Atmosphere prediction analysis system based on pollution source image processing

Abstract

The invention provides an atmosphere prediction analysis system based on pollution source image processing, and the system comprises an image obtaining module which is used for obtaining a pollution source image collected by an image collection device; the image pre-processing module is used for performing image pre-processing on the pollution source image to obtain a pre-processed image; the image information acquisition module is used for carrying out image information matching on the preprocessed image to obtain an image information set corresponding to the preprocessed image; the meteorological data acquisition module is used for acquiring a plurality of pieces of meteorological data and forming a meteorological data set according to the plurality of pieces of meteorological data; and the atmosphere prediction analysis module is used for performing atmosphere prediction analysis according to the image information set and the meteorological data set to obtain an atmosphere prediction analysis result. The technical effects that the prediction accuracy of atmospheric pollution is improved, an accurate atmospheric pollution analysis result is obtained, and then the prevention and treatment effect of atmospheric pollution is improved are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of air quality prediction and analysis, and in particular to an atmosphere prediction and analysis system based on pollution source image processing. Background Art

[0002] Air pollution is a phenomenon in which certain substances enter the atmosphere due to human activities or natural processes, present sufficient concentrations, reach sufficient duration, and thus endanger human comfort, health and welfare or the environment. Therefore, how to prevent and control air pollution is one of the environmental problems that my country urgently needs to solve. However, since air pollution is easily affected by weather and will transfer between different regions, it is very difficult to prevent and control air pollution. At present, air pollution is mainly predicted through meteorological data, and air pollution prevention and control is carried out according to the prediction results. However, this method cannot accurately reflect the scope of air pollution, so it cannot be accurately prevented and controlled.

[0003] Prior art, application publication number: CN114578457A discloses a method for spatiotemporal prediction of atmospheric pollutant concentration based on evolutionary ensemble learning, including: step 1: collect atmospheric pollutant concentration data and perform standardization processing; step 2: collect environmental variable data covering the study area, screen and construct the optimal set of environmental variables; step 3: select spatial prediction algorithms, train and evaluate each sub-model in the initial stage; step 4: construct an initial stage ensemble learning model based on the prediction accuracy of each sub-model; step 5: construct an evolutionary ensemble learning model for atmospheric pollutant concentrations at other time points; step 6: based on Monte Carlo simulation technology, perform uncertainty analysis of prediction results, and use the trained evolutionary ensemble learning model to dynamically output the spatiotemporal distribution map of atmospheric pollutant concentrations. Although this technical solution predicts the concentration of atmospheric pollutants and accurately reflects the impact range of atmospheric pollution, it still needs to rely on existing atmospheric pollutant concentration data when making predictions, rather than directly analyzing the atmospheric pollution sources to obtain results, which may result in inaccurate analysis results.

[0004] Therefore, in order to obtain accurate prediction and analysis results, it is necessary to conduct relevant analysis of pollution sources. Any substance that can make the air quality worse is an atmospheric pollution source, such as carbon monoxide emitted by forest fires and fluorides emitted by industrial production processes. However, there is currently no effective analysis method for pollution sources.

[0005] Therefore, how to improve the prediction accuracy of air pollution by conducting relevant analysis on air pollution sources, obtain accurate air pollution analysis results, and then improve the prevention and control effect of air pollution is one of the problems that urgently need to be solved in the field of air quality prediction and analysis technology. Summary of the invention

[0006] The present invention aims to solve the technical problems in the above-mentioned technologies to at least a certain extent. To this end, the purpose of the present invention is to provide an atmospheric prediction and analysis system based on pollution source image processing, which can improve the prediction accuracy of atmospheric pollution, obtain accurate atmospheric pollution analysis results, and improve the technical effect of atmospheric pollution prevention and control by processing pollution source images and then predicting and analyzing the pollution concentration changes and pollution areas of pollution sources, thereby achieving the technical effect of improving atmospheric pollution prediction accuracy, obtaining accurate atmospheric pollution analysis results, and then improving the prevention and control effect of atmospheric pollution.

[0007] The present invention provides an atmospheric prediction and analysis system based on pollution source image processing, comprising:

[0008] An image acquisition module is used to acquire the pollution source image acquired by the image acquisition device;

[0009] An image preprocessing module is used to perform image preprocessing on the pollution source image to obtain a preprocessed image;

[0010] An image information acquisition module is used to perform image information matching on the preprocessed image to obtain an image information set corresponding to the preprocessed image;

[0011] A meteorological data acquisition module is used to acquire a number of meteorological data and form a meteorological data set according to the several meteorological data;

[0012] The atmospheric prediction and analysis module is used to perform atmospheric prediction and analysis based on the image information set and the meteorological data set to obtain the atmospheric prediction and analysis results.

[0013] Preferably, the atmospheric prediction and analysis system based on pollution source image processing, the image preprocessing module comprises:

[0014] An image sharpening submodule, used for sharpening the pollution source image to obtain a first processed image;

[0015] An image denoising submodule, used for performing denoising on the first processed image to obtain a second processed image;

[0016] The image contour extraction submodule is used to perform a contour extraction operation on the second processed image to obtain a preprocessed image.

[0017] Preferably, the atmospheric prediction and analysis system based on pollution source image processing, the image denoising submodule comprises:

[0018] A gray value acquisition submodule, used to acquire gray values ​​of all pixels of the first processed image;

[0019] A first region determination submodule is used to select any pixel point in the first processed image as a central pixel point, and determine an N*N region with the central pixel point as the region center as the first region;

[0020] The average gray value calculation submodule is used to calculate the average gray value of all pixels in the first area;

[0021] A noise pixel determination submodule is used to compare the grayscale values ​​of all pixels in the first area with the average grayscale value, and when the ratio of the grayscale value of the pixel to the average grayscale value is less than a first predetermined threshold or greater than a second predetermined threshold, the pixel is determined to be a noise pixel, and the above operation is performed on all pixels in the first processed image to determine all noise pixels in the first processed image; wherein the first predetermined threshold is less than the second predetermined threshold;

[0022] The second region determination submodule is used to select any noise pixel point as a pixel point to be processed, and determine an M*M region with the pixel point to be processed as the region center as the second region;

[0023] A similar pixel point determination submodule, used for setting the pixel point in the second area with the smallest distance from the pixel point to be processed as a similar pixel point to the pixel point to be processed;

[0024] A third region determination submodule, used to determine a region consisting of a plurality of pixels in the second region having the same channel type as the adjacent pixels as the third region;

[0025] A weight determination submodule, used to determine a filtering weight of the third area according to the pixel points in the second area and the pixel points in the third area;

[0026] The pixel value correction submodule is used to multiply the filter weight by the pixel value of the pixel to be processed, and use the product as the corrected pixel value of the pixel to be processed. The above operation is performed on all noise pixels to obtain a corrected image, and the corrected image is used as the second processed image.

[0027] Preferably, the atmospheric prediction and analysis system based on pollution source image processing, the weight determination submodule includes:

[0028] A first difference value determination submodule, configured to determine a color difference value between the second area and the third area as a first difference value according to the pixel value vector of the second area and the pixel value vector of the third area;

[0029] A second difference value determination submodule, used to determine a distance difference value between the second area and the third area as a second difference value according to the coordinates of the pixel point to be processed in the second area and the coordinates of the pixel point in the third area farthest from the current pixel point to be processed;

[0030] A first weight calculation submodule, used to determine the color weight of the pixels in the third area according to the first difference value and the Gaussian function curve of the control color as the first weight;

[0031] A second weight calculation submodule, used to determine the distance weight of the pixels in the third area according to the second difference value and the Gaussian function curve of the control distance as the second weight;

[0032] The third weight calculation submodule multiplies the first weight and the second weight to obtain a third weight, and uses the third weight as a filtering weight for the third region.

[0033] Preferably, the atmospheric prediction and analysis system based on pollution source image processing, the image information acquisition module includes:

[0034] An image vector generation submodule is used to generate an image vector corresponding to the preprocessed image;

[0035] The image information determination submodule is used to calculate the distance between the image vector and the image vector corresponding to the image in the preset pollution source image library to obtain the vector distance; when the vector distance is not greater than the preset distance, it is determined that the preprocessed image and the image in the preset pollution source image library are successfully matched; determine the multiple label information carried by the image in the preset pollution source image library that has been successfully matched, and determine the multiple label information as the image information set corresponding to the preprocessed image.

[0036] Preferably, in the atmospheric prediction and analysis system based on pollution source image processing, the meteorological data includes: wind speed data, wind direction data and humidity data.

[0037] Preferably, the atmospheric prediction and analysis system based on pollution source image processing, the atmospheric prediction and analysis module, comprises:

[0038] A first prediction submodule, used to predict the change of pollution concentration generated by the pollution source in the pollution source image according to the initial concentration of the pollution source in the image data set and the meteorological data set, to obtain a first prediction result;

[0039] A second prediction submodule is used to predict the area affected by the pollution source in the pollution source image according to the change of pollution concentration to obtain a second prediction result;

[0040] The prediction result analysis submodule is used to analyze the air pollution situation according to the first prediction result and the second prediction result to obtain the air prediction analysis result.

[0041] Preferably, the atmospheric prediction and analysis system based on pollution source image processing, the first prediction submodule comprises:

[0042] A propagation direction determination submodule is used to determine the propagation direction of pollutants generated by the pollution source according to wind direction data;

[0043] A propagation speed determination submodule is used to determine the propagation speed of pollutants generated by the pollution source based on wind speed data;

[0044] The area determination submodule is used to preliminarily determine the contaminated area of ​​the pollution source according to the initial position, propagation direction and propagation speed of the pollution source in the image data set;

[0045] An attenuation index determination submodule, used to determine the attenuation index of the pollution concentration in the polluted area according to the humidity data;

[0046] A grid division submodule is used to divide the polluted area into a number of grids of preset sizes;

[0047] Pollution concentration prediction submodule, used for:

[0048] Model training is performed based on historical pollution concentration data, historical meteorological data sets, and historical attenuation indexes to obtain a pollution concentration prediction model;

[0049] The current pollution concentration of the grid, the current meteorological data set and the attenuation index are input into the pollution concentration prediction model to obtain the pollution concentration prediction result of the grid;

[0050] The first prediction result determination submodule is used to obtain the pollution concentration change curve of the grid according to several pollution concentration prediction results of the grid, and use several pollution concentration change curves of several grids as the first prediction result.

[0051] Preferably, the second prediction submodule of the atmospheric prediction and analysis system based on pollution source image processing includes:

[0052] The grid determination submodule is used to determine the target edge grid; the target edge grid is a grid whose average pollution concentration prediction result is less than a preset pollution threshold;

[0053] The area correction submodule is used to take the areas determined by the several target edge grids as the target pollution area, correct the pollution area of ​​the pollution source based on the target pollution area, and take the target pollution area as the second prediction result.

[0054] Preferably, the atmospheric prediction and analysis system based on pollution source image processing further includes:

[0055] The result output module is used to output the atmospheric prediction analysis results obtained by the atmospheric prediction analysis module to the terminal interface in a preset output form.

[0056] The present invention provides an atmospheric prediction and analysis system based on pollution source image processing, including: an image acquisition module for acquiring pollution source images acquired by an image acquisition device; an image preprocessing module for performing image preprocessing on pollution source images to obtain preprocessed images; an image information acquisition module for performing image information matching on preprocessed images to obtain image information sets corresponding to preprocessed images; a meteorological data acquisition module for acquiring a number of meteorological data and forming a meteorological data set based on the several meteorological data; an atmospheric prediction and analysis module for performing atmospheric prediction and analysis based on the image information set and the meteorological data set to obtain atmospheric prediction and analysis results. The present invention achieves the technical effect of improving the prediction accuracy of atmospheric pollution, obtaining accurate atmospheric pollution analysis results, and thus improving the prevention and control effect of atmospheric pollution.

[0057] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0058] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0060] Figure 1 A block diagram of an atmospheric prediction and analysis system based on pollution source image processing in an embodiment of the present invention;

[0061] Figure 2 is a block diagram of an optional image denoising submodule in an embodiment of the present invention;

[0062] Figure 3 This is a block diagram of an optional first prediction submodule in an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0064] Reference Figure 1 The embodiment of the present invention provides an atmospheric prediction and analysis system based on pollution source image processing, comprising:

[0065] An image acquisition module 10 is used to acquire the pollution source image acquired by the image acquisition device;

[0066] An image preprocessing module 20 is used to perform image preprocessing on the pollution source image to obtain a preprocessed image;

[0067] An image information acquisition module 30 is used to perform image information matching on the pre-processed image to obtain an image information set corresponding to the pre-processed image;

[0068] A meteorological data acquisition module 40 is used to acquire a plurality of meteorological data and form a meteorological data set according to the plurality of meteorological data;

[0069] The atmosphere prediction and analysis module 50 is used to perform atmosphere prediction and analysis based on the image information set and the meteorological data set to obtain an atmosphere prediction and analysis result.

[0070] In this embodiment, the image acquisition device is a camera, a scanner, a drone with a camera, or a satellite with a camera.

[0071] In this embodiment, the pollution source image may be an image of waste gas emitted by a factory, an image of harmful gas produced by a volcanic eruption, or the like.

[0072] The technical principle of the above technical solution is: an image acquisition module, which is used to acquire the pollution source image acquired by the image acquisition device; an image preprocessing module, which is used to perform image preprocessing on the pollution source image to obtain a preprocessed image; an image information acquisition module, which is used to perform image information matching on the preprocessed image to obtain an image information set corresponding to the preprocessed image; a meteorological data acquisition module, which is used to acquire a number of meteorological data and form a meteorological data set based on the several meteorological data; an atmospheric prediction and analysis module, which is used to perform atmospheric prediction and analysis based on the image information set and the meteorological data set to obtain an atmospheric prediction and analysis result.

[0073] The technical effect of the above technical scheme is: by collecting and processing pollution source images, an image information set is obtained, and atmospheric prediction analysis is performed based on the image information set and the acquired meteorological data set to obtain atmospheric prediction analysis results, thereby achieving the technical effect of improving the prediction accuracy of atmospheric pollution, obtaining accurate atmospheric pollution analysis results, and further improving the prevention and control effects of atmospheric pollution.

[0074] The embodiment of the present invention provides an atmosphere prediction and analysis system based on pollution source image processing, and an image preprocessing module, including:

[0075] An image sharpening submodule, used for sharpening the pollution source image to obtain a first processed image;

[0076] An image denoising submodule, used for performing denoising on the first processed image to obtain a second processed image;

[0077] The image contour extraction submodule is used to perform a contour extraction operation on the second processed image to obtain a preprocessed image.

[0078] In this embodiment, the pollution source image is sharpened to obtain the first processed image. The specific implementation method may be: any pixel point (x, y) in the pollution source image is selected as the central pixel point, and the central pixel point and the eight adjacent pixels form the current processing area; the pixel value of the pixel point (x, y) is F(x, y), and the sharpened pixel value G(x, y) of the central pixel point (x, y) is obtained by using the Sobel operator, G(x, y) = v x |+v y |, where v x With v y The calculation formula is as follows: x =[F(x-1,y-1)+2F(x-1,y)+F(x-1,y+1)]-[F(x+1,y-1)+2F(x+1,y)+F(x+1,y+1)], v y =[F(x-1,y-1)+2F(x,y-1)+F(x+1,y-1)]-[F(x-1,y+1)+2F(x,y+1)+F(x+1,y+1)]; Replace F(x,y) with G(x,y), perform the above operation on all pixels to obtain the first processed image.

[0079] In this embodiment, the contour extraction operation is performed on the second processed image to obtain the specific implementation method of the preprocessed image can be: the second processed image is binarized. For the image after processing, if there is a black pixel in the image and the eight adjacent pixels are all black, the pixel is deleted, and the above processing is performed on all the pixels to obtain the preprocessed image.

[0080] The technical principle of the above technical solution is: an image sharpening submodule is used to sharpen the pollution source image to obtain a first processed image; an image denoising submodule is used to perform denoising on the first processed image to obtain a second processed image; an image contour extraction submodule is used to perform contour extraction on the second processed image to obtain a preprocessed image.

[0081] The technical effect of the above technical solution is: by sharpening, denoising and extracting the contour of the pollution source image, a preprocessed image is obtained, the image clarity is enhanced, the image noise is removed, the image features are highlighted, which is beneficial to improving the accuracy of subsequent image matching.

[0082] Reference Figure 2 The embodiment of the present invention provides an atmospheric prediction and analysis system based on pollution source image processing, and an image denoising submodule, including:

[0083] A gray value acquisition submodule 2021 is used to acquire gray values ​​of all pixels of the first processed image;

[0084] The first region determination submodule 2022 is used to select any pixel point in the first processed image as a central pixel point, and determine an N*N region with the central pixel point as the region center as the first region;

[0085] The average gray value calculation submodule 2023 is used to calculate the average gray value of all pixels in the first area;

[0086] The noise pixel determination submodule 2024 is used to compare the grayscale values ​​of all pixels in the first area with the average grayscale value, and when the ratio of the grayscale value of the pixel to the average grayscale value is less than the first predetermined threshold or greater than the second predetermined threshold, the pixel is determined to be a noise pixel, and the above operation is performed on all pixels in the first processed image to determine all noise pixels in the first processed image; wherein the first predetermined threshold is less than the second predetermined threshold;

[0087] The second region determination submodule 2025 is used to select any noise pixel as a pixel to be processed, and determine an M*M region with the pixel to be processed as the region center as the second region;

[0088] A similar pixel point determination submodule 2026, configured to set a pixel point in the second region having the smallest distance from the pixel point to be processed as a similar pixel point to the pixel point to be processed;

[0089] A third region determination submodule 2027 is used to determine a region consisting of a plurality of pixels in the second region having the same channel type as the adjacent pixels as the third region;

[0090] A weight determination submodule 2028, configured to determine a filtering weight for the third region based on the pixels in the second region and the pixels in the third region;

[0091] The pixel value correction submodule 2029 is used to multiply the filter weight by the pixel value of the pixel to be processed, and use the product as the corrected pixel value of the pixel to be processed. The above operation is performed on all noise pixels to obtain a corrected image, and the corrected image is used as the second processed image.

[0092] In this embodiment, the N*N area can be a 3*3 area, a 5*5 area, etc.

[0093] In this embodiment, the first predetermined threshold may be 0.9, and the second predetermined threshold may be 1.2.

[0094] In this embodiment, the M*M area can be a 3*3 area, a 5*5 area, etc.

[0095] In this embodiment, the number of the plurality of pixels having the same channel type as the adjacent pixel is M*M.

[0096] In this embodiment, the channel types are R channel type, G channel type and B channel type.

[0097] The technical principle of the above technical solution is: a gray value acquisition submodule, used to acquire the gray values ​​of all pixels of the first processed image; a first region determination submodule, used to select any pixel in the first processed image as the central pixel, and determine an N*N region with the central pixel as the region center as the first region; an average gray value calculation submodule, used to calculate the average gray value of all pixels in the first region; a noise pixel determination submodule, used to compare the gray values ​​of all pixels in the first region with the average gray value, and when the ratio of the gray value of the pixel to the average gray value is less than a first predetermined threshold or greater than a second predetermined threshold, the pixel is determined to be a noise pixel, and the above operations are performed on all pixels in the first processed image to determine all noise pixels in the first processed image; wherein the first predetermined threshold is less than the second predetermined threshold; the second region determination submodule A domain determination submodule is used to select any noise pixel point as a pixel point to be processed, and determine an M*M area with the pixel point to be processed as the center of the area as the second area; a similar pixel point determination submodule is used to set the pixel point in the second area with the smallest distance from the pixel point to be processed as the similar pixel point to be processed; a third area determination submodule is used to determine an area composed of several pixels in the second area with the same channel type as the similar pixel point as the third area; a weight determination submodule is used to determine the filtering weight of the third area according to the pixels in the second area and the pixels in the third area; a pixel value correction submodule is used to multiply the filtering weight with the pixel value of the pixel point to be processed, and use the product as the corrected pixel value of the pixel point to be processed, perform the above operations on all noise pixels, obtain a corrected image, and use the corrected image as the second processed image.

[0098] The technical effect of the above technical solution is: by correcting the pixel values ​​of the noise pixels in the first processed image, the denoising process of the first processed image is completed, the clarity of the image is improved, the correctness of the contour extraction is ensured, the features of the image are highlighted, and it is beneficial to improve the accuracy of subsequent image matching.

[0099] The embodiment of the present invention provides an atmospheric prediction and analysis system based on pollution source image processing, and a weight determination submodule, including:

[0100] A first difference value determination submodule, configured to determine a color difference value between the second area and the third area as a first difference value according to the pixel value vector of the second area and the pixel value vector of the third area;

[0101] A second difference value determination submodule, used to determine a distance difference value between the second area and the third area as a second difference value according to the coordinates of the pixel point to be processed in the second area and the coordinates of the pixel point in the third area farthest from the current pixel point to be processed;

[0102] A first weight calculation submodule, used to determine the color weight of the pixels in the third area according to the first difference value and the Gaussian function curve of the control color as the first weight;

[0103] A second weight calculation submodule, used to determine the distance weight of the pixels in the third area according to the second difference value and the Gaussian function curve of the control distance as the second weight;

[0104] The third weight calculation submodule multiplies the first weight and the second weight to obtain a third weight, and uses the third weight as a filtering weight for the third region.

[0105] In this embodiment, the Gaussian function curve for controlling color and the Gaussian function curve for controlling distance can be obtained through existing corresponding functions.

[0106] In this embodiment, a specific implementation method of determining the color difference value between the second area and the third area according to the pixel value vector of the second area and the pixel value vector of the third area may be: determining the pixel value vector of the second area to be Determine the pixel value vector of the third area as The color difference value

[0107] In this embodiment, according to the coordinates of the pixel to be processed in the second area and the coordinates of the pixel farthest from the pixel to be processed in the third area, a specific implementation method of determining the distance difference value between the second area and the third area may be: determining the coordinates of the pixel A to be processed as (a1, b1), determining the coordinates of the pixel B farthest from the pixel to be processed in the third area as (a2, b2), then the distance difference value Dis = (a1-a2) 2 +(b1-b2) 2 .

[0108] In this embodiment, a specific implementation method of determining the color weight of the pixels in the third area according to the first difference value and the Gaussian function curve of the control color may be: determining the average value of the Gaussian function curve of the control color as Csigma, and calculating the color weight as:

[0109] In this embodiment, a specific implementation method of determining the distance weight of the pixels in the third area according to the second difference value and the Gaussian function curve of the control distance may be: determining the average value of the Gaussian function curve of the control distance as Dsigma, and calculating the distance weight as:

[0110] In this embodiment, the filtering weight is C_weight×D_weight.

[0111] The technical principle of the above technical solution is: a first difference value determination submodule is used to determine the color difference value between the second area and the third area according to the pixel value vector of the second area and the pixel value vector of the third area, as the first difference value; a second difference value determination submodule is used to determine the distance difference value between the second area and the third area according to the coordinates of the pixel point to be processed in the second area and the coordinates of the pixel point in the third area that is farthest from the current noise pixel point, as the second difference value; a first weight calculation submodule is used to determine the color weight of the pixel in the third area according to the first difference value and the Gaussian function curve that controls the color, as the first weight; a second weight calculation submodule is used to determine the distance weight of the pixel in the third area according to the second difference value and the Gaussian function curve that controls the distance, as the second weight; a third weight calculation submodule multiplies the first weight and the second weight to obtain a third weight, and uses the third weight as the filtering weight of the third area.

[0112] The technical effect of the above technical solution is: the filtering weight is determined by the color weight and the distance weight, which provides a data basis for the pixel value correction of the noise pixel point, ensures the integrity and correctness of the noise reduction process, and thus ensures that the second processed image can be obtained smoothly.

[0113] The embodiment of the present invention provides an atmosphere prediction and analysis system based on pollution source image processing, and an image information acquisition module, including:

[0114] An image vector generation submodule is used to generate an image vector corresponding to the preprocessed image;

[0115] The image information determination submodule is used to calculate the distance between the image vector and the image vector corresponding to the image in the preset pollution source image library to obtain the vector distance; when the vector distance is not greater than the preset distance, it is determined that the preprocessed image and the image in the preset pollution source image library are successfully matched; determine the multiple label information carried by the image in the preset pollution source image library that has been successfully matched, and determine the multiple label information as the image information set corresponding to the preprocessed image.

[0116] In this embodiment, a specific implementation method of generating an image vector corresponding to the preprocessed image may be: extracting features from the preprocessed image to obtain a feature vector corresponding to the preprocessed image, and using the feature vector as the image vector.

[0117] In this embodiment, the image vector is calculated from the image vector corresponding to the image in the preset pollution source image library, and the specific implementation method of obtaining the vector distance may be: determining the image vector as Determine the image vector corresponding to the image in the preset pollution source image library: The vector distance is

[0118] In this embodiment, the image information set includes the initial concentration of the pollution source, the initial position of the pollution source, the name of the pollution source, and the type of the pollution source.

[0119] The technical principle of the above technical solution is: an image vector generation submodule is used to generate an image vector corresponding to a preprocessed image; an image information determination submodule is used to calculate the distance between the image vector and the image vector corresponding to the image in the preset pollution source image library to obtain the vector distance; when the vector distance is not greater than the preset distance, it is determined that the preprocessed image and the image in the preset pollution source image library are successfully matched; multiple label information carried by the image in the preset pollution source image library that has been successfully matched is determined, and the multiple label information is determined as the image information set corresponding to the preprocessed image.

[0120] The technical effect of the above technical solution is: an image information set of the pollution source image is obtained through image matching, ensuring the accuracy of subsequent atmospheric analysis and prediction results.

[0121] The embodiment of the present invention provides an atmosphere prediction and analysis system based on pollution source image processing, and the meteorological data includes: wind speed data, wind direction data and humidity data.

[0122] The technical principle of the above technical solution is: to obtain wind speed data through wind speed collection equipment, to obtain wind direction data through wind direction collection equipment, and to obtain humidity data through humidity collection equipment.

[0123] The technical effect of the above technical solution is: by acquiring wind speed data, wind direction data and humidity data, the influencing data for atmospheric forecasting and analysis is obtained, thereby ensuring the correctness of the atmospheric analysis and forecasting results.

[0124] The embodiment of the present invention provides an atmosphere prediction and analysis system based on pollution source image processing, and an atmosphere prediction and analysis module, including:

[0125] A first prediction submodule, used to predict the change of pollution concentration generated by the pollution source in the pollution source image according to the initial concentration of the pollution source in the image data set and the meteorological data set, to obtain a first prediction result;

[0126] A second prediction submodule is used to predict the area affected by the pollution source in the pollution source image according to the change of pollution concentration to obtain a second prediction result;

[0127] The prediction result analysis submodule is used to analyze the air pollution situation according to the first prediction result and the second prediction result to obtain the air prediction analysis result.

[0128] In this embodiment, the atmospheric pollution situation is analyzed according to the first prediction result and the second prediction result, and a specific implementation method for obtaining the atmospheric prediction analysis result can be: determining the change in regional pollution concentration according to the first prediction result, determining the pollution range according to the second prediction result, determining the impact of the pollution source according to the change in regional pollution concentration and the pollution range, and obtaining the atmospheric prediction analysis result according to the impact.

[0129] The technical principle of the above technical solution is: a first prediction submodule is used to predict the change of pollution concentration caused by the pollution source in the pollution source image according to the initial concentration of the pollution source in the image data set and the meteorological data set, and obtain a first prediction result; a second prediction submodule is used to predict the area affected by the pollution source in the pollution source image according to the change of pollution concentration, and obtain a second prediction result; a prediction result analysis submodule is used to analyze the atmospheric pollution situation according to the first prediction result and the second prediction result, and obtain an atmospheric prediction analysis result.

[0130] The technical effect of the above technical scheme is: by predicting the changes in pollution concentration caused by pollution sources and predicting the areas affected by pollution sources, the accuracy and rationality of atmospheric forecasting and analysis results are ensured, the accuracy of atmospheric pollution prediction is improved, accurate atmospheric pollution analysis results are obtained, and the technical effect of improving the prevention and control of atmospheric pollution is achieved.

[0131] Reference Figure 3 The embodiment of the present invention provides an atmospheric prediction and analysis system based on pollution source image processing, wherein the first prediction submodule comprises:

[0132] The propagation direction determination submodule 5011 is used to determine the propagation direction of pollutants generated by the pollution source according to the wind direction data;

[0133] The propagation speed determination submodule 5012 is used to determine the propagation speed of pollutants generated by the pollution source according to the wind speed data;

[0134] The area determination submodule 5013 is used to preliminarily determine the pollution area of ​​the pollution source according to the initial position, propagation direction and propagation speed of the pollution source in the image data set;

[0135] The attenuation index determination submodule 5014 is used to determine the attenuation index of the pollution concentration in the polluted area according to the humidity data;

[0136] The grid division submodule 5015 is used to divide the polluted area into a plurality of grids of a preset size;

[0137] The pollution concentration prediction submodule 5016 is used to:

[0138] Model training is performed based on historical pollution concentration data, historical meteorological data sets, and historical attenuation indexes to obtain a pollution concentration prediction model;

[0139] The current pollution concentration of the grid, the current meteorological data set and the attenuation index are input into the pollution concentration prediction model to obtain the pollution concentration prediction result of the grid;

[0140] The first prediction result determination submodule 5017 is used to obtain the pollution concentration change curve of the grid according to the several pollution concentration prediction results of the grid, and use the several pollution concentration change curves of the several grids as the first prediction result.

[0141] In this embodiment, the wind direction is used as the propagation direction of the pollutants.

[0142] In this embodiment, the wind speed is used as the propagation speed of pollutants.

[0143] In this embodiment, the specific implementation method of preliminarily determining the contaminated area of ​​the pollution source based on the initial position, propagation direction and propagation speed of the pollution source in the image data set can be: taking the initial position of the pollution source as the starting point, determining the propagation speed to be Vkm / h, then the contaminated area is a 24V range along the propagation direction.

[0144] In this embodiment, a specific implementation method of determining the attenuation index of the pollution concentration in the polluted area according to the humidity data may be: if the humidity data is S, then the attenuation index is: e -S .

[0145] In this embodiment, the preset size may be 20m*20m.

[0146] The technical principle of the above technical solution is: a propagation direction determination submodule, which is used to determine the propagation direction of pollutants generated by the pollution source according to wind direction data; a propagation speed determination submodule, which is used to determine the propagation speed of pollutants generated by the pollution source according to wind speed data; an area determination submodule, which is used to preliminarily determine the pollution area of ​​the pollution source according to the initial position, propagation direction and propagation speed of the pollution source in the image data set; an attenuation index determination submodule, which is used to determine the attenuation index of the pollution concentration in the pollution area according to humidity data; a grid division submodule, which is used to divide the pollution area into several grids of preset sizes; a pollution concentration prediction submodule, which is used to: perform model training according to historical pollution concentration data, historical meteorological data sets and historical attenuation index to obtain a pollution concentration prediction model; input the current pollution concentration, current meteorological data set and attenuation index of the grid into the pollution concentration prediction model to obtain the pollution concentration prediction result of the grid; a first prediction result determination submodule, which is used to obtain the pollution concentration change curve of the grid according to several pollution concentration prediction results of the grid, and use several pollution concentration change curves of several grids as the first prediction result.

[0147] The technical effect of the above technical solution is: by determining the attenuation index of pollution concentration and regional division, the pollution concentration change curve of each area is obtained, the pollution concentration change prediction is realized, and the accuracy of the pollution concentration change prediction is guaranteed.

[0148] The embodiment of the present invention provides an atmosphere prediction and analysis system based on pollution source image processing, wherein the second prediction submodule comprises:

[0149] The grid determination submodule is used to determine the target edge grid; the target edge grid is a grid whose average pollution concentration prediction result is less than a preset pollution threshold;

[0150] The area correction submodule is used to take the areas determined by the several target edge grids as the target pollution area, correct the pollution area of ​​the pollution source based on the target pollution area, and take the target pollution area as the second prediction result.

[0151] In this embodiment, the average pollution concentration is determined by the pollution concentration variation curve of the target grid, which is the average value of the pollution concentration variation curve.

[0152] In this embodiment, the preset pollution threshold is the average of several average values ​​of several pollution concentration change curves corresponding to several grids of preset sizes. For example, the average value of the pollution concentration change curve of grid 1 is A1, the average value of the pollution concentration change curve of grid 2 is A2, and the average value of the concentration change curve of grid n is An. Then the preset pollution threshold is (A1+A2+…+An) / n.

[0153] In this embodiment, a specific implementation method for taking an area determined by several target edge grids as a target pollution area and correcting the pollution area of ​​a pollution source based on the target pollution area may be: when a current edge grid of the pollution area is included in several target edge grids, the current edge grid is kept as a grid in the target pollution area; when the current edge grid of the pollution area is not included in several target edge grids, the current edge grid is replaced with a grid among several target edge grids that is closest to the current edge grid, and the replaced grid is used as a grid in the target pollution area to correct the pollution area.

[0154] The technical principle of the above technical solution is: a grid determination submodule is used to determine the target edge grid; the target edge grid is a grid whose average pollution concentration prediction result is less than a preset pollution threshold; an area correction submodule is used to use the area determined by several target edge grids as the target pollution area, correct the pollution area of ​​the pollution source based on the target pollution area, and use the target pollution area as the second prediction result.

[0155] The technical effect of the above technical solution is: the polluted area is corrected, the accuracy of the prediction results of the polluted area is improved, and the accuracy of the atmospheric prediction and analysis results is ensured.

[0156] The embodiment of the present invention provides an atmospheric prediction and analysis system based on pollution source image processing, further comprising:

[0157] The result output module is used to output the atmospheric prediction analysis results obtained by the atmospheric prediction analysis module to the terminal interface in a preset output form.

[0158] In this embodiment, the preset output mode may be: chart output mode, report output mode, etc.

[0159] The technical principle of the above technical solution is: the result output module is used to output the atmospheric prediction and analysis results obtained by the atmospheric prediction and analysis module to the terminal interface in a preset output form.

[0160] The technical effect of the above technical solution is: by outputting the atmospheric prediction and analysis results, users can see the atmospheric prediction and analysis results more intuitively, which is beneficial to future atmospheric pollution prevention and control.

[0161] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An atmospheric prediction and analysis system based on pollution source image processing, characterized in that: include: An image acquisition module is used to acquire the pollution source image acquired by the image acquisition device; An image preprocessing module, used for performing image preprocessing on the pollution source image to obtain a preprocessed image; An image information acquisition module, used to perform image information matching on the pre-processed image to obtain an image information set corresponding to the pre-processed image; A meteorological data acquisition module is used to acquire a number of meteorological data and form a meteorological data set according to the several meteorological data; The atmospheric prediction and analysis module is used to perform atmospheric prediction and analysis based on the image information set and the meteorological data set to obtain atmospheric prediction and analysis results.

2. The atmospheric prediction and analysis system based on pollution source image processing according to claim 1, characterized in that: The image preprocessing module comprises: An image sharpening submodule, used for sharpening the pollution source image to obtain a first processed image; An image denoising submodule, configured to perform denoising on the first processed image to obtain a second processed image; The image contour extraction submodule is used to perform a contour extraction operation on the second processed image to obtain a preprocessed image.

3. The atmospheric prediction and analysis system based on pollution source image processing according to claim 2, characterized in that: The image denoising submodule comprises: A gray value acquisition submodule, used to acquire gray values ​​of all pixels of the first processed image; A first region determination submodule, configured to select any pixel point in the first processed image as a central pixel point, and determine an N*N region with the central pixel point as the region center as a first region; An average gray value calculation submodule, used to calculate the average gray value of all pixels in the first area; A noise pixel determination submodule is used to compare the grayscale values ​​of all pixels in the first area with the average grayscale value, and when the ratio of the grayscale value of the pixel to the average grayscale value is less than a first predetermined threshold or greater than a second predetermined threshold, the pixel is determined to be a noise pixel, and the above operation is performed on all pixels in the first processed image to determine all noise pixels in the first processed image; wherein the first predetermined threshold is less than the second predetermined threshold; The second region determination submodule is used to select any noise pixel point as a pixel point to be processed, and determine an M*M region with the pixel point to be processed as the region center as the second region; A similar pixel point determination submodule, used for setting the pixel point in the second area with the smallest distance from the pixel point to be processed as a similar pixel point to the pixel point to be processed; A third region determination submodule, configured to determine a region consisting of a plurality of pixels in the second region having the same channel type as the adjacent pixels as a third region; A weight determination submodule, used to determine a filtering weight of the third area according to the pixel points in the second area and the pixel points in the third area; The pixel value correction submodule is used to multiply the filter weight by the pixel value of the pixel to be processed, and use the product as the corrected pixel value of the pixel to be processed. The above operation is performed on all noise pixels to obtain a corrected image, and the corrected image is used as the second processed image.

4. The atmospheric prediction and analysis system based on pollution source image processing according to claim 3, characterized in that: The weight determination submodule comprises: A first difference value determination submodule, configured to determine a color difference value between the second area and the third area as a first difference value according to the pixel value vector of the second area and the pixel value vector of the third area; A second difference value determination submodule, used to determine a distance difference value between the second area and the third area as a second difference value according to the coordinates of the pixel point to be processed in the second area and the coordinates of the pixel point in the third area farthest from the current pixel point to be processed; A first weight calculation submodule, used to determine the color weight of the pixels in the third area according to the first difference value and the Gaussian function curve of the control color as the first weight; A second weight calculation submodule, used to determine the distance weight of the pixels in the third area according to the second difference value and the Gaussian function curve of the control distance as the second weight; The third weight calculation submodule multiplies the first weight and the second weight to obtain a third weight, and uses the third weight as a filtering weight for the third region.

5. The atmospheric prediction and analysis system based on pollution source image processing according to claim 1, characterized in that: The image information acquisition module comprises: An image vector generation submodule is used to generate an image vector corresponding to the preprocessed image; The image information determination submodule is used to calculate the distance between the image vector and the image vector corresponding to the image in the preset pollution source image library to obtain the vector distance; when the vector distance is not greater than the preset distance, it is determined that the preprocessed image and the image in the preset pollution source image library are successfully matched; determine the multiple label information carried by the image in the preset pollution source image library that has been successfully matched, and determine the multiple label information as the image information set corresponding to the preprocessed image.

6. The atmospheric prediction and analysis system based on pollution source image processing according to claim 1, characterized in that: The meteorological data includes: wind speed data, wind direction data and humidity data.

7. The atmospheric prediction and analysis system based on pollution source image processing according to claim 6, characterized in that: The atmospheric prediction and analysis module comprises: A first prediction submodule, used to predict the change of pollution concentration generated by the pollution source in the pollution source image according to the initial concentration of the pollution source in the image data set and the meteorological data set, to obtain a first prediction result; A second prediction submodule is used to predict the area affected by the pollution source in the pollution source image according to the change of pollution concentration to obtain a second prediction result; The prediction result analysis submodule is used to analyze the air pollution situation according to the first prediction result and the second prediction result to obtain an air prediction analysis result.

8. The atmospheric prediction and analysis system based on pollution source image processing according to claim 7, characterized in that: The first prediction submodule comprises: A propagation direction determination submodule is used to determine the propagation direction of pollutants generated by the pollution source according to wind direction data; A propagation speed determination submodule is used to determine the propagation speed of pollutants generated by the pollution source based on wind speed data; The area determination submodule is used to preliminarily determine the contaminated area of ​​the pollution source according to the initial position, propagation direction and propagation speed of the pollution source in the image data set; An attenuation index determination submodule, used to determine the attenuation index of the pollution concentration in the polluted area according to the humidity data; A grid division submodule, used for dividing the polluted area into a plurality of grids of preset sizes; Pollution concentration prediction submodule, used for: Model training is performed based on historical pollution concentration data, historical meteorological data sets, and historical attenuation indexes to obtain a pollution concentration prediction model; The current pollution concentration of the grid, the current meteorological data set and the attenuation index are input into the pollution concentration prediction model to obtain the pollution concentration prediction result of the grid; The first prediction result determination submodule is used to obtain the pollution concentration change curve of the grid according to several pollution concentration prediction results of the grid, and use several pollution concentration change curves of several grids as the first prediction result.

9. The atmospheric prediction and analysis system based on pollution source image processing according to claim 7, characterized in that: The second prediction submodule comprises: A grid determination submodule is used to determine a target edge grid; the target edge grid is a grid whose average pollution concentration prediction result is less than a preset pollution threshold; The area correction submodule is used to take the areas determined by the several target edge grids as the target pollution area, correct the pollution area of ​​the pollution source based on the target pollution area, and take the target pollution area as the second prediction result.

10. The atmospheric prediction and analysis system based on pollution source image processing according to claim 1, characterized in that: Also includes: The result output module is used to output the atmospheric prediction and analysis results obtained by the atmospheric prediction and analysis module to a terminal interface in a preset output form.

Images