Air pollution inspection method, device, equipment and storage medium

Abstract

This application is applicable to the field of artificial intelligence technology, and provides an air pollution inspection method, including: receiving air quality data, air quality collection point image data and collection positioning information fed back by a drone, wherein the drone collects data according to a pre-planned route and inspection point; matching a pre-built air quality analysis model according to the inspection point, and analyzing and processing the air quality data and the air quality collection point image data based on the air quality analysis model to obtain air quality status information; generating air quality distribution information according to the air quality status information, so as to monitor and alarm according to the air quality distribution information. This application has a fast inspection speed and a wide range, and the management department can obtain data in real time and respond quickly.

Description

Technical Field

[0001] The present application belongs to the field of artificial intelligence technology, and in particular, relates to an air pollution inspection method, device, equipment and storage medium. Background Art

[0002] With the development of social economy, people's working and living environment has gradually improved. However, at the same time, the production and living materials that provide people with convenience have also brought various air pollution, such as dust from construction sites, factories that do not meet exhaust emission standards, motor vehicles with serious substandard emissions on the road, and fumes from restaurants, etc. For these air pollution sources, government departments have also done a lot of prevention and control work and adopted many effective treatment measures. However, these measures generally have high investment costs and low efficiency; some of them have revived after treatment; some pollution sources are quite hidden and difficult to find in the first time, until the masses take photos to collect evidence and complain, and law enforcement officers take the initiative to investigate.

[0003] At present, existing air quality monitoring requires special detection equipment and personnel to perform manual operations. However, for areas with inconvenient transportation or areas that cannot be covered by fixed equipment, a lot of manpower and material resources are wasted to collect air data. In addition, there are often a lot of errors in the analysis of air quality based on human experience, which affects the accuracy of air quality monitoring. Summary of the invention

[0004] In view of the above, the present invention provides an air pollution inspection method, device, equipment and storage medium, which aims to solve the technical problem that there are often a lot of errors in the analysis of air quality based on human experience in the prior art.

[0005] To achieve the above object, the present invention provides an air pollution inspection method, the method comprising:

[0006] Receiving air quality data, air quality collection point image data and collection positioning information fed back by the drone, wherein the drone collects data according to the pre-planned route and inspection points;

[0007] Matching a pre-built air quality analysis model according to the inspection point, and analyzing and processing the air quality data and the image data of the air quality collection point based on the air quality analysis model to obtain air quality status information;

[0008] Air quality distribution information is generated according to the air quality status information, so as to perform monitoring and alarm according to the air quality distribution information.

[0009] Preferably, the receiving of air quality data, air quality collection point image data and collection positioning information fed back by the drone, wherein the drone collects data according to a pre-planned route and inspection point, includes:

[0010] The drone is equipped with a processor, a 5G module, a GPS module, an air quality detector and a camera, and the route and the inspection points are stored in the processor.

[0011] Preferably, the receiving of air quality data, air quality collection point image data and collection positioning information fed back by the drone, wherein the drone collects data according to a pre-planned route and inspection point, further includes:

[0012] The drone sends the positioning information and / or air quality data to the current server in real time, and the server outputs the air quality data so that the monitoring personnel can trigger the image data acquisition instruction;

[0013] After receiving the image data acquisition instruction, the processor in the drone starts the camera to take pictures, and sends the collected image data of the air quality acquisition point to the server.

[0014] Preferably, the air quality data, air quality collection point image data and collection positioning information fed back by the drone are received, wherein before the drone collects data according to the pre-planned route and inspection point, it includes:

[0015] Receive input information of the point to be monitored, obtain a positioning range that matches the monitoring point information, and plan the route in a full coverage manner according to the perimeter and area of ​​the positioning range.

[0016] Preferably, the air quality analysis model is a model constructed based on a combination of at least two machine learning models.

[0017] Preferably, the pre-built air quality analysis model is matched according to the inspection point, and the air quality data and the air quality collection point image data are analyzed and processed based on the air quality analysis model to obtain the air quality status information, including:

[0018] Obtaining historical air quality data and historical image data collected at different inspection points;

[0019] Training the BP neural network prediction model according to the historical air quality data, and completing the training when a preset prediction accuracy is reached;

[0020] The support vector machine model is trained by combining standard images of normal air quality and image data of normal performance in history, and the training is completed when a stable standard model is reached;

[0021] The weights corresponding to the operation results of the BP neural network prediction model and the support vector machine model are configured, and the corresponding relationship between the weights and the BP neural network prediction model and the support vector machine model is combined to obtain the air quality analysis model.

[0022] Preferably, the generating of air quality distribution information according to the air quality status information, and performing monitoring and alarming according to the air quality distribution information, comprises:

[0023] The route and the inspection point of the drone are adjusted according to the air quality status information.

[0024] To achieve the above object, the present invention further provides an air pollution inspection device, the device comprising:

[0025] A receiving module, used to receive air quality data, image data of air quality collection points and collection positioning information fed back by a drone, wherein the drone collects data according to a pre-planned route and inspection points;

[0026] A processing module, used for matching a pre-built air quality analysis model according to the inspection point, and analyzing and processing the air quality data and the image data of the air quality collection point based on the air quality analysis model to obtain air quality status information;

[0027] The monitoring module is used to generate air quality distribution information according to the air quality status information, so as to perform monitoring and alarm according to the air quality distribution information.

[0028] To achieve the above object, the present invention further provides an electronic device, the electronic device comprising:

[0029] at least one processor; and,

[0030] a memory communicatively connected to the at least one processor; wherein,

[0031] The memory stores a program executable by the at least one processor, and the program is executed by the at least one processor so that the at least one processor can perform the air pollution inspection method.

[0032] To achieve the above objectives, the present invention also provides a computer-readable storage medium, which stores an air pollution inspection program. When the air pollution inspection program is executed by a processor, the steps of the air pollution inspection method are implemented.

[0033] The air pollution inspection method, device, equipment and storage medium proposed by the present invention have few devices to maintain, only unmanned aerial vehicles, fast inspection speed and wide range, and management departments can obtain data in real time and respond quickly. The mobile air monitoring inspection method can effectively cover and inspect more areas that cannot be covered by fixed detection equipment and places that are difficult for people to reach due to inconvenient transportation; when inspecting construction sites and catering stores, more pollution sources can be inspected with fewer people, reducing the fatigue of manual inspection personnel; management departments can quickly obtain results online and respond quickly, improving management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 A schematic diagram of a preferred embodiment of the electronic device of the present invention;

[0036] Figure 2 for Figure 1 A schematic diagram of a module of a preferred embodiment of a patrol device for atmospheric pollution;

[0037] Figure 3 A flow chart of a preferred embodiment of the air pollution inspection method of the present invention;

[0038] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0040] Reference Figure 1 FIG. 1 is a schematic diagram of a preferred embodiment of an electronic device 1 of the present invention.

[0041] The electronic device 1 includes but is not limited to: a memory 11, a processor 12, a display 13 and a network interface 14. The electronic device 1 is connected to a network through the network interface 14 to obtain raw data. The network may be a wireless or wired network such as an intranet, the Internet, a global system of mobile communication (GSM), a wideband code division multiple access (WCDMA), a 4G network, a 5G network, Bluetooth, a Wi-Fi calling network, etc.

[0042] Among them, the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a hard disk or memory of the electronic device 1. In other embodiments, the memory 11 can also be an external storage device of the electronic device 1, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped with the electronic device 1. Of course, the memory 11 can also include both the internal storage unit of the electronic device 1 and its external storage device. In this embodiment, the memory 11 is generally used to store the operating system and various application software installed on the electronic device 1, such as the program code of the air pollution inspection program 10, etc. In addition, the memory 11 can also be used to temporarily store various types of data that have been output or are to be output.

[0043] In some embodiments, the processor 12 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 12 is generally used to control the overall operation of the electronic device 1, such as performing control and processing related to data interaction or communication. In this embodiment, the processor 12 is used to run the program code stored in the memory 11 or process data, such as running the program code of the air pollution inspection program 10.

[0044] The display 13 may be referred to as a display screen or a display unit. In some embodiments, the display 13 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an organic light-emitting diode (OLED) touch device. The display 13 is used to display information processed in the electronic device 1 and to display a visual working interface, such as displaying the results of data statistics.

[0045] The network interface 14 may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface), and the network interface 14 is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0046] Figure 1 Only the electronic device 1 having components 11 - 14 and the air pollution inspection program 10 is shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0047] Optionally, the electronic device 1 may further include a target user interface, which may include a display, an input unit such as a keyboard, and an optional target user interface may further include a standard wired interface and a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an organic light-emitting diode (OLED) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visualized target user interface.

[0048] The electronic device 1 may further include a radio frequency (RF) circuit, a sensor, an audio circuit, etc., which will not be described in detail here.

[0049] In the above embodiment, the processor 12 can implement the following steps when executing the air pollution inspection program 10 stored in the memory 11:

[0050] Receiving air quality data, air quality collection point image data and collection positioning information fed back by the drone, wherein the drone collects data according to the pre-planned route and inspection points;

[0051] Matching a pre-built air quality analysis model according to the inspection point, and analyzing and processing the air quality data and the image data of the air quality collection point based on the air quality analysis model to obtain air quality status information;

[0052] Air quality distribution information is generated according to the air quality status information, so as to perform monitoring and alarm according to the air quality distribution information.

[0053] For a detailed description of the above steps, please refer to the following Figure 2 Functional module diagram of an embodiment of an air pollution inspection device 100 and Figure 3 Description of a flowchart of an embodiment of a method for inspecting air pollution.

[0054] Reference Figure 2 , which is a functional module diagram of the air pollution inspection device 100 of the present invention.

[0055] The air pollution inspection device 100 of the present invention can be installed in an electronic device. According to the functions to be implemented, the air pollution inspection device 100 can include a receiving module 110, a processing module 120 and a monitoring module 130. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0056] In this embodiment, the functions of each module / unit are as follows:

[0057] The receiving module 110 is used to receive the air quality data, the image data of the air quality collection points and the collection positioning information fed back by the drone, wherein the drone collects data according to the pre-planned route and inspection points.

[0058] In this embodiment, the drone is equipped with a processor, a 5G module, a GPS module, an air quality detector and a camera. The processor stores the routes and inspection points planned in the server. The inspection points in this application are inspection location points pre-configured according to air quality requirements. When the drone flies along the route, it is positioned in real time based on the GPS module to determine whether it is at the inspection point based on the positioning information.

[0059] It should be noted that in this application, the inspection points are photographed to obtain air quality collection image data, and the air quality detector is started to obtain air quality data. When the route is pre-planned, the positions that need to be monitored in the route can be marked as inspection points in the route. When the positioning information collected based on the GPS module matches this marked position, the processor on the drone starts the air quality detector to collect air quality data, and starts the camera to take pictures to obtain air quality collection point image data. Among them, the air quality collection image data can be multiple pictures taken continuously based on the inspection points, so that when performing analysis and processing, they can be numerically processed to obtain a multi-dimensional vector matrix for calculation.

[0060] In other embodiments, the receiving of air quality data, air quality collection point image data and collection positioning information fed back by a drone, wherein the drone collects data according to a pre-planned route and inspection point, further includes:

[0061] The drone sends the positioning information and / or air quality data to the current server in real time, and the server outputs the air quality data so that the monitoring personnel can trigger the image data acquisition instruction;

[0062] After receiving the image data acquisition instruction, the processor in the drone starts the camera to take pictures, and sends the collected image data of the air quality acquisition point to the server.

[0063] In this embodiment, the monitoring personnel can output the positioning information and / or air quality data received by the server based on the control platform or terminal device. The output can be mapped according to the positioning route trajectory, or displayed according to the curve chart of the air quality data, so that the monitoring personnel can monitor in real time. When the monitoring personnel triggers the image data acquisition instruction through the control platform or terminal device, the server forwards the instruction to the drone processor to enable the drone to start the camera to take pictures, thereby improving the accuracy of obtaining the image data of the air quality collection point.

[0064] In other embodiments, the air quality data, air quality collection point image data and collection positioning information fed back by the drone are received, wherein before the drone collects data according to the pre-planned route and inspection point, the process includes:

[0065] Receive input information of the point to be monitored, obtain a positioning range that matches the monitoring point information, and plan the route in a full coverage manner according to the perimeter and area of ​​the positioning range.

[0066] For example, when the user inputs the name of a chemical plant, the positioning range composed of the positioning coordinates (x1, y1), coordinates (x2, y2), coordinates (x3, y3), and coordinates (x4, y4) of the chemical plant is obtained, and the perimeter of the figure composed of this positioning range is used as the maximum navigation range of the drone route. Then, according to the cross-section of the image taken by the drone camera and the area of ​​the line composed of the positioning range, the route is planned in a full coverage manner, that is, according to the drone route, the shooting of all real scenes in the area is completed. Specifically, for the method of planning the route, the direction parallel to the maximum straight line length in the positioning range is used as the flight direction of the drone's planned route, so that the drone completes the monitoring of the entire inspection point based on the route. If the positioning range formed by the positioning is a point, the planned route only needs to pass through this positioning point.

[0067] The processing module 120 is used to match the pre-built air quality analysis model according to the inspection point, and analyze and process the air quality data and the air quality collection point image data based on the air quality analysis model to obtain air quality status information.

[0068] In this embodiment, since different inspection points are different collection objects of air quality data and image data, combined with the variable characteristics of the air environment, the matching air quality analysis model is pre-trained according to different inspection points. Therefore, after the inspection point is determined, the corresponding air quality analysis model is matched. Specifically, since it is necessary to analyze and process according to the air quality data and the image data of the air quality collection point, one is the detection data and the other is the image data, the air quality analysis model is constructed based on the combination of at least two machine learning models. Correspondingly, during the analysis and processing, the air quality data and the air quality collection image data are respectively used as input data for calculation and processing based on the matching air quality analysis model, and the air quality status information is obtained according to the calculation result obtained by the combination. Among them, the air quality status information is used to characterize the quality of air quality, for example, including excellent, good, poor quality, etc. In addition, the analysis model corresponding to the air quality data in the air quality analysis model is a machine learning model with predictive regression; the recognition method corresponding to the image data adopts the inductive statistical method, and a batch of image information with normal air quality is trained into a stable model and converted into a standard value group, and then the actual image is compared with the standard model.

[0069] In other embodiments, the air quality analysis model pre-built according to the inspection point is matched, and the air quality data and the air quality collection point image data are analyzed and processed based on the air quality analysis model to obtain the air quality status information, including:

[0070] Obtaining historical air quality data and historical image data collected at different inspection points;

[0071] Training the BP neural network prediction model according to the historical air quality data, and completing the training when a preset prediction accuracy is reached;

[0072] The support vector machine model is trained by combining standard images of normal air quality and image data of normal performance in history, and the training is completed when a stable standard model is reached;

[0073] The weights corresponding to the operation results of the BP neural network prediction model and the support vector machine model are configured, and the corresponding relationship between the weights and the BP neural network prediction model and the support vector machine model is combined to obtain the air quality analysis model.

[0074] In the process of training the BP neural network prediction model based on historical air quality data, the image information of the standard air quality and the historical normal images are used as model input for training, and the obtained stable model is converted into a standard value group as the value of the first air quality. Both are added to increase its standardization. Then, the historical images of medium and poor air quality are trained and converted into the second and third air quality value groups. The training process of the support vector machine model is the same as the prior art.

[0075] The training process of the BP neural network prediction model includes: 1. Initializing the connection weights and thresholds and providing the learning mode pair to the network structure; 2. Calculating the input and output of each unit in the middle layer and the output layer, as well as calculating the generalized error of each unit in the output layer and the generalized error of each unit in the middle layer; adjusting the connection weights between the middle layer and the input layer, as well as the thresholds of each unit in the middle layer and the output layer; 3. Updating the learning input mode; 4. After all modes are trained, update the number of learning times. If the training is not completed, continue training; 5. When the error is less than the specified error or the number of learning times is greater than the specified number of learning times, the model training is completed.

[0076] After the training is completed, the first air quality state and the second air quality state are combined according to the weights of the configured different models to obtain air quality state information, for example, a1*first air quality state+a2*second air quality state=air quality state information.

[0077] It should be noted that the first air quality state and the second air quality state are both digitized data during the model processing process, and are matched with different state information until the air quality state information is obtained.

[0078] Specifically, before model training, the captured image data is grouped into vector matrices according to different air quality levels, and the color, texture, fuzzy features and other attributes in the image are digitized to obtain value groups of various levels, so as to be transported as model input during the training process. During the inspection, the image information obtained from the actual inspection is also digitized, and the difference between the obtained image value group and the various standard image value groups is calculated. The closer the value group is, the higher the air quality belongs to, and the judgment result is output.

[0079] The monitoring module 130 is used to generate air quality distribution information according to the air quality status information, so as to perform monitoring and alarm according to the air quality distribution information.

[0080] In this embodiment, since different air quality data corresponds to different image data, and the air quality status information obtained by analysis is different, air quality distribution information can be generated based on the air quality status information. Specifically, matching distribution colors, lines, logos, etc. are configured in advance according to different air quality status information, and the distribution colors, lines or logos corresponding to different air quality status information are rendered based on the route map to generate air quality distribution information. Alarm information is generated according to different air quality distribution information so that monitoring personnel can conduct monitoring and processing.

[0081] In one embodiment, the generating of air quality distribution information according to the air quality status information, and performing monitoring and alarming according to the air quality distribution information, comprises:

[0082] The route and the inspection point of the drone are adjusted according to the air quality status information.

[0083] In this embodiment, the air quality status information is analyzed to see whether it is poor, relatively poor, severe, etc. If so, the inspection points and routes that match the air quality status information are extracted. Since the flight time and flight speed of the drone in the first and last planned inspection points and routes are configured, in order to improve the monitoring of the serious quality range, the routes and inspection points planned by the drone are adjusted, that is, the round-trip status of the route, the flight speed, the number of images taken at the inspection points, etc., so as to adjust the new drone flight routes and inspection points, so that when monitoring is carried out again, the areas with poor air quality status can be monitored with emphasis.

[0084] The air pollution inspection device proposed by the present invention has few equipment to maintain, only drones, fast inspection speed and wide range, and management departments can obtain data in real time and respond quickly. The mobile air monitoring inspection method can effectively cover more areas that cannot be covered by fixed detection equipment and places that are difficult for people to reach due to inconvenient transportation; when inspecting construction sites and catering stores, more pollution sources can be inspected with fewer people, reducing the fatigue of manual inspectors; management departments can quickly obtain results online and respond quickly, improving management efficiency.

[0085] In addition, the present invention also provides a method for inspecting air pollution. Figure 3 , which is a schematic diagram of a method flow of an embodiment of the air pollution inspection method of the present invention. When the processor 12 of the electronic device 1 executes the air pollution inspection program 10 stored in the memory 11, the air pollution inspection method is implemented, including steps S101-S103. Each step is described in detail below.

[0086] Step S101: receiving air quality data, air quality collection point image data and collection positioning information fed back by a drone, wherein the drone collects data according to a pre-planned route and inspection points.

[0087] In this embodiment, the drone is equipped with a processor, a 5G module, a GPS module, an air quality detector and a camera. The processor stores the routes and inspection points planned in the server. The inspection points in this application are inspection location points pre-configured according to air quality requirements. When the drone flies along the route, it is positioned in real time based on the GPS module to determine whether it is at the inspection point based on the positioning information.

[0088] It should be noted that in this application, the inspection points are photographed to obtain air quality collection image data, and the air quality detector is started to obtain air quality data. When the route is pre-planned, the positions that need to be monitored in the route can be marked as inspection points in the route. When the positioning information collected based on the GPS module matches this marked position, the processor on the drone starts the air quality detector to collect air quality data, and starts the camera to take pictures to obtain air quality collection point image data. Among them, the air quality collection image data can be multiple pictures taken continuously based on the inspection points, so that when performing analysis and processing, they can be numerically processed to obtain a multi-dimensional vector matrix for calculation.

[0089] In other embodiments, the receiving of air quality data, air quality collection point image data and collection positioning information fed back by a drone, wherein the drone collects data according to a pre-planned route and inspection point, further includes:

[0090] The drone sends the positioning information and / or air quality data to the current server in real time, and the server outputs the air quality data so that the monitoring personnel can trigger the image data acquisition instruction;

[0091] After receiving the image data acquisition instruction, the processor in the drone starts the camera to take pictures, and sends the collected image data of the air quality acquisition point to the server.

[0092] In this embodiment, the monitoring personnel can output the positioning information and / or air quality data received by the server based on the control platform or terminal device. The output can be mapped according to the positioning route trajectory, or displayed according to the curve chart of the air quality data, so that the monitoring personnel can monitor in real time. When the monitoring personnel triggers the image data acquisition instruction through the control platform or terminal device, the server forwards the instruction to the drone processor to enable the drone to start the camera to take pictures, thereby improving the accuracy of obtaining the image data of the air quality collection point.

[0093] In other embodiments, the air quality data, air quality collection point image data and collection positioning information fed back by the drone are received, wherein before the drone collects data according to the pre-planned route and inspection point, the process includes:

[0094] Receive input information of the point to be monitored, obtain a positioning range that matches the monitoring point information, and plan the route in a full coverage manner according to the perimeter and area of ​​the positioning range.

[0095] For example, when the user inputs the name of a chemical plant, the positioning range composed of the positioning coordinates (x1, y1), coordinates (x2, y2), coordinates (x3, y3), and coordinates (x4, y4) of the chemical plant is obtained, and the perimeter of the figure composed of this positioning range is used as the maximum navigation range of the drone route. Then, according to the cross-section of the image taken by the drone camera and the area of ​​the line composed of the positioning range, the route is planned in a full coverage manner, that is, according to the drone route, the shooting of all real scenes in the area is completed. Specifically, for the method of planning the route, the direction parallel to the maximum straight line length in the positioning range is used as the flight direction of the drone's planned route, so that the drone completes the monitoring of the entire inspection point based on the route. If the positioning range formed by the positioning is a point, the planned route only needs to pass through this positioning point.

[0096] Step S102: matching a pre-built air quality analysis model according to the inspection point, and analyzing and processing the air quality data and the air quality collection point image data based on the air quality analysis model to obtain air quality status information.

[0097] In this embodiment, since different inspection points are different collection objects of air quality data and image data, combined with the variable characteristics of the air environment, the matching air quality analysis model is pre-trained according to different inspection points. Therefore, after the inspection point is determined, the corresponding air quality analysis model is matched. Specifically, since it is necessary to analyze and process according to the air quality data and the image data of the air quality collection point, one is the detection data and the other is the image data, the air quality analysis model is constructed based on the combination of at least two machine learning models. Correspondingly, during the analysis and processing, the air quality data and the air quality collection image data are respectively used as input data for calculation and processing based on the matching air quality analysis model, and the air quality status information is obtained according to the calculation result obtained by the combination. Among them, the air quality status information is used to characterize the quality of air quality, for example, including excellent, good, poor quality, etc. In addition, the analysis model corresponding to the air quality data in the air quality analysis model is a machine learning model with predictive regression; the recognition method corresponding to the image data adopts the inductive statistical method, and a batch of image information with normal air quality is trained into a stable model and converted into a standard value group, and then the actual image is compared with the standard model.

[0098] In other embodiments, the air quality analysis model pre-built according to the inspection point is matched, and the air quality data and the air quality collection point image data are analyzed and processed based on the air quality analysis model to obtain the air quality status information, including:

[0099] Obtaining historical air quality data and historical image data collected at different inspection points;

[0100] Training the BP neural network prediction model according to the historical air quality data, and completing the training when a preset prediction accuracy is reached;

[0101] The support vector machine model is trained by combining standard images of normal air quality and image data of normal performance in history, and the training is completed when a stable standard model is reached;

[0102] The weights corresponding to the operation results of the BP neural network prediction model and the support vector machine model are configured, and the corresponding relationship between the weights and the BP neural network prediction model and the support vector machine model is combined to obtain the air quality analysis model.

[0103] In the process of training the BP neural network prediction model based on historical air quality data, the image information of the standard air quality and the historical normal images are used as model input for training, and the obtained stable model is converted into a standard value group as the value of the first air quality. Both are added to increase its standardization. Then, the historical images of medium and poor air quality are trained and converted into the second and third air quality value groups. The training process of the support vector machine model is the same as the prior art.

[0104] The training process of the BP neural network prediction model includes: 1. Initializing the connection weights and thresholds and providing the learning mode pair to the network structure; 2. Calculating the input and output of each unit in the middle layer and the output layer, as well as calculating the generalized error of each unit in the output layer and the generalized error of each unit in the middle layer; adjusting the connection weights between the middle layer and the input layer, as well as the thresholds of each unit in the middle layer and the output layer; 3. Updating the learning input mode; 4. After all modes are trained, update the number of learning times. If the training is not completed, continue training; 5. When the error is less than the specified error or the number of learning times is greater than the specified number of learning times, the model training is completed.

[0105] After the training is completed, the first air quality state and the second air quality state are combined according to the weights of the configured different models to obtain air quality state information, for example, a1*first air quality state+a2*second air quality state=air quality state information.

[0106] It should be noted that the first air quality state and the second air quality state are both digitized data during the model processing process, and are matched with different state information until the air quality state information is obtained.

[0107] Specifically, before model training, the captured image data is grouped into vector matrices according to different air quality levels, and the color, texture, fuzzy features and other attributes in the image are digitized to obtain value groups of various levels, so as to be transported as model input during the training process. During the inspection, the image information obtained from the actual inspection is also digitized, and the difference between the obtained image value group and the various standard image value groups is calculated. The closer the value group is, the higher the air quality belongs to, and the judgment result is output.

[0108] Step S103: Generate air quality distribution information according to the air quality status information, so as to perform monitoring and alarm according to the air quality distribution information.

[0109] In this embodiment, since different air quality data corresponds to different image data, and the air quality status information obtained by analysis is different, air quality distribution information can be generated based on the air quality status information. Specifically, matching distribution colors, lines, logos, etc. are configured in advance according to different air quality status information, and the distribution colors, lines or logos corresponding to different air quality status information are rendered based on the route map to generate air quality distribution information. Alarm information is generated according to different air quality distribution information so that monitoring personnel can conduct monitoring and processing.

[0110] In one embodiment, the generating of air quality distribution information according to the air quality status information, and performing monitoring and alarming according to the air quality distribution information, comprises:

[0111] The route and the inspection point of the drone are adjusted according to the air quality status information.

[0112] In this embodiment, the air quality status information is analyzed to see whether it is poor, relatively poor, severe, etc. If so, the inspection points and routes that match the air quality status information are extracted. Since the flight time and flight speed of the drone in the first and last planned inspection points and routes are configured, in order to improve the monitoring of the serious quality range, the routes and inspection points planned by the drone are adjusted, that is, the round-trip status of the route, the flight speed, the number of images taken at the inspection points, etc., so as to adjust the new drone flight routes and inspection points, so that when monitoring is carried out again, the areas with poor air quality status can be monitored with emphasis.

[0113] The air pollution inspection method proposed by the present invention requires less equipment to be maintained, only drones, with fast inspection speed and wide range, and the management department can obtain data in real time and respond quickly. The mobile air monitoring inspection method can effectively cover and inspect more areas that cannot be covered by fixed detection equipment and places that are difficult for people to reach due to inconvenient transportation; when inspecting construction sites and catering stores, more pollution sources can be inspected with fewer people, reducing the fatigue of manual inspectors; the management department can quickly obtain results online and respond quickly, improving management efficiency.

[0114] In addition, the embodiment of the present invention further proposes a computer-readable storage medium, which may be any one or any combination of a hard disk, a multimedia card, an SD card, a flash memory card, an SMC, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disk read-only memory (CD-ROM), a USB memory, etc. The computer-readable storage medium includes a data storage area and a program storage area, the data storage area stores data created according to the use of the blockchain node, and the program storage area stores an air pollution inspection program 10. When the air pollution inspection program 10 is executed by the processor, it is applied to the second terminal to implement the following operations:

[0115] Receiving air quality data, air quality collection point image data and collection positioning information fed back by the drone, wherein the drone collects data according to the pre-planned route and inspection points;

[0116] Matching a pre-built air quality analysis model according to the inspection point, and analyzing and processing the air quality data and the image data of the air quality collection point based on the air quality analysis model to obtain air quality status information;

[0117] Air quality distribution information is generated according to the air quality status information, so as to perform monitoring and alarm according to the air quality distribution information.

[0118] The air pollution inspection method, device, equipment and storage medium proposed by the present invention have few devices to maintain, only unmanned aerial vehicles, fast inspection speed and wide range, and management departments can obtain data in real time and respond quickly. The mobile air monitoring inspection method can effectively cover and inspect more areas that cannot be covered by fixed detection equipment and places that are difficult for people to reach due to inconvenient transportation; when inspecting construction sites and catering stores, more pollution sources can be inspected with fewer people, reducing the fatigue of manual inspection personnel; management departments can quickly obtain results online and respond quickly, improving management efficiency.

[0119] In another embodiment, the air pollution inspection method provided by the present invention can further ensure the privacy and security of all the above data, and all the above data can also be stored in a blockchain node. For example, QR codes, identification codes, etc., all of these data can be stored in the blockchain node.

[0120] It should be noted that the blockchain referred to in the present invention is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the blockchain underlying platform, platform product service layer, and application service layer.

[0121] The specific implementation of the computer-readable storage medium of the present invention is substantially the same as the specific implementation of the above-mentioned air pollution inspection method, and will not be described in detail here.

[0122] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof herein are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the existence of other identical elements in the process, device, article or method including the element.

[0123] It should be noted that the above embodiments of the present invention can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0124] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0125] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for enabling an electronic device (which can be a mobile phone, a computer, an electronic device, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0126] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for inspecting air pollution, characterized in that: The method comprises: Receiving air quality data, air quality collection point image data and collection positioning information fed back by the drone, wherein the drone collects data according to the pre-planned route and inspection points; According to the inspection point matching, a pre-constructed air quality analysis model is used, and based on the air quality analysis model, the air quality data and the image data of the air quality collection point are analyzed and processed to obtain air quality status information; wherein the pre-constructed air quality analysis model is obtained by training and matching different inspection points, and is constructed based on a combination of at least two machine learning models; Generate air quality distribution information according to the air quality status information, so as to perform monitoring and alarm according to the air quality distribution information; The receiving of air quality data, air quality collection point image data and collection positioning information fed back by the drone, wherein the drone collects data according to the pre-planned route and inspection point, including: Receive and input information of the point to be monitored, obtain the positioning range that matches the monitoring point information, and plan the route in a full coverage manner according to the perimeter and area of ​​the positioning range; wherein the full coverage manner is: use the perimeter of the figure composed of the positioning range as the maximum navigation range of the UAV route, and according to the cross-section of the image taken by the UAV camera combined with the area of ​​the line composed of the positioning range, complete the shooting of all real scenes within the area according to the UAV route to plan the route.

2. The air pollution inspection method according to claim 1, characterized in that: The receiving of air quality data, air quality collection point image data and collection positioning information fed back by the drone, wherein the drone collects data according to the pre-planned route and inspection points, includes: The drone is equipped with a processor, a 5G module, a GPS module, an air quality detector and a camera, and the route and the inspection points are stored in the processor.

3. The air pollution inspection method according to claim 1, characterized in that: The receiving of air quality data, air quality collection point image data and collection positioning information fed back by the drone, wherein the drone collects data according to a pre-planned route and inspection point, further includes: The drone sends the positioning information and / or air quality data to the current server in real time, and the server outputs the air quality data so that the monitoring personnel can trigger the image data acquisition instruction; After receiving the image data acquisition instruction, the processor in the drone starts the camera to take pictures, and sends the collected image data of the air quality acquisition point to the server.

4. The air pollution inspection method according to claim 1, characterized in that: The air quality analysis model is a model constructed based on a combination of at least two machine learning models.

5. The air pollution inspection method according to claim 4, characterized in that: The method of matching the pre-built air quality analysis model according to the inspection point and analyzing and processing the air quality data and the air quality collection point image data based on the air quality analysis model to obtain the air quality status information includes: Obtaining historical air quality data and historical image data collected at different inspection points; Training the BP neural network prediction model according to the historical air quality data, and completing the training when a preset prediction accuracy is reached; The support vector machine model is trained by combining standard images of normal air quality and image data of normal performance in history, and the training is completed when a stable standard model is reached; The weights corresponding to the operation results of the BP neural network prediction model and the support vector machine model are configured, and the corresponding relationship between the weights and the BP neural network prediction model and the support vector machine model is combined to obtain the air quality analysis model.

6. The air pollution inspection method according to claim 1, characterized in that: After generating air quality distribution information according to the air quality status information and performing monitoring and alarming according to the air quality distribution information, the method further comprises: The route and the inspection point of the drone are adjusted according to the air quality status information.

7. An air pollution inspection device, characterized in that: The device comprises: A receiving module, used to receive air quality data, image data of air quality collection points and collection positioning information fed back by a drone, wherein the drone collects data according to a pre-planned route and inspection points, and is constructed based on a combination of at least two machine learning models; A processing module, used for matching a pre-built air quality analysis model according to the inspection point, and analyzing and processing the air quality data and the image data of the air quality collection point based on the air quality analysis model to obtain air quality status information; wherein the pre-built air quality analysis model is obtained by training and matching different inspection points; A monitoring module, used to generate air quality distribution information according to the air quality status information, so as to perform monitoring and alarm according to the air quality distribution information; Among them, the air quality data, air quality collection point image data and collection positioning information fed back by the drone are received, and before the drone collects data according to the pre-planned route and inspection points, it includes: receiving and entering the information of the point to be monitored, obtaining the positioning range that matches the monitoring point information, and planning the route in a full coverage manner according to the perimeter and area of ​​the positioning range; wherein the full coverage method is: taking the perimeter of the figure composed of the positioning range as the maximum navigation range of the drone route, according to the cross-section of the image taken by the drone camera combined with the area of ​​the line composed of the positioning range, completing the shooting of all real scenes within the area according to the drone route to plan the route.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores a program executable by the at least one processor, and the program is executed by the at least one processor so that the at least one processor can execute the air pollution inspection method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an air pollution inspection program, and when the air pollution inspection program is executed by a processor, the steps of the air pollution inspection method according to any one of claims 1 to 6 are implemented.

Images