Unmanned aerial vehicle intelligent inspection method for particulate matter pollution prevention and control

Abstract

The invention discloses an unmanned aerial vehicle intelligent inspection method for particulate matter pollution prevention and control, and the method comprises the steps: S1, carrying out the aerial photographing to obtain a particulate matter pollution source image, and generating a pollution source image recognition model; s2, aerial photography is carried out to obtain a particulate matter pollution management and control facility image, and a pollution management and control facility image recognition model is generated; s3, collecting related data, and generating a particulate matter pollution source and particulate matter pollution management and control facility list and distribution map; s4, the terminal intelligent control platform plans an unmanned aerial vehicle inspection route; s5, when the particulate matter concentration data of the monitoring site reach an early warning value, starting inspection operation; s6, transmitting the data recorded by the inspection of the unmanned aerial vehicle to the terminal in real time; s7, the terminal identifies the pollution source type, the pollution degree and the pollution control facility state according to the transmitted data to form a pollution source patrol list; and S8, automatically sending the generated list to related personnel. The method is high in target identification precision and accurate in positioning, improves the inspection efficiency of particulate matter pollution prevention and control, and reduces the labor cost.

Description

Technical Field

[0001] The present invention relates to the field of particle pollution prevention and control, and in particular to an unmanned aerial vehicle intelligent inspection method for particle pollution prevention and control. Background Art

[0002] With the development of social economy, the emission of atmospheric pollutants from industrial enterprises is increasing day by day, which not only has an adverse impact on environmental quality, but also has an adverse impact on human health. The public and the government have an urgent need to improve atmospheric pollution prevention and control technologies and equipment. At present, the inspection of atmospheric pollution sources mainly relies on on-site inspection or vehicle cruise inspection. Due to the large number of pollution sources, manual inspection and vehicle cruise inspection are time-consuming and labor-intensive, the inspection area is limited, the degree of intelligence is low, the inspection cost is high, and the timeliness of pollution response and disposal is poor. Summary of the invention

[0003] 1. Technical issues to be resolved

[0004] The purpose of this invention is to address the shortcomings of traditional on-site air pollution inspections, such as high labor and material consumption and low work efficiency, by utilizing an unmanned aerial vehicle intelligent inspection method to save labor consumption in particulate pollution prevention and control work, thereby greatly improving the efficiency and intelligence level of the inspection work.

[0005] (II) Technical solution

[0006] The present invention proposes an unmanned aerial vehicle intelligent inspection method for particulate pollution prevention and control, comprising the following steps: S1: obtaining an image of a particulate pollution source through unmanned aerial vehicle aerial photography, wherein the particulate pollution source includes construction site dust, road dust, bare soil, and open-air burning, denoising, enhancing, adjusting contrast, and adjusting brightness of the pollution source image, extracting image features of construction site dust, road dust, bare soil, and open-air burning based on an edge detection algorithm, establishing an image feature database of construction site dust, road dust, bare soil, and open-air burning, and generating a pollution source image recognition model based on a machine learning algorithm; S2: obtaining an image of a particulate pollution control facility through unmanned aerial vehicle aerial photography, wherein the particulate pollution control facility includes enclosure spraying, fog cannon, and dust prevention. The image of the particulate pollution control facility is processed by denoising, enhancing, adjusting contrast and brightness. Based on the edge detection algorithm, the image features of the shutdown and startup status of the enclosure spraying and fog cannon, as well as the image features of the dust net coverage status, are extracted respectively. An image feature database of the shutdown and startup status of the enclosure spraying and fog cannon, as well as the dust net coverage status, is established. Based on the machine learning algorithm, an image recognition model for the implementation status of the pollution control facility is generated; S3: The particulate pollution source data and the control facility data of the pollution source in the control area are collected, and the collected data are sent to the data analysis system through the data communication system. Corresponding to the information of the pollution source management department and the person in charge of management, the list of particulate pollution sources and pollution control facilities is derived, and the particle The location distribution map of particulate pollution sources and particulate pollution control facilities is visually displayed through a three-dimensional spatial map; S4: The terminal intelligent control platform uses the selected air quality model to calculate the particulate pollution diffusion path based on the particulate pollution source distribution map, terrain elevation, and meteorological environmental parameters in the control area, and collects the particulate matter concentration data of each monitoring point in the area to verify and optimize the air quality model, and plans the drone inspection route according to the three-dimensional map and the optimized pollution diffusion path; S5: When the terminal intelligent control platform detects that the particulate matter concentration data at the monitoring station reaches the set warning value, it will broadcast a high value and issue an inspection instruction to the drone automatic hangar. After receiving the instruction, the drone automatic hangar will automatically receive the warning value from the terminal intelligent control platform. Call the drone inspection route and start the drone inspection operation; S6: During the drone inspection flight, the particle concentration data, meteorological environmental parameters and terrain elevation data in the passing direction are monitored and recorded in real time, and the points where the particle concentration exceeds the standard are automatically taken in all directions and at multiple angles. High-definition shooting, and relevant data are transmitted to the terminal intelligent control platform in real time; S7: The terminal intelligent control platform uses the pollution source image recognition model and the particle pollution control facility operation status image recognition model to identify and determine the type of pollution source, the degree of pollution and the operation status of the pollution control facility based on the data transmitted by the drone, mark the problem points, and form an inspection report; S8: The terminal intelligent control platform automatically sends the generated inspection report to the relevant personnel for processing.

[0007] Preferably, the image recognition model adjusts the network structure, optimizes the training parameters, enhances the image data, introduces the attention module, and extracts the edge features of the image based on the Yolov7 framework, thereby improving the speed and accuracy of image recognition. The image recognition training process specifically includes the following steps: S101: scaling and filling the collected images, and adjusting the size to 640×640; S102: standardizing the images, and normalizing the pixel values ​​of the images to (0,1); S103: denoising, enhancing, adjusting the contrast, and adjusting the brightness of the images, and randomly Crop, rotate, and scale to increase the number of images for training; S104: Use convolutional neural networks to extract image features, combine convolutional layers and pooling layers, dynamically adjust the number of convolutional layers, and refine the extraction of image shape, color, texture, and edge features; S105: Input image data into the Yolov7 model. The batch size of model training starts from 32 and decreases in multiples of 4. The number of training rounds starts from 220 and decreases in multiples of 20. The learning rate is reduced from 0.01 to 0.0001. During the training process, the error between the image recognition result and the true label is dynamically evaluated to gradually establish an optimized image recognition model.

[0008] Preferably, the list information of particulate matter pollution sources and pollution control facilities includes the type of pollution source, geographical location, emission method, pollution control facilities, management department, and person responsible for management.

[0009] Preferably, the terminal intelligent control platform displays the flight trajectory of the drone in real time.

[0010] Preferably, in step S6, the drone takes all-round, multi-angle high-definition photos of the points where the particulate matter concentration exceeds the standard, transmits the relevant data to the terminal intelligent control platform in real time, and then continues to identify and lock new targets to achieve continuous inspection of multiple targets.

[0011] Preferably, the meteorological environment parameters include temperature, humidity, wind direction and wind speed.

[0012] Preferably, the terminal intelligent control platform compares the inspection results of the same pollution source in different time periods, forms a "look back" report, automatically generates a "closed-loop rate" for pollution source treatment, and sends it to relevant personnel.

[0013] (III) Beneficial effects

[0014] The present invention has the following beneficial effects:

[0015] (1) Based on automatic image recognition technology, the pollution sources and control facilities within the inspection area are automatically identified.

[0016] (2) The drone inspection route is targeted, which improves the inspection efficiency and intelligence level and reduces the manpower and material costs during the inspection process.

[0017] (3) The terminal intelligent control platform can automatically generate inspection reports and send them to relevant personnel, improving the response speed and management efficiency of pollution incidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a flow chart of an embodiment of the present invention.

[0019] Figure 2 Inspection route map planned for the embodiment of the present invention DETAILED DESCRIPTION

[0020] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as being limited to the embodiments described herein, but rather these embodiments are provided for a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention. It should be understood that the various steps described in the method implementation of the present invention can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0021] like Figure 1-2 As shown, the present invention proposes an intelligent inspection method of a drone for particulate matter pollution prevention and control, comprising the following steps:

[0022] S1: Use drones to obtain images of particulate pollution sources, including construction site dust, road dust, bare soil, and open-air burning. Perform denoising, enhancement, contrast adjustment, and brightness adjustment on pollution source images. Use edge detection algorithms to extract image features of construction site dust, road dust, bare soil, and open-air burning, respectively. Establish image feature databases of construction site dust, road dust, bare soil, and open-air burning, and generate pollution source image recognition models based on machine learning algorithms.

[0023] S2: Obtain images of particulate pollution control facilities through drone aerial photography. Particulate pollution control facilities include enclosure spraying, fog cannons, and dust nets. De-noise, enhance, adjust contrast, and adjust brightness of the images of particulate pollution control facilities. Based on the edge detection algorithm, extract the image features of the shutdown and startup states of enclosure spraying and fog cannons, as well as the image features of the dust net coverage state. Establish an image feature database of the shutdown and startup states of enclosure spraying, fog cannons, and the dust net coverage state. Generate an image recognition model for the implementation status of pollution control facilities based on a machine learning algorithm.

[0024] S3: Collect the data of particulate pollution sources and the control facilities of the pollution sources within the control area, send the collected data to the data analysis system through the data communication system, correspond to the information of the pollution source management department and the person in charge, and derive a list of particulate pollution sources and pollution control facilities including the type of pollution source, geographical location, emission method, pollution control facilities, management department, and person in charge, draw a distribution map of the location of particulate pollution sources and particulate pollution control facilities, and intuitively display the distribution through a three-dimensional spatial map;

[0025] S4: The terminal intelligent control platform uses the selected air quality model to calculate the particle pollution diffusion path based on the particle pollution source distribution map, terrain elevation, and meteorological environmental parameters in the control area, and collects the particle concentration data of each monitoring point in the area to calibrate and optimize the air quality model. It plans the drone inspection route based on the three-dimensional map and the optimized pollution diffusion path;

[0026] S5: When the terminal intelligent control platform detects that the particulate matter concentration data at the monitoring site reaches the set warning value, it broadcasts a high value and issues an inspection command to the drone automatic hangar. After receiving the command, the drone automatic hangar automatically calls the drone inspection route from the terminal intelligent control platform and starts the drone inspection operation;

[0027] S6: During the inspection flight, the drone monitors and records the particle concentration data, meteorological environmental parameters and terrain elevation data in the direction of the route in real time, automatically takes all-round, multi-angle high-definition photos of the points where the particle concentration exceeds the standard, transmits the relevant data to the terminal intelligent control platform in real time, and continues the inspection operation, identifies and locks on new targets, and realizes continuous inspection of multiple targets.

[0028] S7: The terminal intelligent control platform uses the pollution source image recognition model and the particle pollution control facility operation status image recognition model to identify and judge the type of pollution source, pollution degree and operation status of pollution control facilities based on the data transmitted by the drone, mark the problem points, display the flight trajectory of the drone in real time, and form an inspection report;

[0029] S8: The terminal intelligent control platform automatically sends the generated inspection report to relevant personnel for processing;

[0030] For example, when the concentration of particulate matter monitored by a micro-station in the area exceeds 80% of the limit specified by the station, the platform will issue a high-value reminder, call up a list of particulate matter pollution sources in the area, and use the air quality model to simulate the diffusion trajectory of pollutants in the area based on the local temperature, wind direction, wind speed, humidity and other environmental parameters at that time. After preliminarily determining the area where the pollution source is located, the inspection route is planned, and then the inspection command is issued to the drone. The drone monitors, shoots, and transmits the data of the point where the concentration of particulate matter exceeds the standard in real time according to the route planned by the platform. The platform automatically identifies the image data, marks the type of pollution source, geographical location, emission method, pollution control facilities, management department, and management responsible person information, generates an inspection report, and sends it to the relevant personnel for processing. After the relevant personnel deal with the pollution source, they start the drone on the platform to inspect the treatment effect of the pollution source, use the data transmitted by the drone to judge the treatment effect and the implementation of relevant prevention and control measures, and automatically generate a "look back" conclusion of closed-loop or unclosed-loop treatment.

[0031] An example of the contents of a drone inspection report is shown in Table 1.

[0032] Table 1 UAV inspection report

[0033]

[0034] When the particulate matter concentration data monitored by each station in the area are within the limit, the platform simulates the pollutant diffusion trajectory based on the existing list of particulate matter pollution sources in the area and combines meteorological environmental parameters, plans the drone's preventive inspection route, and starts the drone's regular inspection operations to achieve early warning and control of pollution.

[0035] The terminal intelligent control platform compares the inspection results of the same pollution source in different time periods, further incorporates the frequency of pollution source inspection problem discovery and closed-loop rectification into the dynamic management and control of pollution sources, automatically generates the pollution source treatment "closed-loop rate", and then optimizes the inspection route, realizing the integration, visualization and intelligence of environmental quality monitoring and pollution source inspection and prevention, reducing manual intervention and time costs, and improving the level of refined management and intelligent decision-making of atmospheric environmental protection.

[0036] During the statistical period, the closed-loop rate of pollution source treatment = the number of pollution sources found by "looking back" / the number of pollution sources found by inspection × 100%. The proportion of pollution times in a certain place = the number of pollution sources found by inspection in a certain place / the total number of pollution sources found by inspection in the region × 100%. Examples of statistical reports on inspection and treatment of pollution sources in a region are shown in Table 2.

[0037] Table 2 Pollution source inspection and treatment statistical report in the area

[0038]

[0039]

[0040] This solution is based on automatic image recognition technology to automatically identify pollution sources and control facilities in the inspection area. The drone inspection route is targeted, which improves the inspection efficiency and intelligence level, and reduces the manpower and material costs during the inspection process. The terminal intelligent control platform can automatically generate inspection reports and send them to relevant personnel, improving the response speed and control efficiency of pollution incidents.

Claims

1. A drone intelligent inspection method for particulate matter pollution prevention and control, characterized in that: The following steps are involved: S1: Obtain images of particulate pollution sources through drone aerial photography, the particulate pollution sources include construction site dust, road dust, bare soil, and open-air burning, perform denoising, enhancement, contrast adjustment, and brightness adjustment on the pollution source images, extract image features of construction site dust, road dust, bare soil, and open-air burning based on edge detection algorithms, establish image feature databases of construction site dust, road dust, bare soil, and open-air burning, and generate pollution source image recognition models based on machine learning algorithms; S2: Obtain images of particulate pollution control facilities through drone aerial photography, the particulate pollution control facilities include enclosure spraying, fog cannon, and dust nets, perform denoising, enhancement, contrast adjustment, and brightness adjustment on the images of particulate pollution control facilities, extract image features of enclosure spraying and fog cannon in shutdown and startup states, and image features of dust net coverage states based on edge detection algorithms, establish image feature databases of enclosure spraying and fog cannon in shutdown and startup states, and dust net coverage states, and generate an image recognition model for the implementation status of pollution control facilities based on machine learning algorithms; S3: Collect the data of particulate pollution sources and the control facilities of the pollution sources within the control area, send the collected data to the data analysis system through the data communication system, correspond to the information of the pollution source management department and the person in charge, generate a list of particulate pollution sources and pollution control facilities, draw a location distribution map of particulate pollution sources and particulate pollution control facilities, and intuitively display the distribution through a three-dimensional spatial map; S4: The terminal intelligent control platform uses the selected air quality model to calculate the particle pollution diffusion path based on the particle pollution source distribution map, terrain elevation, and meteorological environmental parameters in the control area, and collects the particle concentration data of each monitoring point in the area to calibrate and optimize the air quality model. It plans the drone inspection route based on the three-dimensional map and the optimized pollution diffusion path; S5: When the terminal intelligent control platform detects that the particulate matter concentration data at the monitoring site reaches the set warning value, it broadcasts a high value and issues an inspection command to the drone automatic hangar. After receiving the command, the drone automatic hangar automatically calls the drone inspection route from the terminal intelligent control platform and starts the drone inspection operation; S6: During the drone inspection flight, it monitors and records the particle concentration data, meteorological environment parameters and terrain elevation data in the passing direction in real time, automatically takes all-round and multi-angle high-definition photos of the points where the particle concentration exceeds the standard, and transmits the relevant data to the terminal intelligent control platform in real time; S7: The terminal intelligent control platform uses the pollution source image recognition model and the particle pollution control facility operation status image recognition model to identify the type of pollution source, the degree of pollution and the operation status of the pollution control facility based on the data transmitted by the drone, mark the problem points and form an inspection report; S8: The terminal intelligent control platform automatically sends the generated inspection report to relevant personnel for processing.

2. According to the method for intelligent inspection of particulate pollution by unmanned aerial vehicles for prevention and control of particulate matter pollution in claim 1, the image recognition model adjusts the network structure, optimizes the training parameters, enhances the image data, introduces the attention module, and extracts the edge features of the image based on the Yolov7 framework, thereby improving the speed and accuracy of image recognition. The image recognition training process specifically includes the following steps: S101: scaling and padding the collected images to adjust the size to 640×640; S102: performing standardization processing on the image, normalizing the pixel values ​​of the image to (0, 1); S103: denoising, enhancing, adjusting contrast, and adjusting brightness of the image, and randomly cropping, rotating, and scaling the image to increase the number of training images; S104: Use convolutional neural network to extract image features, combine convolutional layers and pooling layers, dynamically adjust the number of convolutional layers, and refine the extraction of shape, color, texture, and edge features of the image; S105: The image data is input into the Yolov7 model. The batch size of the model training starts from 32 and decreases by multiples of 4. The number of training rounds starts from 220 and decreases by multiples of 20. The learning rate is reduced from 0.01 to 0.0001. During the training process, the error between the image recognition result and the true label is dynamically evaluated, and an optimized image recognition model is gradually established.

3. According to the method for intelligent inspection by unmanned aerial vehicles for particulate matter pollution prevention and control according to claim 2, in step S3, the list information of particulate matter pollution sources and pollution control facilities includes the type of pollution source, geographical location, emission method, pollution control facilities, management department, and person in charge of management.

4. According to the UAV intelligent inspection method for particulate pollution prevention and control described in claim 3, the terminal intelligent control platform displays the flight trajectory of the UAV in real time.

5. According to the method for intelligent inspection by unmanned aerial vehicles for particulate pollution prevention and control according to claim 4, in step S6, the unmanned aerial vehicle performs all-round, multi-angle high-definition photography of points where particulate matter concentration exceeds the standard, transmits relevant data to the terminal intelligent control platform in real time, and then continues to identify and lock new targets to achieve continuous inspection of multiple targets.

6. The method for intelligent inspection by unmanned aerial vehicles for particulate pollution prevention and control according to claim 5, characterized in that: The meteorological environment parameters include temperature, humidity, wind direction and wind speed.

7. The method for intelligent inspection by unmanned aerial vehicles for particulate pollution prevention and control according to claim 6, characterized in that: The terminal intelligent control platform compares the inspection results of the same pollution source in different time periods, forms a "look back" report, automatically generates a "closed-loop rate" for pollution source treatment, and sends it to relevant personnel.

Images