (i, j) is the density of the rotation center at position (i, j), and s is the radial direction
Quantity of the acceleration in target area.
4. according to the method described in claim 1, wherein, according to the dense mapping of the rotation center and the radial acceleration
After determining heel position of collision, including：
The data of the determining heel position of collision comprising multiple visual angles and walking direction are carried out in several benchmark databases
Assessment, with the determining heel position of collision of detection.
5. according to the method described in claim 1, wherein, the method includes：
The key frame of heel contact point is detected according to the quantity of the radial acceleration in the target area；
According to the fixed heel position of collision of crucial filtering frames for the heel contact point that detection obtains.
6. a kind of computer-readable medium, is stored thereon with computer-readable instruction, the computer-readable instruction can be handled
Device is executed to realize the method as described in any one of claim 1 to 5.
7. the equipment that a kind of heel for Gait Recognition detects, wherein the equipment includes：
One or more processors；And
It is stored with the memory of computer-readable instruction, the computer-readable instruction when executed executes the processor
Such as the operation of any one of claim 1 to 5 the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810322257.7A CN108520242A (en) | 2018-04-11 | 2018-04-11 | The method and apparatus that a kind of heel for Gait Recognition detects |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810322257.7A CN108520242A (en) | 2018-04-11 | 2018-04-11 | The method and apparatus that a kind of heel for Gait Recognition detects |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108520242A true CN108520242A (en) | 2018-09-11 |
Family
ID=63432178
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810322257.7A Pending CN108520242A (en) | 2018-04-11 | 2018-04-11 | The method and apparatus that a kind of heel for Gait Recognition detects |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108520242A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111317480A (en) * | 2020-01-20 | 2020-06-23 | 深圳市丞辉威世智能科技有限公司 | Gait recognition method, device, equipment and readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101548925A (en) * | 2009-05-05 | 2009-10-07 | 浙江大学 | Gait phase detection apparatus with ankle joint angle self-rectification function |
| CN104757976A (en) * | 2015-04-16 | 2015-07-08 | 大连理工大学 | Human gait analyzing method and system based on multi-sensor fusion |
| CN105188530A (en) * | 2013-05-10 | 2015-12-23 | 欧姆龙健康医疗事业株式会社 | Walking posture meter and program |
| CN106821391A (en) * | 2017-03-23 | 2017-06-13 | 北京精密机电控制设备研究所 | Body gait acquisition analysis system and method based on inertial sensor information fusion |
| CN107153461A (en) * | 2017-05-08 | 2017-09-12 | 联想(北京)有限公司 | A kind of method and motion control device for controlling to show the motion mode of object |
-
2018
- 2018-04-11 CN CN201810322257.7A patent/CN108520242A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101548925A (en) * | 2009-05-05 | 2009-10-07 | 浙江大学 | Gait phase detection apparatus with ankle joint angle self-rectification function |
| CN105188530A (en) * | 2013-05-10 | 2015-12-23 | 欧姆龙健康医疗事业株式会社 | Walking posture meter and program |
| CN104757976A (en) * | 2015-04-16 | 2015-07-08 | 大连理工大学 | Human gait analyzing method and system based on multi-sensor fusion |
| CN106821391A (en) * | 2017-03-23 | 2017-06-13 | 北京精密机电控制设备研究所 | Body gait acquisition analysis system and method based on inertial sensor information fusion |
| CN107153461A (en) * | 2017-05-08 | 2017-09-12 | 联想(北京)有限公司 | A kind of method and motion control device for controlling to show the motion mode of object |
Non-Patent Citations (1)
| Title |
|---|
| YAN SUN ET AL: "Detecting heel strikes for gait analysis through acceleration flow", 《IET COMPUTER VISION》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111317480A (en) * | 2020-01-20 | 2020-06-23 | 深圳市丞辉威世智能科技有限公司 | Gait recognition method, device, equipment and readable storage medium |
| CN111317480B (en) * | 2020-01-20 | 2023-05-09 | 深圳市丞辉威世智能科技有限公司 | Gait recognition method, device, equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10506990B2 (en) | Devices and methods for fall detection based on phase segmentation | |
| US9530221B2 (en) | Context aware moving object detection | |
| US20160061582A1 (en) | Scale estimating method using smart device and gravity data | |
| Derpanis et al. | Action spotting and recognition based on a spatiotemporal orientation analysis | |
| US8929588B2 (en) | Object tracking | |
| CN105940430A (en) | Person counting method and device for same | |
| CN111091025B (en) | Image processing method, device and equipment | |
| CN110033472A (en) | A kind of stable objects tracking under the infrared ground environment of complexity | |
| JP2018096709A (en) | Distance measurement device and distance measurement method | |
| US9990857B2 (en) | Method and system for visual pedometry | |
| Min et al. | Human fall detection using normalized shape aspect ratio | |
| Tang et al. | Hybrid blob and particle filter tracking approach for robust object tracking | |
| CN108520242A (en) | The method and apparatus that a kind of heel for Gait Recognition detects | |
| Li et al. | Study on slam algorithm based on object detection in dynamic scene | |
| Shimizu et al. | LIDAR-based body orientation estimation by integrating shape and motion information | |
| CN104346606B (en) | abnormal gait analysis method and system | |
| CN108363974B (en) | Foot root detection method for computer vision gait recognition for social security management | |
| JP5032415B2 (en) | Motion estimation apparatus and program | |
| Madden et al. | A framework for track matching across disjoint cameras using robust shape and appearance features | |
| CN108647589B (en) | Tumble detection method based on regularization morphological ratio | |
| CN113160218A (en) | Method for detecting object motion intensity based on event camera | |
| Marouane et al. | Step and activity detection based on the orientation and scale attributes of the SURF algorithm | |
| Paik et al. | Improving object detection, multi-object tracking, and re-identification for disaster response drones | |
| Sun | Detecting Heel Strikes for Gait Analysis Through Higher-Order Motion Flow | |
| CN114359328B (en) | Motion parameter measuring method utilizing single-depth camera and human body constraint |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180911 |