A method and system for classifying the risk of black and odorous water bodies in a city along a river

Abstract

The application provides a method and system for risk classification of black and odorous water bodies in cities along rivers, and relates to the technical field of black and odorous water body monitoring. The specific steps include: collecting city, water body and environmental parameters from multiple sources, and synchronously and continuously collecting environmental parameters, high-altitude images, visible light images and multispectral images of the city and water body in multiple years, processing to construct time series data, building a black and odorous water body scale prediction model, obtaining the predicted value of the characteristic parameters of the black and odorous water body, generating a predicted risk assessment index through a linear weighting of the scale prediction index, the last year's purification efficiency index and the environmental quality index, and realizing risk grade classification. The application can deeply analyze the characteristics of black and odorous water bodies and consider the dynamic evolution law thereof under complex hydrological conditions. The risk classification result is in line with the actual needs of cities along rivers, and a complete evaluation system is built to evaluate the risk of black and odorous water bodies from multiple dimensions, so that the risk degree of black and odorous water bodies is more reasonably evaluated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of black and odorous water body monitoring, in particular to a method and system for risk classification of black and odorous water bodies in cities along rivers. BACKGROUND

[0002] At present, there are many studies on the risk classification method of black and odorous water bodies. Common technical solutions are mostly based on general water quality indicators, pollution source distribution, population density and other factors to build evaluation models. These methods usually use a unified evaluation system to analyze the pollution degree and source of pollutants of urban water bodies, and divide different levels according to the risk level. However, the existing technology does not fully consider the particularity of cities along rivers, and has obvious limitations. Cities along rivers are built along rivers, and their geographical environment, ecological system and human activity patterns have distinct characteristics. In terms of pollution sources, cities along rivers face unique pollution problems such as oil spillage from ships, direct discharge of domestic sewage from shipping, accumulation of port waste, etc. in addition to conventional urban pollution. High-intensity riverside land development further exacerbates the risk of non-point source pollution, and hydrological conditions such as river tides, seasonal floods significantly affect the diffusion path of pollutants and the self-purification capacity of water bodies.

[0003] In the prior art, a method and system for risk classification of black and odorous water bodies in river network intensive cities are provided in CN118072165B, which includes the following steps: step S1: obtaining a plurality of first remote sensing images and a plurality of first water body data of the target water body; step S2: establishing the numerical relationship between each feature value in the second feature set and the first water body data; step S3: obtaining a plurality of second remote sensing images and obtaining second water body data; step S4: a model creation unit creates a plurality of decision trees; step S5: inputting the third water body data into the plurality of decision trees and outputting the risk classification results; step S6: obtaining the risk level of black and odorous water bodies in river network intensive cities. This method solves the problem of inaccurate risk classification of black and odorous water bodies and classification of black and odorous water bodies in river network intensive cities, and improves the accuracy of risk classification of black and odorous water bodies and classification of black and odorous water bodies in river network intensive cities.

[0004] However, there are still some shortcomings. As can be seen from the above statements, the prior art generally ignores the dynamic correlation between per capita sewage treatment capacity and river flow, the superposition effect of sewage plant scale ratio and water network density, and does not consider the core elements of the city along the river, such as the influence of river flow rate on pollutant diffusion of sewage outlet and the role of ecological shoreline rate on river ecological buffer function. In terms of data collection, it is difficult to fully capture the key driving factors of black and odorous water body risk in the city along the river, as the exclusive data resources such as high-altitude images reflecting topography and multispectral images representing pollutant composition are not systematically integrated. In addition, the existing method for feature analysis of black and odorous water body is limited to conventional morphological parameters, and lacks in-depth research on the dynamic evolution law of black and odorous water body under complex hydrological conditions along the river, resulting in a large deviation between the risk division results and the actual needs of the city along the river.

[0005] The above information disclosed in the background section is only intended to strengthen the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0006] The purpose of the present application is to provide a method and system for dividing the risk of black and odorous water body in a city along a river to solve the problems raised in the background.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0008] A method for dividing the risk of black and odorous water body in a city along a river, comprising the following specific steps:

[0009] S1. Collecting city and water body index parameters of the city along the river in the previous year, and synchronously and continuously collecting environmental parameters, high-altitude images, visible light images and multispectral images of the city along the river in multiple historical years and the current year, preprocessing the high-altitude images, visible light images and multispectral images to obtain fused images;

[0010] S2. Edge detection is performed on the fused images to obtain the water body boundary of the black and odorous water body in the city along the river, and the feature parameters of the black and odorous water body in the city along the river are extracted based on the water body boundary. The feature parameters and environmental parameters of the black and odorous water body in the city along the river are fused to construct the time series data of the black and odorous water body in the city along the river;

[0011] S3. A scale prediction model of the black and odorous water body in the city along the river is constructed based on a long short-term memory network, the time series data of the black and odorous water body in the city along the river in the previous year in multiple historical years is taken as input, and the feature parameters of the black and odorous water body in the city along the river in the next year in multiple historical years are taken as label output, and the scale prediction model of the black and odorous water body in the city along the river is trained;

[0012] S4. Input the time series data of the current year's black and odorous water body in the city along the river into the black and odorous water body size prediction model to obtain the predicted value of the characteristic parameter of the black and odorous water body in the city along the river;

[0013] S5. Based on the predicted value of the characteristic parameter of the black and odorous water body in the city along the river, obtain a size prediction index for comprehensively evaluating the size of the black and odorous water body in the city along the river, based on the city index parameter of the city along the river in the last year, obtain a purification efficiency index for comprehensively evaluating the sewage treatment and self-purification capacity of the city along the river, and based on the water body index parameter of the city along the river in the last year, obtain an environmental quality index for comprehensively evaluating the pollution degree of the water body in the city along the river;

[0014] S6. Based on the size prediction index of the black and odorous water body in the city along the river, the environmental quality index and the purification efficiency index of the city along the river in the last year, obtain a risk assessment index for evaluating the risk degree of the black and odorous water body in the city along the river, compare the risk assessment index with a pre-set threshold value, and according to the comparison result, divide the risk level of the black and odorous water body in the city along the river.

[0015] Further, the city index parameter includes per capita domestic sewage treatment capacity, domestic sewage centralized collection rate, sewage plant size proportion and water network density, the water body index parameter includes water body area index, number of sewage outlets, ecological shoreline rate and water quality comprehensive index, the environmental parameter includes river water flow rate, river water flow and monthly average rainfall, and the characteristic parameter of the black and odorous water body includes area, circumference and circularity.

[0016] Further, edge detection is performed on the fused image to obtain the water body boundary of the black and odorous water body in the city along the river, and the specific steps are as follows:

[0017] The fused image is converted into a gray image, and for an RGB image, the conversion is performed through the following formula:

[0018] ;

[0019] wherein, is the gray value of the pixel, , , is the pixel value of the red channel, the green channel and the blue channel;

[0020] The Sobel operator is used to calculate the gradient in the horizontal direction and the vertical direction:

[0021] ;

[0022] ;

[0023] wherein, is the gradient in the horizontal direction, is the gradient in the vertical direction, is a fused image, represents a convolution operation;

[0024] The gradient magnitude and direction are respectively:

[0025] ;

[0026] ;

[0027] wherein, is the gradient magnitude, is the gradient direction;

[0028] The gradient magnitude and two threshold values are compared to identify the edge pixels in the image, and the specific process is as follows:

[0029] When , the pixel is marked as a strong edge pixel;

[0030] When , the pixel is marked as a weak edge pixel;

[0031] When , the pixel is marked as a non-edge pixel;

[0032] wherein, is a high threshold value, is a low threshold value;

[0033] Starting from the strong edge pixels, connectivity analysis is performed, and then a closed curve is found by using a contour extraction algorithm. Finally, the largest contour is preliminarily screened out by area calculation, and the contour is the water body boundary of the black and odorous water body in the city along the river.

[0034] Further, based on the predicted value of the characteristic parameter of the black and odorous water body in the city along the river, a scale prediction index for comprehensively evaluating the scale of the black and odorous water body in the city along the river is obtained, and the formula is as follows:

[0035] ;

[0036] wherein, is the scale prediction index, which is used to combine the area prediction value, the perimeter prediction value and the circularity prediction value to comprehensively evaluate the scale of the black and odorous water body in the city along the river;

[0037] In the formula, is the area prediction value of the black and odorous water body, is the circularity prediction value of the black and odorous water body, is the perimeter prediction value of the black and odorous water body.

[0038] Further, based on the urban index parameters of the cities along the river in the last year, the purification efficiency index in the last year is obtained, and the formula is as follows:

[0039] ;

[0040] wherein, is the purification efficiency index in the last year, and the purification efficiency index is used to combine the per capita domestic sewage treatment capacity, the domestic sewage centralized collection rate, the sewage plant scale proportion and the water network density to comprehensively evaluate the sewage treatment and self-purification capacity of the cities along the river;

[0041] in the formula, is the per capita domestic sewage treatment capacity in the last year, is the domestic sewage centralized collection rate in the last year, is the sewage plant scale proportion in the last year, is the water network density in the last year;

[0042] in the formula, is the weight coefficient of the per capita domestic sewage treatment capacity, is the weight coefficient of the domestic sewage centralized collection rate, is the weight coefficient of the sewage plant scale proportion, is the weight coefficient of the water network density, and on the basis of , let .

[0043] Further, based on the water body index parameters of the cities along the river in the last year, the environmental quality index in the last year is obtained, and the formula is as follows:

[0044] ;

[0045] wherein, is the environmental quality index in the last year, and the environmental quality index is used to combine the water body area index, the number of sewage outlets, the ecological shoreline rate and the water quality comprehensive index to comprehensively evaluate the water pollution degree;

[0046] in the formula, is the water body area index in the last year, is the number of sewage outlets in the last year, is the ecological shoreline rate in the last year, is the water quality comprehensive index in the last year;

[0047] in the formula, is the weight coefficient of the water body area index, is the weight coefficient of the number of sewage outlets, is the weight coefficient of the ecological shoreline rate, and on the basis of , let .

[0048] Further, based on the scale prediction index of the black and odorous water body in the city along the river, the environmental quality index of the city along the river in the last year and the purification efficiency index, the risk assessment index is obtained, and the formula is as follows:

[0049] ;

[0050] Wherein, The risk assessment index is used to combine the scale prediction index, the environmental quality index of the last year and the purification efficiency index to comprehensively evaluate the risk degree of the black and odorous water body in the city along the river.

[0051] In the formula, The scale prediction index is, The environmental quality index of the last year is, The purification efficiency index of the last year is;

[0052] , , The weight coefficients of the scale prediction index, the environmental quality index and the purification efficiency index are respectively, and , , The specific values of the weight coefficients are determined by the analytic hierarchy process.

[0053] Further, the risk assessment index and the threshold value set in advance are compared to divide the risk grade of the black and odorous water body in the city along the river, and the specific process is as follows:

[0054] When the risk assessment index is less than the threshold value, the predicted black and odorous water body in the city along the river is in a low risk grade; When the risk assessment index is greater than the threshold value, the predicted black and odorous water body in the city along the river is in a medium risk grade;

[0055] When the risk assessment index is greater than the threshold value, the predicted black and odorous water body in the city along the river is in a high risk grade;

[0056] Wherein, The threshold value for dividing the low risk and the medium risk is, The threshold value for dividing the medium risk and the high risk is.

[0057] To achieve the above purpose, the application also provides the following technical schemes:

[0058] A risk division system for the black and odorous water body in the city along the river, which is used for executing the risk division method for the black and odorous water body in the city along the river, and comprises:

[0059] A risk division system for the black and odorous water body in the city along the river, which is used for executing the risk division method for the black and odorous water body in the city along the river, and comprises: ​​

[0060] a data collection module, configured to collect city and water body index parameters of a previous year of a city along a river, and synchronously and continuously collect environmental parameters, high-altitude images, visible light images and multispectral images of the city along the river in multiple historical years and a current year, and perform preprocessing on the high-altitude images, the visible light images and the multispectral images to obtain fused images;

[0061] a time series data construction module, configured to perform edge detection on the fused images to obtain a water body boundary of a black and odorous water body of the city along the river, extract feature parameters of the black and odorous water body of the city along the river based on the water body boundary, and perform data fusion on the feature parameters of the black and odorous water body of the city along the river and the environmental parameters to construct time series data of the black and odorous water body of the city along the river;

[0062] a scale prediction model construction module, configured to construct a scale prediction model of the black and odorous water body of the city along the river based on a long short-term memory network, take time series data of the black and odorous water body of the city along the river in a previous year in the multiple historical years as input, take feature parameters of the black and odorous water body of the city along the river in a next year in the multiple historical years as label output, and train the scale prediction model of the black and odorous water body of the city along the river;

[0063] a simulation module, configured to input the time series data of the black and odorous water body of the city along the river in the current year into the scale prediction model of the black and odorous water body to obtain a predicted value of the feature parameters of the black and odorous water body of the city along the river;

[0064] a data processing module, configured to obtain a scale prediction index for comprehensively evaluating a scale of the black and odorous water body of the city along the river based on the predicted value of the feature parameters of the black and odorous water body of the city along the river, obtain a purification efficiency index for comprehensively evaluating a sewage treatment and self-cleaning capability of the city along the river based on city index parameters of the city along the river in the previous year, and obtain an environmental quality index for comprehensively evaluating a pollution degree of a water body of the city along the river based on water body index parameters of the city along the river in the previous year;

[0065] a grade division module, configured to obtain a risk evaluation index for evaluating a risk degree of the black and odorous water body of the city along the river based on the scale prediction index of the black and odorous water body of the city along the river, the environmental quality index and the purification efficiency index of the city along the river in the previous year, compare the risk evaluation index with a preset threshold value, and divide a risk grade of the black and odorous water body of the city along the river according to a comparison result.

[0066] Compared with the prior art, the present application has the following beneficial effects:

[0067] The present application collects city and water body index parameters of a previous year of a city along a river, considers core elements of the city along the river such as an influence of a river flow rate on pollutant diffusion of a sewage outlet and an effect of an ecological shoreline rate on an ecological buffer function of the river, makes index selection more comprehensive and scientific, and more accurately reflects an actual situation of a black and odorous water body of the city along the river.

[0068] The present application synchronously and continuously collects environmental parameters, high-altitude images, visible light images and multispectral images of cities and water bodies in multiple historical years and the current year of the city along the river, and pre-processes to obtain fusion images, more systematically collects and integrates data, provides more abundant and comprehensive data support for subsequent analysis, and helps to more accurately analyze the risk of black and odorous water bodies.

[0069] The present application not only performs edge detection on the fusion image to obtain the water body boundary and extract feature parameters, but also constructs a black and odorous water body scale prediction model based on a long short-term memory network, trains the model with time series data of multiple historical years, can deeply analyze the characteristics of black and odorous water bodies, and considers the dynamic evolution law under complex hydrological conditions, so that the understanding of black and odorous water bodies is more accurate and in-depth, and the risk division result is more in line with the actual needs of the city along the river.

[0070] The present application respectively obtains a scale prediction index for comprehensive evaluation of the scale of black and odorous water bodies, a purification efficiency index for evaluation of the sewage treatment and self-purification capacity of the city, and an environmental quality index for evaluation of the pollution degree of the water body, and then obtains a risk assessment index based on these indexes to divide the risk grade of black and odorous water bodies, and constructs a complete evaluation system to evaluate the risk of black and odorous water bodies in the city along the river from multiple dimensions. Compared with the single or one-sided evaluation method of the prior art, the risk degree of black and odorous water bodies can be more reasonably evaluated. BRIEF DESCRIPTION OF DRAWINGS

[0071] Figure 1 The present application is a schematic diagram of the overall method flow;

[0072] Figure 2 The present application is a block diagram of the module composition. DETAILED DESCRIPTION

[0073] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below in combination with specific embodiments.

[0074] It should be noted that the technical terms or scientific terms used in the present application should be understood as the general meaning understood by those skilled in the art to which the present application belongs, unless otherwise defined. The terms "first", "second" and the like used in the present application do not represent any order, number or importance, but are only used to distinguish different components. The terms "include" or "contain" and the like mean that the elements or objects before the terms cover the elements or objects listed after the terms and their equivalents, and do not exclude other elements or objects. The terms "connect" or "connected" and the like are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The terms "up", "down", "left", "right" and the like are only used to represent relative positional relationships, and when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0075] Embodiment 1:

[0076] Referring to Figure 1 The present application provides a technical solution:

[0077] A method for dividing the risk of black and odorous water bodies in a city along a river, the specific steps comprising:

[0078] S1. According to the formation mechanism of black and odorous water bodies and the geographical and ecological characteristics of the city along the river, the city and water body index parameters of the previous year are collected, and the environmental parameters of the city and water body of multiple historical years and the current year are synchronously and continuously collected. High-resolution satellite remote sensing images are used to continuously collect high-altitude images of black and odorous water bodies of multiple historical years and the current year. Low-altitude photography is performed by means of a drone. Visible light images and multispectral images of multiple historical years and the current year are continuously collected. GIS (Geographic Information System) software is used to preprocess the collected high-altitude images, visible light images and multispectral images. Through radiation correction, geometric correction and other operations, noise and errors in the images are eliminated. Then image fusion is performed to obtain a fused image.

[0079] On the basis of the above embodiment, the high-altitude images collect the topographic and geomorphic features of the mountains, hills and plains along the river, which have an impact on the flow direction and pollutant diffusion path.

[0080] The visible light images can directly observe the color change of the water surface. Black and odorous water bodies usually show different colors from normal water bodies, such as black, gray or abnormal textures with floating scum, etc. These information are helpful for preliminarily judging the distribution range of black and odorous water bodies.

[0081] The multispectral image contains spectral information of multiple different wavebands, and different substances have unique reflection, absorption and transmission characteristics for light of different wavebands. For black and odorous water bodies, the multispectral image can identify the pollutant composition, concentration and eutrophication degree of the water body by analyzing the reflectivity difference of the water body in specific wavebands such as red light, near-infrared light and short-wave infrared light.

[0082] On the basis of the above-mentioned embodiments, the method for obtaining the fusion image is as follows:

[0083] The spatial resolution and geographical range of the high-altitude image, the visible light image and the multispectral image are set to be consistent, the corresponding pixel values of the high-altitude image, the visible light image and the multispectral image are weighted and summed by using the weighted average method, a weight is assigned to each image, and the corresponding pixels are weighted and summed.

[0084] On the basis of the above-mentioned embodiments, the urban index parameters include per capita domestic sewage treatment capacity, domestic sewage centralized collection rate, sewage plant scale proportion and water network density, and the water body index parameters include water body area index, number of sewage outlets, ecological shoreline rate and water quality comprehensive index.

[0085] The collection method of the per capita domestic sewage treatment capacity, the domestic sewage centralized collection rate, the sewage plant scale proportion and the water network density of the previous year is as follows:

[0086] From the local urban drainage management department or the sewage treatment plant, the total amount of domestic sewage treatment data of the previous year is obtained, and at the same time, the number of urban permanent population of the previous year is obtained from the local government statistical department, and the per capita domestic sewage treatment capacity of the previous year is obtained by the formula “per capita domestic sewage treatment capacity = annual total amount of domestic sewage treatment / number of urban permanent population”.

[0087] From the local urban drainage management department, the amount of domestic sewage that has been included in the centralized collection pipe network of the previous year is obtained, and combined with the total amount of domestic sewage discharge of the previous year, the domestic sewage centralized collection rate of the previous year is obtained by the formula “domestic sewage centralized collection rate = amount of domestic sewage that has been included in the centralized collection pipe network / total amount of domestic sewage discharge × 100%”.

[0088] From the operation report of the local sewage treatment plant, the total treatment capacity of the local sewage treatment plant of the previous year is obtained, and at the same time, combined with the number of local urban population of the previous year and the per capita sewage generation amount, the total amount of sewage that needs to be treated in theory is estimated, and the sewage plant scale proportion of the previous year is obtained by the formula “sewage plant scale proportion = total treatment capacity of sewage treatment plant / total amount of sewage that needs to be treated in theory × 100%”.

[0089] Using geographic information system software, import the local city water system vector map, calculate the total length of the water system in the previous year through the software, and then obtain the total area of the local city in the previous year. Through the formula "water network density = total length of water system / total area of city", the water network density in the previous year is obtained.

[0090] The collection method of the water body area index, the number of sewage outlets, the ecological shoreline rate, and the water quality comprehensive index in the previous year is as follows:

[0091] By consulting maps and geographic information, the administrative region, geographic coordinates and other information of the black and odorous water body in the previous year are determined. The position coordinates of the black and odorous water body can also be measured on site with the aid of a global positioning system device. The area prone to black and odorous water is divided into three categories: urban-rural junction, old city area, and new city area. Different risk values are assigned according to different risks. The urban-rural junction is assigned a value of 3, the old city area is assigned a value of 2, and the new city area is assigned a value of 1. The lower the value, the lower the risk of water body blackening and odor.

[0092] Let the area of the urban-rural junction be , the area of the old city area be , and the area of the new city area be . The water body area index in the previous year is calculated as . The formula is as follows:

[0093] ;

[0094] Investigate the black and odorous water body on site, observe and record, and count the number of sewage outlets directly discharging sewage into the black and odorous water body. At the same time, reference can be made to the sewage outlet registration data of the environmental protection department to obtain the number of sewage outlets in the previous year.

[0095] Use high-resolution satellite remote sensing images or unmanned aerial vehicle aerial images, combined with GIS software, to extract the total length of the shoreline of the black and odorous water body in the previous year. Then, through field research, determine the length of the ecological shoreline in the previous year. Through the formula "ecological shoreline rate = ecological shoreline length / total length of black and odorous water body shoreline x 100%", the ecological shoreline rate in the previous year is obtained.

[0096] Set multiple sampling points at different positions of the black and odorous water body, collect water samples, and send them to a professional water quality testing laboratory for testing of chemical oxygen demand, ammonia nitrogen, dissolved oxygen and other water quality indicators. Compare the measured concentration values of each water quality indicator with the corresponding standard values to obtain the sub-index of each indicator. The formula is: , wherein is the sub-index of the th water quality indicator, is the measured concentration value of the th water quality indicator, is the standard value of the th water quality indicator. As the index of water quality index, the Nemerow index is calculated by using comprehensive water quality evaluation methods, such as the Nemerow index method. ,in, is the Nemerow index, is the maximum value among all sub-indices, The Nemerow index is the average value of each sub-indices, and the value of the Nemerow index is used as the comprehensive water quality index of the previous year.

[0097] Based on the above embodiment, the environmental parameters include river flow velocity, river flow rate and average monthly rainfall.

[0098] The above-mentioned "river water" refers to the water of rivers that are adjacent to black and odorous water bodies and have a direct impact on them. Taking cities along rivers as an example, large rivers such as the Yangtze River and the Yellow River, when cities are built along their main streams or larger tributaries, the water of these rivers is the "river water" of concern. Taking the cities in the Yangtze River Basin as an example, if a city has black and odorous water bodies close to the main stream of the Yangtze River, then the water of the Yangtze River is the "river water" in this method, and its flow rate, flow and other parameters will have an effect on the black and odorous water bodies.

[0099] River water affects black and odorous water bodies in various ways. The velocity and flow of river water are related to the degree of water exchange between black and odorous water bodies and river water. Fast velocity and large flow can accelerate the diffusion and dilution of pollutants in black and odorous water bodies, affecting their scale and scope of pollution. Conversely, insufficient exchange will cause pollutants to accumulate easily. The average monthly rainfall indirectly affects black and odorous water bodies by affecting the river water level and runoff. When the rainfall is heavy, the river water level rises and may flow back into the black and odorous water bodies, changing their water quality indicators such as salinity and dissolved oxygen.

[0100] The methods for collecting river flow velocity, river flow, and monthly average rainfall are as follows:

[0101] Obtain river flow velocity data for the corresponding time period from the local hydrological monitoring station.

[0102] The river flow data for the corresponding time period is obtained from the hydrological monitoring station. The hydrological station usually uses the velocity-area method to obtain the river flow by measuring the river flow velocity and the cross-sectional area of ​​the water.

[0103] Get the average monthly rainfall data for the 30 days before the current time from the local weather station.

[0104] On the basis of the above embodiment, after collecting the per capita domestic sewage treatment volume, the centralized collection rate of domestic sewage, the proportion of sewage treatment plant scale, the water network density, the water body area index, the number of sewage outlets, the ecological shoreline rate, and the comprehensive water quality index, these parameters are respectively subjected to maximum-minimum normalization processing, and then the normalized data is used for subsequent analysis and processing, so that in the subsequent analysis and processing process, various data are analyzed and processed under the same dimension, avoiding the problem of some data being ignored due to different dimensions.

[0105] S2. Perform edge detection on the fused image to obtain the water body boundary of the black and odorous water body, extract the characteristic parameters of the black and odorous water body based on the water body boundary, perform data fusion on the characteristic parameters and the environmental parameters of the black and odorous water body, and construct time series data;

[0106] On the basis of the above embodiment, edge detection is performed on the fused image to obtain the water body boundary of the black and odorous water body, and the specific steps are as follows:

[0107] The fused image is converted into a gray-scale image. For an RGB image, conversion is performed through the following formula:

[0108] ;

[0109] wherein, is the gray-scale value of a pixel, , , is the pixel value of the red channel, the green channel and the blue channel;

[0110] The Sobel operator is used to calculate the gradient in the horizontal direction and the vertical direction:

[0111] ;

[0112] ;

[0113] wherein, is the gradient in the horizontal direction, is the gradient in the vertical direction, is the fused image, denotes a convolution operation,

[0114] The gradient amplitude and direction are respectively:

[0115] ;

[0116] ;

[0117] wherein, is the gradient amplitude, is the gradient direction;

[0118] The gradient amplitude is compared with two threshold values to identify the edge pixels in the image, and the specific process is as follows:

[0119] When , the pixel is marked as a strong edge pixel;

[0120] When , the pixel is marked as a weak edge pixel;

[0121] When , the pixel is marked as a non-edge pixel;

[0122] wherein, is a high threshold value, is a low threshold value;

[0123] Starting from the strong edge pixels, a connectivity analysis is performed, a closed curve is found using a contour extraction algorithm, and finally the largest contour is preliminarily screened out through area calculation. The contour is the water body boundary of the black and odorous water body.

[0124] wherein, the connectivity analysis and the contour extraction algorithm are prior art and will not be described in detail here.

[0125] On the basis of the above embodiment, the characteristic parameters of the black and odorous water body include area, perimeter and circularity.

[0126] wherein, the collection method of the area, the perimeter and the circularity is as follows:

[0127] With the help of professional image analysis software, the image containing the black and odorous water body boundary is binarized, so that the black and odorous water body boundary region appears white and the background is black. The software automatically counts the number of white pixel points, converts the number of pixels into the actual area according to the image resolution, and obtains the area of the black and odorous water body.

[0128] After the edge detection algorithm identifies the edge pixels of the black and odorous water body boundary, the number of edge pixels is counted, and the distance between adjacent edge pixels is calculated according to the image resolution. The distance of all pixels is added to obtain the perimeter of the black and odorous water body.

[0129] After successfully extracting the area and perimeter data, the circularity of the black and odorous water body is obtained according to the formula "circularity = 4π×area ÷ perimeter squared".

[0130] On the basis of the above embodiment, after collecting the river water flow rate, the river water flow rate, the monthly average rainfall, the area, the perimeter and the circularity, the maximum-minimum normalization processing is performed on these parameters respectively, and then the normalized data is used for analysis and processing in the following text. In the analysis and processing process in the following text, various data are analyzed and processed under the same dimension, so as to avoid the problem that some data are ignored due to different dimensions.

[0131] S3. Constructing a black and odorous water body scale prediction model based on LSTM (Long Short Term Memory Network), taking the time series data of the black and odorous water body in the previous year in multiple historical years as input, and taking the characteristic parameters of the black and odorous water body in the next year in multiple historical years as label output, training the black and odorous water body scale prediction model;

[0132] On the basis of the above embodiments, when building the LSTM model, a deep learning framework needs to be selected, the LSTM model is composed of one or more LSTM layers and a fully connected layer, the LSTM layer can effectively learn the long-term dependence in the time series data and capture the trend of data change over time, and the fully connected layer generates the final prediction result according to the output result of the LSTM layer. In this process, the number of neurons of the LSTM layer needs to be determined, the number of neurons will affect the learning ability and complexity of the model; the time step is determined, that is, the number of consecutive time periods considered by the model when analyzing the data; the number of features of the input data is determined, that is, the number of feature parameters and environmental parameters of the black and odorous water body in the previous year in multiple historical years; the number of features of the output data is determined, that is, the number of feature parameters of the black and odorous water body in the next year in multiple historical years.

[0133] In order to enable the model to effectively learn the rules in the data, a suitable loss function and optimizer need to be selected, the commonly used loss functions are mean square error and mean absolute error, which are used to measure the difference between the model prediction result and the true label, and the optimizer is responsible for adjusting the parameters of the model to minimize the loss function, and the commonly used optimizers are Adam (adaptive moment estimation), SGD (stochastic gradient descent) and the like.

[0134] The trained model is trained using the training set, in the training process, appropriate training rounds and batch size need to be set, the training rounds determine the number of times the model learns the training data, and the batch size determines the number of data samples used each time; at the same time, the performance of the model is monitored using the validation set, and the hyperparameters are adjusted in time to prevent the model from overfitting.

[0135] S4. Inputting the time series data of the black and odorous water body of the current year into the black and odorous water body scale prediction model to obtain the predicted value of the feature parameter of the black and odorous water body;

[0136] S5. Based on the predicted value of the feature parameter of the black and odorous water body, a scale prediction index for comprehensively evaluating the scale of the black and odorous water body is obtained, based on the urban index parameter of the previous year, a purification efficiency index for comprehensively evaluating the sewage treatment and self-purification ability of the city is obtained, and based on the water body index parameter of the previous year, an environmental quality index for comprehensively evaluating the pollution degree of the water body is obtained;

[0137] On the basis of the above embodiments, based on the predicted value of the feature parameter of the black and odorous water body, a scale prediction index for comprehensively evaluating the scale of the black and odorous water body is obtained, and the formula is as follows:

[0138] ;

[0139] wherein, The scale prediction index is used to combine the area prediction value, the perimeter prediction value and the circularity prediction value to comprehensively evaluate the scale of the black and odorous water body, and the larger the scale index is, the larger the scale of the black and odorous water body is.

[0140] In the formula, The area prediction value of the black and odorous water body is The circularity prediction value of the black and odorous water body is The perimeter prediction value of the black and odorous water body is

[0141] On this basis, it should be noted that the area size indicates the space occupied by the black and odorous water body in the plane, and when the area prediction value increases, the area is larger, the range covered by the water body is wider, and the scale of the black and odorous water body is larger, so that the predicted scale index increases; the circularity describes the regularity of the shape of the black and odorous water body, and when the circularity prediction value increases, the shape is closer to a circle, and the circularity prediction value tends to 1, at this time, under the same perimeter, the water body has the largest area, and in a natural state, the more compact the shape of the water body is, the larger the scale of the black and odorous water body is, and the predicted scale index increases; the perimeter reflects the length of the boundary of the black and odorous water body, and when the perimeter prediction value increases, it means that the boundary of the water body becomes longer, and when the area and the circularity are unchanged, the longer perimeter indicates that the shape of the water body is more dispersed, the water body area corresponding to the unit perimeter decreases, the scale of the black and odorous water body decreases, and the predicted scale index decreases.

[0142] Therefore, the predicted scale index is positively correlated with the area prediction value and the circularity prediction value , and the scale index is negatively correlated with the perimeter prediction value .

[0143] In the formula, The multiplication form can reflect the common influence of the two factors on the scale prediction index. When the area prediction value increases, if the circularity is also high, the multiplication of the two will make the scale index increase more greatly, which conforms to the actual situation. Because a black and odorous water body with a large area and a relatively regular shape (high circularity) is obviously larger than a black and odorous water body with the same area but a very irregular shape (low circularity).

[0144] Conversely, if the area prediction value remains unchanged, the circularity changes, and the scale index can also be adjusted accordingly through the multiplication operation, reflecting the role of the shape factor on the scale assessment of black and odorous water bodies. For example, when the area is the same, the water body with high circularity may be more concentrated and have more scale in visual and actual impact range, while the water body with low circularity may be more dispersed, and its contribution to the overall scale is relatively small. The multiplication method can better reflect this difference.

[0145] Therefore, the above weighted sum formula is used to represent the predicted scale index and the functional relationship between the area prediction value , the circularity prediction value , and the perimeter prediction value .

[0146] On the basis of the above embodiment, based on the urban index parameters of the previous year, the purification efficiency index for comprehensively evaluating the urban sewage treatment and self-purification capacity is obtained, and the formula is as follows:

[0147] ;

[0148] Among them, is the purification efficiency index of the previous year, and the purification efficiency index is used to combine per capita domestic sewage treatment capacity, domestic sewage centralized collection rate, sewage plant scale proportion and water network density four indexes to comprehensively evaluate the urban sewage treatment and self-purification capacity, and the larger the purification efficiency index, the higher the urban sewage treatment and self-purification capacity;

[0149] In the formula, is the per capita domestic sewage treatment capacity of the previous year, is the domestic sewage centralized collection rate of the previous year, is the sewage plant scale proportion of the previous year, is the water network density of the previous year;

[0150] On this basis, it should be noted that the per capita domestic sewage treatment capacity reflects the city's treatment capacity for sewage generated by residents on a daily basis. When the per capita domestic sewage treatment capacity increases, it means that the city has increased investment in sewage treatment facilities and optimized the process, which can handle more sewage generated by residents, more sewage is properly treated, reducing the direct discharge of sewage into natural water bodies, effectively preventing the deterioration of the water quality of the surrounding water bodies due to pollution, and the urban sewage treatment and self-purification capacity is naturally enhanced, so that the purification efficiency index increases;

[0151] When the domestic sewage centralized collection rate increases, meaning that the city can more comprehensively collect domestic sewage generated by residents and enterprises. In the past, part of the sewage may be directly discharged into the surrounding environment due to imperfect collection systems, causing damage to the ecology. Now, with the increase in collection rate, more sewage enters the formal treatment system, reducing the risk of environmental pollution by sewage and creating conditions for subsequent centralized treatment by sewage treatment plants, significantly improving the city's sewage treatment and self-purification capacity, and increasing the purification efficiency index ;

[0152] When the sewage plant scale ratio is larger, it indicates that the city's sewage treatment facility scale is larger and the treatment capacity is stronger. Large-scale sewage treatment plants can use more advanced treatment technologies for deep treatment of sewage, not only removing conventional pollutants but also treating some difficult-to-degrade harmful substances to ensure that the discharged water quality meets standards. At the same time, the scale effect reduces the unit sewage treatment cost and improves the treatment efficiency, greatly enhancing the city's overall sewage treatment and self-purification capacity, and increasing the purification efficiency index ;

[0153] When the water network density is larger, the flow and exchange of water bodies are better, which can dilute pollutants and accelerate the decomposition and transformation of pollutants. In addition, aquatic plants and microorganisms in the water network can absorb and degrade pollutants, further purifying water quality. Rich water networks can also improve the city's ecological environment and regulate local climate, providing more favorable ecological conditions for urban sewage treatment and self-purification, thereby enhancing the city's overall sewage treatment and self-purification capacity and increasing the purification efficiency index .

[0154] Therefore, the purification efficiency index is positively correlated with per capita domestic sewage treatment capacity , domestic sewage centralized collection rate , sewage plant scale ratio , and water network density .

[0155] In addition, the four indicators above have the same influence direction on the city's sewage treatment and self-purification capacity, which is that the larger the value, the higher the city's sewage treatment and self-purification capacity. Linear addition can accumulate the positive effects of these factors and comprehensively reflect the comprehensive improvement effect of these factors on purification efficiency. For example, the increase in per capita domestic sewage treatment capacity means that the amount of sewage treated per person increases, the rise in domestic sewage centralized collection rate allows more sewage to be centrally treated, the increase in sewage plant scale ratio enhances sewage treatment capacity, and the increase in water network density is conducive to the natural purification and circulation of water bodies. Linear addition of these factors can reasonably reflect the positive influence of the combined action of these factors on the city's sewage treatment and self-purification capacity.

[0156] In summary, the above-mentioned weighted summation formula is used to represent the purification efficiency index and per capita domestic sewage treatment capacity , domestic sewage centralized collection rate , sewage plant scale proportion , water network density between the functional relationship.

[0157] In the formula, is the weight coefficient of per capita domestic sewage treatment capacity, is the weight coefficient of domestic sewage centralized collection rate, is the weight coefficient of sewage plant scale proportion, is the weight coefficient of water network density;

[0158] In the case of large and stable urban domestic sewage discharge, sewage plants can concentrate and efficiently treat sewage with professional treatment processes and equipment, ensuring that the effluent quality meets the standards. Compared with the self-purification of water network, which is restricted by natural factors such as season and climate, the treatment capacity of sewage plants is more stable and reliable. For example, in cities with developed industry, a large amount of industrial wastewater and domestic sewage is generated, and the expansion of sewage plant scale and the upgrading of treatment technology play a decisive role in maintaining urban water environmental quality, so the weight is the largest.

[0159] Even though sewage plants bear the main treatment task, water network, as an important part of the urban ecosystem, still has an important self-purification capacity, and aquatic plants and microorganisms in the water network can absorb and degrade pollutants in sewage, playing an auxiliary purification role. In cities with good ecological environment and dense water network, the self-purification function of water network helps to improve the local water environment and enhance the overall ecological self-purification capacity of the city, so the weight .

[0160] In some cities, the sewage collection pipe network has been relatively complete, and the domestic sewage centralized collection rate is at a high level. At this time, it is difficult to further improve the collection rate, and the space for improving the overall sewage treatment and self-purification capacity is limited. In contrast, the expansion of sewage plant scale and the development of water network self-purification capacity have a more significant effect on improving urban water environmental quality, so the weight is relatively low, so .

[0161] The per capita domestic sewage treatment capacity is affected by sewage collection rate, sewage treatment plant capacity and other factors. In the case of incomplete sewage collection or insufficient sewage treatment plant capacity, the per capita domestic sewage treatment capacity is difficult to be effectively improved. Moreover, simply increasing the per capita treatment capacity cannot fundamentally solve the problem of urban sewage treatment and self-purification capacity, and the entire treatment system needs to be coordinated, so the weight of the per capita treatment capacity is the smallest, so .

[0162] Based on the above, let .

[0163] As an embodiment, the value range of a is 0.1-0.2, the value range of b is 0.2-0.25, the value range of c is 0.35-0.4, the value range of d is 0.25-0.35, and the specific value is set by the technical personnel according to the actual situation, which is not limited here.

[0164] On the basis of the above embodiment, based on the water body index parameters of the last year, the environmental quality index for comprehensively evaluating the water pollution degree is obtained, and the formula is as follows:

[0165] ;

[0166] Among them, is the environmental quality index of the last year, which is used to combine the water area index, the number of sewage outlets, the ecological shoreline rate and the water quality comprehensive index to comprehensively evaluate the water pollution degree, and the greater the environmental quality index, the higher the water pollution degree;

[0167] In the formula, is the water area index of the last year, is the number of sewage outlets of the last year, is the ecological shoreline rate of the last year, is the water quality comprehensive index of the last year;

[0168] On this basis, it should be noted that the water area index usually represents the risk of black and odorous water. When the water area index increases, it means that the risk of black and odorous water increases, the water pollution capacity may decrease, or the potential pollution source in the region increases, the pollution accumulation increases, which makes the water more easily polluted, and the water pollution degree significantly improves. Since the environmental quality index is an index for comprehensively evaluating the water pollution degree, and the greater the index, the higher the water pollution degree, when the actual water pollution degree intensifies, the environmental quality index​ The sewage outlet is the direct way for pollutants to enter the water body. When the pollution source entering the water body increases, the pollution source also increases accordingly. The sewage continuously discharged from each sewage outlet contains various pollutants. These substances continue to accumulate in the water body, destroying the original ecological balance of the water body, causing the degree of water pollution to continue to increase, thus the environmental quality index Increase;

[0169] Ecological shorelines have a natural purification effect on water bodies. Aquatic plants growing on the shorelines can absorb nutrients such as nitrogen and phosphorus in the water, and microorganisms can decompose organic pollutants. When the ecological shoreline rate is When the water quality index increases, the self-purification capacity of the water body is enhanced, which can effectively degrade and remove pollutants, reduce the concentration of pollutants in the water body, thereby reducing the degree of water pollution and making the environmental quality index reduce;

[0170] The comprehensive water quality index is an intuitive reflection of the current pollution status of the water body. It quantifies the degree of water pollution by integrating multiple water quality indicators. When it increases, it means that the concentration of pollutants in the water increases, and problems such as eutrophication and reduced dissolved oxygen occur, which leads to deterioration of water quality and aggravation of water pollution, thus making the Environmental Quality Index Increase.

[0171] Therefore, the environmental quality index and water body regional index , Number of sewage outlets , comprehensive water quality index are positively correlated, and the environmental quality index and ecological shoreline rate is negatively correlated.

[0172] in, The comprehensive water quality index reflects the potential pollution risk based on the regional characteristics of the water body and the number of sewage outlets. It is a direct measure of the actual current pollution status of a water body. The potential pollution risk needs to be reflected in the actual water quality to reflect its ultimate impact on the degree of water pollution. The two are interrelated and work together. Multiplication can reflect this correlation. That is, when the potential pollution risk is high and the actual water quality is poor, the degree of water pollution will increase significantly. If the potential pollution risk is high but the actual water quality is good, the multiplication result will be relatively small, reflecting that the current degree of water pollution has not reached the level of potential risk, which is consistent with the actual situation.

[0173] In summary, the above weighted summation formula is used to represent the environmental quality index and water body regional index the number of pollution outlets the ecological shoreline rate the water quality comprehensive index a function relationship between them.

[0174] In the formula, is the weight coefficient of the water body area index, is the weight coefficient of the number of pollution outlets, is the weight coefficient of the ecological shoreline rate;

[0175] In the industrial concentration area, although the pollution outlet is the direct channel for pollutants to enter the water body, the environmental protection department strictly controls the number and discharge behavior of the pollution outlet, so that the growth and disorderly discharge of the pollution outlet are curbed. In contrast, the high-intensity pollution brought by the region itself makes the water body area have a greater impact on water pollution, resulting in the contribution of the number of pollution outlets to water pollution being inferior to the overall pollution risk of the region, so .

[0176] The large-scale development and construction of the industrial concentration area often accompanies the destruction of the natural ecological environment, and the ecological shoreline is the first to be affected. The reclamation and the construction of wharfs and other projects greatly reduce the area of the ecological shoreline, destroying the natural structure and function of the ecological shoreline. The rich aquatic plants and microbial communities originally owned by the ecological shoreline are sharply reduced, resulting in a significant decline in its self-purification capacity for the water body. Even though the ecological shoreline still has a certain purification effect, its influence in the entire pollution control system is far inferior to the regional pollution risk and the number of pollution outlets, resulting in .

[0177] In summary, on the basis of , let .

[0178] As an embodiment, the value range of is 0.5-0.7, the value range of is 0.2-0.3, the value range of is 0.1-0.2, and the specific value is set by the technical personnel according to the actual situation, which is not limited herein.

[0179] S6. Based on the scale prediction index, the environmental quality index of the previous year, and the purification efficiency index, obtain a risk assessment index for assessing the risk degree of the black and odorous water body. Compare the risk assessment index with a pre-set threshold value, and according to the comparison result, divide the risk level of the black and odorous water body.

[0180] On the basis of the above embodiment, based on the scale prediction index, the environmental quality index of the previous year, and the purification efficiency index, obtain a risk assessment index, and the formula is as follows:

[0181] ;

[0182] in, The risk assessment index is used to comprehensively assess the risk level of black and odorous water bodies by combining the scale prediction index, the environmental quality index of the previous year, and the purification efficiency index. The larger the risk assessment index, the higher the risk level of black and odorous water bodies.

[0183] On this basis, it should be noted that:

[0184] The scale prediction index represents the coverage of black and odorous water bodies. When the number of black and odorous water bodies increases, it means that the scale of black and odorous water bodies will expand and the total amount of pollutants contained in them will increase, which will not only aggravate the damage to the surrounding ecological environment, but also increase the difficulty of governance, thereby increasing the risk level of black and odorous water bodies and making the risk assessment index Increase; Environmental Quality Index reflects the degree of water pollution;

[0185] When the environmental quality index When it increases, it indicates that the concentration of pollutants in the water body increases and the water quality deteriorates, which means that the risk of black and odorous water bodies increases, making the risk assessment index Increase;

[0186] The purification efficiency index measures the ability of water to remove pollutants. When the water purification capacity increases, it can effectively reduce the content of pollutants in the water, reduce the pollution level of black and odorous water bodies, and thus reduce the risk level of black and odorous water bodies, making the risk assessment index reduce.

[0187] Therefore, the risk assessment index and scale forecast index , Environmental Quality Index are all positively correlated, and the risk assessment index and purification efficiency index Is negatively correlated.

[0188] Since the scale prediction index , Environmental Quality Index The impact on the risk level of black and odorous water bodies is not isolated, but synergistic. When the scale of black and odorous water bodies expands, that is, When the water volume increases, the surface area and volume of the water body increase, which not only accommodates more pollutants, but also provides more space for the diffusion and accumulation of pollutants. Also elevated, indicating that the degree of water pollution intensified, more pollutants into the water, the interaction of large-scale, greatly enhance the risk of black and odorous water, by multiplying can directly reflect the synergistic effect, compared with the simple addition operation, multiplication can more prominent two common effect on the risk of amplification effect.

[0189] In summary, therefore, the weighted sum formula above represents the risk assessment index and scale prediction index , environmental quality index , purification efficiency index between the functional relationship.

[0190] In the formula, , , respectively, the scale prediction index, environmental quality index, purification efficiency index weight coefficient, and , , The specific value is determined by the analytic hierarchy process, the specific logic as follows:

[0191] The scale prediction index, environmental quality index, purification efficiency index 3 indexes are marked, the relative importance of the values between each other is determined by the nine scale method, the judgment matrix is constructed, wherein, the index of scale prediction index is marked as 1, the index of environmental quality index is marked as 2, the index of purification efficiency index is marked as 3, the judgment matrix constructed is:

[0192] ;

[0193] Among them, , Indicate the index of the index, and , , the index of the index relative to the index of the index, the importance of the risk assessment index, The specific value is determined by the relevant experts using 1-9 scoring method, Indicates that the index of the index is more important than the index of the index, Indicates that the index of the index is very important than the index of the index; Indicates that the index of the index is very important than the index of the index;

[0194] ​​Divide each element value in the judgment matrix by the sum of its columns to obtain a normalized judgment matrix. Calculate the mean of the element values ​​in each row of the normalized judgment matrix, and use the mean of the element values ​​in the first row as the proportional coefficient of the scale prediction index, the mean of the element values ​​in the second row as the proportional coefficient of the environmental quality index, and the mean of the element values ​​in the third row as the proportional coefficient of the purification efficiency index. With the constraint that the sum of the scaled values ​​is equal to 1, scale the three proportional coefficients equally, and use the scaled values ​​as the weights of the corresponding indexes.

[0195] Based on the above embodiment, the risk assessment index is compared with the preset threshold value to classify the black and odorous water risk level. The specific process is as follows:

[0196] When the risk assessment index When , the predicted black and odorous water body is at a low risk level;

[0197] When the risk assessment index When , the predicted black and odorous water body is at a medium risk level;

[0198] When the risk assessment index When , the predicted black and odorous water bodies are at a high risk level;

[0199] in, To divide the critical value between low risk and medium risk, The critical value for dividing medium risk and high risk.

[0200] 、 Setting method: Determine the risk assessment index by statistically analyzing a large amount of black and odorous water monitoring data The distribution of The threshold value can be set based on the statistical indicators such as the maximum value, minimum value, mean value and standard deviation of is set to a certain standard deviation below the mean, and It can be set to a certain standard deviation higher than the mean to define the low, medium and high risk intervals in the black and odorous water risk assessment, and to achieve scientific judgment of different risk levels.

[0201] See also Figure 2 , the present invention also provides a technical solution:

[0202] A system for classifying black and odorous water bodies in riverside cities, the system being used to implement any of the above-mentioned methods for classifying black and odorous water bodies in riverside cities, comprising:

[0203] The data acquisition module is configured to acquire urban and water body index parameters of a previous year, and synchronously and continuously acquire environmental parameters, high-altitude images, visible light images and multispectral images of the urban and water bodies of multiple historical years and a current year, pre-process the high-altitude images, the visible light images and the multispectral images, and obtain a fusion image;

[0204] The time series data construction module is configured to perform edge detection on the fusion image, acquire a water body boundary of the black and odorous water body, extract a feature parameter of the black and odorous water body based on the water body boundary, perform data fusion on the feature parameter of the black and odorous water body and the environmental parameter, and construct time series data.

[0205] The scale prediction model construction module is configured to construct a black and odorous water body scale prediction model based on a long short-term memory network, take time series data of a black and odorous water body of a previous year in multiple historical years as input, take a feature parameter of a black and odorous water body of a next year in the multiple historical years as label output, and train the black and odorous water body scale prediction model.

[0206] The simulation module is configured to input time series data of a black and odorous water body of a current year into the black and odorous water body scale prediction model, and obtain a predicted value of the feature parameter of the black and odorous water body.

[0207] The data processing module is configured to obtain a scale prediction index for comprehensively evaluating a scale of the black and odorous water body based on the predicted value of the feature parameter of the black and odorous water body, obtain a purification efficiency index for comprehensively evaluating a sewage treatment and self-purification capability of the city based on the urban index parameter of the previous year, and obtain an environmental quality index for comprehensively evaluating a pollution degree of the water body based on the water body index parameter of the previous year.

[0208] The grade division module is configured to obtain a risk evaluation index for evaluating a risk degree of the black and odorous water body based on the scale prediction index, the environmental quality index and the purification efficiency index of the previous year, compare the risk evaluation index with a preset threshold value, and divide a risk grade of the black and odorous water body according to a comparison result.

[0209] The above formulas are all dimensionless values calculated, the formulas are obtained by software simulation of a large amount of data to obtain a formula of a recent real situation, and the preset parameters in the formulas are set by a person skilled in the art according to an actual situation.

[0210] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of the examples described in connection with the embodiments disclosed herein can be realized by electronic hardware, or a combination of computer software and electronic hardware. Whether the functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solutions.

[0211] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, and can be located in one place or distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiments according to actual needs.

[0212] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application.

Claims

1. A method for risk classification of black and odorous water bodies in a river city, characterized in that: The specific steps include: S1. Collecting the city and water body index parameters of the city along the river in the previous year, and synchronously collecting the environmental parameters, high-altitude images, visible light images and multispectral images of the city and water body in multiple historical years and the current year of the city along the river, preprocessing the high-altitude images, visible light images and multispectral images to obtain a fusion image; S2. Edge detection is performed on the fusion image to obtain the water body boundary of the black and odorous water body of the city along the river, and the characteristic parameters of the black and odorous water body of the city along the river are extracted based on the water body boundary, the characteristic parameters of the black and odorous water body of the city along the river are data fused with the environmental parameters, and the time series data of the black and odorous water body of the city along the river is constructed; S3. A scale prediction model of the black and odorous water body of the city along the river is constructed based on a long short-term memory network, the time series data of the black and odorous water body of the city along the river in the previous year in multiple historical years is taken as input, the characteristic parameters of the black and odorous water body of the city along the river in the next year in multiple historical years are taken as label output, and the scale prediction model of the black and odorous water body of the city along the river is trained; S4. The time series data of the black and odorous water body of the city along the river in the current year is input into the black and odorous water body scale prediction model to obtain the predicted value of the characteristic parameters of the black and odorous water body of the city along the river; S5. Based on the predicted value of the characteristic parameters of the black and odorous water body of the city along the river, a scale prediction index for comprehensively evaluating the scale of the black and odorous water body of the city along the river is obtained, based on the city index parameters of the city along the river in the previous year, a purification efficiency index for comprehensively evaluating the sewage treatment and self-purification capacity of the city along the river is obtained, and based on the water body index parameters of the city along the river in the previous year, an environmental quality index for comprehensively evaluating the pollution degree of the water body of the city along the river is obtained; S6. Based on the scale prediction index of the black and odorous water body of the city along the river, the environmental quality index and the purification efficiency index of the city along the river in the previous year, a risk assessment index for evaluating the risk degree of the black and odorous water body of the city along the river is obtained, the risk assessment index is compared with a pre-set threshold value, and according to the comparison result, the risk level of the black and odorous water body of the city along the river is divided.

2. The method for classifying black and odorous water bodies along the river according to claim 1 is characterized by: The city index parameters include per capita domestic sewage treatment capacity, domestic sewage centralized collection rate, sewage plant scale proportion and water network density, the water body index parameters include water body area index, number of sewage outlets, ecological shoreline rate and water quality comprehensive index, the environmental parameters include river water flow rate, river water flow and monthly average rainfall, and the characteristic parameters of the black and odorous water body include area, circumference and circularity.

3. The method for classifying black and odorous water bodies along the river according to claim 1 is characterized by: Edge detection is performed on the fusion image to obtain the water body boundary of the black and odorous water body of the city along the river, and the specific steps are as follows: The fusion image is converted into a gray image, and for an RGB image, the conversion is performed through the following formula: ； wherein, is a gray value of a pixel, , , is a pixel value of a red channel, a green channel, a blue channel. The Sobel operator is used to calculate the gradient in the horizontal and vertical directions: ； ； wherein is a gradient in the horizontal direction, is a gradient in the vertical direction, is a fused image, denotes a convolution operation; The gradient amplitude and direction are respectively: ； ； wherein is the gradient magnitude, is the gradient direction; The gradient magnitude and two threshold values, which identifies the edge pixels in the image, in particular as follows: When , the pixel is marked as a strong edge pixel; When , the pixel is marked as a weak edge pixel; When , the pixel is marked as a non-edge pixel; wherein is a high threshold value, is a low threshold value; Starting from the strong edge pixels, connectivity analysis is performed, then a closed curve is found by using a contour extraction algorithm, and finally the largest area contour is preliminarily screened out by area calculation, which is the water body boundary of the black and odorous water body of the city along the river.

4. The method for classifying black and odorous water bodies along the river according to claim 2 is characterized by: Based on the predicted value of the characteristic parameters of the black and odorous water body of the city along the river, a scale prediction index for comprehensively evaluating the scale of the black and odorous water body of the city along the river is obtained, and the formula is as follows: ； wherein, is a scale prediction index, the scale prediction index is used to combine the area prediction value, the circumference prediction value and the circularity prediction value to comprehensively evaluate the scale of the black and odorous water body in the city along the river; In the formula, is a predicted value of the area of the black and odorous water body, is a predicted value of the circularity of the black and odorous water body, is a predicted value of the circumference of the black and odorous water body.

5. The method according to claim 2, wherein the method is characterized by: Based on the urban index parameters of the cities along the river in the last year, the purification efficiency index in the last year is obtained, and the formula is as follows: ； wherein, is the purification efficiency index of the previous year, which is used to combine the per capita domestic sewage treatment capacity, the centralized collection rate of domestic sewage, the proportion of sewage plant scale, and the water network density to comprehensively evaluate the sewage treatment and self-purification capacity of cities along the river. In the formula, is the per capita quantity of domestic sewage treatment in the previous year, is the centralized collection rate of domestic sewage in the previous year, is the proportion of sewage plant scale in the previous year, is the water network density in the previous year; In the formula, is the weight coefficient of per capita domestic sewage treatment capacity, is the weight coefficient of domestic sewage centralized collection rate, is the weight coefficient of sewage plant scale proportion, is the weight coefficient of water network density, and is the weight coefficient of water network density, and .

6. The method according to claim 2, wherein the method is characterized by: Based on the water body index parameters of the cities along the river in the last year, the environmental quality index in the last year is obtained, and the formula is as follows: ； wherein, is the environmental quality index of the previous year, which is used to comprehensively evaluate the degree of water pollution in combination with the water body area index, the number of sewage outlets, the ecological shoreline rate, and the water quality comprehensive index. In the formula, is the water area index of the previous year, is the number of sewage outlets of the previous year, is the ecological shoreline rate of the previous year, is the water quality comprehensive index of the previous year; In the formula, is a weight coefficient of the water body area index, is a weight coefficient of the number of sewage outlets, is a weight coefficient of the ecological shoreline rate, and is a weight coefficient of the water quality index, .

7. The method according to claim 2, wherein the method is characterized by: Based on the scale prediction index of the black and odorous water body of the city along the river, the environmental quality index and the purification efficiency index of the city along the river in the last year, the risk assessment index is obtained, and the formula is as follows: ； wherein, is the risk assessment index, which is used to combine the scale prediction index, the environmental quality index of the previous year, and the purification efficiency index to comprehensively assess the risk degree of the black and odorous water body in the city along the river; In the formula, is a scale prediction index, is an environmental quality index of the previous year, is a purification efficiency index of the previous year; , , are weight coefficients of scale prediction index, environmental quality index and purification efficiency index respectively, and , , The specific values of are determined by the analytic hierarchy process.

8. The risk classification method of the black and odorous water body in the city along the river according to claim 7, characterized in that: The risk assessment index is compared with the threshold value set in advance, and the risk level of the black and odorous water body of the city along the river is divided, and the specific process is as follows: When the risk assessment index is less than 0.5, the predicted black and odorous water body in the city along the river is at a low risk level; When the risk assessment index is greater than 0.5 and less than 0.7, it is predicted that the black and odorous water body in the city along the river is at a high risk level; When the risk assessment index is greater than 0.5, the predicted black and odorous water body in the city along the river is at a high risk level; wherein, is the critical value to divide low risk from medium risk, is the critical value to divide medium risk from high risk.

9. A system for classifying the risk of black and odorous water bodies in a city along a river, the system being used to perform the method for classifying the risk of black and odorous water bodies in a city along a river according to any one of claims 1-8, characterized in that: It includes: The data acquisition module is used to collect the urban and water body index parameters of the cities along the river in the last year, and synchronously collect the environmental parameters, high-altitude image, visible light image and multispectral image of the cities along the river in multiple historical years and the current year, and preprocess the high-altitude image, visible light image and multispectral image to obtain the fusion image; The time series data construction module is used to perform edge detection on the fusion image to obtain the water body boundary of the black and odorous water body of the city along the river, extract the feature parameters of the black and odorous water body of the city along the river based on the water body boundary, and fuse the feature parameters of the black and odorous water body of the city along the river with the environmental parameters to construct the time series data of the black and odorous water body of the city along the river; The scale prediction model construction module is used to construct the scale prediction model of the black and odorous water body of the city along the river based on the long short-term memory network, take the time series data of the black and odorous water body of the city along the river in the previous year in multiple historical years as input, and take the feature parameters of the black and odorous water body of the city along the river in the next year in multiple historical years as label output, and train the scale prediction model of the black and odorous water body of the city along the river; The simulation module is used to input the time series data of the black and odorous water body of the city along the river in the current year into the black and odorous water body scale prediction model to obtain the predicted value of the feature parameters of the black and odorous water body of the city along the river; The data processing module is used to obtain the scale prediction index for comprehensively evaluating the scale of the black and odorous water body of the city along the river based on the predicted value of the feature parameters of the black and odorous water body of the city along the river, obtain the purification efficiency index for comprehensively evaluating the sewage treatment and self-cleaning capacity of the city along the river based on the urban index parameters of the city along the river in the last year, and obtain the environmental quality index for comprehensively evaluating the pollution degree of the water body of the city along the river based on the water body index parameters of the city along the river in the last year; The grade division module is used to obtain the risk assessment index for evaluating the risk degree of the black and odorous water body of the city along the river based on the scale prediction index of the black and odorous water body of the city along the river, the environmental quality index and the purification efficiency index of the city along the river in the last year, and compare the risk assessment index with the threshold value set in advance, and divide the risk level of the black and odorous water body of the city along the river according to the comparison result.

Images