CN109951866B - People flow monitoring method based on hidden Markov model - Google Patents
People flow monitoring method based on hidden Markov model Download PDFInfo
- Publication number
- CN109951866B CN109951866B CN201910204501.4A CN201910204501A CN109951866B CN 109951866 B CN109951866 B CN 109951866B CN 201910204501 A CN201910204501 A CN 201910204501A CN 109951866 B CN109951866 B CN 109951866B
- Authority
- CN
- China
- Prior art keywords
- data
- hidden markov
- markov model
- real
- radio frequency
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000012544 monitoring process Methods 0.000 title claims abstract description 20
- 239000013598 vector Substances 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000007476 Maximum Likelihood Methods 0.000 claims abstract description 7
- 238000004891 communication Methods 0.000 claims abstract description 7
- 238000002474 experimental method Methods 0.000 claims abstract description 4
- 238000003672 processing method Methods 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 230000000737 periodic effect Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims 1
- 238000003325 tomography Methods 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 5
- 238000007635 classification algorithm Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000009434 installation Methods 0.000 description 1
- 230000001795 light effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Landscapes
- Alarm Systems (AREA)
- Selective Calling Equipment (AREA)
Abstract
The invention discloses a pedestrian flow monitoring method based on a hidden Markov model, which comprises the following steps of (1) establishing an experimental environment: a wireless radio frequency tomography network node is used as a communication basis of the network; constructing a complete wireless sensor network communication system by using network nodes; (2) collecting and processing experimental data: the processing method of the wireless sensor network data is 'transmitting-receiving-storing'; calculating an attenuation value of a link according to the strength of the signal received by the experiment, and using the attenuation value as a characteristic vector to train a hidden Markov model of the mixed Gaussian; (3) the real-time personnel counting method comprises the following steps: and receiving data in real time to calculate a link attenuation value, comparing with GMM-HMM models with different parameters, taking a model class with the maximum likelihood as a corresponding people number class, and counting the passing people number. The invention has simple structure, is convenient and feasible in implementation and is suitable for most scenes.
Description
Technical Field
The invention relates to the technical field of wireless sensor networks, in particular to a pedestrian flow monitoring method based on a hidden Markov model.
Background
The current people flow monitoring problem becomes a social problem, people flow information can be known in time in many public places to ensure the normal order of the public places, the number of people and the change trend of the places can be known by users according to the personnel density, the potential conditions can be processed in time, and meanwhile, the people flow condition can be mastered in real time to provide the maximum benefit for the users. At present, there are two main methods for monitoring the flow of people: contact and contactless. The touch-type mainly comprises an inlet mechanical railing device and a pedal pressure sensor. The non-contact type mainly comprises an infrared detection technology and a video monitoring technology. The inlet mechanical railing device is convenient and accurate in counting, but the method is complex in equipment and high in required cost, has certain influence on the traveling of people, and is not beneficial to forming good customer impression; the pedal pressure sensor is mainly completed through the pressure sensor, the method is simple in structure, convenient and quick, requirements for conditions are strict, a large number of pressure sensors are needed, all people need to pass through an area where the pressure sensors are distributed when entering and exiting, and obvious limitation is caused when multiple people enter and exit the method. The infrared detection technology has very high detection accuracy rate for a single person, but when a plurality of persons or obstacles pass through the infrared detection technology, the accuracy of the infrared technology is greatly reduced; the video statistics technology acquires images according to a camera and then is matched with a computer to recognize human body targets. The method is visual and simple and has high accuracy, but a large amount of image data is obtained by using the method, so that the data processing work becomes more complicated, and in an actual environment, various light effects also have great influence on the image processing effect. The method is also costly due to the high requirements on the image acquisition and processing equipment.
Disclosure of Invention
The invention aims to solve the problems that: the pedestrian flow monitoring method based on the hidden Markov model is simple in structure, convenient and feasible in implementation and suitable for most scenes.
The technical scheme provided by the invention for solving the problems is as follows: a pedestrian flow monitoring method based on a hidden Markov model comprises the following steps,
(1) and building an experimental environment: a wireless sensing network node is used as a communication basis of the network; constructing a complete wireless sensing network communication system by utilizing network nodes;
(2) collecting and processing experimental data: the processing method of the wireless sensing network data is 'transmitting-receiving-storing'; calculating an attenuation value of a link according to the strength of the signal received by the experiment, and using the attenuation value as a characteristic vector to train a hidden Markov model of the mixed Gaussian;
(3) the real-time personnel counting method comprises the following steps: and receiving data in real time to calculate a link attenuation value, comparing with GMM-HMM models with different parameters, taking a model class with the maximum likelihood as a corresponding people number class, and counting the passing people number.
Preferably, the wireless sensing network system comprises a door frame, sensor nodes, sink nodes and a PC.
Preferably, the "transmission-reception-storage" mode is as follows:
(1) the sensing nodes on the door frame periodically transmit the radio frequency signal strength of the links formed by the nodes in a polling mode when the power supply is switched on;
(2) and the periodic radio frequency signal sent by the sensor node is received by a sink node connected to the PC and is transmitted to the PC through the serial assistant to be stored as a txt text.
Preferably, the hidden markov model classification algorithm of the mixed gaussians is as follows:
(1) and data preprocessing: deleting incomplete periods before and after each data text to obtain a plurality of complete periods of the obtained data, converting hexadecimal into decimal, and subtracting an offset to obtain a real RSS value;
(2) and extracting a characteristic value: extracting the average radio frequency signal intensity value of each effective link of each complete period of each data sample under the condition of human existence and unmanned condition, converting the average radio frequency signal intensity value into an Mx 1 characteristic vector, and carrying out difference on the radio frequency signal intensity characteristic vectors of the effective links under the condition of human existence and unmanned condition to obtain an attenuation characteristic vector of the characteristic signal intensity;
(3) modeling and parameter solving: and taking the attenuation characteristic vector obtained by processing as training data for inputting the GMM-HMM model, training by a Baum-Welch algorithm, and repeatedly iterating until convergence to obtain model parameters under the conditions of different people numbers.
Preferably, the people flow statistics is performed by continuously performing real-time monitoring, collecting real-time data, calculating a real-time attenuation value, comparing the attenuation value with each model, and taking the model class with the maximum likelihood as the corresponding people number class.
Compared with the prior art, the invention has the advantages that: the work applies radio frequency signals to people flow rate monitoring, uses a novel wireless sensing network node as a communication node, monitors people flow rate by utilizing the characteristic that the radio frequency signals can be influenced by human bodies and combining a machine learning method, is simple in structure, convenient and feasible in implementation, and is suitable for most scenes.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a diagram of the monitoring system configuration of the present invention;
FIG. 2 is a computational flow diagram of the hidden Markov model classification algorithm of the hybrid Gaussian of the present invention;
FIG. 3 is a schematic representation of the hidden Markov model classification algorithm classification using a hybrid Gaussian according to the present invention;
Detailed Description
The following detailed description of the embodiments of the present invention will be provided with reference to the accompanying drawings and examples, so that how to implement the embodiments of the present invention by using technical means to solve the technical problems and achieve the technical effects can be fully understood and implemented.
A pedestrian flow monitoring method based on a hidden Markov model comprises the following steps,
(1) and building an experimental environment: a wireless sensing network node is used as a communication basis of the network; constructing a complete wireless perception network system by utilizing network nodes; the node is a low-power-consumption wireless sensing network node platform suitable for deployment in smaller scenes such as intrusion detection, people flow statistics and falling monitoring. The low-power-consumption wireless radio frequency network node platform takes a system-on-chip (SoC) chip CC2530 as a core, and 2 programmable LED lamps and 1 key are expanded; the onboard voltage stabilizing module is compatible with the most common 3.3-5V power supply; and the JTAG interface of the programming program is designed as a non-standard interface with only 5 pins (which can supply power to the nodes) so as to reduce the size of the nodes as much as possible, namely, only 24 x 31 mm. Therefore, the low-power-consumption wireless sensing network node platform has the characteristics of low cost, small size, light weight, low power consumption, high reliability and the like, and is very suitable for indoor installation and deployment of wireless sensing networks facing smart homes and intelligent sensing. As shown in the attached figure 1 of the specification, each node is uniformly arranged on the door frame to form a wireless sensing network, and a tester passing through the node network can affect the wireless sensing network, so that certain data change is generated.
(2) Collecting and processing experimental data: the processing method of the wireless sensing network data is 'transmitting-receiving-storing'; calculating an attenuation value of a link according to the strength of the signal received by the experiment, and using the attenuation value as a characteristic vector to train a hidden Markov model of the mixed Gaussian;
(3) the real-time personnel counting method comprises the following steps: and receiving data in real time to calculate a link attenuation value, comparing with GMM-HMM models with different parameters, taking a model class with the maximum likelihood as a corresponding people number class, and counting the passing people number.
The wireless network system comprises a door frame, sensor nodes, a sink node and a PC.
The transmission-receiving-storing mode is as follows:
(1) the power supply is switched on, and the sensor nodes periodically transmit the radio frequency signal intensity of the link formed by the nodes in a polling mode;
(2) and the periodic radio frequency signal sent by the sensor node is received by a sink node connected to the PC and is transmitted to the PC through the serial assistant to be stored as a txt text.
1. The hidden Markov model classification algorithm of the mixed Gaussian comprises the following steps:
(1) and data preprocessing: deleting incomplete periods before and after each data text to obtain a plurality of complete periods of data, converting hexadecimal into decimal, and subtracting an offset to obtain a real RSS value;
(2) and extracting a characteristic value: extracting the average radio frequency signal intensity value of each effective link of each complete period of each data sample under the condition of human existence and unmanned condition, converting the average radio frequency signal intensity value into an Mx 1 characteristic vector, and carrying out difference on the radio frequency signal intensity characteristic vectors of the effective links under the condition of human existence and unmanned condition to obtain an attenuation characteristic vector of the characteristic signal intensity;
(3) modeling and parameter solving: and taking the attenuation characteristic vector obtained by processing as training data for inputting the GMM-HMM model, training by a Baum-Welch algorithm, and repeatedly iterating until convergence to obtain model parameters under the conditions of different people numbers.
The algorithm is realized according to the following principle: the resonance frequency of the radio frequency signal is consistent with that of water at 2.4GHz, and the human body can generate great influence on the radio frequency signal due to the fact that most of the human body consists of water, and the algorithm is provided.
And continuously monitoring in real time, collecting real-time data, calculating a real-time attenuation value, comparing the attenuation value with each model, and taking the model class with the maximum likelihood as the corresponding people number class to carry out people flow statistics. The beneficial effects of the invention are:
the foregoing is merely illustrative of the preferred embodiments of the present invention and is not to be construed as limiting the claims. The present invention is not limited to the above embodiments, and the specific structure thereof is allowed to vary. All changes which come within the scope of the invention as defined by the appended claims are intended to be embraced therein.
Claims (4)
1. A people flow monitoring method based on a hidden Markov model is characterized by comprising the following steps:
(1) and building an experimental environment: a wireless sensing network node with a CC2530 chip as a core is adopted as a communication basis of a network; on a door-shaped frame, 5 nodes are respectively installed on two door frames at equal intervals, 8 nodes are installed on a door beam at equal intervals, and a wireless sensing network system consisting of N =18 nodes is constructed;
(2) collecting and processing experimental data: the processing method of the wireless sensor network data is 'transmitting-receiving-storing'; deleting incomplete periods before and after each stored data text to enable the obtained data to be a plurality of complete sampling periods, then converting hexadecimal into decimal and subtracting offset to obtain a real RSS value; then extracting the average radio frequency signal intensity value of each effective link of each complete period of each data sample under the condition of existence and non-existence from the RSS value, converting the average radio frequency signal intensity value into an M multiplied by 1 feature vector, wherein M = N (N-1)/2, and subtracting the difference value of the feature vector under the condition of existence from the radio frequency signal intensity feature vector of the effective link under the condition of non-existence to obtain an attenuation feature vector of the feature signal intensity; then, the attenuation feature vector is used as training data for inputting a GMM-HMM model, training is carried out through a Baum-Welch algorithm, and repeated iteration is carried out until convergence is achieved, so that hidden Markov model parameters of mixed gaussians under the condition of different people numbers are obtained;
(3) the real-time personnel counting method comprises the following steps: and receiving data in real time to calculate a link attenuation value, comparing with GMM-HMM models with different parameters, taking a model class with the maximum likelihood as a corresponding people number class, and counting the passing people number.
2. The pedestrian flow monitoring method based on the hidden markov model according to claim 1, wherein: the wireless sensing network system comprises an experiment door frame, sensor nodes, a sink node and a PC.
3. The hidden markov model-based human traffic monitoring method according to claim 1, wherein the "transmission-reception-storage" mode is:
(1) when the power supply is switched on, 18 wireless sensing network nodes on the gate-shaped frame periodically transmit the radio frequency signal intensity of a link formed by each node in a polling mode;
(2) and the periodic radio frequency signal sent by the sensor node is received by a sink node connected to the PC and is transmitted to the PC through the serial assistant to be stored as a txt text.
4. The pedestrian flow monitoring method based on the hidden markov model according to claim 1, wherein: and continuously monitoring in real time, collecting real-time data, calculating a real-time attenuation value, comparing the attenuation value with each model, and taking the model class with the maximum likelihood as the corresponding people number class to carry out people flow statistics.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910204501.4A CN109951866B (en) | 2019-03-18 | 2019-03-18 | People flow monitoring method based on hidden Markov model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910204501.4A CN109951866B (en) | 2019-03-18 | 2019-03-18 | People flow monitoring method based on hidden Markov model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109951866A CN109951866A (en) | 2019-06-28 |
| CN109951866B true CN109951866B (en) | 2022-05-17 |
Family
ID=67008899
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910204501.4A Expired - Fee Related CN109951866B (en) | 2019-03-18 | 2019-03-18 | People flow monitoring method based on hidden Markov model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109951866B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112967353A (en) * | 2021-04-02 | 2021-06-15 | 福州大学 | Sparse radio frequency tomography method based on Gaussian mixture model |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104217592A (en) * | 2014-09-24 | 2014-12-17 | 福建星网锐捷网络有限公司 | People flow volume statistical method, apparatus and system |
| CN207650903U (en) * | 2017-12-25 | 2018-07-24 | 南京信息工程大学 | A kind of multiple port classroom passenger number statistical system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845318B (en) * | 2015-12-03 | 2019-06-21 | 杭州海康威视数字技术股份有限公司 | Passenger flow information acquisition method and device, passenger flow information processing method and processing device |
| US10346688B2 (en) * | 2016-01-12 | 2019-07-09 | Hitachi Kokusai Electric Inc. | Congestion-state-monitoring system |
-
2019
- 2019-03-18 CN CN201910204501.4A patent/CN109951866B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104217592A (en) * | 2014-09-24 | 2014-12-17 | 福建星网锐捷网络有限公司 | People flow volume statistical method, apparatus and system |
| CN207650903U (en) * | 2017-12-25 | 2018-07-24 | 南京信息工程大学 | A kind of multiple port classroom passenger number statistical system |
Non-Patent Citations (3)
| Title |
|---|
| "A Statistical Method for People Counting in Crowded Environments";M. Bozzoli, L. Cinque and E. Sangineto;《14th International Conference on Image Analysis and Processing (ICIAP 2007)》;20071029;全文 * |
| 客流量监测的WiFi嗅探网格化方法;佟慧姣等;《电子测量技术》;20180423;第41卷(第08期);全文 * |
| 监控场景中人数统计算法的研究与应用;马海军;《硕士论文》;20161231;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109951866A (en) | 2019-06-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109686109B (en) | Parking lot safety monitoring management system and method based on artificial intelligence | |
| CN108921039A (en) | The forest fire detection method of depth convolution model based on more size convolution kernels | |
| CN108764059B (en) | Human behavior recognition method and system based on neural network | |
| CN109635483A (en) | A kind of motor and failure of pump hypothesis analysis system based on electromagnetic detection | |
| CN105100724A (en) | Remote and safe intelligent household monitoring method and device based on visual analysis | |
| CN106372576A (en) | Deep learning-based intelligent indoor intrusion detection method and system | |
| CN107403154A (en) | A kind of gait recognition method based on dynamic visual sensor | |
| CN110070530A (en) | A kind of powerline ice-covering detection method based on deep neural network | |
| CN109600758B (en) | RSS-based people flow monitoring method | |
| CN113642403B (en) | Crowd abnormal intelligent safety detection system based on edge calculation | |
| CN102254394A (en) | Antitheft monitoring method for poles and towers in power transmission line based on video difference analysis | |
| CN114423034B (en) | Indoor personnel action recognition method, system, medium, equipment and terminal | |
| Saleem et al. | Smart spaces: occupancy detection using adaptive back-propagation neural network | |
| CN108334902A (en) | A kind of track train equipment room smog fireproof monitoring method based on deep learning | |
| CN104159088A (en) | System and method of remote monitoring of intelligent vehicle | |
| CN109583499A (en) | A kind of transmission line of electricity target context categorizing system based on unsupervised SDAE network | |
| CN109951866B (en) | People flow monitoring method based on hidden Markov model | |
| CN113627326B (en) | Behavior recognition method based on wearable equipment and human skeleton | |
| CN105160285A (en) | Method and system for recognizing human body tumble automatically based on stereoscopic vision | |
| CN110428617A (en) | A kind of traffic object recognition methods based on 5G Portable intelligent terminal and MEC | |
| Lin et al. | Posting techniques in indoor environments based on deep learning for intelligent building lighting system | |
| CN116443682B (en) | Intelligent elevator control system | |
| CN117520664A (en) | Public opinion detection method and system based on graphic neural network | |
| Yaman et al. | New approach for intelligent street lights using computer vision and wireless sensor networks | |
| CN116931524A (en) | Intelligent monitoring system and process for building |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220517 |