CN107766930B - Equivalent ROM distance calculation method based on DTW cluster fuzzy clustering SOM neurons - Google Patents
Equivalent ROM distance calculation method based on DTW cluster fuzzy clustering SOM neurons Download PDFInfo
- Publication number
- CN107766930B CN107766930B CN201710795124.7A CN201710795124A CN107766930B CN 107766930 B CN107766930 B CN 107766930B CN 201710795124 A CN201710795124 A CN 201710795124A CN 107766930 B CN107766930 B CN 107766930B
- Authority
- CN
- China
- Prior art keywords
- rom
- community
- distance
- equivalent
- cluster
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/043—Architecture, e.g. interconnection topology based on fuzzy logic, fuzzy membership or fuzzy inference, e.g. adaptive neuro-fuzzy inference systems [ANFIS]
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
- G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Automation & Control Theory (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Fuzzy Systems (AREA)
- Computational Mathematics (AREA)
- User Interface Of Digital Computer (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a DTW cluster-based fuzzy clustering SOM neuron ROM equivalent distance calculation method, which comprises the steps of establishing data after secondary Gaussian covariance white noise of an initial Kalman filtering algorithm is performed on a feature vector, and obtaining a sample sequence for a dynamic time warping algorithm; converting the sample sequence into the amplitude values of the X axis and the Y axis of the acceleration amplitude value SVM of the acceleration signal, and dividing the SVM amplitude value into different communities by using a fuzzy clustering neuron algorithm; combining the two axial community values of the SVM (x) and the SVM (y) and the cluster center point and the gravity center point corresponding to the relative community values; and combining the cluster gravity center and the center to obtain a ROM equivalent numerical value to calculate the distance. The invention can find out the space deviation distance of different community sensors and reduce the path time of the end-to-end body state sensor.
Description
Technical Field
The invention relates to the technical field of sensor information processing and gesture recognition, in particular to a ROM equivalent distance calculation method of a fuzzy clustering SOM neuron based on dynamic time normalization.
Background
The intelligent bracelet has characteristics such as wearable, small, multisensor, wireless transmission, can abstract into the node and be used for human perception network, can constitute limited wireless sensor network through two bracelets or a plurality of wrist strap sensors promptly. The data are transmitted in real time through the wireless transmission module, and the wireless sensor network is combined with a neuron topological structure, so that the mobility and the rhythmicity of network nodes can be dynamically monitored to process the human body multi-dimensional posture perception information. The acceleration sensor and the angular velocity sensor of the wrist strap sensor are used for multi-body modeling, a behavior track collaborative operation model of the wrist strap sensor of a user is established, the posture behaviors of the human body are recorded, and the human body posture can be tested and subjected to scene analysis.
In the prior art, most of gesture recognition is based on the change of the direction of an angle and acceleration as a judgment basis. But the angle change quantization processing is not carried out, and the gesture recognition cannot be subjected to quantization criterion and tracking.
In order to overcome the defects of the prior art, the invention provides a ROM equivalent distance calculation method of a fuzzy clustering SOM neuron based on dynamic time warping. The invention networks the sensor nodes, and dynamically detects the sensor nodes through the SOM neural network element, so that the network topology is in dynamic change. And (4) recognizing the gesture direction according to dynamic time arrangement and recognizing the gesture offset angle according to the ROM equivalent distance. The optimal rhythm distance detection of human body dynamic perception is realized by combining sensor data modeling and multi-channel intelligent control transmission. The human posture detection efficiency is improved.
Disclosure of Invention
The invention provides an equivalent ROM distance calculation method based on DTW cluster fuzzy clustering SOM neurons, which comprises the following steps:
a. the method comprises the steps that a sensor network is built by at least two motion sensors, and motion signals of human body motion are synchronously acquired, wherein the motion signals comprise three-dimensional acceleration signals and three-dimensional angular velocity signals;
b. establishing a feature vector of the motion signal, wherein the feature vector is subjected to secondary Gaussian covariance white noise removal by an initial state Kalman filtering unit to obtain a sample sequence for a dynamic time warping algorithm;
c. converting the sample sequence into a first acceleration amplitude and a second acceleration amplitude, and clustering the first acceleration amplitude and the second acceleration amplitude by using a fuzzy clustering neuron computing unit to form a corresponding first cluster and a second cluster;
d. combining the community values of the first community and the second community to obtain a first community center point and a first community center point of gravity of the first community, and a second community center point of the second community;
e. and calculating to obtain an equivalent ROM distance value according to the first colony center point, the second colony center point and the second colony center point, wherein the equivalent ROM distance value is used for guiding the human body to dynamically sense the optimal rhythm distance detection selection.
In the equivalent ROM distance calculation method based on DTW cluster fuzzy clustering SOM neurons provided by the present invention, after calculating the equivalent ROM distance value, the method further comprises:
f. obtaining weights according to SOM neuron criteria; the equivalent ROM distance value with large weight is a priority output value;
g. obtaining the node track length according to the weight value;
h. and outputting the node track length and the equivalent ROM distance value.
In the equivalent ROM distance calculation method based on the DTW cluster fuzzy clustering SOM neurons provided by the invention, the weight value is expressed as the following formula:
wherein the content of the first and second substances,
wherein H (i) represents the current colony deviation amplitude,
representing an equivalent distance mean; ROM (i, k) represents an equivalent ROM distance value; t represents the current time; i represents the current community; and k represents the current community node.
In the equivalent ROM distance calculation method based on the DTW cluster fuzzy clustering SOM neurons provided by the invention, the node track length is expressed as the following formula:
S(t,i)＝L(t,i)×ROM(t,(i,k))×F(t,(i,k))×NET(k)；
in the formula, L (t, i) represents the distance between the center point fcm (i) of the community i and the origin, net (k) represents the weight, and F (t, (i, k)) is criterion information.
In the equivalent ROM distance calculation method based on the DTW cluster fuzzy clustering SOM neurons, the original azimuth angle and the original mobile criterion are subjected to quantitative processing by screening and extracting the angle change characteristic value of the sensor, quantitative analysis and distance conversion of the angle quantity value during equivalent ROM distance calculation.
In the equivalent ROM distance calculation method based on the DTW cluster fuzzy clustering SOM neurons provided by the invention, after the initial state Kalman filtering is performed, the required attitude identification samples can be screened through the DTW, and the screened sample sequence is helpful for grouping the refined attitude again through the fuzzy clustering SOM algorithm.
According to the invention, on the basis that a wrist strap sensor acquires a motion signal, the motion signal is filtered through a primary KALMAN and a secondary Gaussian covariance white noise matrix and then processed to obtain azimuth motion characteristic vector information of a human wrist, the dynamic time integral mapping of the characteristic vector information is carried out to obtain gesture azimuth attitude information D (i), then a node ROM (i) value is obtained by aiming at wireless network node equivalent fuzzy clustering based on the wrist strap sensor, and then the equivalent distance information of the human hand motion is obtained by combining with criteria of wrist attitude information D (i) and ROM (i) by using an SOM neuron algorithm. According to the invention, the required attitude identification samples can be screened through DTW, the screened sample sequence is helpful for grouping the refined attitude again through the fuzzy clustering SOM algorithm, the cluster analysis efficiency can be improved, the dimension of the screened multivariate variables can be reduced, and the angle quantization processing can be accelerated. Compared with the traditional method for tracking the human body posture, the method needs video stream analysis posture information, does not need judgment after video stream acquisition and processing, and reduces hardware cost.
In the invention, DTW refers to Dynamic Time Warping and is called Dynamic Time Warping; ROM refers to angle tracking, which is called Range of Motion; the SOM is a Self-Organizing (competitive) neural network, all known as Self-Organizing Maps.
Drawings
FIG. 1 is a diagram of the overall system architecture of the fuzzy clustering SOM neuron ROM equivalent distance algorithm based on dynamic time warping.
FIG. 2 is a dual node sensor ROM computation graph in accordance with the present invention.
FIG. 3 is a multi-node sensor ROM computation graph in accordance with the present invention.
FIG. 4 is a SOM neuron order diagram in the present invention.
Detailed Description
The wireless network node data acquisition of the invention fuses accelerometer and gyroscope sensor information. According to the method, a rotation matrix (rolling, pitching and heading) is calculated on the basis of the Euler angle, and acceleration data from a person can be obtained through conversion of the rotation matrix.
The wrist strap sensor processes data information obtained after data processing of the acceleration sensor for the first time, and performs quantitative processing on the data, which is realized by software and has feasibility. And processing the obtained information through polymorphic Kalman filtering according to the collected motion signals, detecting the motion of the wrist of the upper limb of the human body according to the preliminarily collected motion signals, and further obtaining the azimuth motion information of the hand of the human body. Kalman filtering is a filtering and prediction technique for a linear gaussian system. It consists of a processing model and a measuring model. The data processing principle is mainly the fusion of dual-channel data, and an intelligent remote sensing control system and a method thereof based on the fusion of double hand-ring sensors are continued, and the sensor data are fused by a technology for detecting and estimating the system state. To continuously estimate the hand motion state space model is as follows:
x(k-1)＝Mz(k-1)+u(k-1) (1)
and performing space state allocation processing during further data processing. x (k-1) represents the output result when the first step of data processing is carried out, M is a defined quantization mapping table, z (k-1) is the raw output data of the sensor, and u (k-1) is the noise of the system. The invention continues an intelligent remote sensing control system based on the fusion of two hand ring sensors and a method thereof to obtain a fusion mapping table, wherein y (k-1) ═ x1(k-1),x2(k-1)) indicating a two-channel information set. And y (k-1) obtains the gesture motion direction after fusing the mapping tables. And controlling the motion direction of the gesture, wherein BDFLR respectively represents back (B represents back), down (D represents down), front (F represents forward), left (L represents left) and right (R represents right). And setting that the control character is valid only when the control character in the following state is sent by two channels, and otherwise, the control character is invalid. For example, only two bracelets are used for controlling, and the character state conversion factor R is judged by integrating posture processingi(i-12, 3,4) is
As shown in fig. 1, the ROM equivalent fuzzy clustering SOM neuron algorithm based on dynamic time warping: the three-axis accelerometer of the wrist strap sensor samples the acceleration of a human body, and converts an acceleration signal obtained by sampling into an acceleration signal amplitude value as a sample sequence with regular dynamic time.
D (i, j) ═ D (i, j) + min (D (i-1, j), D (i, j-1), D (i-1, j)) (3.1) wherein:
in the above formula, D (i, j) is the minimum cumulative distance among the continuous acceleration signal amplitudes, and D (i, j) is the euclidean distance among the continuous acceleration signal amplitudes.
Correlation coefficient:
wherein epsilon (0,2) is that the adjustable parameters are different according to different values of the sensor 1, the sensor 2, the sensor 3 and the sensor 4.
Acceleration signal acceleration amplitude value SVM of the sensor 1, the sensor 2, the sensor 3 and the sensor 4 has serious loss, and the swing of the Z axis does not influence the amplitude value trends of the X axis and the Y axis, so that only two values of SVM (X) and SVM (Y) are recorded, the acceleration signal amplitude value (SVM) is subjected to end point detection, and the motion state of the wrist is predicted.
As shown in fig. 2, the data of the sensor 1 and the sensor 2 are divided into 3 groups by the DTW correlation coefficient using the fuzzy clustering neuron algorithm, and the center points of the groups (Fcm1, Fcm2) are obtained. The cluster center point is found for each group of sensors in scenario 1 and its center of gravity position is found (GM1, GM 2).
As shown in fig. 3, the data of the sensor 1, the sensor 2, the sensor 3, and the sensor 4 are divided into six groups by the DTW correlation coefficient using the fuzzy clustering neuron algorithm, and the mass center points (Fcm1, Fcm2, Fcm3, Fcm4) are obtained. The cluster center point is found for each group of sensors in scenario 2 and its center of gravity position is found (GM1, GM2, GM3, GM 4).
As shown in fig. 2, inverse triangles of svm (x) and svm (y) values of the barycentric position are calculated, respectively:
subtracting the two to obtain an equivalent ROM:
ROM(1,1)＝θG(1+1)-θG(1) (6)
as shown in fig. 3, inverse triangles of svm (x) and svm (y) values corresponding to the barycentric positions (GM1, GM2, GM3, GM4) are calculated:
subtracting the two to obtain an equivalent ROM (i):
ROM(i,k)＝θG(k+i)-θG(k) (9)
differentiating the equivalent ROM (t, (i, k)) and ROM (t + Δ t, (i, k)) of consecutive sequences obtained consecutively, and thresholding to obtain:
u(t,(i,k))＝ROM(t+Δt,(i,k))-ROM(t,(i,k)) (10)
ξ represents the difference threshold. F (t, (i, k)) is criterion information.
The SOM is a self-organizing (competitive) neural network, and the structure of the self-organizing (competitive) neural network and the learning rule thereof have own characteristics compared with other neural networks. In terms of network structure, it is generally a two-layer network composed of an input layer and a competition layer; and each neuron between the two layers realizes bidirectional connection.
As shown in fig. 4, consists of an input layer, a competition layer, and an output layer. The input layers are the positions of the center of gravity (GM1, GM2, GM3, GM4) different wristband sensors acquire the post-processing state. The network neurons compete with each other in order to be activated, with the result that only one output neuron is activated at a time. This activated neuron is called the competition winning neuron, while the state of the other neurons is suppressed. NET (t, k) will be preferentially output if the weight is large or the fluctuation is large. And sequentially displaying the body state change by using the first output value and the third output value as the first output value of the body state detection.
The relative distance S (t, i) represents the wristband sensor node trace length:
S(t,i)＝L(t,i)×ROM(t,(i,k))×F(t,(i,k))×NET(k) (14)
l (t, i) represents the distance from the origin, such as Fcm (i) in FIG. 3 or FIG. 4.
The process of obtaining the criterion information by the ROM equivalent fuzzy clustering SOM neuron algorithm based on the dynamic time integration specifically comprises the following steps:
step 1: after the feature vector of each bracelet is subjected to secondary Gaussian covariance white noise of an initial state Kalman filtering algorithm, obtaining a sample sequence for a dynamic time warping algorithm;
step 2: sampling the acceleration of a human body by a three-axis accelerometer of a bracelet sensor, and converting an acceleration signal obtained by sampling into an acceleration signal acceleration amplitude SVM;
and step 3: acceleration signal acceleration amplitude values SVM of the sensor 1, the sensor 2, the sensor 3 and the sensor 4 are recorded, and the amplitude trends of the X axis and the Y axis are not influenced by the swing of the Z axis with serious loss, so that only two values of SVM (X) and SVM (Y) are recorded;
and 4, step 4: dividing data of the sensor 1, the sensor 2, the sensor 3 and the sensor 4 into six groups by using a fuzzy clustering neuron algorithm, and respectively calculating a center point Fcm (i) (namely Fcm1, Fcm2, Fcm3 and Fcm4) of the group and a distance L (t, i) between the center point and an origin;
and 5: the clustered center points were found for each set of sensors in scenario 1 and their center of gravity positions were found (GA1, GA2, GA3, GA 4). Finding a cluster center point for each group of sensors in scenario 2 and finding the position of the center of gravity (GB1, GB2, GB3, GB 4);
step 6: calculating the gravity center position to obtain ROM (i) equivalent value;
and 7: obtaining weights NET (t, k) according to the SOM neuron criterion;
and 8: obtaining the node track length S (t, i);
and step 9: and transmitting the obtained S (t, i) and ROM (t (i, k)) equivalent values to the mobile terminal and the computer terminal through the serial port.
The protection of the present invention is not limited to the above embodiments. Variations and advantages that may occur to those skilled in the art may be incorporated into the invention without departing from the spirit and scope of the inventive concept, and the scope of the appended claims is intended to be protected.
Claims (6)
1. An equivalent ROM distance calculation method based on DTW cluster fuzzy clustering SOM neurons is characterized by comprising the following steps:
a. the method comprises the steps that a sensor network is built by at least two motion sensors, and motion signals of human body motion are synchronously acquired, wherein the motion signals comprise three-dimensional acceleration signals and three-dimensional angular velocity signals;
b. establishing a feature vector of the motion signal, wherein the feature vector is subjected to secondary Gaussian covariance white noise removal by an initial state Kalman filtering unit to obtain a sample sequence for a dynamic time warping algorithm;
c. converting the sample sequence into a first acceleration amplitude and a second acceleration amplitude, and clustering the first acceleration amplitude and the second acceleration amplitude by using a fuzzy clustering neuron computing unit to form a corresponding first cluster and a second cluster;
d. combining the community values of the first community and the second community to obtain a first community center point and a first community center point of gravity of the first community, and a second community center point of the second community;
e. and calculating to obtain an equivalent ROM distance value according to the first colony center point, the second colony center point and the second colony center point, wherein the equivalent ROM distance value is used for guiding the human body to dynamically sense the optimal rhythm distance detection selection.
2. The method of claim 1, wherein the step of calculating the equivalent ROM distance value further comprises:
f. obtaining weights according to SOM neuron criteria; the equivalent ROM distance value with large weight is a priority output value;
g. obtaining the node track length according to the weight value;
h. and outputting the node track length and the equivalent ROM distance value.
3. The method of claim 2, wherein the weight values are expressed as follows:
wherein the content of the first and second substances,
wherein H (i) represents the current colony deviation amplitude,
representing an equivalent distance mean; ROM (i, k) represents an equivalent ROM distance value; t represents the current time; i represents the current community; and k represents the current community node.
4. The DTW-clustering-fuzzy-clustering-based SOM neuron equivalent ROM distance computation method of claim 2, wherein the node trajectory length is expressed as the following formula:
S(t,i)＝L(t,i)×ROM(t,(i,k))×F(t,(i,k))×NET(k)；
in the formula, L (t, i) represents the distance between the center point fcm (i) of the community i and the origin, net (k) represents the weight, and F (t, (i, k)) is criterion information.
5. The DTW-cluster-fuzzy-clustering-SOM-neuron-based equivalent ROM distance calculation method of claim 1, wherein the equivalent ROM distance calculation is performed by extracting a sensor angle change characteristic value through screening, performing quantitative analysis, converting an angle value into a distance, and performing quantitative processing on an original azimuth angle and a moving criterion.
6. The equivalent ROM distance calculation method based on the DTW clustered fuzzy clustering SOM neurons of claim 1, wherein the required attitude recognition samples can be screened through DTW after the initial state Kalman filtering is performed, and the screened sample sequence is helpful for grouping the refined attitude again through the fuzzy clustering SOM algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710795124.7A CN107766930B (en) | 2017-09-06 | 2017-09-06 | Equivalent ROM distance calculation method based on DTW cluster fuzzy clustering SOM neurons |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710795124.7A CN107766930B (en) | 2017-09-06 | 2017-09-06 | Equivalent ROM distance calculation method based on DTW cluster fuzzy clustering SOM neurons |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107766930A CN107766930A (en) | 2018-03-06 |
| CN107766930B true CN107766930B (en) | 2021-03-26 |
Family
ID=61265414
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710795124.7A Active CN107766930B (en) | 2017-09-06 | 2017-09-06 | Equivalent ROM distance calculation method based on DTW cluster fuzzy clustering SOM neurons |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107766930B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108712760B (en) * | 2018-03-29 | 2019-11-19 | 北京邮电大学 | High-throughput relay selection method based on random Learning Automata and fuzzy algorithmic approach |
| CN110095120A (en) * | 2019-04-03 | 2019-08-06 | 河海大学 | Biology of the Autonomous Underwater aircraft under ocean circulation inspires Self-organizing Maps paths planning method |
| CN110319829B (en) * | 2019-07-08 | 2022-11-18 | 河北科技大学 | Unmanned aerial vehicle flight path planning method based on self-adaptive polymorphic fusion ant colony algorithm |
| CN110415336B (en) * | 2019-07-12 | 2021-12-14 | 清华大学 | High-precision human body posture reconstruction method and system |
| CN113065604B (en) * | 2021-04-15 | 2022-10-21 | 北京理工大学 | Air target grouping method based on DTW-DBSCAN algorithm |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7855684B2 (en) * | 2007-11-15 | 2010-12-21 | Samsung Electronics Co., Ltd. | Method and system for locating sensor node in sensor network using distance determining algorithm |
| WO2013096954A1 (en) * | 2011-12-23 | 2013-06-27 | The Trustees Of Dartmouth College | Wearable computing device for secure control of physiological sensors and medical devices, with secure storage of medical records, and bioimpedance biometric |
| CN103543826A (en) * | 2013-07-30 | 2014-01-29 | 广东工业大学 | Method for recognizing gesture based on acceleration sensor |
| CN106500695A (en) * | 2017-01-05 | 2017-03-15 | 大连理工大学 | A kind of human posture recognition method based on adaptive extended kalman filtering |
-
2017
- 2017-09-06 CN CN201710795124.7A patent/CN107766930B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7855684B2 (en) * | 2007-11-15 | 2010-12-21 | Samsung Electronics Co., Ltd. | Method and system for locating sensor node in sensor network using distance determining algorithm |
| WO2013096954A1 (en) * | 2011-12-23 | 2013-06-27 | The Trustees Of Dartmouth College | Wearable computing device for secure control of physiological sensors and medical devices, with secure storage of medical records, and bioimpedance biometric |
| CN103543826A (en) * | 2013-07-30 | 2014-01-29 | 广东工业大学 | Method for recognizing gesture based on acceleration sensor |
| CN106500695A (en) * | 2017-01-05 | 2017-03-15 | 大连理工大学 | A kind of human posture recognition method based on adaptive extended kalman filtering |
Non-Patent Citations (3)
| Title |
|---|
| Content-Based Retrieval for Human Motion Data;Chih-Yi Chiu et al.;《Journal of Visual Communication and Image Representation》;20041231;第1-36页 * |
| Localization with Single Stationary Anchor for Mobile Node in Wireless Sensor Networks;Haipeng Qu et al.;《Hindawi》;20131231;第1-12页 * |
| 加权DTW距离的自动步态识别;张浩 等;《中国图象图形学报》;20100531;第15卷(第5期);第830-836页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107766930A (en) | 2018-03-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107766930B (en) | Equivalent ROM distance calculation method based on DTW cluster fuzzy clustering SOM neurons | |
| AU2020273327B2 (en) | Systems and methods of swimming analysis | |
| CN112153736B (en) | Personnel action identification and position estimation method based on channel state information | |
| CN105184325B (en) | Mobile intelligent terminal | |
| US10061389B2 (en) | Gesture recognition system and gesture recognition method | |
| Furukawa et al. | Noise correlation matrix estimation for improving sound source localization by multirotor UAV | |
| US20190172232A1 (en) | Method and system for sensor data recognition using data enrichment for the learning process | |
| CN112114665B (en) | Hand tracking method based on multi-mode fusion | |
| CN102640085A (en) | System and method for recognizing gestures | |
| CN108245880A (en) | Body-sensing detection method for visualizing and system based on more wearing annulus sensor fusions | |
| Ahmed et al. | An approach to classify human activities in real-time from smartphone sensor data | |
| Javeed et al. | Body-worn hybrid-sensors based motion patterns detection via bag-of-features and Fuzzy logic optimization | |
| WO2017185222A1 (en) | System and method for motion trajectory collection and analysis based on ball games | |
| CN110929242B (en) | Method and system for carrying out attitude-independent continuous user authentication based on wireless signals | |
| CN109190762B (en) | Mobile terminal information acquisition system | |
| Zhang et al. | Positioning method of pedestrian dead reckoning based on human activity recognition assistance | |
| CN107392106B (en) | Human activity endpoint detection method based on double thresholds | |
| CN110929766B (en) | Self-adaptive pedestrian mobile phone attitude identification method based on Gaussian mixture clustering algorithm | |
| Bao et al. | Mobile phone-based internet of things human action recognition for E-health | |
| Sung et al. | Motion quaternion-based motion estimation method of MYO using K-means algorithm and Bayesian probability | |
| Shu et al. | Research on human motion recognition based on Wi-Fi and inertial sensor signal fusion | |
| CN116009695A (en) | Dynamic gesture recognition method and system based on data glove | |
| CN110926467B (en) | Self-adaptive pedestrian mobile phone attitude identification method based on mean value clustering algorithm | |
| Bao et al. | Operation action recognition using wearable devices with inertial sensors | |
| CN117290773B (en) | Amphibious personalized gesture recognition method and recognition system based on intelligent data glove |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP02 | Change in the address of a patent holder |
Address after: 200241 No. 500, Dongchuan Road, Shanghai, Minhang District Patentee after: EAST CHINA NORMAL University Patentee after: SHANGHAI MINGFENG INFORMATION TECHNOLOGY Co.,Ltd. Address before: 200062 No. 3663, Putuo District, Shanghai, Zhongshan North Road Patentee before: EAST CHINA NORMAL University Patentee before: SHANGHAI MINGFENG INFORMATION TECHNOLOGY Co.,Ltd. |
|
| CP02 | Change in the address of a patent holder |