CN112880660B - Fusion positioning system and method for WiFi and infrared thermal imager of intelligent building - Google Patents
Fusion positioning system and method for WiFi and infrared thermal imager of intelligent building Download PDFInfo
- Publication number
- CN112880660B CN112880660B CN202110027946.7A CN202110027946A CN112880660B CN 112880660 B CN112880660 B CN 112880660B CN 202110027946 A CN202110027946 A CN 202110027946A CN 112880660 B CN112880660 B CN 112880660B
- Authority
- CN
- China
- Prior art keywords
- wifi
- mobile
- personnel
- space
- positioning
- 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
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/005—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 with correlation of navigation data from several sources, e.g. map or contour matching
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
- G01C21/206—Instruments for performing navigational calculations specially adapted for indoor navigation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/33—Services specially adapted for particular environments, situations or purposes for indoor environments, e.g. buildings
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
The invention discloses a fusion positioning system and a fusion positioning method for WiFi and infrared thermal imagers of intelligent buildings. The method overcomes the defects that a single positioning system is inaccurate in positioning accuracy and cannot identify identity information by adopting a fusion positioning system, and can carry out real-time positioning and identity identification on mobile personnel in an intelligent building by using human body temperature information and WiFi access information on the premise of not invading privacy. The system disclosed by the invention is simple to realize, low in calculation complexity, has the advantages of practical application to sensors which are independent of complicated types and enter a building, establishes an intelligent building mobile personnel identification and positioning system, and ensures the identification and safety of mobile personnel in the building to a certain extent.
Description
Technical Field
The invention belongs to the field of intelligent buildings, and particularly relates to a fusion positioning system and method of WiFi and infrared thermal imager of an intelligent building.
Background
As the demand for location-based services in indoor environments continues to increase, the development of an indoor mobile personnel location system becomes an essential prerequisite for building intelligence. Under the great trend of continuous intellectualization of cities and buildings, the indoor positioning situation is more and more complex, the requirements for positioning and navigation are more and more increased in more occasions, and the timely positioning of indoor mobile personnel is more and more important for the life of human beings. The position of mobile personnel in the building is more and more important for people at present, and the mobile personnel can provide timely positioning and navigation services for consumers in a supermarket, utilize a positioning technology and provide equivalent marketing services based on the market. In hospitals, it is convenient to find nearby medical equipment in real time, which is a practical way to call quickly when necessary, and it is desirable to perform on-site monitoring for a particular patient to prevent accidents. High-risk chemical facilities are responsible for managing the field and preventing accidents and the like. Indoor mobile personnel location technology has been widely used in various industries. The traditional indoor positioning technology can face the privacy problems of unstable signals, inaccurate positioning precision and mobile personnel in buildings.
Disclosure of Invention
The invention aims to provide a fusion positioning system and a fusion positioning method of WiFi and infrared thermal imager of an intelligent building, which are used for solving the problems of unstable signals, low positioning precision and privacy of mobile personnel in the building in the prior art.
In order to achieve the purpose, the invention adopts the technical scheme that:
a fusion positioning method of WiFi and infrared thermal imager of an intelligent building comprises the following steps:
s1, dividing a space in a building into a static space and a dynamic space, wherein the dynamic space comprises a free space and a constrained space;
s2, acquiring a probe request packet sent by the mobile equipment and the fixed equipment and a probe response packet sent by the routing equipment for providing the communication service in the building by using the intelligent router, and calculating an initial position M (x, y) of the mobile personnel obtained by the WiFi positioning module according to the acquired probe request packet and the acquired probe response packet, wherein the initial position M obeys an expectation of mu 1 、μ 2 Variance is σ 1 2 、σ 2 2 A gaussian distribution P (x, y);
acquiring thermometer lattice data by using an infrared thermal imager, detecting moving personnel with complex background, performing morphological processing on the image, extracting the moving personnel, and acquiring the position P of the moving personnel obtained by a positioning module of the infrared thermal imager 2 (x,y)；
The position information and the acquisition time of the mobile personnel are acquired by using the infrared sensor, and the position P of the mobile personnel obtained by the infrared sensor positioning module is acquired 3 (x,y)；
S3, according to the static space and the dynamic space divided by the S1, obtaining M (x, y) and P by the S2 2 (x, y) and P 3 (x, y) obtaining accurate position information and moving rail of moving personnel in the building through particle filter fusionAnd (4) tracing.
Further, step S1 includes the steps of:
s101, selecting all doors, stairs and turning places in the space in the building as landmarks for space division according to the movement capability of moving personnel in different spaces, and dividing the space in the building into a static space and a dynamic space;
s102, dividing a teaching and research room, an office and a toilet into static spaces;
s103, dividing the stairs, the room doorways, the corridors and the vacant spaces into dynamic spaces; the dynamic space is divided into a constraint space and a free space, wherein stairs, room doorways and corridors are divided into the constraint space; the open space region is divided into free space.
Further, in step S2, calculating the initial position M (x, y) of the mobile personnel obtained by the WiFi positioning module includes the following steps:
S2A1, data acquisition: acquiring a probe request packet transmitted by a mobile device and a fixed device and a probe response packet transmitted by a routing device for providing communication service in the building through a plurality of intelligent routers arranged in an intelligent building, and analyzing currently acquired time information, a Mac address of the mobile device or the fixed device and a WiFi signal strength RSSI (received signal strength indicator) from the probe request packet;
S2A2, feature extraction and model training: calculating a WiFi signal trend index STI and a signal interval according to the WiFi signal strength RSSI acquired in the S2A1, taking the WiFi signal trend index STI, the signal interval and the WiFi signal strength RSSI as feature vectors, training a recurrent neural network model, and constructing a fingerprint database of the WiFi signal strength RSSI of the mobile equipment;
S2A3, establishing an indoor mobile personnel positioning model of human-scene interaction: the method comprises the steps of establishing a regression neural network model between WiFi signal strength RSSI and a physical position based on a fingerprint database of WiFi signal strength RSSI of mobile equipment established by S2A2, establishing a two-dimensional coordinate system by taking the southwest corner of a plane graph of an intelligent building as the origin of the coordinate system, the length of the plane graph as the x axis and the width as the y axis, taking the position of a mobile person in the coordinate system, namely the physical position of the person in the intelligent building, taking a group of WiFi signal strength RSSIs of the same mobile equipment or fixed equipment acquired by a plurality of intelligent routers as input, bringing the input RSSI into the trained regression neural network model, and outputting the initial position M (x, y) of the mobile person obtained by a WiFi positioning module.
Further, in S2A2,
the method for calculating the signal tendency index STI is as follows:
1) The method comprises the steps of recording WiFi signal strength RSSI (received signal strength indicator) of the same mobile equipment or fixed equipment collected by a plurality of intelligent routers into a group, and calculating an average value s of the RSSI avg ，
Wherein s is i For i groups of wifi signal strengths RSSI, s avg The average value of the i groups of wifi signal strength RSSI;
2) Calculating the Signal Trend index STI of the ith data i ，
STI i ＝||s i -s avg || (2)
The calculation method of the data interval is as follows:
interval i ＝||s i -s i+1 || (3)
wherein s is i+1 I +1 sets of wifi signal strengths RSSI.
Further, before S2A2, the abnormal value of the signal strength RSSI obtained in step S2A1 is deleted.
Further, in S2, a position P of the moving person obtained by the infrared thermal imager positioning module is obtained 2 The procedure for (x, y) is as follows:
S2B1, data acquisition: acquiring temperature chart data of indoor corridors of the intelligent building at intervals of set time by an infrared thermal imager which is arranged in the intelligent building in advance;
S2B2, detecting a moving person with a complex background: drawing a thermal image based on the thermometer lattice data obtained in the S2B1, performing morphological processing on the image to remove noise, and performing mobile personnel detection and extraction on people with complex backgrounds by using a Gaussian mixture model and a temperature mask to obtain complete mobile personnel individuals;
S2B3, acquiring position coordinates of the person: tracking and identifying the individual of the mobile personnel by using Kalman filtering and Hungarian algorithm; after the movable personnel individual is marked by a rectangular frame, the position of the movable personnel individual marked by the rectangular frame is used as the position coordinate of an image pixel point, a pixel coordinate system takes a pixel as a unit, the origin of the coordinate is at the upper left corner, in the actual space, the southwest corner of a plane graph of the intelligent building is used as the origin of the coordinate system, the length of the plane graph is used as an x axis, and the width of the plane graph is used as a y axis to establish a two-dimensional coordinate system; calculating the position P of the moving personnel obtained by the positioning module of the infrared thermal imager through the mapping relation between the pixel coordinate value and the actual coordinate value 2 (x,y)。
Further, S2B2 includes the steps of:
S2B21, extracting the foreground and the background by using a background subtraction method based on a Gaussian mixture model: extracting mobile personnel from the complex background, and marking the extracted mobile personnel by using a rectangular frame;
S2B22, processing the thermodynamic diagram by adopting a mathematical morphology method, preserving the basic style appearance of the graph by extracting the boundary, filling the region and extracting the connected components, deleting irrelevant elements in the image and adjusting the boundary;
S2B23, the individual characteristics of the mobile personnel are extracted in a perfecting way by combining the temperature information recorded by the infrared thermal imager, and the method specifically comprises the following situations:
when the extracted figure shape is divided into a plurality of parts, the temperature information of the previous and the next frames is used for judging, the temperature of the current frame is calculated and binarized, namely a threshold value T is set, if the temperature of the current frame is more than T, the threshold value T is set as 1, the same operation is carried out on the current frame, and the result after two frames of binarization is subjected to AND operation for judgment;
when a single person walks in a building and has a reflection phenomenon, whether the rectangular frame mark is a human figure or a suspected human reflection is judged according to the coordinate condition of the human figure and the reflection in an extraction result: aiming at left and right wall surface reflections, judging transverse side-by-side rectangles and deleting rectangles at two ends; if only one end of one person walks and has reflection, deleting the rectangle with the lowest height; aiming at ground reflection, when a reflection is connected with a mobile person, taking the upper half part of the integral rectangle; when the inverted image is separated from the mobile personnel, finding a rectangle meeting the requirement, and deleting a rectangle with a large ordinate;
when two or more persons are removed and inverted images appearing in the images are removed, the images are traversed from top to bottom, the maximum and minimum temperature range in the rectangular frame, the average temperature in the frame and the temperature variance of the persons in the frame are calculated for the passing rectangular frame, a temperature distribution histogram of the position of the central part is drawn, the temperature range difference between the real person and the reflected person is obtained and is used as a threshold value, inverted images are selected when the temperature range is smaller than the threshold value, and otherwise, the temperature distribution histogram is used as a human body frame; defaulting to a human body frame when only one frame passes; and extracting individual characteristics of the mobile personnel by combining the background subtraction method based on the Gaussian mixture model with the temperature information recorded by the infrared thermal imager.
Further, S3 includes the steps of:
s301, aiming at the initial position M (x, y) of the mobile personnel obtained by the WiFi positioning module when the mobile personnel are positioned in different spaces and the position P of the mobile personnel obtained by the infrared thermal imager positioning module 2 (x, y) and the position P of the moving person obtained by the infrared sensor positioning module 3 (x, y) initializing particle weights randomly, fusing by using particle filtering, detecting whether the particles penetrate through the wall or not, setting the particle weights to be 0 if the particles penetrate through the wall, and setting the particle weights to be 1 if the particles do not penetrate through the wall;
1) If the mobile personnel only move in the static space, only the RSSI value of WiFi can be detected, the standard deviation sigma is set to widen the distribution amplitude, and N candidate positions are selected t Optimum position P of initial position of individual moving person 1 At this time, the optimum position of the initial position of the mobile person is P 1 The final positioning position G (x, y) of the mobile personnel;
2) If the mobile personnel move in the static space and pass through the constraint space of the dynamic space in the moving process, the position P of the mobile personnel obtained by the infrared sensor positioning module is obtained through the jth infrared sensor 3 ，P 3 I.e. the mounting position R of the jth out-of-line sensor t (x, y), judgment of P 3 Whether or not at P 1 Internal: if P 3 At P 1 Inner, then will be away from P 3 L 1 Setting the weight of the particles in the range to be 1, setting the weight of the particles outside the range to be 0, replacing the particles with the weight of 1 with the particles with the weight of 0, and calculating the final positioning position G (x, y) of the moving person through weighted average according to the positions and the weights of the particles; if P 3 Out of P 1 Inner, P 3 The position is the final positioning position G (x, y) of the mobile personnel;
if P 3 Is not at P 1 In the interior, get P 3 A final positioning position G (x, y) for the moving person;
3) If the mobile personnel are only located in the free space of the dynamic space, obtaining N through the WiFi positioning module t Optimum position P of initial position of individual moving person 1 And obtaining the position P of the moving personnel through the positioning module of the infrared thermal imager 2 Judgment of P 2 Whether or not at P 1 Internal:
if P is 2 At P 1 If so, the distance P 2 L 2 Setting the weight of the particles in the range to be 1, setting the weight of the particles outside the range to be 0, replacing the particles with the weight of 1 with the particles with the weight of 0, and calculating the final positioning position G (x, y) of the moving person through weighted average according to the positions and the weights of the particles;
if P 2 Is not at P 1 In the interior, get P 2 A final positioning position G (x, y) for the moving person;
4) If the mobile personnel move in the free space in the dynamic space and pass through the constrained space of the dynamic space in the moving process, the position P of the mobile personnel obtained by the infrared sensor positioning module is obtained through the t-th infrared sensor 3 ，P 3 I.e. the mounting position R of the t-th out-of-line sensor t (x, y), judgment of P 3 Whether or not at P 1 Internal: if P is 3 At P 1 If so, the distance P 3 L 3 The weight of the particles in the range is set to 1, the distance P 2 L 2 The weight of the particles in the range is set to 0.5, the weight of the particles outside the range is set to 0, the particles are screened and deleted according to the resampling principleDividing the particles with the weight value of 0, copying other particles, and carrying out weighted average on the particles to obtain the final positioning position G (x, y) of the moving personnel;
if P is 3 Is not at P 1 In, then take P 2 For the final positioning position G (x, y) of the moving person, by P 3 To correct P 2 An error of (2);
and S302, extracting the final positioning positions G (x, y) of the mobile personnel corresponding to the same Mac address based on the positions of all the mobile personnel in the building obtained in S301, arranging the final positioning positions G (x, y) of the mobile personnel according to time, and drawing a track graph in sequence.
A fusion positioning system of WiFi and infrared thermal imager of an intelligent building comprises: the system comprises a region division module, a WiFi positioning module, an infrared thermal imager positioning module, an infrared sensor module and a fusion module;
the output end of the area division module is connected with the WiFi positioning module, the infrared thermal imager positioning module and the input end of the infrared sensor system, and the area division module is used for dividing the indoor space of the building into a static space and a dynamic space from the perspective of an intelligent building;
the WiFi positioning module comprises a router arranged in an intelligent building and is used for acquiring a probe request packet sent by mobile equipment and fixed equipment, providing a probe response packet sent by the routing equipment of communication service in the building and acquiring the initial position M (x, y) of the mobile personnel obtained by the WiFi positioning module according to the acquired information;
infrared thermal imager orientation module, including arranging the infrared thermal imager in intelligent building for gather thermometer grid data, and obtain the removal personnel position P that infrared thermal imager orientation module obtained according to thermometer grid data 2 (x,y)；
The infrared sensor positioning module comprises a infrared sensors arranged in an intelligent building and is used for acquiring the position P of the mobile personnel obtained by the infrared sensor positioning module 3 (x, y) and recording the passage time of the moving person, P 3 (x, y) is the installation position of the infrared sensor;
the fusion module is used for collecting the initial position M (x, y) of the mobile personnel obtained by the WiFi positioning module and the position P of the mobile personnel obtained by the infrared thermal imager positioning module 2 (x, y) and the position P of the moving person obtained by the infrared sensor positioning module 3 And (x, y) obtaining accurate position information and movement track of the moving personnel by a particle filtering method.
A fusion positioning system of WiFi and ir imager in an intelligent building, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.
Compared with the prior art, the invention at least comprises the following beneficial effects:
a fusion positioning method of an intelligent building WiFi and an infrared thermal imager can collect data of indoor mobile personnel of a building on a large scale to analyze arrival, departure and movement modes of the mobile personnel in multiple rooms of the building regularly through fusion of a WiFi positioning module, an infrared positioning module and an infrared sensor module on the premise of non-intruding privacy of other people. The fusion positioning system not only overcomes the problem that signals of a WIFI positioning module are easily interfered, but also solves the problem that the infrared thermal imager positioning module cannot identify the identity of a mobile person, utilizes the infrared sensor to fuse and correct positioning results, and utilizes the advantage that particle filtering can work under nonlinear and non-Gaussian noises, thereby greatly improving the positioning precision of indoor mobile persons.
Furthermore, before the feature extraction and the model training, an abnormal value of the signal strength RSSI is filtered, the data quality is improved, and the positioning accuracy is improved.
Further, the position P of the moving personnel obtained by the positioning module of the infrared thermal imager is obtained 2 (x, y), performing moving person detection of complex background: drawing a thermal image based on the thermometer lattice data obtained in the step S2B1, performing morphological processing on the image to remove noise, performing mobile personnel detection and extraction on people with complex backgrounds by using a Gaussian mixture model and a temperature mask,and a complete individual mobile personnel is obtained, and the detection accuracy is improved.
The utility model provides an intelligence building wiFi and infrared thermal imager's integration positioning system, includes regional division module, wiFi orientation module, infrared thermal imager orientation module, infrared sensor module and fuses the module, under the non-invasive prerequisite of not invading other people's privacy, fuses through wiFi orientation module, infrared orientation module and infrared sensor module, and then obtains indoor personnel's position and removal orbit.
Drawings
FIG. 1 is a block diagram of a fusion positioning system of WiFi and infrared thermal imager of an intelligent building according to the present invention;
FIG. 2 is a flow chart of the information physical fusion of the fusion positioning system of the intelligent building WiFi and the infrared thermal imager of the invention;
FIG. 3 is a schematic diagram of a region partitioning module;
FIG. 4 is a flow chart of a WiFi positioning module;
FIG. 5 is a flow chart of an infrared thermal imager positioning module;
FIG. 6 is a flow chart of an infrared sensor positioning module;
FIG. 7 is a flow diagram of a fusion module;
FIG. 8 is a diagram of the results of mobile personnel detection of a complex background in the infrared thermal imager positioning module;
fig. 9 is a result diagram of the fusion module obtaining the movement trajectories of different moving persons.
Detailed Description
The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings. It should be noted that, in the present application, the embodiments and features of the embodiments may be combined with each other without conflict. The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the invention and, together with the description, serve to explain the invention and not to limit the invention.
The following detailed description is exemplary in nature and is intended to provide further explanation of the invention as claimed. Unless otherwise defined, all technical terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention.
Referring to fig. 1, the present invention provides a fusion positioning system for WiFi and infrared thermal imager of an intelligent building, which includes a region division module, a WiFi positioning module, an infrared thermal imager positioning module, and a fusion module.
The area division module is used for dividing the indoor space of the building into a static space and a dynamic space from the perspective of an intelligent building, and positioning indoor mobile personnel by combining a positioning algorithm according to space scene characteristics of different spaces and equipment behavior characteristics in the spaces.
The WiFi positioning module is used for acquiring a probe request packet sent by mobile equipment (a mobile phone, a notebook computer and the like carried by mobile personnel in a building) and fixed equipment (a desktop, a projector and an industrial personal computer) and the like and a probe response packet sent by the routing equipment for providing communication service in the building through arranging an intelligent router in the intelligent building, and analyzing the current time, the Mac address of the mobile equipment or the fixed equipment and the WiFi signal strength RSSI of the same mobile equipment or the fixed equipment acquired by a plurality of intelligent routers. Analyzing and preprocessing the RSSI of the WiFi, extracting a characteristic vector, training a recurrent neural network model, and acquiring the initial position M (x, y) of the mobile personnel obtained by the WiFi positioning module, wherein the initial position M obeys an expectation of mu 1 、μ 2 Variance is σ 1 2 、σ 2 2 P (x, y).
The infrared thermal imager positioning module collects thermometer grid data every second by arranging the infrared thermal imager, draws a thermometer grid into a thermodynamic diagram, detects moving personnel with a complex background, morphologically processes an image, adds a temperature mask to remove impurities, extracts the moving personnel, and acquires a moving personnel position P obtained by the infrared thermal imager positioning module 2 (x,y)。
Infrared sensor positioning module by constraint in dynamic spaceA infrared sensors are arranged in a space, namely a region of fixed obstacles such as stairs, doors, corridors and the like to acquire time information and position information, and the installation position record R = { R } of each infrared sensor is recorded 1 (x,y),R 2 (x,y),…R a (x, y) }, when the mobile personnel pass through the jth infrared sensor, recording the passing time of the mobile personnel and the position P of the mobile personnel obtained by the infrared sensor positioning module 3 ，P 3 (x, y) mounting position R of jth infrared sensor j (x, y) wherein 1<j<a。
And the fusion module is used for obtaining the position information and the moving track of the moving personnel in the building through particle filter fusion by three types of position information obtained by the area division module, the WiFi positioning module, the infrared thermal imager positioning module and the infrared sensor positioning module.
Referring to fig. 2, the process of positioning the information physical fusion by the fusion module includes the following steps:
arranging DS006N dual-network card intelligent routers in an intelligent building in advance to form an intelligent routing system which is a physical hardware part of a WiFi positioning module; arranging an Oupishi PI300 infrared thermal imager in free space to form an infrared thermal imager system which is a physical hardware part of an infrared thermal imager positioning module; arranging infrared sensors in constrained spaces, namely stairs, corridors and the like to form an infrared sensor system which is a hardware part of an infrared sensor positioning module; analyzing time information, a Mac address of user equipment and WiFi signal strength RSSI (received signal strength indicator) through an intelligent routing system, acquiring time information and thermometer form data through an infrared thermal imager system, acquiring position information of a user through an infrared sensor system and recording acquisition time, wherein the position information of the user is the installation position of an infrared sensor; analyzing, processing and calculating the data to respectively obtain three types of positions P 1 ,P 2 ,P 3 : obtaining optimal position P of initial position of mobile personnel by using WiFi positioning module 1 Position P of moving person obtained by infrared thermal imager positioning module 2 Position P obtained by infrared sensor positioning module 3 . Using data lines and all DS006N pairsThe network card intelligent router, the infrared thermal imager and the infrared sensor are connected, three types of position information are fused through a particle filter program loaded in the fusion module, the position information at the next moment is transmitted through feedback and corrected, and finally the position coordinates of the moving personnel and the indoor moving track are displayed through the visual platform. Taking a floor plan of an intelligent building as an example, P1 to P19 represent different rooms, different symbols represent different moving persons, and a dotted line represents a moving track of the moving person.
A fusion positioning method of WiFi and infrared thermal imager of an intelligent building comprises the following steps:
s1, dividing an indoor space into a static space and a dynamic space, wherein the dynamic space comprises a free space and a constraint space;
s2, acquiring a probe request packet sent by mobile equipment (a mobile phone, a notebook computer and the like carried by mobile personnel in a building) and fixed equipment (a desktop computer, a projector, an industrial personal computer and the like) and a probe response packet sent by routing equipment for providing communication service in the building by arranging an intelligent router, analyzing and preprocessing the RSSI (received signal strength indicator) of WiFi (wireless fidelity), extracting a characteristic vector, training a recurrent neural network model and obtaining the position of the mobile personnel;
s3, acquiring thermometer grid data by arranging an infrared thermal imager, detecting moving personnel with a complex background, performing morphological processing on the image, adding a temperature mask to remove impurities, extracting the moving personnel, and acquiring the position of the moving personnel;
s4, acquiring time information and position information by installing a infrared sensors in a constrained space of a dynamic space, namely, a region of fixed obstacles such as stairs, doors, corridors and the like, taking the acquired time information and position information as a calibration standard, and recording the installation positions R = { R } of the infrared sensors 1 (x,y),R 2 (x,y),…R a (x, y) }, when the mobile personnel passes through the jth infrared sensor, recording the passing time of the mobile personnel and the position P of the mobile personnel, which is obtained by the infrared sensor positioning module 3 ，P 3 (x, y) mounting position R of jth infrared sensor j (x, y) wherein 1<j<a；
And S5, based on the three types of position information obtained by the WiFi positioning module, the infrared thermal imager positioning module and the infrared sensor positioning module, obtaining accurate position information and moving tracks of moving personnel in the building through particle filter fusion.
Referring to fig. 3, S1 includes the following steps:
s101, from the perspective of an intelligent building, according to the movement capacity of mobile personnel in different spaces, all doors, stairs and turning places in an indoor space are selected as landmarks for space division, and the indoor space of the building is divided into a static space and a dynamic space.
S102, dividing places with weak motion capability, such as teaching and research rooms, offices, toilets and other mobile personnel, into static spaces;
s103, dividing places with strong motion capability of moving personnel such as stairs, corridors and open spaces into dynamic spaces. The dynamic space is divided into a constraint space and a free space; the region of fixed obstacles such as stairs, doors, corridors and the like is divided into a constraint space; the region such as the open space is divided into a free space.
Referring to fig. 4, S2 includes the following steps:
s201, data acquisition. A plurality of intelligent routers arranged in an intelligent building are used for actively acquiring probe request packets sent by mobile equipment (mobile phones, notebooks and the like carried by mobile personnel in the building) and fixed equipment (desktops, projectors, industrial personal computers) and the like and probe response packets sent by routing equipment for providing communication services in the building, and analyzing currently acquired time information, mac addresses of the mobile equipment or the fixed equipment and WiFi signal strength RSSI (received signal strength indicator).
S202, data analysis and pretreatment. Preprocessing is performed based on the signal strength RSSI obtained in step S201: and the abnormal value is deleted through the abnormal value processing, so that the data quality is improved. The processed data form is shown in table 1.
S203, feature extraction and model training. And calculating a WiFi signal trend index STI and a signal interval, training a recurrent neural network model by taking the WiFi signal strength RSSI measured by the data acquisition module as a characteristic vector, and constructing a fingerprint database of the WiFi signal strength RSSI of the mobile equipment. Since the Mac address of each mobile device is unique and fixed, the identity of the mobile personnel is identified by the Mac address.
The signal trend index STI is calculated by the following method:
1) The WiFi signal strength RSSI of the same mobile equipment or fixed equipment collected by a plurality of intelligent routers is recorded as a group, the average value of the RSSI is obtained,
wherein s is i For i group wifi signal strength RSSI, s avg The average value of the i groups of wifi signal strength RSSI;
2) Calculating the Signal Trend index STI of the ith data i ，
STI i ＝||s i -s avg || (2)
Wherein s is i For i group wifi signal strength RSSI, s avg The average value of the i groups of wifi signal strength RSSI;
the calculation method of the data interval is as follows:
interval i ＝||s i -s i+1 || (3)
wherein s is i+1 The signal strength RSSI of the i +1 group wifi signal strength RSSI;
s204, establishing an indoor mobile personnel positioning model for interaction between people and a scene. Based on the fingerprint database of the WiFi signal strength RSSI of the mobile device constructed in S203, a recurrent neural network model between the WiFi signal strength RSSI and the physical location is established. The southwest angle of the plane graph of the intelligent building is used as the origin of a coordinate system, the length of the plane graph is used as an x axis, the width of the plane graph is used as a y axis to establish a two-dimensional coordinate system, and the coordinate position of the mobile personnel is the physical position of the person in the intelligent building. A group of WiFi signal strength RSSI of the same mobile equipment or fixed equipment, which is acquired by a plurality of intelligent routers, is taken as input and is brought into a trained recurrent neural network model, the initial position M (x, y) of the mobile personnel, which is obtained by a WiFi positioning module, is output,subject to a desire of mu 1 、μ 2 Variance is σ 1 2 、σ 2 2 P (x, y).
Referring to fig. 5, S3 includes the following steps:
and S301, collecting data. The method comprises the steps of collecting temperature table data of the indoor corridor of the intelligent building every other one second through an infrared thermal imager which is arranged on the ceiling of the indoor corridor of the intelligent building in advance, wherein the value of each point in a table is the temperature of the pixel point. The thermometer grid data is named by the acquisition time.
And S302, detecting the moving personnel with complex background. Based on the thermometer lattice data obtained in step S301, a thermal image is drawn, the image is morphologically processed to remove noise and other phenomena, and then a gaussian mixture model and a temperature mask are used to perform mobile personnel detection and extraction on a person with a complex background, so as to obtain a complete individual feature of the mobile personnel, as shown in fig. 8.
S3021, extracting the foreground and the background by using a background subtraction method based on a Gaussian mixture model. The method comprises the steps of extracting moving personnel from a complex background, marking the extracted moving personnel by using a rectangular frame, and identifying the shadow of reflection as a real human phenomenon because the human body is a natural heating source and can reflect to the left wall, the right wall, the ceiling and the ground. Therefore, the rectangle frame has the shadow or sundries of a suspected human shape besides the real human body.
S3022, processing the thermodynamic diagram by adopting a mathematical morphology method, keeping the basic pattern appearance of the graph through boundary extraction, region filling, connected component extraction and the like, deleting irrelevant elements in the image, and adjusting the boundary.
S3023, improving the extracted individual characteristics of the mobile personnel by combining the temperature information recorded by the infrared thermal imager, wherein the method specifically comprises the following steps:
1. when the extracted character shape is divided into several parts, it may be that the mobile person carries a bag or holds an umbrella. A temperature mask Tmask is newly added, and low-temperature objects outside the background, such as door foreign objects and partial reflection shadows, are removed by setting a temperature threshold value (T > 27.6). When a moving person carries a schoolbag and a human body is segmented when the moving person wears an umbrella, the human body segmentation phenomenon is judged by utilizing temperature information of front and rear frames, the temperature of a current frame is calculated and binarized, namely a threshold value T is set, the temperature of the current frame is greater than T and is set to be 1, the same operation is carried out on the current frame, and the results after binarization of the two frames are subjected to AND operation to judge, so that the human body segmentation phenomenon is improved to a certain extent, and a reflecting area can be further eliminated.
2. When a single person walks in a building and has a reflection phenomenon, whether the mark of the rectangular frame is a human figure or a suspected human reflection is roughly judged according to the coordinate condition of the human figure and the reflection in the extraction result. Aiming at the reflection of the left and right wall surfaces, the reflection formed by the walls is always at the left and right ends of the picture, so that the rectangles transversely arranged side by side are judged and the rectangles at the two ends are deleted. If only one end of one person walks and has reflection, deleting the rectangle with the lowest height; aiming at ground reflection, no matter how far or near the mobile personnel are, complete human-shaped reflection can be formed on the ground capable of reflecting infrared rays, and when the reflection is connected with the mobile personnel, the upper half part of the whole rectangle is taken. The inverted image is separated from the moving personnel, because the abscissa of the two rectangles is extremely close, and the distance between the bottom edge of the upper rectangle and the top edge of the lower rectangle is shorter, the rectangle meeting the requirements is searched, and the ordinate is deleted.
3. And removing two or more people and removing the people, and when the reflection occurs, simultaneously considering the attribute characteristics of the temperature data acquired by the infrared thermal imager, traversing the graph from top to bottom, and calculating the maximum and minimum temperature range in the rectangular frame, the average temperature in the rectangular frame and the temperature variance of the people in the rectangular frame for the crossed rectangular frame. Drawing a temperature distribution histogram of the position of the central part to obtain that the temperature range difference between a real human body and a reflected human body is larger, selecting the range difference as a threshold value, and regarding the range difference smaller than the threshold value as a reverse image, otherwise, regarding the range difference as a human body frame; only one frame is passed through by default as a body frame. And extracting individual characteristics of the mobile personnel by combining the background subtraction method based on the Gaussian mixture model with the temperature information recorded by the infrared thermal imager.
And S303, acquiring the position coordinates of the mobile personnel. Background subtraction method based on Gaussian mixture model, morphological processing and temperature mask adding(Tmask) extracting individual characteristics of the mobile personnel, and tracking and identifying the people by using Kalman filtering and Hungarian algorithm. After the movable personnel are marked by the rectangular frame, the position of the movable personnel marked by the rectangular frame is taken as the position coordinates of the pixel points of the image, the pixel coordinate system takes the pixel as a unit, the origin of the coordinate is at the upper left corner, in the actual space, the southwest corner of the plane graph of the intelligent building is taken as the origin of the coordinate system, the length of the plane graph is taken as an x axis, and the width of the plane graph is taken as a y axis to establish a two-dimensional coordinate system. Calculating the position P of the moving personnel obtained by the positioning module of the infrared thermal imager through the mapping relation between the pixel coordinate value and the actual coordinate value 2 (x,y)。
Referring to fig. 6, S4 includes the following steps:
s401, mounting a infrared sensors in a constrained space of a dynamic space, namely, a stair, a door, a corridor and other fixed obstacle areas;
s402, recording the installation position of each infrared sensor as R = { R = { R 1 (x,y),R 2 (x,y),…,R a (x,y)}；
S403, when the moving person passes through the jth infrared sensor, recording the passing time of the moving person and the position P of the moving person, obtained by the infrared sensor positioning module 3 (x,y)，P 3 (x, y) mounting position R of jth infrared sensor j (x, y) wherein 0<j<a。
Referring to fig. 7, S5 includes the following steps:
s501, obtaining the initial position M (x, y) of the mobile personnel obtained by the WiFi positioning module based on S204, wherein the initial position M (x, y) is subject to one expectation of mu 1 、μ 2 Variance is σ 1 2 、σ 2 2 Based on the position P (x, y) of the moving person obtained by the positioning module of the S303 infrared thermal imager 2 (x, y) and the position P of the mobile person obtained based on the S403 infrared sensor positioning module 3 (x, y), randomly initializing particle weights according to three types of position information of people in different spaces, fusing by using particle filtering, detecting whether the particles penetrate through the wall or not, setting the particle weights to be 0 if the particles penetrate through the wall, and setting the particle weights to be 1 if the particles penetrate through the wall.
1. WiFi Intelligent deploymentCan build, obtain the initial position M (x, y) of the mobile personnel through the WiFi positioning module, which follows the gaussian distribution P (x, y). If the mobile personnel only move in the static space, only the RSSI value of WiFi can be detected, the standard deviation sigma is set to widen the distribution amplitude, and N candidate positions are selected t Optimum position P of initial position of individual moving person 1 Optimum position P of initial position of moving person 1 I.e. the final positioning position G (x, y) of the moving person.
2. If the mobile personnel move in the static space and pass through the constraint space of the dynamic space in the moving process, the position P of the mobile personnel obtained by the infrared sensor positioning module is obtained through the jth infrared sensor 3 ，P 3 I.e. the mounting position R of the jth out-of-line sensor j (x, y). Judgment of P 3 Whether or not at P 1 In, if P 3 At P 1 Inner, then will be away from P 3 The particle weight within the 0.5m range is set to 1, and the particle weight outside the range is set to 0. In order to prevent the degradation of the particles, the total number of particles is kept constant, the particles with weight 1 are substituted for the particles with weight 0, and the final positioning position G (x, y) of the person is moved by a weighted average according to the positions and weights of the particles. The Mac address extracted by the WiFi positioning module provides identity recognition for the mobile personnel, and the position P of the mobile personnel is obtained through the infrared sensor positioning module 3 And correcting the recurrent neural network model in the WiFi positioning module. If P 3 Out of P 1 Inner, P 3 I.e. the final positioning position G (x, y) of the moving person.
3. If the mobile personnel are only located in the free space of the dynamic space, obtaining N through the WiFi positioning module t Optimum position P of initial position of individual moving person 1 And obtaining the position P of the moving personnel through the positioning module of the infrared thermal imager 2 Judgment of P 2 Whether or not at P 1 In, if P 2 At P 1 If so, the distance P 2 The weight of the particles in the range of 0.8m is set to 1, the weight of the particles outside the range is set to 0, in order to prevent the degradation of the particles, the total number of the particles is kept unchanged, the particles with the weight of 1 are substituted for the particles with the weight of 0, and the weight is weighted according to the positions and the weights of the particlesAnd averagely calculating the final positioning position G (x, y) of the current moving person. The Mac address extracted by the WiFi positioning module provides the identification of the mobile personnel for the positioning, if P 2 Is not at P 1 In, then take P 2 The final position G (x, y) for the moving person is determined.
4. If the mobile personnel move in the free space in the dynamic space and pass through the constrained space of the dynamic space in the moving process, the position P of the mobile personnel obtained by the infrared sensor positioning module is obtained through the tth infrared sensor 3 ，P 3 I.e. the mounting position R of the t-th out-of-line sensor t (x, y). Judgment of P 3 Whether or not at P 1 In, if P 3 At P 1 If so, the distance P 3 The weight of the particles in the range of 0.5m is set to 1, the distance P 2 The particle weight within the 0.8m range is set to 0.5, and the particle weight outside the range is set to 0. And screening the particles according to a resampling principle, deleting the particles with lower weight, and copying the particles with higher weight in order to prevent the particles from degrading. The final positioning position G (x, y) of the moving person is obtained by weighted averaging of the particles. The Mac address extracted by the WiFi positioning module provides identity recognition for the mobile personnel, and the position P of the mobile personnel is obtained through the infrared sensor positioning module 3 And correcting the positioning error caused by the size of the infrared thermal imager positioning module and the regression neural network model in the WiFi positioning module. If P is 3 Out of P 1 In, then take P 2 For the final positioning position G (x, y) of the moving person, by P 3 Make a correction of P 2 The error of (2).
And S502, extracting the final positioning positions G (x, y) of the mobile personnel corresponding to the same Mac address based on the positions of all the mobile personnel in the building obtained in the S301, arranging the final positioning positions G (x, y) of the mobile personnel according to time, and sequentially connecting and drawing a track graph, as shown in FIG. 9.
Table 1 data format after data analysis and processing of WiFi positioning module of the present invention
Table 1 shows a data format reserved after data analysis and processing by the WiFi positioning module, where a first row _ id is a serial number, a second row i is a number of a mobile device or a fixed device, the same value of i indicates the same device acquired by multiple intelligent routers and is regarded as a group, a third row is time information during acquisition, a fourth row Mac is a Mac address of the device, and a last row s is a row i Representing the i-th group WiFi signal strength RSSI.
Table 2 positioning of the present invention relative to WiFi positioning, infrared thermal imager positioning alone
Table 2 shows the positioning accuracy of the WiFi positioning module, the infrared thermal imager positioning module, and the fusion positioning system of the WiFi positioning module and the infrared thermal imager positioning module within 1, 2, 3, and 4 meters. As can be seen from the table, the fusion positioning system of the WiFi and the infrared thermal imager of the intelligent building has higher precision, reaches 83% of accuracy at 2m, and has 96% of accuracy at 4 m.
It will be appreciated by those skilled in the art that the invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The embodiments disclosed above are therefore to be considered in all respects as illustrative and not restrictive. All changes which come within the scope of or equivalence to the invention are intended to be embraced therein.
Claims (10)
1. A fusion positioning method of WiFi and infrared thermal imager of an intelligent building is characterized by comprising the following steps:
s1, dividing a space in a building into a static space and a dynamic space, wherein the dynamic space comprises a free space and a constrained space;
s2, acquiring a probe request packet sent by the mobile equipment and the fixed equipment and a probe response packet sent by the routing equipment for providing the communication service in the building by using the intelligent router, and acquiring the probe request packet and the probe response packet according to the acquired probe response packetThe probe request packet and the probe response packet, the initial position M (x, y) of the mobile personnel obtained by the WiFi positioning module is calculated, and the initial position M (x, y) is subject to an expectation of mu 1 、μ 2 Variance is σ 1 2 、σ 2 2 A gaussian distribution P (x, y);
acquiring thermometer lattice data by using an infrared thermal imager, detecting moving personnel with complex background, performing morphological processing on the image, extracting the moving personnel, and acquiring the position P of the moving personnel obtained by a positioning module of the infrared thermal imager 2 (x,y)；
The position information and the acquisition time of the mobile personnel are acquired by using the infrared sensor, and the position P of the mobile personnel obtained by the infrared sensor positioning module is acquired 3 (x,y)；
S3, according to the static space and the dynamic space area divided by the S1, obtaining M (x, y) and P by the S2 2 (x, y) and P 3 And (x, y) obtaining accurate position information and movement tracks of the mobile personnel in the building through particle filter fusion.
2. The fusion positioning method of WiFi and infrared thermal imaging cameras of intelligent buildings according to claim 1, wherein the S1 comprises the following steps:
s101, selecting all doors, stairs and turning places in the space in the building as landmarks for space division according to the movement capacity of moving personnel in different spaces, and dividing the space in the building into static space and dynamic space;
s102, dividing a teaching and research room, an office and a toilet into static spaces;
s103, dividing the stairs, the room doorways, the corridors and the vacant spaces into dynamic spaces; the dynamic space is divided into a constraint space and a free space, wherein stairs, room doorways and corridors are divided into the constraint space; the open space region is divided into free space.
3. The fusion positioning method of WiFi and infrared thermal imager of intelligent building of claim 1, wherein in S2, calculating the initial position M (x, y) of the mobile personnel obtained by WiFi positioning module includes the following steps:
S2A1, data acquisition: acquiring a probe request packet transmitted by a mobile device and a fixed device and a probe response packet transmitted by a routing device for providing communication service in the building through a plurality of intelligent routers arranged in an intelligent building, and analyzing currently acquired time information, a Mac address of the mobile device or the fixed device and a WiFi signal strength RSSI (received signal strength indicator) from the probe request packet;
S2A2, feature extraction and model training: calculating a WiFi signal trend index STI and a signal interval according to the WiFi signal strength RSSI acquired in the S2A1, taking the WiFi signal trend index STI, the signal interval and the WiFi signal strength RSSI as feature vectors, training a recurrent neural network model, and constructing a fingerprint database of the WiFi signal strength RSSI of the mobile equipment;
S2A3, establishing an indoor mobile personnel positioning model of human-scene interaction: the method comprises the steps of establishing a regression neural network model between WiFi signal strength RSSI and a physical position based on a fingerprint database of WiFi signal strength RSSI of mobile equipment established by S2A2, establishing a two-dimensional coordinate system by taking the southwest corner of a plane graph of an intelligent building as the origin of the coordinate system, the length of the plane graph as the x axis and the width as the y axis, taking the position of a mobile person in the coordinate system, namely the physical position of the person in the intelligent building, taking a group of WiFi signal strength RSSIs of the same mobile equipment or fixed equipment acquired by a plurality of intelligent routers as input, bringing the input RSSI into the trained regression neural network model, and outputting the initial position M (x, y) of the mobile person obtained by a WiFi positioning module.
4. The method for fusion positioning of WiFi and infrared thermal imaging camera of intelligent building according to claim 3, wherein in S2A2,
the method for calculating the signal tendency index STI is as follows:
1) The method comprises the steps of recording WiFi signal strength RSSI (received signal strength indicator) of the same mobile equipment or fixed equipment collected by a plurality of intelligent routers into a group, and calculating an average value s of the RSSI avg ，
Wherein s is i For i group wifi signal strength RSSI, s avg The average value of the wifi signal strength RSSI of the i groups is obtained;
2) Calculating the Signal Trend index STI of the ith data i ，
STI i ＝||s i -s avg || (2)
The calculation method of the data interval is as follows:
interval i ＝||s i -s i+1 || (3)
wherein s is i+1 I +1 sets of wifi signal strength RSSI.
5. The method for fusion positioning of WiFi and infrared thermal imager of an intelligent building according to claim 3, wherein before the step S2A2, the abnormal value of RSSI obtained in the step S2A1 is deleted.
6. The fusion positioning method of WiFi and infrared thermal imaging cameras of intelligent building according to claim 1, wherein in S2, the position P of the mobile personnel obtained by the positioning module of the infrared thermal imaging cameras is obtained 2 The procedure for (x, y) is as follows:
S2B1, data acquisition: acquiring temperature chart data of indoor corridors of the intelligent building at intervals of set time through an infrared thermal imager which is arranged in the intelligent building in advance;
S2B2, detecting the moving personnel with complex background: drawing a thermal image based on thermometer lattice data obtained in the S2B1, performing morphological processing on the image to remove noise, and then performing mobile personnel detection and extraction of people with complex backgrounds by using a Gaussian mixture model and a temperature mask to obtain complete mobile personnel individuals;
S2B3, acquiring position coordinates of the person: tracking and identifying individual mobile personnel by using Kalman filtering and Hungarian algorithm; after the individual of the mobile personnel is marked by the rectangular frame, the mobile personnel marked by the rectangular frameThe individual position is the position coordinate of the image pixel point, the pixel coordinate system takes the pixel as the unit, the origin of the coordinate is at the upper left corner, and in the actual space, the southwest corner of the plane graph of the intelligent building is taken as the origin of the coordinate system, the length of the plane graph is taken as the x axis, and the width is taken as the y axis to establish a two-dimensional coordinate system; calculating the position P of the moving personnel obtained by the positioning module of the infrared thermal imager through the mapping relation between the pixel coordinate value and the actual coordinate value 2 (x,y)。
7. The fusion positioning method of WiFi and infrared thermal imager of intelligent building according to claim 6, wherein said S2B2 comprises the following steps:
S2B21, extracting the foreground and the background by using a background subtraction method based on a Gaussian mixture model: extracting mobile personnel from the complex background, and marking the extracted mobile personnel by using a rectangular frame;
S2B22, processing the thermodynamic diagram by adopting a mathematical morphology method, preserving the basic style appearance of the graph by extracting the boundary, filling the region and extracting the connected components, deleting irrelevant elements in the image and adjusting the boundary;
S2B23, improving the extracted individual characteristics of the mobile personnel by combining the temperature information recorded by the infrared thermal imager, wherein the method specifically comprises the following conditions:
when the extracted figure shape is divided into a plurality of parts, the temperature information of the previous and the next frames is used for judging, the temperature of the current frame is calculated and binarized, namely a threshold value T is set, if the temperature of the current frame is more than T, the threshold value T is set as 1, the same operation is carried out on the current frame, and the result after two frames of binarization is subjected to AND operation for judgment;
when a single person walks in a building, a reflection phenomenon appears, and whether the figure of the person or the inverted image of a suspected person is at the mark position of the rectangular frame is judged according to the coordinate condition of the figure and the inverted image in the extraction result: aiming at left and right wall surface reflections, judging transverse side-by-side rectangles and deleting rectangles at two ends; if only one end of one person walks and has reflection, deleting the rectangle with the lowest height; aiming at ground reflection, when a reflection is connected with a moving person, taking the upper half part of the integral rectangle; when the inverted image is separated from the mobile personnel, finding a rectangle meeting the requirement, and deleting the rectangle with a large ordinate;
when two or more persons are removed and inverted images appearing in the images are removed, the images are traversed from top to bottom, the maximum and minimum temperature range in the rectangular frame, the average temperature in the frame and the temperature variance of the persons in the frame are calculated for the passing rectangular frame, a temperature distribution histogram of the position of the central part is drawn, the temperature range difference between the real person and the reflected person is obtained and is used as a threshold value, inverted images are selected when the temperature range is smaller than the threshold value, and otherwise, the temperature distribution histogram is used as a human body frame; defaulting to a human body frame when only one frame passes; and extracting individual characteristics of the mobile personnel by combining the background subtraction method based on the Gaussian mixture model with the temperature information recorded by the infrared thermal imager.
8. The fusion positioning method of WiFi and infrared thermal imaging cameras of intelligent buildings according to claim 1, wherein the S3 comprises the following steps:
s301, aiming at initial positions M (x, y) of the mobile personnel obtained by WiFi positioning modules of different spaces where the mobile personnel are located, and the positions P of the mobile personnel obtained by infrared thermal imager positioning modules 2 (x, y) and the position P of the moving person obtained by the infrared sensor positioning module 3 (x, y) initializing particle weights randomly, fusing by using particle filtering, detecting whether the particles penetrate through the wall or not, setting the particle weights to be 0 if the particles penetrate through the wall, and setting the particle weights to be 1 if the particles do not penetrate through the wall;
1) If the mobile personnel only move in a static space, only the RSSI value of WiFi can be detected, the standard deviation sigma is set, the distribution amplitude is widened, and N candidate positions are selected from N candidate positions t Optimum position P of initial position of individual moving person 1 At this time, the optimum position of the initial position of the mobile person is P 1 I.e. the final positioning position G (x, y) of the moving person;
2) If the mobile personnel move in the static space and pass through the constraint space of the dynamic space in the moving process, the position P of the mobile personnel obtained by the infrared sensor positioning module is obtained through the jth infrared sensor 3 ，P 3 I.e. the mounting position R of the jth out-of-line sensor t (x, y), judgmentP breaking 3 Whether or not at P 1 Internal: if P 3 At P 1 Inner, then, will be away from P 3 L 1 Setting the weight of the particles in the range to be 1, setting the weight of the particles out of the range to be 0, replacing the particles with the weight of 1 with the particles with the weight of 0, and calculating the final positioning position G (x, y) of the moving person through weighted average according to the positions and the weights of the particles; if P is 3 Is not at P 1 Inner, P 3 The position is the final positioning position G (x, y) of the mobile personnel;
if P 3 Is not at P 1 In, then take P 3 A final positioning position G (x, y) for the moving person;
3) If the mobile personnel are only located in the free space of the dynamic space, obtaining N through the WiFi positioning module t Optimum position P of initial position of individual moving person 1 And obtaining the position P of the moving personnel through the positioning module of the infrared thermal imager 2 Judgment of P 2 Whether or not it is at P 1 Internal:
if P is 2 At P 1 Inner, then the distance P 2 L 2 Setting the weight of the particles in the range to be 1, setting the weight of the particles out of the range to be 0, replacing the particles with the weight of 1 with the particles with the weight of 0, and calculating the final positioning position G (x, y) of the moving person through weighted average according to the positions and the weights of the particles;
if P is 2 Is not at P 1 In, then take P 2 A final positioning position G (x, y) for the mobile person;
4) If the mobile personnel move in the free space in the dynamic space and pass through the constrained space of the dynamic space in the moving process, the position P of the mobile personnel obtained by the infrared sensor positioning module is obtained through the tth infrared sensor 3 ，P 3 I.e. the mounting position R of the t-th out-of-line sensor t (x, y), judgment of P 3 Whether or not at P 1 Internal: if P is 3 At P 1 If so, the distance P 3 L 3 Particle weight in range set to 1, distance P 2 L 2 Setting the weight of the particles within the range to 0.5 and the weight of the particles outside the range to 0, screening the particles according to the resampling principle, and deleting the weightCopying other particles for the particle of 0, and obtaining the final positioning position G (x, y) of the moving person by carrying out weighted average on the particles;
if P is 3 Is not at P 1 In, then take P 2 For the final positioning position G (x, y) of the moving person, by P 3 Make a correction of P 2 An error of (2);
and S302, extracting the final positioning positions G (x, y) of the mobile personnel corresponding to the same Mac address based on the positions of all the mobile personnel in the building obtained in the S301, arranging the final positioning positions G (x, y) of the mobile personnel according to time, and drawing a track graph in sequence.
9. The utility model provides an intelligence building wiFi and infrared thermal imager's integration positioning system which characterized in that includes: the system comprises a region division module, a WiFi positioning module, an infrared thermal imager positioning module, an infrared sensor positioning module and a fusion module;
the output end of the area dividing module is connected with the WiFi positioning module, the infrared thermal imager positioning module and the input end of the infrared sensor system, and the area dividing module is used for dividing the indoor space of the building into a static space and a dynamic space from the perspective of an intelligent building;
the WiFi positioning module comprises a router arranged in an intelligent building and is used for acquiring a probe request packet sent by mobile equipment and fixed equipment, providing a probe response packet sent by the routing equipment of communication service in the building and acquiring an initial position M (x, y) of a mobile person obtained by the WiFi positioning module according to the acquired information;
infrared thermal imager orientation module, including arranging the infrared thermal imager in intelligent building for gather thermometer grid data, and obtain the removal personnel position P that infrared thermal imager orientation module obtained according to thermometer grid data 2 (x，y)；
The infrared sensor positioning module comprises a infrared sensors arranged in an intelligent building and is used for acquiring the position P of a mobile person obtained by the infrared sensor positioning module 3 (x, y) and recording the passage time of the moving person, P 3 (x, y) i.e. redThe mounting position of the external sensor;
the fusion module is used for collecting the initial position M (x, y) of the mobile personnel obtained by the WiFi positioning module and the position P of the mobile personnel obtained by the infrared thermal imager positioning module 2 (x, y) and the position P of the moving person obtained by the infrared sensor positioning module 3 And (x, y), obtaining accurate position information and movement track of the moving personnel by a particle filtering method according to the divided static space and dynamic space areas.
10. A fusion positioning system of WiFi and infrared thermal imager in an intelligent building, comprising a memory and a processor, wherein the memory stores a computer program running on the processor, and the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110027946.7A CN112880660B (en) | 2021-01-09 | 2021-01-09 | Fusion positioning system and method for WiFi and infrared thermal imager of intelligent building |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110027946.7A CN112880660B (en) | 2021-01-09 | 2021-01-09 | Fusion positioning system and method for WiFi and infrared thermal imager of intelligent building |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112880660A CN112880660A (en) | 2021-06-01 |
| CN112880660B true CN112880660B (en) | 2022-10-25 |
Family
ID=76047475
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110027946.7A Active CN112880660B (en) | 2021-01-09 | 2021-01-09 | Fusion positioning system and method for WiFi and infrared thermal imager of intelligent building |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112880660B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116090796B (en) * | 2023-04-06 | 2023-06-16 | 威海海洋职业学院 | Hotel intelligent management system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109525935A (en) * | 2018-11-23 | 2019-03-26 | 武汉大学 | A kind of intelligent floor cognitive method and control system for pervasive indoor environment |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0813040A3 (en) * | 1996-06-14 | 1999-05-26 | Xerox Corporation | Precision spatial mapping with combined video and infrared signals |
| CN107786938B (en) * | 2017-09-15 | 2021-04-16 | 南京轩世琪源软件科技有限公司 | Mobile user position coordinate correction method based on WLAN indoor positioning |
| CN108875716A (en) * | 2018-09-07 | 2018-11-23 | 杭州善贾科技有限公司 | A kind of human body motion track trace detection camera system |
| CN110031904B (en) * | 2019-04-19 | 2024-01-26 | 福州大学 | Indoor personnel presence detection system based on low-resolution infrared thermal imaging |
| CN111596259A (en) * | 2020-05-19 | 2020-08-28 | 沈宇桐 | Infrared positioning system, positioning method and application thereof |
-
2021
- 2021-01-09 CN CN202110027946.7A patent/CN112880660B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109525935A (en) * | 2018-11-23 | 2019-03-26 | 武汉大学 | A kind of intelligent floor cognitive method and control system for pervasive indoor environment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112880660A (en) | 2021-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020215961A1 (en) | Personnel information detection method and system for indoor climate control | |
| US8254633B1 (en) | Method and system for finding correspondence between face camera views and behavior camera views | |
| US8179440B2 (en) | Method and system for object surveillance and real time activity recognition | |
| Boltes et al. | Collecting pedestrian trajectories | |
| CN110246112B (en) | Laser scanning SLAM indoor three-dimensional point cloud quality evaluation method based on deep learning | |
| CN109190508B (en) | Multi-camera data fusion method based on space coordinate system | |
| Benezeth et al. | Towards a sensor for detecting human presence and characterizing activity | |
| KR100612858B1 (en) | Method and apparatus for tracking human using robot | |
| JP4970195B2 (en) | Person tracking system, person tracking apparatus, and person tracking program | |
| CN109598794B (en) | Construction method of three-dimensional GIS dynamic model | |
| Aghaei et al. | Single image human proxemics estimation for visual social distancing | |
| US20060028552A1 (en) | Method and apparatus for stereo, multi-camera tracking and RF and video track fusion | |
| WO2008070206A2 (en) | A seamless tracking framework using hierarchical tracklet association | |
| CN109163731A (en) | A kind of semanteme map constructing method and system | |
| CN109272554A (en) | A kind of method and system of the coordinate system positioning for identifying target and semantic map structuring | |
| CN112257549B (en) | Floor danger detection early warning method and system based on computer vision | |
| CN112880660B (en) | Fusion positioning system and method for WiFi and infrared thermal imager of intelligent building | |
| CN111967443A (en) | Image processing and BIM-based method for analyzing interested area in archive | |
| CN108229421A (en) | A kind of falling from bed behavior real-time detection method based on deep video information | |
| CN107862713A (en) | Video camera deflection for poll meeting-place detects method for early warning and module in real time | |
| CN113240678B (en) | Plane information detection method and system | |
| CN114511592A (en) | Personnel trajectory tracking method and system based on RGBD camera and BIM system | |
| CN112513870A (en) | System and method for detecting, tracking and counting human objects of interest with improved altimetry | |
| CN116486480A (en) | Human body falling detection method and device based on point cloud | |
| CN112784703A (en) | Multispectral-based personnel action track determination method |
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 |