CN108495095B - Haze diffusion monitoring system based on unmanned aerial vehicle - Google Patents

Haze diffusion monitoring system based on unmanned aerial vehicle Download PDF

Info

Publication number
CN108495095B
CN108495095B CN201810439091.7A CN201810439091A CN108495095B CN 108495095 B CN108495095 B CN 108495095B CN 201810439091 A CN201810439091 A CN 201810439091A CN 108495095 B CN108495095 B CN 108495095B
Authority
CN
China
Prior art keywords
haze
module
image
data
monitoring
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
Application number
CN201810439091.7A
Other languages
Chinese (zh)
Other versions
CN108495095A (en
Inventor
匡卫红
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan City University
Original Assignee
Hunan City University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hunan City University filed Critical Hunan City University
Priority to CN201810439091.7A priority Critical patent/CN108495095B/en
Publication of CN108495095A publication Critical patent/CN108495095A/en
Application granted granted Critical
Publication of CN108495095B publication Critical patent/CN108495095B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10032—Satellite or aerial image; Remote sensing
    • G06T2207/10044—Radar image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30232—Surveillance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30241—Trajectory

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Emergency Alarm Devices (AREA)

Abstract

The invention belongs to the technical field of haze monitoring, and discloses a haze diffusion monitoring system based on an unmanned aerial vehicle, which comprises: the haze monitoring system comprises a video acquisition module, a haze monitoring module, a single-chip microcomputer control module, a wireless communication module, a haze image analysis module, a diffusion path drawing module, a data storage module and an alarm module. The haze image analysis module can be used for carrying out detailed analysis on the haze image to obtain accurate data information of haze, and the diffusion path drawing module can be used for drawing the diffusion path of the haze, so that timely protection on haze diffusion is facilitated; can in time report to the police haze monitoring abnormal data through the alarm, be favorable to doing timely safeguard measure.

Description

Haze diffusion monitoring system based on unmanned aerial vehicle
Technical Field
The invention belongs to the technical field of haze monitoring, and particularly relates to a haze diffusion monitoring system based on an unmanned aerial vehicle.
Background
Haze, which is a combination word of fog and haze. Haze is common in cities. In China, the fog is merged into the haze to be used as a disaster weather phenomenon for early warning and forecasting, and the phenomenon is called as haze weather. Haze is the result of specific climatic conditions interacting with human activity. Economic and social activities of high-density population inevitably discharge a large amount of fine particulate matters (PM 2.5), once the discharge exceeds the atmospheric circulation capacity and the bearing capacity, the concentration of the fine particulate matters is continuously accumulated, and at the moment, if the influence of calm weather and the like is caused, the haze in a large range is extremely easy to appear. However, the data acquired by the existing haze diffusion monitoring is inaccurate, and the diffusion path cannot be judged; meanwhile, if the haze is serious, the alarm cannot be sent out in time.
In summary, the problems of the prior art are as follows: the existing haze diffusion monitoring method has the disadvantages that the obtained data is inaccurate, and the diffusion path cannot be judged; meanwhile, if the haze is serious, the alarm cannot be sent out in time.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a haze diffusion monitoring system based on an unmanned aerial vehicle.
The invention is realized in this way, a haze diffusion monitoring system based on unmanned aerial vehicle includes:
the video acquisition module is connected with the single-chip microcomputer control module and used for acquiring haze video images through the camera;
the video acquisition module performs polar coordinate transformation on the image by taking the current visual attention focus as an original point, and searches in a polar coordinate space;
let P denote the set of all pixels in the polar boundary map, and L ═ {0,1} is the set of possible labels for each pixel in the polar boundary map;
a label function is sought on the probability boundary graph from the set P to the set L, which satisfies the following energy equation:
Figure BDA0001655444560000021
in the formula, delta (l)p,lq) Is a Kronecker symbol, Up(lp) Is an energy function data item;
the position of the focus is obtained by an Itti visual attention model, polar coordinate transformation is carried out on the probability boundary graph by taking the focus as a coordinate origin to obtain a polar coordinate probability boundary graph, an optimal boundary is obtained according to the formula, and inverse polar coordinate transformation is carried out on the optimal boundary to obtain a closed region, namely a current focus region;
focal region Ai,Aj，ρ(Ai,Aj) Is a focal region Ai,AjBased on the similarity measurement of Bhattacharyya coefficient, let Hi,HjAre respectively Ai,AjAnd (3) normalizing the histogram of the region, wherein the Bhattacharyya coefficient between the two regions is as follows:
Figure BDA0001655444560000022
in the formula, puRepresents a histogram HiThe u-th dimension of (a), quRepresents a histogram HjThe u-th dimension of (1);
the Bhattacharyya coefficient is practically equal to the calculation unit vector
Figure BDA0001655444560000023
And
Figure BDA0001655444560000024
the cosine distance between;
setting a focus area Ai,AjAre mutually crossed, and Ai∩AjIf is equal to S
Figure BDA0001655444560000025
Then A will bei,AjDirectly merging; if it is
Figure BDA0001655444560000026
Then it cannot merge and S is driven from ai,AjRemoving any one of them;
for the focal region Ai,Aj，AiIs contained in AjOr AjIs contained in AiFirst from AjMiddle deletion of and AiSimilar superpixels, set to
Figure BDA0001655444560000027
And then calculating according to the Bhattacharyya coefficient
Figure BDA0001655444560000028
And AiIf the similarity is lower than a predetermined threshold, then A is determinedjNeutral with AiSimilar superpixel from AjIs removed as AiThe super pixel of (2);
the number of the adjusted focus areas is the number of the final segmentation areas, and the superpixels in the focus areas are seed points of the segmentation areas;
the haze monitoring module is connected with the single-chip microcomputer control module and used for monitoring haze data information through a haze monitor;
the haze monitoring module takes the super pixels of the image to be segmented as nodes to form a graph G, and defines a Laplace matrix of the graph G:
Figure BDA0001655444560000031
Figure BDA0001655444560000032
is a pole SiDegree of (c) is defined as all and the pole SiThe sum of the weights between the connected poles;
the final number of the focus areas is K, the area marking variable is t, and t is more than or equal to 1 and less than or equal to K;
for a certain partition area, all nodes in the graph are divided into two types: set of marked points VMAnd a set of unmarked points VU，VM∪VUIs equal to V and VM∩VUPhi, according to the different sets to which the nodes belong, the Laplace matrix is written as:
Figure BDA0001655444560000033
node setting
Figure BDA0001655444560000034
The probability of reaching the seed point marked t is
Figure BDA0001655444560000035
Defining a marking function for seed points of a current focus area
Figure BDA0001655444560000036
For all VMThe nodes in (2) are:
Figure BDA0001655444560000037
for unlabeled VUThe probability of the node in (1) for the seed point marked as t is solved according to the Direchlett boundary condition and the following formula:
LUX＝-BTM
finally according to LUDetermining which segmentation area the super-pixel seed point belongs to according to the probability from each unmarked point to each super-pixel seed point;
the single-chip microcomputer control module is connected with the video acquisition module, the haze monitoring module, the wireless communication module, the haze image analysis module, the diffusion path drawing module, the data storage module and the alarm module, and is used for processing and analyzing data information acquired by the video acquisition module and the haze monitoring module and scheduling each module to normally work;
the wireless communication module is connected with the single chip microcomputer control module and is used for controlling the single chip microcomputer control module in a wireless mode;
the haze image analysis module is connected with the single-chip microcomputer control module and used for processing and analyzing haze video images collected by the video collection module;
the diffusion path drawing module is connected with the single-chip microcomputer control module and used for drawing the haze diffusion path;
the data storage module is connected with the single-chip microcomputer control module and used for storing haze monitoring data;
and the alarm module is connected with the singlechip control module and used for alarming the detected abnormal data through the alarm.
Further, the haze image analysis module comprises the following analysis method:
firstly, determining the time interval of image acquisition, and acquiring a monitoring video image in real time;
secondly, removing invalid images, extracting image contrast change, gradient change and visibility information, and determining haze indexes of the haze images;
and then, comparing the current image with the haze-free reference image in the sample library to determine the dust-haze grade. If the haze level of the image at the same place changes in a short time, judging that the image is a fog image and not a haze image, and deleting the fog image;
and finally, labeling the video haze index, the grade, the time and the position information of the analyzed image and storing the labeled video haze index, the grade, the time and the position information into a database.
Further, the diffusion path drawing module draws the following method:
firstly, valuable monitoring video position and time information is selected, and specific position information and time information in a haze diffusion path are obtained by combining the haze index of a monitoring network point and the trend of the level information changing along with time.
Secondly, carrying out overall haze trajectory analysis by using the remote sensing image: analyzing the haze remote sensing image by using a HYSPUT mode, and tracing a pollution source on a large scale;
then, drawing a haze diffusion path: and drawing a dust-haze diffusion path changing along with time according to the obtained position, dust-haze index, grade information and remote sensing analysis data with the time series characteristics. The drawing time can be in units of hours, days and weeks;
finally, determination of the source grid: and finding the source grids of the dust haze according to the dust haze diffusion path and the dust haze grade change process along with time, thereby reducing the range of the source main body and determining the responsibility main body.
Further, the PM2.5 measurement module analyzes as follows:
is provided with: the PM2.5 measuring instrument is embedded on the shell and is connected with the data acquisition unit and the singlechip through a wire; the data acquisition unit is connected with the liquid crystal display screen through a wire; the liquid crystal display screen is embedded in the shell; the range finder is connected with the data acquisition unit and is connected with the singlechip through a lead; the casing is fixed in on the unmanned aerial vehicle through the fixing clip.
Further, the alarm module analyzes as follows:
is provided with: shell body, novel singlechip, wireless voice emitter, speaker, converter. The outer shell is fixed on the unmanned aerial vehicle through a staple bolt; the novel single chip microcomputer is embedded in the outer shell; the wireless voice transmitter is connected with the novel single chip microcomputer through the wireless receiver and used for collecting data and transmitting the data; the converter is connected with the novel single chip microcomputer and used for converting signal formats; the loudspeaker is connected with the converter and used for alarming and warning.
The invention has the advantages and positive effects that: the haze image analysis module can be used for carrying out detailed analysis on the haze image to obtain accurate data information of haze, and the diffusion path drawing module can be used for drawing the diffusion path of the haze, so that timely protection on haze diffusion is facilitated; the PM2.5 measuring instrument transmits infrared rays to the measuring instrument, the measured data are transmitted to the data collector for recording after the infrared rays return, the distance meter returns the longitude and latitude of the measuring area to the data collector for recording, the collected data are transmitted to the liquid crystal display screen after the information processing of the data collector, the PM2.5 concentration of the determined coordinate position can be displayed in real time, workers can conveniently master the haze condition of each place, and the device is simpler and clearer and is convenient to control and deal with; the wireless voice transmitter collects the collected information, compresses the information data and transmits the data to a novel singlechip below the wireless voice transmitter through wireless transmission; then converting the signal into an electric signal under the action of a converter; at last, alarm information is transmitted through a loudspeaker, and workers below can be accurately informed of the specific haze condition, so that the system is efficient and convenient. Can in time report to the police haze monitoring abnormal data through the alarm, be favorable to doing timely safeguard measure.
Drawings
Fig. 1 is a block diagram of a haze diffusion monitoring system based on an unmanned aerial vehicle according to an embodiment of the present invention.
In the figure: 1. a video acquisition module; 2. a haze monitoring module; 3. a single chip microcomputer control module; 4. a wireless communication module; 5. a haze image analysis module; 6. a diffusion path drawing module; 7. a data storage module; 8. and an alarm module.
Fig. 2 is a block diagram of a structure of a PM2.5 measuring device of a haze diffusion monitoring system based on an unmanned aerial vehicle according to an embodiment of the present invention;
in the figure: 9. a housing; 10. a PM2.5 measuring instrument; 11. a range finder; 12. a data acquisition unit; 13. and a liquid crystal display screen.
Fig. 3 is a structural block diagram of an alarm device of the haze diffusion monitoring system based on the unmanned aerial vehicle according to the embodiment of the invention;
in the figure: 14. an outer housing; 15. a novel single chip microcomputer; 16. a wireless voice transmitter; 17. a speaker; 18. a converter.
Detailed Description
In order to further understand the contents, features and effects of the present invention, the following embodiments are illustrated and described in detail with reference to the accompanying drawings.
The structure of the present invention will be described in detail below with reference to the accompanying drawings.
As shown in fig. 1, the haze diffusion monitoring system based on the unmanned aerial vehicle provided by the invention comprises: the haze monitoring system comprises a video acquisition module 1, a haze monitoring module 2, a single-chip microcomputer control module 3, a wireless communication module 4, a haze image analysis module 5, a diffusion path drawing module 6, a data storage module 7 and an alarm module 8.
The video acquisition module 1 is connected with the singlechip control module 3 and is used for acquiring haze video images through a camera;
the haze monitoring module 2 is connected with the single-chip microcomputer control module 3 and used for monitoring haze data information through a haze monitor;
the single-chip microcomputer control module 3 is connected with the video acquisition module 1, the haze monitoring module 2, the wireless communication module 4, the haze image analysis module 5, the diffusion path drawing module 6, the data storage module 7 and the alarm module 8, and is used for processing and analyzing data information acquired by the video acquisition module 1 and the haze monitoring module 2 and scheduling each module to normally work;
the wireless communication module 4 is connected with the single-chip microcomputer control module 3 and is used for controlling the single-chip microcomputer control module 3 in a wireless mode;
the haze image analysis module 5 is connected with the single-chip microcomputer control module 3 and used for processing and analyzing haze video images acquired by the video acquisition module 1;
the diffusion path drawing module 6 is connected with the single-chip microcomputer control module 3 and used for drawing a haze diffusion path;
the data storage module 7 is connected with the single chip microcomputer control module 3 and used for storing haze monitoring data;
and the alarm module 8 is connected with the singlechip control module 3 and used for alarming the detected abnormal data through an alarm.
The video acquisition module performs polar coordinate transformation on the image by taking the current visual attention focus as an original point, and searches in a polar coordinate space;
let P denote the set of all pixels in the polar boundary map, and L ═ {0,1} is the set of possible labels for each pixel in the polar boundary map;
a label function is sought on the probability boundary graph from the set P to the set L, which satisfies the following energy equation:
Figure BDA0001655444560000071
in the formula, delta (l)p,lq) Is a Kronecker symbol, Up(lp) Is an energy function data item;
the position of the focus is obtained by an Itti visual attention model, polar coordinate transformation is carried out on the probability boundary graph by taking the focus as a coordinate origin to obtain a polar coordinate probability boundary graph, an optimal boundary is obtained according to the formula, and inverse polar coordinate transformation is carried out on the optimal boundary to obtain a closed region, namely a current focus region;
focal region Ai,Aj，ρ(Ai,Aj) Is a focal region Ai,AjBased on the similarity measurement of Bhattacharyya coefficient, let Hi,HjAre respectively Ai,AjAnd (3) normalizing the histogram of the region, wherein the Bhattacharyya coefficient between the two regions is as follows:
Figure BDA0001655444560000081
in the formula, puRepresents a histogram HiThe u-th dimension of (a), quRepresents a histogram HjThe u-th dimension of (1);
the Bhattacharyya coefficient is practically equal to the calculation unit vector
Figure BDA0001655444560000082
And
Figure BDA0001655444560000083
the cosine distance between;
setting a focus area Ai,AjAre mutually crossed, and Ai∩AjIf is equal to S
Figure BDA0001655444560000084
Then A will bei,AjDirectly merging; if it is
Figure BDA0001655444560000085
Then it cannot merge and S is driven from ai,AjRemoving any one of them;
for the focal region Ai,Aj，AiIs contained in AjOr AjIs contained in AiFirst from AjMiddle deletion of and AiSimilar superpixels, set to
Figure BDA0001655444560000086
And then calculating according to the Bhattacharyya coefficient
Figure BDA0001655444560000087
And AiIf the similarity is lower than a predetermined threshold, then A is determinedjNeutral with AiSimilar superpixel from AjIs removed as AiThe super pixel of (2);
the number of the adjusted focus areas is the number of the final segmentation areas, and the superpixels in the focus areas are seed points of the segmentation areas;
the haze monitoring module takes the super pixels of the image to be segmented as nodes to form a graph G, and defines a Laplace matrix of the graph G:
Figure BDA0001655444560000088
Figure BDA0001655444560000089
is a pole SiDegree of (c) is defined as all and the pole SiThe sum of the weights between the connected poles;
the final number of the focus areas is K, the area marking variable is t, and t is more than or equal to 1 and less than or equal to K;
for a certain partition area, all nodes in the graph are divided into two types: set of marked points VMAnd a set of unmarked points VU，VM∪VUIs equal to V and VM∩VUPhi, according to the different sets to which the nodes belong, the Laplace matrix is written as:
Figure BDA0001655444560000091
node setting
Figure BDA0001655444560000092
The probability of reaching the seed point marked t is
Figure BDA0001655444560000093
Defining a marking function for seed points of a current focus area
Figure BDA0001655444560000094
For all VMThe nodes in (2) are:
Figure BDA0001655444560000095
for unlabeled VUThe probability of the node in (1) for the seed point marked as t is solved according to the Direchlett boundary condition and the following formula:
LUX＝-BTM
finally according to LUDetermining which segmentation area the super-pixel seed point belongs to according to the probability from each unmarked point to each super-pixel seed point;
as shown in fig. 2, the device for measuring PM2.5 of the unmanned aerial vehicle-based haze diffusion monitoring system provided by the invention comprises: the device comprises a shell 9, a PM2.5 measuring instrument 10, a range finder 11, a data acquisition unit 12 and a liquid crystal display 13.
The PM2.5 measuring instrument 10 is embedded on the shell 9 and is connected with the data acquisition unit 12 and the single chip through a wire; the data acquisition device 12 is connected with the liquid crystal display screen 13 through a wire; the liquid crystal display screen 13 is embedded in the shell 9; the distance measuring instrument 11 is connected to the data acquisition device 12 and is connected with the singlechip 14 through a lead; casing 9 is fixed in on the unmanned aerial vehicle through the fixing clip.
As shown in fig. 3, the alarm device of the haze diffusion monitoring system based on the unmanned aerial vehicle provided by the invention comprises: outer casing 14, novel singlechip 15, wireless voice emitter 16, speaker 17, converter 18. The outer shell 14 is fixed on the unmanned aerial vehicle through a staple; the novel single chip microcomputer 15 is embedded in the outer shell 14; the wireless voice transmitter 16 is connected with the novel single chip microcomputer 15 through a wireless receiver and used for collecting data and transmitting the data; the converter is connected with the novel single chip microcomputer 15 of 18 and used for converting signal formats; the speaker 17 is connected to the transducer 18 for alarm warning.
The haze image analysis module 5 provided by the invention has the following analysis method:
firstly, determining the time interval of image acquisition, and acquiring a monitoring video image in real time;
secondly, removing invalid images, extracting image contrast change, gradient change and visibility information, and determining haze indexes of the haze images;
and then, comparing the current image with the haze-free reference image in the sample library to determine the dust-haze grade. If the haze level of the image at the same place changes in a short time, judging that the image is a fog image and not a haze image, and deleting the fog image;
and finally, labeling the video haze index, the grade, the time and the position information of the analyzed image and storing the labeled video haze index, the grade, the time and the position information into a database.
The drawing method of the diffusion path drawing module 6 provided by the invention is as follows:
firstly, valuable monitoring video position and time information is selected, and specific position information and time information in a haze diffusion path are obtained by combining the haze index of a monitoring network point and the trend of the level information changing along with time.
Secondly, carrying out overall haze trajectory analysis by using the remote sensing image: analyzing the haze remote sensing image by using a HYSPUT mode, and tracing a pollution source on a large scale;
then, drawing a haze diffusion path: and drawing a dust-haze diffusion path changing along with time according to the obtained position, dust-haze index, grade information and remote sensing analysis data with the time series characteristics. The drawing time can be in units of hours, days and weeks;
finally, determination of the source grid: and finding the source grids of the dust haze according to the dust haze diffusion path and the dust haze grade change process along with time, thereby reducing the range of the source main body and determining the responsibility main body.
In the monitoring process, a haze video image is collected through the video collecting module 1; monitoring haze data information through a haze monitoring module 2; the singlechip control module 3 processes and analyzes data information acquired by the video acquisition module 1 and the haze monitoring module 2; the staff controls the singlechip control module 3 through the wireless communication module 4; the single chip microcomputer control module 3 dispatches the haze image analysis module 5 to process and analyze the haze video image acquired by the video acquisition module 1; a haze diffusion path is drawn through a diffusion path drawing module 6; haze monitoring data are stored through a data storage module 7; and if the detected abnormal data exist, alarming is carried out through an alarming module 8. The PM2.5 measuring instrument 10 transmits infrared rays to the measurement device, the measured data is transmitted to the data acquisition device 12 to be recorded after the infrared rays return, the distance measuring instrument 11 returns the longitude and latitude of the measuring area to the data acquisition device 12 to be recorded, the collected data is transmitted to the liquid crystal display 13 after the information processing of the data acquisition device 12, the PM2.5 concentration of the determined coordinate position can be displayed in real time, workers can conveniently master the haze condition of each place, and the device is simpler and clearer and is convenient to control and respond; the wireless voice transmitter 16 collects the collected information, compresses the information data, and transmits the data to the novel singlechip 15 below through wireless transmission; the signal is then converted into an electrical signal by the action of the transducer 18; at last, alarm information is transmitted through the loudspeaker 17, the specific haze condition of the workers below can be accurately notified, and the method is efficient and convenient.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and all simple modifications, equivalent changes and modifications made to the above embodiment according to the technical spirit of the present invention are within the scope of the technical solution of the present invention.

Claims (3)

1. The utility model provides a haze diffusion monitored control system based on unmanned aerial vehicle, a serial communication port, haze diffusion monitored control system based on unmanned aerial vehicle includes:
the video acquisition module is connected with the single-chip microcomputer control module and used for acquiring haze video images through the camera;
the video acquisition module performs polar coordinate transformation on the image by taking the current visual attention focus as an original point, and searches in a polar coordinate space;
let P denote the set of all pixels in the polar boundary map, and L ═ {0,1} is the set of possible labels for each pixel in the polar boundary map;
a label function is sought on the probability boundary graph from the set P to the set L, which satisfies the following energy equation:
Figure FDA0003015933620000011
Figure FDA0003015933620000012
in the formula, delta (l)p,lq) Is a Kronecker symbol, Up(lp) Is an energy function data item;
the position of the focus is obtained by an Itti visual attention model, polar coordinate transformation is carried out on the probability boundary graph by taking the focus as a coordinate origin to obtain a polar coordinate probability boundary graph, an optimal boundary is obtained according to the formula, and inverse polar coordinate transformation is carried out on the optimal boundary to obtain a closed region, namely a current focus region;
focal region Ai,Aj，ρ(Ai,Aj) Is a focal region Ai,AjBased on the similarity measurement of Bhattacharyya coefficient, let Hi,HjAre respectively Ai,AjAnd (3) normalizing the histogram of the region, wherein the Bhattacharyya coefficient between the two regions is as follows:
Figure FDA0003015933620000013
in the formula, puRepresents a histogram HiThe u-th dimension of (a), quRepresents a histogram HjThe u-th dimension of (1);
the Bhattacharyya coefficient is practically equal to the calculation unit vector
Figure FDA0003015933620000014
And
Figure FDA0003015933620000015
the cosine distance between;
setting a focus area Ai,AjAre mutually crossed, and Ai∩AjIf is equal to S
Figure FDA0003015933620000021
Then A will bei,AjDirectly merging; if it is
Figure FDA0003015933620000022
Then it cannot merge and S is driven from ai,AjRemoving any one of them;
for the focal region Ai,Aj，AiIs contained in AjFirst from AjMiddle deletion of and AiSimilar superpixels, set to
Figure FDA0003015933620000023
And then calculating according to the Bhattacharyya coefficient
Figure FDA0003015933620000024
And AiIf the similarity is lower than a predetermined threshold, then A is determinedjNeutral with AiSimilar superpixel from AjIs removed as AiThe super pixel of (2);
the number of the adjusted focus areas is the number of the final segmentation areas, and the superpixels in the focus areas are seed points of the segmentation areas;
the haze monitoring module is connected with the single-chip microcomputer control module and used for monitoring haze data information through a haze monitor;
the haze monitoring module takes the super pixels of the image to be segmented as nodes to form a graph G, and defines a Laplace matrix of the graph G:
Figure FDA0003015933620000025
Figure FDA0003015933620000026
is a pole SiDegree of (c) is defined as all and the pole SiThe sum of the weights between the connected poles;
the final number of the focus areas is K, the area marking variable is t, and t is more than or equal to 1 and less than or equal to K;
for a certain partition area, all nodes in the graph are divided into two types: set of marked points VMAnd a set of unmarked points VU，VM∪VUIs equal to V and VM∩VUPhi, according to the different sets to which the nodes belong, the Laplace matrix is written as:
Figure FDA0003015933620000027
node setting
Figure FDA0003015933620000028
The probability of reaching the seed point marked t is
Figure FDA0003015933620000029
Defining a marking function for seed points of a current focus area
Figure FDA00030159336200000210
For all VMThe nodes in (2) are:
Figure FDA00030159336200000211
for unlabeled VUThe probability of the node in (1) for the seed point marked as t is solved according to the Direchlett boundary condition and the following formula:
LUX＝-BTM
finally according to LUDetermining which segmentation area the super-pixel seed point belongs to according to the probability from each unmarked point to each super-pixel seed point;
the single-chip microcomputer control module is connected with the video acquisition module, the haze monitoring module, the wireless communication module, the haze image analysis module, the diffusion path drawing module, the data storage module and the alarm module, and is used for processing and analyzing data information acquired by the video acquisition module and the haze monitoring module and scheduling each module to normally work;
the wireless communication module is connected with the single chip microcomputer control module and is used for controlling the single chip microcomputer control module in a wireless mode;
the haze image analysis module is connected with the single-chip microcomputer control module and used for processing and analyzing haze video images collected by the video collection module;
the diffusion path drawing module is connected with the single-chip microcomputer control module and used for drawing the haze diffusion path;
the data storage module is connected with the single-chip microcomputer control module and used for storing haze monitoring data;
the alarm module is connected with the singlechip control module and used for alarming the detected abnormal data through an alarm;
haze monitoring module is provided with: the PM2.5 measuring instrument is embedded on the shell and is connected with the data collector and the singlechip control module through a wire; the data acquisition unit is connected with the liquid crystal display screen through a wire; the liquid crystal display screen is embedded in the shell; the range finder is connected to the data acquisition unit and is connected with the singlechip control module through a lead; the shell is fixed on the unmanned aerial vehicle through the fixing card, the PM2.5 measuring instrument transmits infrared rays to the measuring area, the infrared rays transmit the measured data to the data acquisition unit for recording after returning, the distance measuring instrument returns the longitude and latitude of the measuring area to the data acquisition unit for recording, the collected data are transmitted to the liquid crystal display screen through information processing of the data acquisition unit, and the PM2.5 concentration of the determined coordinate position is displayed in real time;
the alarm module is provided with: the device comprises an outer shell, a single chip microcomputer, a wireless voice emitter, a loudspeaker and a converter, wherein the outer shell is fixed on the unmanned aerial vehicle through a staple bolt; the singlechip is embedded on the shell; the wireless voice transmitter is connected with the singlechip through the wireless receiver and is used for collecting data and transmitting the data; the converter is connected with the single chip microcomputer and used for converting signal formats; the loudspeaker is connected with the converter and used for alarming and warning; the wireless voice transmitter collects the collected information, compresses the information data and transmits the data to the singlechip below through wireless transmission; then converting the signal into an electric signal under the action of a converter; and finally, the alarm information is transmitted out through a loudspeaker.
2. The unmanned-aerial-vehicle-based haze diffusion monitoring system of claim 1, wherein the haze image analysis module analysis method is as follows:
firstly, determining the time interval of image acquisition, and acquiring a monitoring video image in real time;
secondly, removing invalid images, extracting image contrast change, gradient change and visibility information, and determining haze indexes of the haze images;
then, comparing the current image with a haze-free reference image in a sample library, determining a haze level, if the haze level of the image at the same place changes in a short time, judging that the image is a fog image and not a haze image, and deleting the fog image;
and finally, labeling the video haze index, the grade, the time and the position information of the analyzed image and storing the labeled video haze index, the grade, the time and the position information into a database.
3. The unmanned aerial vehicle-based haze diffusion monitoring system of claim 1, wherein the diffusion path drawing module draws the following method:
firstly, selecting valuable monitoring video position and time information, and obtaining specific position information and time information in a haze diffusion path by combining the haze index of a monitoring network point and the trend of the level information changing along with time;
secondly, carrying out overall haze trajectory analysis by using the remote sensing image: analyzing the haze remote sensing image by using a HYSPUT mode, and tracing a pollution source on a large scale;
then, drawing a haze diffusion path: drawing a dust-haze diffusion path which changes along with time according to the obtained position, dust-haze index, grade information and remote sensing analysis data with time series characteristics; the drawing time is in units of hours, days or weeks;
finally, determination of the source grid: and finding the source grids of the dust haze according to the dust haze diffusion path and the dust haze grade change process along with time, thereby reducing the range of the source main body and determining the responsibility main body.
CN201810439091.7A 2018-05-09 2018-05-09 Haze diffusion monitoring system based on unmanned aerial vehicle Active CN108495095B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810439091.7A CN108495095B (en) 2018-05-09 2018-05-09 Haze diffusion monitoring system based on unmanned aerial vehicle

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810439091.7A CN108495095B (en) 2018-05-09 2018-05-09 Haze diffusion monitoring system based on unmanned aerial vehicle

Publications (2)

Publication Number Publication Date
CN108495095A CN108495095A (en) 2018-09-04
CN108495095B true CN108495095B (en) 2021-08-03

Family

ID=63352722

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810439091.7A Active CN108495095B (en) 2018-05-09 2018-05-09 Haze diffusion monitoring system based on unmanned aerial vehicle

Country Status (1)

Country Link
CN (1) CN108495095B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108872040A (en) * 2018-09-30 2018-11-23 徐州工业职业技术学院 A kind of city haze monitoring system
CN109916791A (en) * 2019-04-16 2019-06-21 中国海洋大学 Gray haze vertical structure survey meter
CN113588507B (en) * 2021-07-28 2022-07-05 沭阳新辰公路仪器有限公司 Automatic haze weather detection and warning system and device for road traffic
CN114842621A (en) * 2022-05-31 2022-08-02 北京万云科技开发有限公司 Sand-dust weather early warning method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106355166A (en) * 2016-10-13 2017-01-25 北京师范大学 Monitoring video and remote sensing image-based dust-haze spreading path drawing and source determination method
CN106454241A (en) * 2016-10-13 2017-02-22 北京师范大学 Dust haze diffusion path drawing and source determining method based on monitoring video and social network data
CN106556558A (en) * 2015-09-28 2017-04-05 东莞前沿技术研究院 Haze monitoring system
CN107341576A (en) * 2017-07-14 2017-11-10 河北百斛环保科技有限公司 A kind of visual air pollution of big data is traced to the source and trend estimate method
CN107976220A (en) * 2017-12-24 2018-05-01 安徽省环境科学研究院 Based on Atmospheric components synchronization detecting system and method under fixed point different height

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104834912B (en) * 2015-05-14 2017-12-22 北京邮电大学 A kind of weather recognition methods and device based on image information detection

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106556558A (en) * 2015-09-28 2017-04-05 东莞前沿技术研究院 Haze monitoring system
CN106355166A (en) * 2016-10-13 2017-01-25 北京师范大学 Monitoring video and remote sensing image-based dust-haze spreading path drawing and source determination method
CN106454241A (en) * 2016-10-13 2017-02-22 北京师范大学 Dust haze diffusion path drawing and source determining method based on monitoring video and social network data
CN107341576A (en) * 2017-07-14 2017-11-10 河北百斛环保科技有限公司 A kind of visual air pollution of big data is traced to the source and trend estimate method
CN107976220A (en) * 2017-12-24 2018-05-01 安徽省环境科学研究院 Based on Atmospheric components synchronization detecting system and method under fixed point different height

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于视觉注意的随机游走图像分割;王富治 等;《仪器仪表学报》;20170731;第38卷(第7期);第1772-1777页 *

Also Published As

Publication number Publication date
CN108495095A (en) 2018-09-04

Similar Documents

Publication Publication Date Title
CN108495095B (en) Haze diffusion monitoring system based on unmanned aerial vehicle
CN108216252B (en) Subway driver vehicle-mounted driving behavior analysis method, vehicle-mounted terminal and system
CN109407177B (en) Machine learning and conventional meteorological observation-based fog identification system and application method
CN110930048A (en) Crop drought risk assessment system and method based on disaster mechanism process
EP4177803A1 (en) Vegetation management system and vegetation management method
Kuang et al. A remote sensing monitoring method for alpine grasslands desertification in the eastern Qinghai-Tibetan Plateau
Sun et al. Remote estimation of grafted apple tree trunk diameter in modern orchard with RGB and point cloud based on SOLOv2
CN114518143A (en) Intelligent environment sensing system
CN210183356U (en) Sponge city rainstorm detection early warning system
CN113593191A (en) Visual urban waterlogging monitoring and early warning system based on big data
KR20120132985A (en) Search system for spatiotemporal distribution of disaster weather when disturbance approaches including typhoon
CN109195126B (en) Pipeline information acquisition system
CN116704366A (en) Storm drop zone identification early warning method and device based on transform deep learning model
CN117197978A (en) Forest fire monitoring and early warning system based on deep learning
CN116413740A (en) Laser radar point cloud ground detection method and device
CN116702588A (en) Wind power plant key weather factor forecasting method and system based on multi-source data
CN110765900A (en) DSSD-based automatic illegal building detection method and system
JP5563683B1 (en) ENVIRONMENTAL INFORMATION PROVIDING DEVICE AND PROGRAM
CN115311591A (en) Early warning method and device for abnormal behaviors and intelligent camera
Guillermo et al. A Comparative Study on Satellite Image Analysis for Road Traffic Detection using YOLOv3-SPP, Keras RetinaNet and Full Convolutional Network
CN115100551A (en) Land use attribute monitoring method based on remote sensing image technology
CN114005041A (en) Road disease identification control method and equipment based on UAVRS and BIM
CN113111790A (en) Intelligent community management system, method and storage medium based on time sequence database
CN111611872A (en) Novel binocular vision vehicle detection method and system
CN116739357B (en) Multi-mode fusion perception city existing building wide area monitoring and early warning method and device

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