CN110705471A - Passive posture recognition method based on short-time Fourier and principal component analysis method - Google Patents
Passive posture recognition method based on short-time Fourier and principal component analysis method Download PDFInfo
- Publication number
- CN110705471A CN110705471A CN201910940724.7A CN201910940724A CN110705471A CN 110705471 A CN110705471 A CN 110705471A CN 201910940724 A CN201910940724 A CN 201910940724A CN 110705471 A CN110705471 A CN 110705471A
- Authority
- CN
- China
- Prior art keywords
- vector
- received signal
- short
- nodes
- central server
- 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
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000012847 principal component analysis method Methods 0.000 title claims abstract description 14
- 238000004891 communication Methods 0.000 claims abstract description 9
- 239000013598 vector Substances 0.000 claims description 48
- 238000000513 principal component analysis Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000036544 posture Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
Images
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/309—Measuring or estimating channel quality parameters
- H04B17/318—Received signal strength
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Quality & Reliability (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The invention relates to a passive gesture recognition method based on a short-time Fourier and principal component analysis method, and the system implementation comprises the following steps: wireless sensor nodes and a central server; the wireless sensor nodes adopt a multi-hop network for communication, the wireless sensor nodes arranged on indoor walls are set as routing nodes, and the wireless signal strength detected by the routing nodes in each communication is uniformly transmitted to the nodes connected with the central server; and the node connected with the central server is set as a coordinator node, and the central server is used for processing the received signal and recognizing the posture of the target.
Description
Technical Field
The invention relates to a passive gesture recognition method based on a short-time Fourier and principal component analysis method, and belongs to the technical field of target gesture recognition.
Background
Along with the construction and planning of smart cities in China, the importance and the universality of indoor scene perception technology are more and more concerned by people. The common context awareness technology requires a user to carry an electronic device such as a mobile phone and a sports bracelet, and therefore the context awareness technology based on the self-carried device is inconvenient and has certain limitations. The invention aims to realize a passive technology which can realize user gesture identification without carrying any electronic communication equipment. The invention provides a passive posture identification method based on a short-time Fourier and principal component analysis method, aiming at overcoming the defects of indoor multipath, non-line-of-sight and the like of radio frequency signals. The technology can realize the recognition of the gesture of the user, and control the equipment with the function of the Internet of things to provide services for the user according to the prejudgment of the gesture of the user.
Through the search of patent applicants, the current domestic invention patents related to document image processing mainly focus on the recognition of target gestures by adopting a deep neural network algorithm [1-3], but the deep neural network needs large computing resources and storage resources for processing received signals, and is high in cost and long in operation time. The invention adopts the principal component analysis method, so that the dimensionality of data can be effectively reduced, the influence of environmental noise on a received signal is reduced, and the accuracy and the operation speed of gesture recognition can be effectively improved.
Reference documents:
[1]Wang J,Zhang X,Gao Q,et al.Device-free Wireless Localization andActivity Recognition:A Deep Learning Approach[J].IEEE Transactions onVehicular Technology,2017,66(7):6258-6267.
[2]Huang X,Dai M.Indoor Device-Free Activity Recognition Based onRadio Signal[J].IEEE Transactions on Vehicular Technology,2016:1-1.
[3]Gu Y,Zhan J,Ji Y,et al.MoSense:A RF-based Motion Detection Systemvia Off-the-Shelf WiFi Devices[J].IEEE Internet of Things Journal,2017,PP(99):1-1.
disclosure of Invention
The invention aims to: a passive gesture recognition method based on short-time Fourier and principal component analysis is provided. The method utilizes the influence of human body on radio frequency signals and the shielding effect of Fresnel regions to estimate the posture of the target. And reducing the dimension of the signal by a principal component analysis method, extracting signal characteristics caused by different postures, and realizing accurate identification of the different postures by a fingerprint library matching method. The technical scheme of the invention is as follows:
a passive gesture recognition method based on short-time Fourier and principal component analysis comprises the following steps: wireless sensor nodes and a central server. The wireless sensor nodes adopt a multi-hop network for communication, the wireless sensor nodes installed on indoor walls are set as routing nodes, and the wireless signal strength detected by the routing nodes in each communication is uniformly transmitted to the nodes connected with the central server. And the node connected with the central server is set as a coordinator node, and the central server is used for processing the received signal and recognizing the posture of the target. The method comprises the following steps:
step 1: the received signal strength vector is set to z (t) ═ a (t) + n (t), where a ═ a (θ)1),...,a(θK)],n(t)=[n1(t),...,nL(t)]A is the arrival direction vector to the received signal, a (θ)i) K is the respective angle of arrival, λ is the wavelength of the signal, d is the distance between the two nodes, T represents the matrix transpose, and n (T) is the noise vector.
Step 2: the vector after short-time Fourier transform of the received signal vector is STFT (t, z (t))
And step 3: processing the vector of the received signal vector by a principal component analysis method and performing short-time Fourier transform to obtain a vector STFT (t, z (t)), and acquiring a feature vector v of the STFT (t, z (t)) vectoractAnd v isactAnd storing the corresponding attitude tag into a fingerprint library.
And 4, step 4: in the attitude recognition stage, a feature vector v of a vector STFT (t, z (t)) obtained by short-time Fourier transform of a received signal vector is calculateddet。
And 5: calculating a feature vector corresponding to the gesture to be recognized andthe feature differences between the feature vectors corresponding to poses in the fingerprint library,wherein v isactAnd _iis a feature vector corresponding to the ith gesture stored in the fingerprint database.
The invention provides a passive posture identification method based on a short-time Fourier and principal component analysis method, aiming at overcoming the defects of indoor multipath, non-line-of-sight and the like of radio frequency signals. The technology can realize the recognition of the gesture of the user. The invention adopts the principal component analysis method, so that the dimensionality of data can be effectively reduced, the influence of environmental noise on a received signal is reduced, and the accuracy and the operation speed of gesture recognition can be effectively improved.
Drawings
FIG. 1 is a system scene diagram of the passive gesture recognition method based on short-time Fourier and principal component analysis methods of the present invention.
FIG. 2 is a block flow diagram of a method of target gesture recognition provided by the present invention.
Detailed Description
The invention is further described in detail in the following with reference to the drawings, which are only illustrative of one embodiment of the invention and do not represent a limitation to the scope of the invention.
As shown in fig. 1, which is a scene schematic diagram of the passive gesture recognition method based on short-time fourier and principal component analysis methods of the present invention, a plurality of wireless sensor nodes are installed on an indoor wall.
A passive gesture recognition method based on short-time Fourier and principal component analysis comprises the following steps: wireless sensor nodes and a central server. The wireless sensor nodes adopt a multi-hop network for communication. The wireless sensor nodes can be randomly installed at any position in a room, and the number of the wireless nodes can be increased according to the precision requirement. The wireless sensor nodes installed on the indoor wall are set as routing nodes, and the wireless signal strength detected by the routing nodes in each communication with each other is uniformly transmitted to the nodes connected with the central server. And the node connected with the central server is set as a coordinator node, and the central server is used for processing the received signal and recognizing the posture of the target. The specific steps are shown in fig. 2 and described as follows:
step 1: the received signal strength vector is set to z (t) ═ a (t) + n (t), where a ═ a (θ)1),...,a(θK)],n(t)=[n1(t),...,nL(t)]A is the arrival direction vector to the received signal, a (θ)i) K is the respective angle of arrival, λ is the wavelength of the signal, d is the distance between the two nodes, T represents the matrix transpose, and n (T) is the noise vector.
Step 2: the vector after short-time Fourier transform of the received signal vector is STFT (t, z (t))
And step 3: processing the vector of the received signal vector by a principal component analysis method and performing short-time Fourier transform to obtain a vector STFT (t, z (t)), and acquiring a feature vector v of the STFT (t, z (t)) vectoractAnd v isactAnd storing the corresponding attitude tag into a fingerprint library.
And 4, step 4: in the attitude recognition stage, a feature vector v of a vector STFT (t, z (t)) obtained by short-time Fourier transform of a received signal vector is calculateddet。
And 5: calculating the characteristic difference between the characteristic vector corresponding to the gesture to be recognized and the characteristic vector corresponding to the gesture in the fingerprint database,wherein v isactI is the characteristic corresponding to the ith gesture stored in the fingerprint databaseAnd (5) vector quantity.
Claims (1)
1. A passive gesture recognition method based on short-time Fourier and principal component analysis comprises the following steps: wireless sensor nodes and a central server; the wireless sensor nodes adopt a multi-hop network for communication, the wireless sensor nodes arranged on indoor walls are set as routing nodes, and the wireless signal strength detected by the routing nodes in each communication is uniformly transmitted to the nodes connected with the central server; the node connected with the central server is set as a coordinator node, and the central server is used for processing the received signal and identifying the posture of the target; the method comprises the following steps:
step 1: the received signal strength vector is set to z (t) ═ a (t) + n (t), where a ═ a (θ)1),...,a(θK)],n(t)=[n1(t),...,nL(t)]A is the arrival direction vector to the received signal, a (θ)i) K is the respective angle of arrival, λ is the wavelength of the signal, d is the distance between the two nodes, T represents the matrix transpose, n (T) is the noise vector;
step 2: the vector after short-time Fourier transform of the received signal vector is STFT (t, z (t))
And step 3: processing the vector of the received signal vector by a principal component analysis method and performing short-time Fourier transform to obtain a vector STFT (t, z (t)), and acquiring a feature vector v of the STFT (t, z (t)) vectoractAnd v isactStoring the corresponding attitude tag into a fingerprint library;
and 4, step 4: in the postureA state identification stage for calculating a feature vector v of a vector STFT (t, z (t)) obtained by short-time Fourier transform of a received signal vectordet;
And 5: calculating the characteristic difference between the characteristic vector corresponding to the gesture to be recognized and the characteristic vector corresponding to the gesture in the fingerprint database,wherein v isactThe _iis a characteristic vector corresponding to the ith posture stored in the fingerprint database;
And 7: the gesture recognition function isThe m value is the number of the corresponding gesture.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910940724.7A CN110705471A (en) | 2019-09-30 | 2019-09-30 | Passive posture recognition method based on short-time Fourier and principal component analysis method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910940724.7A CN110705471A (en) | 2019-09-30 | 2019-09-30 | Passive posture recognition method based on short-time Fourier and principal component analysis method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110705471A true CN110705471A (en) | 2020-01-17 |
Family
ID=69197967
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910940724.7A Pending CN110705471A (en) | 2019-09-30 | 2019-09-30 | Passive posture recognition method based on short-time Fourier and principal component analysis method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110705471A (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103971108A (en) * | 2014-05-28 | 2014-08-06 | 北京邮电大学 | Wireless communication-based human body posture recognition method and device |
CN105682048A (en) * | 2016-03-16 | 2016-06-15 | 重庆邮电大学 | Subspace match indoor fingerprint positioning method based on PCA under cellular network environment |
CN106941718A (en) * | 2017-04-07 | 2017-07-11 | 南京邮电大学 | A kind of mixing indoor orientation method based on signal subspace fingerprint base |
CN108805194A (en) * | 2018-06-04 | 2018-11-13 | 上海交通大学 | A kind of hand-written recognition method and system based on WIFI channel state informations |
CN109655790A (en) * | 2018-12-18 | 2019-04-19 | 天津大学 | Multi-target detection and identification system and method based on indoor LED light source |
CN109711251A (en) * | 2018-11-16 | 2019-05-03 | 天津大学 | A kind of directionally independent gait recognition method based on commercial Wi-Fi |
CN110073301A (en) * | 2017-08-02 | 2019-07-30 | 强力物联网投资组合2016有限公司 | The detection method and system under data collection environment in industrial Internet of Things with large data sets |
-
2019
- 2019-09-30 CN CN201910940724.7A patent/CN110705471A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103971108A (en) * | 2014-05-28 | 2014-08-06 | 北京邮电大学 | Wireless communication-based human body posture recognition method and device |
CN105682048A (en) * | 2016-03-16 | 2016-06-15 | 重庆邮电大学 | Subspace match indoor fingerprint positioning method based on PCA under cellular network environment |
CN106941718A (en) * | 2017-04-07 | 2017-07-11 | 南京邮电大学 | A kind of mixing indoor orientation method based on signal subspace fingerprint base |
CN110073301A (en) * | 2017-08-02 | 2019-07-30 | 强力物联网投资组合2016有限公司 | The detection method and system under data collection environment in industrial Internet of Things with large data sets |
CN108805194A (en) * | 2018-06-04 | 2018-11-13 | 上海交通大学 | A kind of hand-written recognition method and system based on WIFI channel state informations |
CN109711251A (en) * | 2018-11-16 | 2019-05-03 | 天津大学 | A kind of directionally independent gait recognition method based on commercial Wi-Fi |
CN109655790A (en) * | 2018-12-18 | 2019-04-19 | 天津大学 | Multi-target detection and identification system and method based on indoor LED light source |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wu et al. | Passive indoor localization based on csi and naive bayes classification | |
Nessa et al. | A survey of machine learning for indoor positioning | |
Kunhoth et al. | Indoor positioning and wayfinding systems: a survey | |
Ibrahim et al. | CNN based indoor localization using RSS time-series | |
Abdelnasser et al. | SemanticSLAM: Using environment landmarks for unsupervised indoor localization | |
US20200142045A1 (en) | Fingerprint positioning method and system in smart classroom | |
Milioris et al. | Low-dimensional signal-strength fingerprint-based positioning in wireless LANs | |
CN104394588B (en) | Indoor orientation method based on Wi Fi fingerprints and Multidimensional Scaling | |
Milioris et al. | Empirical evaluation of signal-strength fingerprint positioning in wireless LANs | |
Turgut et al. | Deep learning in indoor localization using WiFi | |
Xiao et al. | Abnormal behavior detection scheme of UAV using recurrent neural networks | |
Hashemifar et al. | Augmenting visual SLAM with Wi-Fi sensing for indoor applications | |
Liu et al. | Large-scale deep learning framework on FPGA for fingerprint-based indoor localization | |
Farid et al. | Hybrid Indoor‐Based WLAN‐WSN Localization Scheme for Improving Accuracy Based on Artificial Neural Network | |
BelMannoubi et al. | Deep neural networks for indoor localization using WiFi fingerprints | |
Ding et al. | Multiview features fusion and Adaboost based indoor localization on Wifi platform | |
Ng et al. | A kernel method to nonlinear location estimation with RSS-based fingerprint | |
Tarekegn et al. | Applying long short-term memory (LSTM) mechanisms for fingerprinting outdoor positioning in hybrid networks | |
Mantoro et al. | Extreme learning machine for user location prediction in mobile environment | |
Subakti et al. | Indoor Localization with Fingerprint Feature Extraction | |
Zhou et al. | IMLours: Indoor mapping and localization using time-stamped WLAN received signal strength | |
Cai et al. | SAP: A novel stationary peers assisted indoor positioning system | |
CN110705471A (en) | Passive posture recognition method based on short-time Fourier and principal component analysis method | |
Kim et al. | InFo: indoor localization using fusion of visual information from static and dynamic cameras | |
Ayinla et al. | SALLoc: An Accurate Target Localization In Wifi-Enabled Indoor Environments Via Sae-Alstm |
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 |
Application publication date: 20200117 |
|
RJ01 | Rejection of invention patent application after publication |