CN111166341A - Tumble identification method based on acceleration impact energy clustering and wearable system - Google Patents

Tumble identification method based on acceleration impact energy clustering and wearable system Download PDF

Info

Publication number
CN111166341A
CN111166341A CN202010011004.5A CN202010011004A CN111166341A CN 111166341 A CN111166341 A CN 111166341A CN 202010011004 A CN202010011004 A CN 202010011004A CN 111166341 A CN111166341 A CN 111166341A
Authority
CN
China
Prior art keywords
acceleration
clustering
teager
formula
impact energy
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
Application number
CN202010011004.5A
Other languages
Chinese (zh)
Inventor
李勇
邢改兰
周邵萍
郑浩然
陈浩
管衍栋
资瑞卿
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
East China University of Science and Technology
Original Assignee
East China University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by East China University of Science and Technology filed Critical East China University of Science and Technology
Priority to CN202010011004.5A priority Critical patent/CN111166341A/en
Publication of CN111166341A publication Critical patent/CN111166341A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording 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, mobility of a limb
    • A61B5/1116—Determining posture transitions
    • A61B5/1117—Fall detection
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording 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, mobility of a limb
    • A61B5/1126—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb using a particular sensing technique
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/74—Details of notification to user or communication with user or patient ; user input means
    • A61B5/746—Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
    • G01S19/01—Satellite radio beacon positioning systems transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
    • G01S19/13—Receivers
    • G01S19/14—Receivers specially adapted for specific applications
    • G01S19/17—Emergency applications

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Veterinary Medicine (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Surgery (AREA)
  • Molecular Biology (AREA)
  • Data Mining & Analysis (AREA)
  • Biomedical Technology (AREA)
  • Pathology (AREA)
  • Biophysics (AREA)
  • Physiology (AREA)
  • Theoretical Computer Science (AREA)
  • Dentistry (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • General Physics & Mathematics (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Emergency Alarm Devices (AREA)

Abstract

The invention relates to a tumble identification method based on acceleration impact energy clustering and a wearable system, wherein the tumble identification method comprises the following steps: 1) acquiring three-way acceleration and positioning information of an identified object in real time; 2) fusing the three-direction acceleration to obtain corresponding Teager local energy and random impact characteristics; 3) recording two-dimensional scattered points consisting of Teager local energy and random impact characteristics in a set time period, and clustering the two-dimensional scattered points; 4) and acquiring the current action of the identification object based on the clustering result, and generating an alarm when judging that the current action is violent. Compared with the prior art, the method has the advantages of high accuracy and the like.

Description

Tumble identification method based on acceleration impact energy clustering and wearable system
Technical Field
The invention belongs to the technical field of wearable electronic equipment and dynamic motion signal processing, relates to a tumble identification method and device, and particularly relates to a tumble identification method based on acceleration impact energy clustering and a wearable system.
Background
With the improvement of living and medical level, the daily real-time monitoring of children and the elderly has become a major problem for kindergarten, children hospital, nursing home and other institutions. The existing fall monitoring system mainly comprises mainstream methods such as image video, environment variable and wearable method. Due to the limitations of cost, monitoring area, privacy, etc., a miniature wearable device based on three-way acceleration monitoring is more acceptable. However, in each link of monitoring implementation, the accurate identification of the characteristic information reflecting the motion state of the subject in the acceleration signal is the key for successful implementation of monitoring. In the prior art, the judgment is triggered through a threshold value on the basis of the vector amplitude of the characteristic direction acceleration and the three-direction acceleration. Although the method is easy to implement, the method is greatly influenced by noise, and the problem of high false alarm rate is easily caused.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a tumble identification method based on acceleration impact energy clustering and a wearable system.
The purpose of the invention can be realized by the following technical scheme:
a tumble identification method based on acceleration impact energy clustering comprises the following steps:
1) acquiring three-way acceleration and positioning information of an identified object in real time;
2) fusing the three-direction acceleration to obtain corresponding Teager local energy and random impact characteristics;
3) recording two-dimensional scattered points consisting of Teager local energy and random impact characteristics in a set time period, and clustering the two-dimensional scattered points;
4) and acquiring the current action of the identification object based on the clustering result, and generating an alarm when judging that the current action is violent.
Further, the three-direction acceleration fusion specifically includes:
Figure BDA0002357159240000021
in the formula, ax(t)、ay(t)、azAnd (t) are three-way acceleration information respectively, and a (t) is fusion information.
Further, the Teager local energy acquisition formula is as follows:
Figure BDA0002357159240000022
in the formula, a (T) is fusion information, and T (T) is Teager local energy.
Further, the random impact characteristic obtaining formula is as follows:
D(t)＝w(t)*a(t)
in the formula, w (t) is an inverse filter, a (t) is fusion information, and represents convolution.
Further, the random impact characteristics are obtained through an iterative optimization mode, and the method specifically comprises the following steps:
101) initializing the inverse filter w(0)(t) where all elements are 1, let i equal 1;
102) calculating D (t) ═ w(i-1)(t)*a(t)；
103) Computing
Figure BDA0002357159240000023
Wherein
Figure BDA0002357159240000024
l is lag time, N total time length;
104) calculating w(i)＝A-1b(i)Wherein the matrix A is an autocorrelation matrix of the sequence a (t);
105) if it is not
Figure BDA0002357159240000025
less than a given threshold ξ, the iteration is stopped, otherwise i is incremented by 1 and the procedure returns to step 102).
Further, in the step 3), an objective function adopted in clustering is as follows:
Figure BDA0002357159240000026
wherein m is the cluster number of the cluster, i represents the ith sample, N represents the total number of samples, j represents the jth class, C represents the total number of classes,
Figure BDA0002357159240000027
represents a sample piDegree of membership belonging to class j, cjRepresents the cluster center of class j, | x | is a distance measure.
Further, the update formula of the membership degree is as follows:
Figure BDA0002357159240000028
the update formula of the cluster center is as follows:
Figure BDA0002357159240000031
further, the generating an alarm includes generating a buzzer alarm signal.
Further, the generating of the alarm comprises issuing a danger signal and identifying a position signal of the object.
The invention also provides a wearable system for realizing the tumble identification method based on the acceleration impact energy clustering, which comprises wearable equipment and a computer end, wherein the wearable equipment realizes the step 1), and the computer end realizes the steps 2) to 4).
Compared with the prior art, the invention has the following beneficial effects:
1) the invention realizes the classification judgment of the actions through clustering processing, does not need to set a threshold value, is easy to implement, has small noise influence and high accuracy of the monitoring result.
2) The invention carries out fusion processing on the three-way acceleration information, can better reflect the actual situation of a wearer, reduces the false alarm rate, realizes the precise analysis of the whole action, and realizes the precise identification of the falling state and other violent actions.
Drawings
FIG. 1 is a schematic view of a fall recognition process according to the present invention;
FIG. 2 is a schematic structural diagram of a wearable system of the present invention;
FIG. 3 is a X, Y, Z three-way acceleration signal;
FIG. 4 is a fused acceleration signal;
FIG. 5 is a Teager local energy index T (T);
FIG. 6 is a random impact signature D (t);
FIG. 7 is a discrete scatter plot;
FIG. 8 is a schematic diagram of clustering results;
FIG. 9 is the local energy returned to Teager for the third type of scatter;
FIG. 10 is a third type of scatter return and random impact signature.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
Example 1
As shown in fig. 1, the present embodiment provides a tumble identification method based on acceleration impact energy clustering, including the following steps:
in step S1, three-directional acceleration and positioning information of the recognition object are acquired in real time. Fig. 3 shows the acceleration raw signals of X, Y, Z obtained by the test in three directions.
In step S2, the three-directional acceleration is precisely analyzed, specifically including information fusion and obtaining corresponding Teager local energy and random impact characteristics.
The three-direction acceleration fusion is specifically as follows:
Figure BDA0002357159240000041
in the formula, ax(t)、ay(t)、az(t) are three-way acceleration information, a (t) is fusion information, and fig. 4 is an acceleration signal after fusion.
The Teager local energy acquisition formula is as follows:
Figure BDA0002357159240000042
in the formula, a (T) is fusion information, and T (T) is Teager local energy. FIG. 5 is a calculated Teager local energy index.
The random impact characteristic obtaining formula is as follows:
Figure BDA0002357159240000043
in the formula, w (t) is an inverse filter, a (t) is fusion information, represents convolution, and L is the length of the inverse filter.
The random impact characteristics are obtained in an iterative optimization mode, and the optimal random impact index D (t) is obtained through optimization updating of an inverse filter, and the method specifically comprises the following steps:
101) initializing the inverse filter w(0)(t) where all elements are 1, let i equal 1;
102) calculating D (t) ═ w(i-1)(t)*a(t)；
103) Computing
Figure BDA0002357159240000044
Wherein
Figure BDA0002357159240000045
l is lag time, N total time length;
104) calculating w(i)＝A-1b(i)Wherein the matrix A is an autocorrelation matrix of the sequence a (t);
105) if it is not
Figure BDA0002357159240000046
less than a given threshold ξ, the iteration is stopped, otherwise i is incremented by 1 and the procedure returns to step 102).
The random impact characteristics D (t) extracted according to the above steps are shown in fig. 6.
In step S3, two-dimensional scatter points consisting of Teager local energy and random impact characteristics in a set time period are recorded, and the two-dimensional scatter points are recorded as p in a scattered form1[T(1),D(1)],p2[T(2),D(2)],…pN[T(N),D(N)]Fig. 7 is a discrete scattergram. And clustering the two-dimensional scatter points.
The clustering process specifically comprises the following steps:
based on the scattered point coordinates, an objective function of cluster center search is established as follows:
Figure BDA0002357159240000051
wherein m is the cluster number of the cluster, i represents the ith sample, N represents the total number of samples, j represents the jth class, C represents the total number of classes,
Figure BDA0002357159240000052
represents a sample piDegree of membership belonging to class j, cjRepresents the cluster center of class j, | x | is a distance measure.
In this embodiment, m is taken as 3 types, and the first type includes: slow movements such as standing, walking, bending down, lying down, the second category includes: jogging, light jumping, etc., and the third category includes: sudden falls, bumps, and other transient violent actions. And finally, judging the current state of the wearer according to the membership degree, and combining the synchronously acquired GPS position information to acquire the specific position of the potential danger.
The membership u is continuously updated by iteration by the following equationijAnd cluster center cjUntil the result is reachedRequire that
Figure BDA0002357159240000053
Figure BDA0002357159240000054
Wherein the above-mentioned iteration termination condition is
Figure BDA0002357159240000055
Fig. 8 shows the results of clustering 3 cluster centers obtained by iterative computation and the scatter points according to the maximum membership, where the first type is shown by the star points in the figure, the second type is shown by the small circle points in the figure, and the third type is shown by the black circle points in the figure.
FIG. 9 shows the result of returning the third type scatter dots (e.g., the black dots in FIG. 8, indicating a violent action such as a fall) to the Teager's local energy index, and FIG. 10 shows the result of returning the third type scatter dots to the random impact signature. It can be seen from the figure that the violent motion (such as a sudden fall, a bump, etc.) causes the features to be accurately classified and identified (the violent motion occurs at the position shown by the black dots in fig. 9 and 10).
In step S4, the current motion of the recognition target is acquired based on the clustering result, and when it is determined that the current motion is a violent motion, an alarm is generated. Generating the alarm comprises generating a buzzing alarm signal or simultaneously issuing a danger signal and a position signal of the identification object.
Example 2
The embodiment provides a wearable system for implementing the tumble identification method based on acceleration impact energy clustering as described in embodiment 1, and the wearable system includes a wearable device and a computer terminal.
As shown in fig. 2, the wearable device includes an acceleration and positioning real-time monitoring module, a data wireless transmission module, a power module, and an alarm module, wherein the acceleration and positioning real-time monitoring module acquires three-way acceleration and positioning information of an identification object in real time; the data wireless transmission module transmits three-way acceleration and positioning information to the computer end, the power module supplies electric energy to the wearable device, and the alarm module gives out a buzzing alarm. When the wearable equipment is worn, a Cartesian coordinate system is established by taking a wearing point of a body of a wearer as an origin, and the vertical direction of a horizontal plane is the Z direction of a gyroscope; the horizontal plane is XY direction, the gyroscope X direction is right in front of the wearer, and the direction pointed by the shoulders is the gyroscope Y direction. In the real-time monitoring process, the wearable device needs to synchronously test the three-direction acceleration information and the real-time GPS positioning information of the wearer during action. In the implementation process, the test process comprises the action simulation of standing, walking, jogging, falling, impacting and the like.
The computer terminal comprises a tumble identification algorithm module based on impact energy clustering, and is an operation platform for data processing, and finally, the display and the alarm of the current state and the position information of the wearer are realized.
The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions that can be obtained by a person skilled in the art through logic analysis, reasoning or limited experiments based on the prior art according to the concept of the present invention should be within the protection scope determined by the present invention.

Claims (10)

1. A tumble identification method based on acceleration impact energy clustering is characterized by comprising the following steps:
1) acquiring three-way acceleration and positioning information of an identified object in real time;
2) fusing the three-direction acceleration to obtain corresponding Teager local energy and random impact characteristics;
3) recording two-dimensional scattered points consisting of Teager local energy and random impact characteristics in a set time period, and clustering the two-dimensional scattered points;
4) and acquiring the current action of the identification object based on the clustering result, and generating an alarm when judging that the current action is violent.
2. The acceleration impact energy clustering-based fall identification method according to claim 1, wherein the three-way acceleration fusion is specifically:
Figure FDA0002357159230000011
in the formula, ax(t)、ay(t)、azAnd (t) are three-way acceleration information respectively, and a (t) is fusion information.
3. The method of claim 1, wherein the Teager local energy acquisition formula is:
Figure FDA0002357159230000012
in the formula, a (T) is fusion information, and T (T) is Teager local energy.
4. The method of claim 1, wherein the random impact characteristic acquisition formula is:
D(t)＝w(t)*a(t)
in the formula, w (t) is an inverse filter, a (t) is fusion information, and represents convolution.
5. The method for fall recognition based on acceleration impact energy clustering of claim 4, wherein the random impact signature is obtained by means of iterative optimization, comprising in particular the following steps:
101) initializing the inverse filter w(0)(t) where all elements are 1, let i equal 1;
102) calculating D (t) ═ w(i-1)(t)*a(t)；
103) Computing
Figure FDA0002357159230000013
Wherein
Figure FDA0002357159230000014
l is lag time, N total time length;
104) calculating w(i)＝A-1b(i)Wherein the matrix A is an autocorrelation matrix of the sequence a (t);
105) if it is not
Figure FDA0002357159230000021
less than a given threshold ξ, the iteration is stopped, otherwise i is incremented by 1 and the procedure returns to step 102).
6. The method for identifying a fall based on acceleration impact energy clustering of claim 1, wherein in the step 3), the objective function used for clustering is as follows:
Figure FDA0002357159230000022
wherein m is the cluster number of the cluster, i represents the ith sample, N represents the total number of samples, j represents the jth class, C represents the total number of classes,
Figure FDA0002357159230000023
represents a sample piDegree of membership belonging to class j, cjRepresents the cluster center of class j, | x | is a distance measure.
7. The method of claim 6, wherein the membership update formula is:
Figure FDA0002357159230000024
the update formula of the cluster center is as follows:
Figure FDA0002357159230000025
8. the method of claim 1, wherein the generating an alarm comprises generating a beeping alarm signal.
9. The method of claim 1, wherein the generating of the alarm comprises issuing a danger signal and a location signal identifying the location of the object.
10. A wearable system implementing the acceleration impact energy clustering-based fall recognition method of claim 1, comprising a wearable device and a computer, wherein the wearable device implements step 1), and the computer implements steps 2) -4).
CN202010011004.5A 2020-01-06 2020-01-06 Tumble identification method based on acceleration impact energy clustering and wearable system Pending CN111166341A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010011004.5A CN111166341A (en) 2020-01-06 2020-01-06 Tumble identification method based on acceleration impact energy clustering and wearable system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010011004.5A CN111166341A (en) 2020-01-06 2020-01-06 Tumble identification method based on acceleration impact energy clustering and wearable system

Publications (1)

Publication Number Publication Date
CN111166341A true CN111166341A (en) 2020-05-19

Family

ID=70647432

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010011004.5A Pending CN111166341A (en) 2020-01-06 2020-01-06 Tumble identification method based on acceleration impact energy clustering and wearable system

Country Status (1)

Country Link
CN (1) CN111166341A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113378682A (en) * 2021-06-03 2021-09-10 山东省科学院自动化研究所 Millimeter wave radar fall detection method and system based on improved clustering algorithm

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105989854A (en) * 2015-02-27 2016-10-05 佳能株式会社 Impact sound detection device and impact sound detection method
CN106061369A (en) * 2013-11-04 2016-10-26 帝国创新技术有限公司 Biomechanical activity monitoring
CN110470475A (en) * 2019-09-04 2019-11-19 中国人民解放军空军工程大学航空机务士官学校 A kind of aero-engine intershaft bearing early-stage weak fault diagnostic method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106061369A (en) * 2013-11-04 2016-10-26 帝国创新技术有限公司 Biomechanical activity monitoring
CN105989854A (en) * 2015-02-27 2016-10-05 佳能株式会社 Impact sound detection device and impact sound detection method
CN110470475A (en) * 2019-09-04 2019-11-19 中国人民解放军空军工程大学航空机务士官学校 A kind of aero-engine intershaft bearing early-stage weak fault diagnostic method

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
华仙等: "基于人体加速度多特征融合和K近邻算法的跌倒检测", 《中国康复理论与实践》 *
张玉学等: "基于EEMD与FCM聚类的自动机故障诊断", 《中国测试》 *
李彤等: "基于时差辨识的机翼模型光纤冲击定位方法", 《光电子？激光》 *
王晗等: "最小熵解卷积法轮对轴承故障诊断", 《中国测试》 *
高强等: "基于增强聚类分割与L-峭度的Teager能量算子解调诊断轴向柱塞泵故障", 《机械工程学报》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113378682A (en) * 2021-06-03 2021-09-10 山东省科学院自动化研究所 Millimeter wave radar fall detection method and system based on improved clustering algorithm

Similar Documents

Publication Publication Date Title
Wang et al. Fall detection based on dual-channel feature integration
CN103886341A (en) Gait behavior recognition method based on feature combination
CN108334827B (en) Gait identity authentication method based on intelligent shoe and intelligent shoe
CN110287825A (en) It is a kind of that motion detection method is fallen down based on crucial skeleton point trajectory analysis
CN114676956A (en) Old man's risk early warning system that tumbles based on multidimensional data fusion
Sun et al. Fall detection algorithm for the elderly based on human posture estimation
CN110456320A (en) A kind of ULTRA-WIDEBAND RADAR personal identification method based on free space gait temporal aspect
CN114495267A (en) Old people falling risk assessment method based on multi-dimensional data fusion
Kumar et al. A unified grid-based wandering pattern detection algorithm
Yang et al. Human posture recognition in intelligent healthcare
Wang et al. A2dio: Attention-driven deep inertial odometry for pedestrian localization based on 6d imu
Rana et al. Markerless gait classification employing 3D IR-UWB physiological motion sensing
Qu et al. Convolutional neural network for human behavior recognition based on smart bracelet
CN115205891A (en) Personnel behavior recognition model training method, behavior recognition method and device
Lin et al. Adaptive multi-modal fusion framework for activity monitoring of people with mobility disability
CN111166341A (en) Tumble identification method based on acceleration impact energy clustering and wearable system
CN117158955A (en) User safety intelligent monitoring method based on wearable monitoring equipment
CN112883922A (en) Sign language identification method based on CNN-BiGRU neural network fusion
Zhao et al. Wear‐free indoor fall detection based on RFID and deep residual networks
Ni et al. Sports dance action recognition system oriented to human motion monitoring and sensing
CN115393956A (en) CNN-BilSTM fall detection method for improving attention mechanism
CN114764580A (en) Real-time human body gesture recognition method based on no-wearing equipment
CN114022956A (en) Method for multi-dimensional intelligent study and judgment of body-building action and movement effect
Yangjiaozi et al. Research on fall detection based on improved human posture estimation algorithm
Xia et al. Real-time recognition of human daily motion with smartphone sensor

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
CB03 Change of inventor or designer information

Inventor after: Li Yong

Inventor after: Zheng Haoran

Inventor after: Chen Hao

Inventor after: Guan Yandong

Inventor after: Ziruiqing

Inventor after: Xing Gailan

Inventor after: Zhou Shaoping

Inventor before: Li Yong

Inventor before: Xing Gailan

Inventor before: Zhou Shaoping

Inventor before: Zheng Haoran

Inventor before: Chen Hao

Inventor before: Guan Yandong

Inventor before: Ziruiqing

CB03 Change of inventor or designer information
RJ01 Rejection of invention patent application after publication

Application publication date: 20200519

RJ01 Rejection of invention patent application after publication