CN110751115B - Non-contact human behavior identification method and system - Google Patents
Non-contact human behavior identification method and system Download PDFInfo
- Publication number
- CN110751115B CN110751115B CN201911019219.5A CN201911019219A CN110751115B CN 110751115 B CN110751115 B CN 110751115B CN 201911019219 A CN201911019219 A CN 201911019219A CN 110751115 B CN110751115 B CN 110751115B
- Authority
- CN
- China
- Prior art keywords
- behavior
- human
- cloud platform
- human body
- behaviors
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
Abstract
The invention discloses a non-contact human behavior identification method and a system, wherein the method comprises the following steps: continuously receiving the channel state information data value sent by the behavior recognition device, filtering the channel state information data value with a preset time length, and filtering out an interference data value; and performing pre-classification operation on the filtered channel state information data values to obtain a data set corresponding to the human behavior, and matching the data set by using a wireless fingerprint database set to obtain the human behavior. According to the scheme of the invention, the human behavior can be recognized without using any video image equipment or wearable sensor equipment, the human behavior recognition is realized without being limited by factors such as user privacy and illumination conditions, the user does not need to wear the sensor at any time, and the human behavior recognition is realized without being influenced by the problems of high energy consumption and resource waste under frequent information interaction, the complex operation is not needed, the convenience is realized, and the user experience is greatly improved.
Description
Technical Field
The invention relates to the field of intelligent equipment, in particular to a non-contact human behavior identification method and a non-contact human behavior identification system.
Background
With the continuous development of smart homes, developers want smart home devices in homes to automatically recognize human behavior states of users (such as nobody, walking, standing, sitting, sleeping, falling, running, etc.) and predict next user behaviors according to the behaviors of the users at the moment, for example: the intelligent temperature controller can identify the behavior state of a user around the equipment at the moment, so that intelligent services of turning on an air conditioner, turning off the air conditioner, increasing or reducing the air speed and temperature of the air conditioner and the like are provided for the user.
Currently, the method generally adopted for judging the behavior state of the user is as follows: determining the user behavior state is achieved by using a video image device or a wearable sensor device. However, the behavior recognition technology based on the video image is limited by factors such as user privacy and illumination conditions, and the scheme based on the wearable sensor not only requires the user to wear the sensor at any time, but also is affected by the problems of high energy consumption and resource waste under frequent information interaction, and is very inconvenient.
Disclosure of Invention
In view of the above problems, the present invention provides a method and a system for contactless human behavior recognition, which solves the above problems.
In order to solve the above problem, an embodiment of the present invention provides a contactless human behavior identification method, which is applied to a cloud platform, where the cloud platform is connected to a behavior identification device, the cloud platform is configured with a wireless fingerprint database set, and the wireless fingerprint database set is a database set corresponding to human behaviors generated according to big data, and the method includes:
continuously receiving a channel state information data value sent by the behavior recognition device, wherein the channel state information data value is used for representing data corresponding to human body behaviors;
filtering the channel state information data value with a preset time length to filter out an interference data value;
performing a pre-classification operation on the filtered channel state information data values to obtain a data set corresponding to human behaviors, wherein the pre-classification operation is an operation of dividing human body similar behaviors into a same behavior state group;
and matching the data set by utilizing the wireless fingerprint database set to obtain the human body behavior.
Optionally, the human behavior comprises: the behavior of no person, walking, standing, sitting, sleeping, falling down and running is pre-classified according to the channel state information data value after filtering processing, and a data set corresponding to the human body behavior is obtained, and the method comprises the following steps:
dividing the filtered channel state information data values into subsequences with preset number elements, and calculating standard variance values of the subsequences;
and pre-classifying by using the standard variance value, dividing sleeping and falling behaviors into one group, dividing running and walking behaviors into one group, and dividing standing, sitting and unmanned behaviors into one group, thereby obtaining a data set corresponding to human behaviors.
Optionally, matching the data set by using the wireless fingerprint database set to obtain human body behaviors, including:
subdividing the standard variance values of all the groups by using a classification algorithm, and accurately identifying a data set corresponding to the human body behaviors obtained by pre-classification;
and matching the data set obtained by accurate identification by utilizing the wireless fingerprint database set so as to accurately obtain the human body behavior.
Optionally, the cloud platform is further configured with a human behavior law database, the human behavior law database is a database set corresponding to human behavior laws generated according to big data, and after the data set is matched by using the wireless fingerprint database set to obtain human behaviors, the method further includes:
and obtaining the next behavior of the human body through the currently identified human body behavior, the peripheral parameters and the human body behavior rule database, wherein the peripheral parameters are parameters corresponding to the data of the environment, the position and the time of the human body.
And sending a control instruction to the related equipment according to the next behavior of the human body so as to control the running state of the related equipment which needs to be operated according to the next behavior of the human body.
Optionally, obtaining a next behavior of the human body through the currently recognized human body behavior, the peripheral parameters, and the human body behavior rule database, where the next behavior of the human body includes:
and associating the human body behavior with the current time, the human body position and the environment state by using the classification algorithm, and combining the human body behavior rule database after association to obtain the next behavior of the human body.
Optionally, after the data set is matched by using the wireless fingerprint database set to obtain the human behavior, the method further includes:
and sending an alarm signal to the behavior recognition device under the condition that the obtained human body behavior is the preset alarm behavior, so that the behavior recognition device sends alarm information.
The embodiment of the invention also provides a non-contact human behavior recognition method, which is applied to a behavior recognition device, wherein the behavior recognition device is connected with a cloud platform, and the method comprises the following steps:
the behavior recognition device acquires a channel state information data value in a wireless signal after transmitting the wireless signal, wherein the channel state information data value is used for representing data corresponding to human body behaviors;
and the behavior recognition device sends the channel state information data value to the cloud platform so as to enable the cloud platform to recognize human body behaviors.
Optionally, the acquiring, by the behavior recognition device, a channel state information data value in the wireless signal after the wireless signal is transmitted includes:
the behavior recognition device collects wireless signals after transmitting the wireless signals, performs perimeter division on the collected wireless signals, and selects wireless signals in a preset area;
the behavior recognizing device acquires a channel state information data value in a wireless signal of a selected preset area.
Optionally, after the behavior recognition device sends the channel state information data value to the cloud platform, so that the cloud platform performs human behavior recognition, the method further includes:
and the behavior recognition device receives the alarm signal of the cloud platform and sends alarm information to a preset emergency number.
The embodiment of the invention also provides a non-contact human behavior recognition system, which comprises: the cloud platform is connected with the behavior recognition device, the cloud platform is used for executing any one of the methods, and the behavior recognition device is used for executing any one of the methods.
By adopting the non-contact human behavior identification method provided by the invention, the cloud platform continuously receives the channel state information data value acquired by the behavior identification device, filters the channel state information data value with the preset time length, filters out the interference data value, performs pre-classification operation on the filtered channel state information data value to obtain a data set corresponding to the human behavior, and finally matches the data set by utilizing the wireless fingerprint database set to obtain the human behavior. The invention can realize the identification of the human behavior without using any video image equipment or wearable sensor equipment, has no problems generated by the prior art, can realize the identification of the human behavior only by using one behavior identification device indoors, does not need complex operation, is very convenient, and greatly improves the use feeling of users.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a flow chart of a method for contactless human behavior recognition according to an embodiment of the present invention;
FIG. 2 is a flow chart of another method for contactless human behavior recognition according to an embodiment of the present invention;
fig. 3 is an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below. It should be understood that the specific embodiments described herein are merely illustrative of the invention, but do not limit the invention to only some, but not all embodiments.
The inventor finds that the existing smart home devices recognize the human behavior state of the user (such as nobody, walking, standing, sitting, sleeping, falling, running, etc.) and need to be realized by using a video image device or a wearable sensor device, but the traditional recognition schemes have respective disadvantages, such as: the behavior recognition technology based on the video images is limited by factors such as user privacy and illumination conditions, and the scheme based on the wearable sensor not only requires a user to wear the sensor at any time, but also is affected by the problems of high energy consumption and resource waste under frequent information interaction, so that the behavior recognition technology is very inconvenient and the experience of the user is not good.
Based on the problems, the inventor creatively provides a method for combining the behavior recognition device and the cloud platform big data through deep research, and realizes that the human body behavior can be recognized only by using one behavior recognition device indoors and matching with the cloud platform. The invention is explained and illustrated in detail below.
Referring to fig. 1, a flow chart of the non-contact human behavior recognition method of the present invention is shown. The method is applied to a cloud platform, the cloud platform is connected with a behavior recognition device, the cloud platform is configured with a wireless fingerprint database set, the wireless fingerprint database set is a database set corresponding to the human behavior generated according to big data, and the non-contact human behavior recognition method specifically comprises the following steps:
step 101: and continuously receiving the channel state information data value sent by the behavior recognition device, wherein the channel state information data value is used for representing data corresponding to the human body behavior.
In the embodiment of the invention, a cloud platform and behavior recognition devices perform data interaction through a network, wherein the behavior recognition devices are installed indoors, a user can select one or more behavior recognition devices to be installed at proper positions respectively according to the requirements of the user, the behavior recognition devices mainly play the roles of transmitting wireless (WiFi) signals and collecting WiFi signals, acquiring Channel State Information (CSI) data values in the collected WiFi signals and sending the CSI data values to the cloud platform, the CSI data values are used for representing data corresponding to human body behaviors, the CSI can be changed by the indoor behaviors of the human body, and corresponding action information can be extracted from the CSI. The cloud platform is configured with a wireless fingerprint database set, wherein the wireless fingerprint database set is a database set corresponding to human body behaviors generated by the cloud platform according to big data, namely the wireless fingerprint database set is a database set formed by CSI data values corresponding to various behaviors of a human body.
When the behavior recognition device starts to work, the obtained CSI data value is continuously sent to the cloud platform, and the cloud platform continuously receives the CSI data value sent by the behavior recognition device.
Step 102: and filtering the channel state information data value with the preset time length to filter out an interference data value.
In the embodiment of the invention, after the cloud platform continuously receives the CSI data value for the preset time, the CSI data value with the preset time length is filtered, and the interference data value is filtered.
Because human body behaviors are generally formed by a series of consecutive actions, if the CSI data value at each time point (for example, one time point every 1 second) is filtered and subsequently processed, a huge data computation amount is caused, which results in lower working efficiency of the cloud platform, and the method of processing once every preset time length is adopted, which meets the requirement of subsequent data processing work and improves the working efficiency of the cloud platform.
Step 103: and performing pre-classification operation on the filtered channel state information data values to obtain a data set corresponding to human behaviors, wherein the pre-classification operation is an operation of dividing human body similar behaviors into the same behavior state group.
In the embodiment of the invention, after filtering, the cloud platform performs a pre-classification operation on the filtered CSI data values to obtain a data set corresponding to human behaviors, wherein the pre-classification operation is an operation of dividing the human body similar behaviors into the same behavior state group, and because the CSI data values corresponding to the human body similar behaviors are relatively close, the CSI data values corresponding to the similar behaviors are divided into the same behavior state group within a preset time length, and the CSI data values corresponding to different behaviors form a plurality of behavior state groups.
Optionally, the human behavior comprises: the behavior of no person, walking, standing, sitting, sleeping, falling and running, step 103 specifically includes:
step s 1: and dividing the filtered channel state information data values into subsequences with preset number elements, and calculating the standard variance value of the subsequences.
In the embodiment of the invention, the human body behaviors comprise: the behavior of no person, walking, standing, sitting, sleeping, falling, and running, which are different corresponding to CSI data values.
The simple example is as follows: assuming that the CSI data value corresponding to the unmanned behavior is 100, the CSI data value corresponding to the station behavior is 90, the CSI data value corresponding to the walking behavior is 60, and the CSI data value corresponding to the running behavior is 30, if the human body performs the behavior from the station to the running within the preset time, the corresponding CSI data value is changed from 90 to 30. It should be noted that, because the human body behavior is continuous, the CSI data value corresponding to the behavior from station to run may be a smooth decreasing change, a smooth plus linear change, a linear decreasing change, or the like, so that there are many different CSI data values obtained within a preset time, and these data need to be processed.
The specific processing method of the cloud platform for the data is as follows: dividing the filtered CSI data value into subsequences with a preset number of elements, and calculating a standard variance value of the subsequences, for example, dividing the filtered CSI data value into subsequences with 50 elements, and then calculating the standard variance value of the subsequences, where the smaller the variance of the calculation result, the CSI data values corresponding to the 50 subsequences may be naturally divided into the same behavior state group, and of course, if more accurate behavior recognition is required, more element subsequences may be divided, and the corresponding cloud platform computation amount may also become higher.
Step s 2: the standard variance values are used for pre-classification, sleeping and falling behaviors are divided into one group, running and walking behaviors are divided into one group, standing, sitting and unmanned behaviors are divided into one group, and therefore a data set corresponding to human behaviors is obtained.
In the embodiment of the invention, after the standard variance value of the CSI data values of the preset time length is calculated, the calculation results are utilized to perform pre-classification, the CSI data values corresponding to sleeping and falling human behaviors are divided into one group, the CSI data values corresponding to walking and falling human behaviors are divided into one group, the CSI data values corresponding to standing, sitting and unmanned human behaviors are divided into one group, and thus the data sets corresponding to human behaviors are obtained. Of course, such grouping may be adjustable, and the embodiment of the present invention does not limit this.
Step 104: and matching the data sets by using a wireless fingerprint database set to obtain the human body behaviors.
In the embodiment of the invention, after the data set corresponding to the human behavior is obtained, the cloud platform matches the data set corresponding to the human behavior by using the existing wireless fingerprint database set so as to obtain the state of the human behavior.
Optionally, step 104 specifically includes:
step t 1: subdividing the standard variance values of all the groups by using a classification algorithm, and accurately identifying a data set corresponding to the human body behaviors obtained by pre-classification;
in the embodiment of the invention, after the data set corresponding to the human body behaviors is obtained, the cloud platform subdivides the standard variance values of all the groups in the data set by using the classification algorithm so as to achieve the purpose of accurately identifying the data set corresponding to the human body behaviors obtained through pre-classification, namely, the classification algorithm is used for calculating the standard variance values of all the groups in the data set again, and the CSI data values roughly grouped during pre-classification are further accurately subdivided into specific CSI data values. For example: the human body behaviors of standing, sitting and unmanned people are divided into one group during pre-classification, although the corresponding CSI data values are close to each other and still have certain differences, the specific corresponding human body behaviors of the CSI data values in the group can be definitely determined by using a classification algorithm, and of course, the group may only correspond to one human body behavior or to a plurality of human body behaviors.
Step t 2: and matching the data set obtained by accurate identification by utilizing a wireless fingerprint database set so as to accurately obtain the human body behavior.
In the embodiment of the invention, after the specific CSI data values in each group are obtained, the cloud platform matches the data sets obtained by accurate identification by using the wireless fingerprint database set, namely, the cloud platform matches the CSI data values subdivided by each group by using the wireless fingerprint database set, and finally the state of human behavior is accurately obtained.
In addition, the identification method of the embodiment of the invention is further provided with an alarm function, for example, the fallen human behavior is preset as an alarm behavior for sending alarm information, and then when the cloud platform identifies that the human behavior is the preset alarm behavior: and under the condition of falling, the cloud platform sends an alarm signal to the behavior recognition device so that the behavior recognition device sends alarm information.
Optionally, the cloud platform is further configured with a human behavior law database, the human behavior law database is a database set corresponding to human behavior laws generated according to big data, and after the data set is matched by using the wireless fingerprint database set to obtain human behaviors, the non-contact human behavior identification method further includes:
step u 1: and obtaining the next behavior of the human body through the currently identified human body behavior, the peripheral parameters and the human body behavior rule database, wherein the peripheral parameters are parameters corresponding to the data of the environment, the position and the time of the human body.
In the embodiment of the invention, the cloud platform is also provided with a human behavior law database, the human behavior law database is a database set corresponding to the human behavior laws generated according to the big data, and after the current human behavior is identified, the next behavior of the human body is obtained through prediction by combining the peripheral parameters and the human behavior law database. Specifically, the classification algorithm is used for correlating the human body behavior with the current time, the human body position and the environment state, and the next behavior of the human body is obtained by combining the human body behavior rule database after the correlation.
As an example: if the person arrives at home after 19:00 a working day in summer, the person stands in the entrance hall, opens the door wardrobe, hangs clothes, moves to the living room and turns on the television and the air conditioner. After a certain time, the cloud platform generates a human behavior rule database of A, and if the cloud platform recognizes that A appears and stands in the entrance hall in a working day 19:08, the cloud platform: a lobby; the current time: 19: 08; the environment is as follows: summer; and a human behavior rule database of A, predicting that the next human behavior of A is to move to the living room.
Step u 2: and sending a control instruction to the related equipment according to the next behavior of the human body so as to control the running state of the related equipment which needs to be operated according to the next behavior of the human body.
In the embodiment of the invention, after the next behavior of the human body is predicted, the cloud platform sends the control instruction to the relevant equipment so as to control the running state of the relevant equipment which needs to be operated by the next behavior of the human body.
Following the above example: and D, the cloud platform predicts that the next human behavior of the A is to move to the living room, and then the cloud platform sends a starting instruction to the television and the air conditioner of the living room, so that the television and the air conditioner start to operate.
It should be noted that if the workday 19:08 cloud platform recognizes that a appears and sits in the lobby, then according to the position of a: a lobby; the current time: 19: 08; the environment is as follows: summer; and the human behavior rule database of A, the cloud platform can not predict the next human behavior of A, then it can not send any instruction, but only continue to identify the human behavior of A, if A has the preset alarm behavior in the subsequent human behavior: and if the vehicle falls down, the cloud platform sends an alarm signal to the behavior recognition device so that the behavior recognition device sends alarm information.
Referring to fig. 2, another contactless human behavior recognition method according to an embodiment of the present invention is shown, and is applied to a behavior recognition device, where the behavior recognition device is connected to a cloud platform, and the contactless human behavior recognition method includes:
step 201: the behavior recognition device acquires a channel state information data value in the wireless signal after transmitting the wireless signal, wherein the channel state information data value is used for representing data corresponding to human body behaviors;
step 202: and the behavior recognition device sends the channel state information data value to the cloud platform so that the cloud platform can recognize human body behaviors.
In the embodiment of the invention, the behavior recognition device transmits WiFi signals, then WiFi signals reflected by objects or human bodies are collected, and then CSI data values are obtained from the collected WiFi signals and sent to the cloud platform, so that the cloud platform can recognize human body behaviors.
Wherein, step 201 specifically includes:
step X1: the behavior recognition device collects the wireless signals after transmitting the wireless signals, performs perimeter division on the collected wireless signals, and selects the wireless signals in a preset area;
step X2: the behavior recognizing device acquires a channel state information data value in the wireless signal of the selected preset area.
In the embodiment of the invention, the behavior recognition device transmits the wireless signals and then collects the wireless signals, and the collected wireless signals are subjected to perimeter division to select the wireless signals in the preset area.
Because the behavior recognition device is installed indoors, but is limited to factors such as self power, it is impossible to achieve full house coverage, especially in the case of shelters. For example: the user A lives in a house with four rooms and two rooms, the installed behavior recognition device of the living room cannot collect WiFi signals reflected in the bedroom, but if the behavior recognition device or other WiFi transmitting equipment is also installed in the bedroom, the behavior recognition device of the living room may collect WiFi signals transmitted by the behavior recognition device of the bedroom or other WiFi transmitting equipment, interference influence is generated on the behavior recognition device of the living room by the WiFi signals, therefore, after the behavior recognition device of the living room collects the WiFi signals, perimeter division needs to be carried out on the collected WiFi signals firstly, the WiFi signals in the area of the living room are selected, interference signals are eliminated, and accuracy of recognition of human behavior by the behavior recognition device of the living room is guaranteed. Similarly, the behavior recognition device installed in a room such as a bedroom or a restaurant needs to be divided into a plurality of zones.
After the WiFi signals in the preset area are selected by the behavior recognition device, the CSI data values in the WiFi signals are acquired by the behavior recognition device, and then the CSI data values are sent to the cloud platform.
Of course, it can be understood that after the behavior recognition device receives the alarm signal of the cloud platform, an alarm sound may be sent out, meanwhile, an emergency number, for example, a telephone number of an emergency contact person, may be preset in the behavior recognition device, and after the behavior recognition device receives the alarm signal of the cloud platform, an alarm message may be sent out to the emergency number, and the emergency contact person may receive the alarm message and perform subsequent processing.
In summary, in the embodiment of the present invention, the behavior recognition device transmits the WiFi signals, and then collects the WiFi signals reflected by the object or the human body, first performs perimeter division on the collected WiFi signals, selects the WiFi signals in the preset area, then obtains the CSI data values in the WiFi signals, and then continuously sends the CSI data values to the cloud platform.
The cloud platform can continuously receive the CSI data value sent by the behavior recognition device, after the CSI data value is continuously received for a preset time, the CSI data value with the preset time length is filtered, the interference data value is filtered, and after the filtering, the cloud platform performs a pre-classification operation on the filtered CSI data value: dividing the filtered channel state information data values into subsequences with preset number elements, calculating standard variance values of the subsequences, after the standard variance values of the CSI data values of preset time length are calculated, performing pre-classification by using a calculation result, dividing the CSI data values corresponding to sleeping and falling human behaviors into a group, dividing the CSI data values corresponding to running and walking human behaviors into a group, and dividing the CSI data values corresponding to standing, sitting and unmanned human behaviors into a group, so that a data set corresponding to the human behaviors is obtained.
After the data set corresponding to the human body behaviors is obtained, the cloud platform subdivides the standard variance values of all groups in the data set by using a classification algorithm so as to achieve the purpose of accurately identifying the data set corresponding to the human body behaviors obtained through pre-classification, and after the specific CSI data values in all groups are obtained, the cloud platform matches the CSI data values subdivided by all groups by using a wireless fingerprint database set, so that the state of the human body behaviors is accurately obtained finally.
If the fallen human body behavior is preset as the alarm behavior for sending alarm information, when the cloud platform identifies that the human body behavior is the preset alarm behavior: and when the mobile terminal falls down, the cloud platform sends an alarm signal to the behavior recognition device, after the behavior recognition device receives the alarm signal of the cloud platform, the mobile terminal can send an alarm sound and an alarm message to the emergency number, and the emergency contact can receive the alarm message and perform subsequent processing.
In addition, after the data sets are matched by using the wireless fingerprint database set to obtain the human body behaviors, the human body behaviors are associated with the current time, the human body position and the environment state by using a classification algorithm, the next behavior of the human body is obtained by combining the human body behavior rule database after the association, and the cloud platform sends a control instruction to the related equipment after predicting the next behavior of the human body so as to control the running state of the related equipment which needs to be operated by the next behavior of the human body.
Referring to fig. 3, a contactless human behavior recognition system according to another embodiment of the present invention is shown, and the system includes: cloud platform and action recognition device, the cloud platform is connected with action recognition device, and the cloud platform includes: the system comprises a behavior identification module and a behavior analysis module, wherein the behavior identification module is used for identifying the state of the specific human behavior of the user, the behavior analysis module is used for predicting the behavior of the user after the current human behavior, and the cloud platform is used for executing any one of the methods; the behavior recognition apparatus includes: the wireless signal transmitting and receiving module is used for transmitting a wireless WiFi signal, collecting the WiFi signal to obtain a CSI data value and sending the CSI data value to the cloud platform, the abnormal behavior alarming module is used for sending alarming information, and the behavior recognition device is used for executing any one of the methods.
Through the embodiment, the non-contact human body behavior identification method and the non-contact human body behavior identification system, the behavior identification device collects the CSI data value of the WiFi signal and continuously sends the CSI data value to the cloud platform, the cloud platform continuously receives the CSI data value collected by the behavior identification device, filters the CSI data value with the preset time length, filters out interference data values, performs pre-classification operation on the filtered CSI data value to obtain a data set corresponding to the human body behavior, finally matches the data set by using the wireless fingerprint database set to obtain the human body behavior, and sends an alarm signal to the behavior identification device when the human body behavior is identified to be the preset alarm behavior. In addition, the cloud platform obtains the next behavior of the human body through the currently recognized human body behavior, the peripheral parameters and the human body behavior rule database, and sends a control instruction to the relevant equipment according to the next behavior of the human body so as to control the running state of the relevant equipment which needs to be operated by the next behavior of the human body.
The human behavior recognition method and the human behavior recognition system can realize the recognition of the human behavior without using any video image equipment or wearable sensor equipment, are not limited by factors such as user privacy, illumination conditions and the like, do not need to wear sensors at any time, are not influenced by the problems of high energy consumption and resource waste under frequent information interaction, can realize the recognition of the human behavior, do not need complex operation, are very convenient, and greatly improve the experience of users.
In addition, the behavior recognition device of the embodiment of the invention can be expanded, and a behavior recognition module and a behavior analysis module can be arranged on the behavior recognition device, so that the behavior recognition device can directly complete human behavior recognition in a local area network without being connected with a cloud platform. In addition, the human behavior recognition scheme of the embodiment of the invention can also be matched with the existing traditional human behavior recognition scheme for interaction, thereby achieving a more ideal recognition effect and providing better service for users.
It is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, or article that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, or article.
The foregoing detailed description of the embodiments of the present invention has been presented for purposes of illustration and description, and is intended to be exemplary only and is not intended to be exhaustive or to limit the invention to the precise forms disclosed; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (9)
1. A non-contact human behavior identification method is applied to a cloud platform, the cloud platform is connected with a behavior identification device, the cloud platform is configured with a wireless fingerprint data base set, the wireless fingerprint data base set is a data base set corresponding to human behaviors generated according to big data, and the method comprises the following steps:
continuously receiving a channel state information data value sent by the behavior recognition device, wherein the channel state information data value is used for representing data corresponding to human body behaviors;
filtering the channel state information data value with a preset time length to filter out an interference data value;
dividing the filtered channel state information data values into subsequences with preset number elements, calculating standard variance values of the subsequences, and dividing the channel state information data values corresponding to the subsequences into the same behavior state group if the variance of the calculation result is smaller;
performing a pre-classification operation by using the standard variance value to obtain a data set corresponding to human behaviors, wherein the pre-classification operation is an operation of dividing human similar behaviors into a same behavior state group;
matching the data set by using the wireless fingerprint database set to obtain human body behaviors;
wherein the human behavior comprises: the behavior of nobody, walking, standing, sitting, sleeping, falling down and running, the standard variance value is utilized to carry out the pre-classification operation, and a data set corresponding to the human behavior is obtained, and the method comprises the following steps:
pre-classifying by using the standard variance value, dividing sleeping and falling behaviors into one group, dividing running and walking behaviors into one group, and dividing standing, sitting and unmanned behaviors into one group so as to obtain a data set corresponding to human behaviors;
and matching the data set by utilizing the wireless fingerprint database set to obtain human body behaviors, wherein the steps comprise:
and subdividing the standard variance values of all the groups by using a classification algorithm, and accurately identifying a data set corresponding to the human body behaviors obtained by pre-classification.
2. The method of claim 1, wherein matching the data set using the wireless fingerprint database set to obtain human behavior comprises:
and matching the data set obtained by accurate identification by utilizing the wireless fingerprint database set so as to accurately obtain the human body behavior.
3. The method of claim 1, wherein the cloud platform is further configured with a human behavior law database, the human behavior law database is a database set corresponding to human behavior laws generated according to big data, and after the data set is matched by using the wireless fingerprint database set to obtain human behaviors, the method further comprises:
obtaining the next behavior of the human body through the currently identified human body behavior, the peripheral parameters and the human body behavior rule database, wherein the peripheral parameters are parameters corresponding to data of the environment, the position and the time of the human body;
and sending a control instruction to the related equipment according to the next behavior of the human body so as to control the running state of the related equipment which needs to be operated according to the next behavior of the human body.
4. The method of claim 1, wherein obtaining the next behavior of the human body from the currently identified human body behavior, the peripheral parameters and the human body behavior rule database comprises:
and associating the human body behavior with the current time, the human body position and the environment state by using the classification algorithm, and combining the human body behavior rule database after association to obtain the next behavior of the human body.
5. The method of claim 1, wherein after matching the data set using the wireless fingerprint database set to obtain human behavior, the method further comprises:
and sending an alarm signal to the behavior recognition device under the condition that the obtained human body behavior is the preset alarm behavior, so that the behavior recognition device sends alarm information.
6. A non-contact human behavior recognition method is applied to a behavior recognition device, the behavior recognition device is connected with a cloud platform, and the method comprises the following steps:
the behavior recognition device acquires a channel state information data value in a wireless signal after transmitting the wireless signal, wherein the channel state information data value is used for representing data corresponding to human body behaviors;
the behavior recognition device sends the channel state information data value to the cloud platform so that the cloud platform can recognize human body behaviors;
wherein, the cloud platform disposes wireless fingerprint database set, wireless fingerprint database set is the database set that the human behavior that generates according to the big data corresponds, the cloud platform carries out human behavior identification, include:
the cloud platform continuously receives the channel state information data value sent by the behavior recognition device;
the cloud platform filters the channel state information data value with a preset time length to filter out an interference data value;
the cloud platform divides the filtered channel state information data values into subsequences with preset number elements, calculates standard variance values of the subsequences, and divides the channel state information data values corresponding to the subsequences into the same behavior state group if the variance of the calculation result is smaller;
the cloud platform performs a pre-classification operation by using the standard variance value to obtain a data set corresponding to human behaviors, wherein the pre-classification operation is an operation of dividing human similar behaviors into a same behavior state group;
the cloud platform matches the data set by using the wireless fingerprint database set to obtain human body behaviors;
wherein the human behavior comprises: unmanned, walk, stand, sit, sleep, fall and the action of running, the cloud platform utilizes standard variance value carries out the operation of presorting, obtains the data set that human action corresponds, includes:
the cloud platform performs pre-classification by using the standard variance value, divides sleeping and falling behaviors into one group, divides running and walking behaviors into one group, and divides standing, sitting and unmanned behaviors into one group, so as to obtain a data set corresponding to human behaviors;
the cloud platform utilizes the wireless fingerprint database set to match the data set to obtain human body behaviors, and the method comprises the following steps:
the cloud platform subdivides the standard variance values of all the groups by using a classification algorithm, and accurately identifies the data sets corresponding to the human behaviors obtained through pre-classification.
7. The method of claim 6, wherein obtaining the channel state information data value in the wireless signal after the behavior recognition device transmits the wireless signal comprises:
the behavior recognition device collects wireless signals after transmitting the wireless signals, performs perimeter division on the collected wireless signals, and selects wireless signals in a preset area;
the behavior recognizing device acquires a channel state information data value in a wireless signal of a selected preset area.
8. The method according to claim 6, wherein after the behavior recognition device transmits the channel state information data value to the cloud platform so that the cloud platform performs human behavior recognition, the method further comprises:
and the behavior recognition device receives the alarm signal of the cloud platform and sends alarm information to a preset emergency number.
9. A contactless human behavior recognition system, the system comprising: a cloud platform connected to the behavior recognition device, the cloud platform being configured to perform any of the methods of claims 1-5, and the behavior recognition device being configured to perform any of the methods of claims 6-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911019219.5A CN110751115B (en) | 2019-10-24 | 2019-10-24 | Non-contact human behavior identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911019219.5A CN110751115B (en) | 2019-10-24 | 2019-10-24 | Non-contact human behavior identification method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110751115A CN110751115A (en) | 2020-02-04 |
| CN110751115B true CN110751115B (en) | 2021-01-01 |
Family
ID=69279713
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911019219.5A Active CN110751115B (en) | 2019-10-24 | 2019-10-24 | Non-contact human behavior identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110751115B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112733930B (en) * | 2021-01-07 | 2022-10-18 | 北京邮电大学 | Human behavior sensing system, method and storage medium |
| CN113177442B (en) * | 2021-04-12 | 2024-01-30 | 广东省科学院智能制造研究所 | Human behavior detection method and device based on edge calculation |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104951757A (en) * | 2015-06-10 | 2015-09-30 | 南京大学 | Action detecting and identifying method based on radio signals |
| CN106066995A (en) * | 2016-05-25 | 2016-11-02 | 西安交通大学 | A kind of wireless unbundling human body behavioral value algorithm |
| CN107479393A (en) * | 2017-08-17 | 2017-12-15 | 北京天平检验行有限公司 | A kind of intelligent domestic system based on big data |
| CN108875584A (en) * | 2018-05-23 | 2018-11-23 | 西北工业大学 | A kind of highly reliable user behavior recognition method based on wireless aware |
| CN108875597A (en) * | 2018-05-30 | 2018-11-23 | 浙江大学城市学院 | A kind of two layers of movable clustering recognition method towards large-scale dataset |
| CN108988968A (en) * | 2018-07-27 | 2018-12-11 | 河北工程大学 | Human behavior detection method, device and terminal device |
| CN109728863A (en) * | 2018-12-26 | 2019-05-07 | 河北工程大学 | Personnel activity's duration estimation method, device and terminal device |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106685590B (en) * | 2016-12-08 | 2019-10-29 | 浙江工业大学 | A kind of indoor human body based on channel state information and KNN is towards recognition methods |
| CN106792808A (en) * | 2016-12-08 | 2017-05-31 | 南京邮电大学 | Los path recognition methods under a kind of indoor environment based on channel condition information |
| CN106658590B (en) * | 2016-12-28 | 2023-08-01 | 南京航空航天大学 | Design and implementation of multi-person indoor environment state monitoring system based on WiFi channel state information |
| CN106875629B (en) * | 2017-03-07 | 2023-02-10 | 吉林省家人帮信息服务有限公司 | Home-based endowment system based on somatosensory network and control method thereof |
| CN109635837A (en) * | 2018-11-10 | 2019-04-16 | 天津大学 | A kind of carefree fall detection system of scene based on commercial wireless Wi-Fi |
| CN109670434B (en) * | 2018-12-13 | 2022-09-27 | 南京理工大学 | Sitting posture duration detection method based on wireless channel state information |
-
2019
- 2019-10-24 CN CN201911019219.5A patent/CN110751115B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104951757A (en) * | 2015-06-10 | 2015-09-30 | 南京大学 | Action detecting and identifying method based on radio signals |
| CN106066995A (en) * | 2016-05-25 | 2016-11-02 | 西安交通大学 | A kind of wireless unbundling human body behavioral value algorithm |
| CN107479393A (en) * | 2017-08-17 | 2017-12-15 | 北京天平检验行有限公司 | A kind of intelligent domestic system based on big data |
| CN108875584A (en) * | 2018-05-23 | 2018-11-23 | 西北工业大学 | A kind of highly reliable user behavior recognition method based on wireless aware |
| CN108875597A (en) * | 2018-05-30 | 2018-11-23 | 浙江大学城市学院 | A kind of two layers of movable clustering recognition method towards large-scale dataset |
| CN108988968A (en) * | 2018-07-27 | 2018-12-11 | 河北工程大学 | Human behavior detection method, device and terminal device |
| CN109728863A (en) * | 2018-12-26 | 2019-05-07 | 河北工程大学 | Personnel activity's duration estimation method, device and terminal device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110751115A (en) | 2020-02-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9618918B2 (en) | System and method for estimating the number of people in a smart building | |
| US10074383B2 (en) | Sound event detection | |
| CN109076310A (en) | The autonomous semantic marker of physical location | |
| CN110751115B (en) | Non-contact human behavior identification method and system | |
| CN109595757B (en) | Control method and device of air conditioner and air conditioner with control device | |
| JP2004164282A (en) | Personal behavior detection system | |
| JP4754800B2 (en) | Method and system for modeling user movement in an environment, and method and system for modeling the geometry of a sensor network in an environment | |
| CN113339965A (en) | Method and device for air conditioner control and air conditioner | |
| AU2019333044B2 (en) | Assisted creation of video rules via scene analysis | |
| US11501618B1 (en) | Security device with user-configurable motion detection settings | |
| US11734932B2 (en) | State and event monitoring | |
| CN113867171A (en) | Intelligent home system based on ZigBee wireless sensor network | |
| CN106325136A (en) | Linkage control method, linkage control system and air conditioner | |
| CN110805995A (en) | Control method, device, controller and storage medium for air conditioning equipment | |
| JP2018078428A (en) | Control information output system, device control system, device control method, and program | |
| CN110017573A (en) | A kind of automation HVAC system | |
| KR101340287B1 (en) | Intrusion detection system using mining based pattern analysis in smart home | |
| US11550276B1 (en) | Activity classification based on multi-sensor input | |
| CN110941189A (en) | Intelligent household system and control method thereof and readable storage medium | |
| CN110674694B (en) | Activity signal separation method based on commercial WiFi | |
| CN107179699B (en) | Intelligent household appliance system | |
| WO2018154711A1 (en) | Apparatus control device and apparatus control method | |
| CN111625794A (en) | Input method, operation control module, household appliance, system and storage medium | |
| CN105026900A (en) | Spectrometer for personal context | |
| WO2020212644A1 (en) | A home automation system |
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 |