Surface water body trail judgment method and pollution identification method based on flow field matrix expansion

Abstract

The present invention discloses a method for judging the trailing of surface water bodies based on the expansion of flow field matrix, comprising the following steps: obtaining water body image data; performing grayscale transformation on the water body image data, calculating the edge curve of pollutants on the water body surface, and obtaining the speed of movement of the center of gravity of pollutants; taking the speed of movement of the center of gravity of pollutants as a restriction condition, based on the velocity field matrix correction transformation, obtaining the expanded water body surface flow field velocity matrix; based on the expanded water body surface flow field velocity matrix, selecting the judgment basis from the water body image data, performing trailing judgment on the water body pollutants, and obtaining the trailing judgment result. The present invention also provides a method for identifying water body pollution including the method for judging the trailing of surface water bodies based on the expansion of flow field matrix. The method of the present invention can detect the wastewater discharge treatment device generated by non-ferrous metal smelting in real time, thereby improving the monitoring accuracy of sudden water pollution.

Description

Technical Field

[0001] The invention belongs to the field of rapid detection of nonferrous metal smelting water treatment, and specifically relates to a surface water body trailing image judgment method and a pollution identification method based on flow field matrix expansion. Background Art

[0002] Nonferrous metals refer to all metals except iron, manganese, chromium and iron-based alloys, which can be divided into heavy metals, light metals, precious metals and rare metals. Nonferrous metals in a broad sense also include nonferrous alloys, which are alloys composed of nonferrous metals as the matrix and one or more other elements. Nonferrous metals are the basic materials for the development of the national economy. Most industries such as aviation, aerospace, automobiles, machinery manufacturing, electricity, communications, construction, and home appliances are based on nonferrous metals. With the rapid development of modern chemical industry, agriculture, and science and technology, nonferrous metals are becoming more and more important in human development.

[0003] Non-ferrous metal smelting can be roughly divided into full pyrometallurgy, full hydrometallurgy and hydrometallurgy combined pyrometallurgy. The non-ferrous metal smelting process will produce a large amount of wastewater, which needs to be treated to meet the emission standards before it can be discharged. However, long-term high-intensity production will inevitably lead to leakage failures in wastewater pipelines, valves and other devices. Wastewater leakage will cause great harm to the environment, so it is necessary to monitor the pollutants at the sewage outlet of non-ferrous metal smelting wastewater in real time to ensure that leakage does not occur. Previously, manual inspections of equipment were generally used, which not only consumed a certain amount of manpower, but also could not detect in real time whether a leak occurred. There are also methods for real-time monitoring of sewage through image recognition technology, but due to problems such as image smearing, the recognition accuracy is not high, and there are problems of false alarms and missed alarms. Summary of the invention

[0004] One of the purposes of the present invention is to provide a method for determining the trailing image of surface water bodies based on flow field matrix expansion for high-precision real-time monitoring.

[0005] The second object of the present invention is to provide a water pollution identification method including the surface water body trailing judgment method based on flow field matrix expansion.

[0006] The present invention provides a method for judging the trailing image of a surface water body based on flow field matrix expansion, comprising the following steps:

[0007] S1. Acquire water body image data;

[0008] S2. grayscale transformation of the water body image data obtained in step S1, calculating the edge curve of pollutants on the water body surface based on the grayscale transformation result, and obtaining the moving speed of the center of gravity of the pollutants through the edge curve;

[0009] S3. Taking the moving speed of the center of gravity of the pollutant obtained in step S2 as the constraint condition, based on the velocity field matrix correction transformation, the expanded water surface flow field velocity matrix is ​​obtained;

[0010] S4. Based on the expanded water body surface flow field velocity matrix obtained in step S3, a judgment basis is selected from the water body image data obtained in step S1, and a smear judgment is performed on the water body pollutants to obtain a smear judgment result.

[0011] In step S1, the water body image data is captured at the sewage outlet using an industrial camera and divided into grids.

[0012] Step S2 includes the following steps:

[0013] Performing grayscale transformation on the water body image data obtained in step S1;

[0014] Based on the second-order partial derivatives of the grayscale transformation results in the x-direction and the y-direction, the edge curve of the pollutant on the water surface is calculated;

[0015] Design pixel variance correction factor based on water image saturation, and design pixel exposure correction factor based on ambient brightness;

[0016] Based on the pixel variance correction factor and the pixel exposure correction coefficient, the edge curve of the pollutant on the water surface is corrected to obtain the corrected edge curve;

[0017] According to the corrected edge curve, the moving speed of the center of gravity of the pollutant is calculated.

[0018] Step S2 specifically includes the following steps:

[0019] The water body image data obtained in step S1 is subjected to grayscale transformation and is calculated using the following formula: in, is the blue component of the pixel with the horizontal coordinate x and the vertical coordinate y in the Nth image of the water body image data; The green component of the pixel point with the horizontal coordinate x and the vertical coordinate y in the water body image data; The red component of the pixel point with the horizontal coordinate x and the vertical coordinate y in the water body image data; is the result of grayscale transformation of the Nth image in the water body image data;

[0020] Based on the second-order partial derivatives of the grayscale transformation results in the x and y directions, the edge curve of the pollutant on the water surface is calculated, and the pollutant contour line is obtained by solving the following formula: in, is the sum of the second-order partial derivatives of the image in the x-direction and the y-direction; is the gray value of the image;

[0021] The pixel variance correction factor is designed based on the saturation of the water image and calculated using the following formula: in, is the saturation of the Nth image in the water body image data;

[0022] Design the pixel exposure correction factor based on the ambient brightness and calculate it using the following formula: in, is the ambient brightness parameter;

[0023] Based on the pixel variance correction factor and the pixel exposure correction coefficient, the edge curve of the pollutant on the water surface is corrected to obtain the corrected edge curve, which is expressed by the following formula: in, is the corrected edge curve; is the pixel variance correction factor; is the pixel exposure correction factor.

[0024] According to the corrected edge curve, the moving speed of the center of gravity of the pollutant is calculated by the following formula: in, is the velocity of the water flow in the x direction at the center of gravity of the pollutant; is the velocity of the water flow in the y direction at the center of gravity of the pollutant; a is the corrected edge curve The projection line segment in the x direction; b is the corrected edge curve The projection line segment in the y direction; t is the time difference of the camera taking pictures when sampling the water body image data.

[0025] Step S3 includes the following steps:

[0026] Using the moving speed of the center of gravity of the pollutant obtained in step S2 as a constraint condition, construct a two-dimensional flow field equation group;

[0027] Solve the two-dimensional flow field equations and obtain the result matrix;

[0028] Expand and modularize the result matrix;

[0029] Calculate the signal covariance matrix in the x-direction and y-direction respectively;

[0030] Based on the signal covariance matrix in the x-direction and the y-direction, an expansion matrix is ​​further designed;

[0031] Insert the extended matrix into the two-dimensional flow field equations to obtain the expanded water surface flow field velocity matrix.

[0032] Step S3 specifically includes the following steps:

[0033] Taking the moving speed of the center of gravity of the pollutant obtained in step S2 as the constraint condition, a two-dimensional flow field equation group is constructed and expressed by the following formula: in, is the water density; is the velocity of the water in the x direction; v is the velocity of the water in the y direction; is the viscosity coefficient of water; P is the micro-element pressure of fluid; is the volume force in the x direction; is the volume force in the y direction; The coordinates are Velocity vector of position;

[0034] To solve the two-dimensional flow field equations , v as the basis, the calculation result matrix of the flow field in the x direction is U, and the calculation result matrix of the flow field in the y direction is V;

[0035] The result matrix is ​​expanded and modularized, and the expansion matrix module process is expressed by the following formula: in, is the front of the matrix U OK; is the front of the matrix V OK; is the matrix U OK; is the matrix V OK; is the first row of the matrix U; is the first row of the matrix V; is the Mth row of matrix U; is the Mth row of matrix V;

[0036] The signal covariance matrices in the x-direction and y-direction are calculated respectively, and the signal covariance matrices in the x-direction and y-direction are expressed by the following formulas: Among them, R is the covariance matrix; H is the conjugate transpose operator of the matrix; E is the mathematical expectation operator;

[0037] Based on the signal covariance matrix in the x-direction and y-direction, the extended matrix is ​​further designed and expressed as follows: Among them, the matrix K is the expansion matrix.

[0038] Insert the extended matrix into the two-dimensional flow field equations to obtain the expanded water surface flow field velocity matrix, which is expressed by the following formula: Among them, the matrix KK is the velocity matrix of the water surface flow field after interpolation and expansion; The coordinates of the expanded elements of the velocity matrix.

[0039] Step S4 includes the following steps:

[0040] Based on the expanded water body surface flow field velocity matrix obtained in step S3, selecting a determination basis from the water body image data obtained in step S1;

[0041] Based on the judgment basis, the water pollutants are subjected to a trailing judgment to obtain a trailing judgment result.

[0042] Step S4 specifically includes the following steps:

[0043] Based on the expanded water surface flow field velocity matrix obtained in step S3, select the coordinates of any point on the edge of the pollutant The following formula is used to express the judgment based on several images before and after the target image in the time axis within t seconds: Among them, ZZ is the coordinate The position is the state of a circle whose center is taken as the radius approaching infinitesimal; For coordinates The instantaneous velocity of the position; It is the position state of the pollutant after it moves, which is the sum of its initial position and velocity vector;

[0044] Based on the judgment basis, the water pollutants are judged by dragging, which is expressed by the following formula: in, The left limit of the initial state of the coordinate position is 0, and the right limit is 1; ST is the result of the smear judgment; is the state without drag; T is the state of drag; when the pollutant moves, the left and right limits of the position are If the initial states of the coordinate positions are the same, it is determined that no smearing occurs; otherwise, it is determined that a smearing occurs.

[0045] The present invention also provides a surface water pollution identification method including the surface water smear judgment method based on flow field matrix expansion, comprising the following steps:

[0046] Performing a smear judgment according to the surface water body smear judgment method based on flow field matrix expansion to obtain a smear judgment result;

[0047] According to the obtained afterimage judgment result, if it is determined that the pollutants in the water body image data do not have afterimage status, the water body pollution is judged and the pollution situation is estimated according to the coverage area of ​​the pollutants to obtain the identification result of the water body pollutants; otherwise, the water body pollution identification is not performed on the image.

[0048] The present invention discloses a surface water body trailing image judgment method and a pollution identification method based on flow field matrix expansion. The water body condition at the sewage outlet is monitored by image, and a model is constructed to judge the trailing image degree of pollutants in the water body, thereby qualitatively judging the water body pollution state and improving the monitoring accuracy of sudden water pollution. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a flow chart of the surface water body trailing image judging method based on flow field matrix expansion of the present invention;

[0050] Figure 2 A schematic flow chart of a water pollution identification method including the surface water body trailing image judgment method based on flow field matrix expansion according to the present invention;

[0051] Figure 3 The grayscale transformation result image of the water body detection image in the embodiment of the present invention;

[0052] Figure 4 The edge detection processing result of the water body detection image in the embodiment of the present invention;

[0053] Figure 5 The contour line recognition result of the water body detection image in the embodiment of the present invention;

[0054] Figure 6 The pollutant rectangle recognition result of the water body detection image in the embodiment of the present invention;

[0055] Figure 7 Schematic diagram of the movement of pollutants in water body detection samples in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The present invention provides a method for judging the surface water body trail and identifying pollution based on the expansion of the flow field matrix, and its flow chart is as follows: Figure 1 As shown, the following steps are included:

[0057] S1. Acquire water body image data;

[0058] The water body image data is captured at a sewage outlet using an industrial camera and divided into grids.

[0059] S2. grayscale transformation of the water body image data obtained in step S1, calculating the edge curve of pollutants on the water body surface based on the grayscale transformation result, and obtaining the moving speed of the center of gravity of the pollutants through the edge curve;

[0060] Step S2 includes the following steps:

[0061] Performing grayscale transformation on the water body image data obtained in step S1;

[0062] Based on the second-order partial derivatives of the grayscale transformation results in the x-direction and the y-direction, the edge curve of the pollutant on the water surface is calculated;

[0063] Design pixel variance correction factor based on water image saturation, and design pixel exposure correction factor based on ambient brightness;

[0064] Based on the pixel variance correction factor and the pixel exposure correction coefficient, the edge curve of the pollutant on the water surface is corrected to obtain the corrected edge curve;

[0065] According to the corrected edge curve, the moving speed of the center of gravity of the pollutant is calculated.

[0066] Step S2 specifically includes the following steps:

[0067] The water body image data obtained in step S1 is subjected to grayscale transformation and is calculated using the following formula: in, is the blue component of the pixel with the horizontal coordinate x and the vertical coordinate y in the Nth image of the water body image data; The green component of the pixel point with the horizontal coordinate x and the vertical coordinate y in the water body image data; The red component of the pixel point with the horizontal coordinate x and the vertical coordinate y in the water body image data; is the result of grayscale transformation of the Nth image in the water body image data;

[0068] Based on the second-order partial derivatives of the grayscale transformation results in the x and y directions, the edge curve of the pollutant on the water surface is calculated, and the pollutant contour line is obtained by solving the following formula: in, is the sum of the second-order partial derivatives of the image in the x-direction and the y-direction; is the gray value of the image;

[0069] The pixel variance correction factor is designed based on the saturation of the water image and calculated using the following formula: in, is the saturation of the Nth image in the water body image data;

[0070] Design the pixel exposure correction factor based on the ambient brightness and calculate it using the following formula: in, is the ambient brightness parameter;

[0071] Based on the pixel variance correction factor and the pixel exposure correction coefficient, the edge curve of the pollutant on the water surface is corrected to obtain the corrected edge curve, which is expressed by the following formula: in, is the corrected edge curve; is the pixel variance correction factor; is the pixel exposure correction factor.

[0072] According to the corrected edge curve, the moving speed of the center of gravity of the pollutant is calculated by the following formula: in, is the velocity of the water flow in the x direction at the center of gravity of the pollutant; is the velocity of the water flow in the y direction at the center of gravity of the pollutant; a is the corrected edge curve The projection line segment in the x direction; b is the corrected edge curve The projection line segment in the y direction; t is the time difference of the camera taking pictures when sampling the water body image data.

[0073] S3. Taking the moving speed of the center of gravity of the pollutant obtained in step S2 as the constraint condition, based on the velocity field matrix correction transformation, the expanded water surface flow field velocity matrix is ​​obtained;

[0074] Step S3 includes the following steps:

[0075] Using the moving speed of the center of gravity of the pollutant obtained in step S2 as a constraint condition, construct a two-dimensional flow field equation group;

[0076] Solve the two-dimensional flow field equations and obtain the result matrix;

[0077] Expand and modularize the result matrix;

[0078] Calculate the signal covariance matrix in the x-direction and y-direction respectively;

[0079] Based on the signal covariance matrix in the x-direction and the y-direction, an expansion matrix is ​​further designed;

[0080] Insert the extended matrix into the two-dimensional flow field equations to obtain the expanded water surface flow field velocity matrix.

[0081] Step S3 specifically includes the following steps:

[0082] Taking the moving speed of the center of gravity of the pollutant obtained in step S2 as the constraint condition, a two-dimensional flow field equation group is constructed and expressed by the following formula: in, is the water density; is the velocity of the water in the x direction; v is the velocity of the water in the y direction; is the viscosity coefficient of water; P is the micro-element pressure of fluid; is the volume force in the x direction; is the volume force in the y direction; The coordinates are Velocity vector of position;

[0083] To solve the two-dimensional flow field equations , v as the basis, the calculation result matrix of the flow field in the x direction is U, and the calculation result matrix of the flow field in the y direction is V;

[0084] The result matrix is ​​expanded and modularized, and the expansion matrix module process is expressed by the following formula: in, is the front of the matrix U OK; is the front of the matrix V OK; is the matrix U OK; is the matrix V OK; is the first row of the matrix U; is the first row of the matrix V; is the Mth row of matrix U; is the Mth row of matrix V;

[0085] The signal covariance matrices in the x-direction and y-direction are calculated respectively, and the signal covariance matrices in the x-direction and y-direction are expressed by the following formulas: Among them, R is the covariance matrix; H is the conjugate transpose operator of the matrix; E is the mathematical expectation operator;

[0086] Based on the signal covariance matrix in the x-direction and y-direction, the extended matrix is ​​further designed and expressed as follows: Among them, the matrix K is the expansion matrix.

[0087] Insert the extended matrix into the two-dimensional flow field equations to obtain the expanded water surface flow field velocity matrix, which is expressed by the following formula: Among them, the matrix KK is the velocity matrix of the water surface flow field after interpolation and expansion; The coordinates of the expanded elements of the velocity matrix.

[0088] S4. Based on the expanded water surface flow field velocity matrix obtained in step S3, a judgment basis is selected from the water image data obtained in step S1, and a smear judgment is performed on the water pollutants to obtain a smear judgment result;

[0089] Step S4 includes the following steps:

[0090] Based on the expanded water body surface flow field velocity matrix obtained in step S3, selecting a determination basis from the water body image data obtained in step S1;

[0091] Based on the judgment basis, the water pollutants are subjected to a trailing judgment to obtain a trailing judgment result.

[0092] Step S4 specifically includes the following steps:

[0093] Based on the expanded water surface flow field velocity matrix obtained in step S3, select the coordinates of any point on the edge of the pollutant The following formula is used to express the judgment based on several images before and after the target image in the time axis within t seconds: Among them, ZZ is the coordinate The position is the state of a circle whose center is taken as the radius approaching infinitesimal; For coordinates The instantaneous velocity of the position; It is the position state of the pollutant after it moves, which is the sum of its initial position and velocity vector;

[0094] Based on the judgment basis, the water pollutants are judged by dragging, which is expressed by the following formula: in, The left limit of the initial state of the coordinate position is 0, and the right limit is 1; ST is the result of the smear judgment; is the state without drag; T is the state of drag; when the pollutant moves, the left and right limits of the position are If the initial states of the coordinate positions are the same, it is determined that no smearing occurs; otherwise, it is determined that a smearing occurs.

[0095] The present invention also provides a surface water pollution identification method including the surface water smear judgment method based on flow field matrix expansion, and its flow chart is as follows: Figure 2 As shown, the following steps are included:

[0096] Performing a smear judgment according to the surface water body smear judgment method based on flow field matrix expansion to obtain a smear judgment result;

[0097] According to the obtained afterimage judgment result, if it is determined that there is no afterimage in the water body image data, the water body pollution is judged and the pollution situation is estimated according to the pollutant coverage area to obtain the identification result of the water body pollutant; otherwise, the water body pollution identification is not performed on the image.

[0098] The method of the present invention is further described below in conjunction with an embodiment:

[0099] The sampling device of a smelter's sewage outlet was selected as the research object. During the sampling process, the industrial camera was set to complete one shot at an interval of 0.02 seconds. The shutter time seconds, the initial grid intersections are 200, and the number of grid intersections after expansion is 10,000.

[0100] After the system is started, it runs continuously. Take the 12512th photo processing and comparison process as an example, perform grayscale transformation, and obtain the grayscale transformation result of the photo, as shown in Figure 3 shown. Figure 3 Image size in pixels It is easy to see from the figure that there are point pollutants in the water body being tested, and there are also influencing factors such as overexposure and light source reflection.

[0101] The edge curve is further calculated from the grayscale transformation result to perform edge detection. The result is as follows: Figure 4 As shown. Figure 4It can be seen that the water body image after pollutant edge detection can clearly characterize point pollutants, but on the other hand, factors such as ripples that are not easily perceived in the measured water body are also magnified.

[0102] The edge curve is further modified and the pollutant contour is marked. The result is as follows Figure 5 As shown. Figure 5 It can be seen that after contour line marking correction, the main pollutants are marked and interference factors such as ripples are eliminated.

[0103] Further, in The projection line segment in the axis direction is a, and the projection line segment in the x-axis direction is b. The minimum rectangle with a length of a and a width of b that satisfies the coverage of the edge curve of the pollutant is obtained. The result is as follows Figure 6 As shown. Figure 5 As shown, the detected water pollutants are marked by rectangular frames, where Figure 6 The two pollutant points in the lower right corner are close to each other and are identified as a single pollutant point. Considering that the pollutants themselves may have a certain degree of adhesion, and the external fluid properties of the aggregated pollutants also change accordingly, it can be considered that the pollutants adhered together are a single individual and do not affect the judgment of the water body detection system.

[0104] By calculating the distribution of the extended flow field, the schematic diagram of the movement of pollutants in the water body detection sample is obtained, such as Figure 7 shown. Figure 7 In the calculation process, the center of the rectangular frame, i.e., the intersection of the diagonal lines, is taken as the center of gravity of the pollution point. The flow field influence vector is added to the mark with the center of gravity as the starting point. The direction of the line is the flow field direction, and the length is the force magnitude of the pollution point. Compare the left and right limits of the position of the pollution after moving under the flow field condition with The coordinate position is different from the initial state, and the smear situation is determined according to step S4 of the method of the present invention.

[0105] Then, according to the smear judgment result, the existing method is used to identify the pollution level of the surface layer of the water body to obtain the identification result.

[0106] This series of samples were taken from the sewage outlet of the smelter. The system was run continuously for 5 days, and the pollutant smearing judgment results showed that 60% of the sampled photos did not have smearing. The system can be used to judge the surface pollution of water bodies and estimate the pollution level. The accuracy rate of manual on-site investigation and judgment is over 95%.

Claims

1. A method for judging the surface water body smear based on flow field matrix expansion, characterized in that: The following steps are involved: S1. Acquire water body image data; S2. grayscale transformation of the water body image data obtained in step S1, calculating the edge curve of pollutants on the water body surface based on the grayscale transformation result, and obtaining the moving speed of the center of gravity of the pollutants through the edge curve; S3. Taking the moving speed of the center of gravity of the pollutant obtained in step S2 as the constraint condition, based on the velocity field matrix correction transformation, the expanded water surface flow field velocity matrix is ​​obtained; S4. Based on the expanded water surface flow field velocity matrix obtained in step S3, a judgment basis is selected from the water image data obtained in step S1, and a smear judgment is performed on the water pollutants to obtain a smear judgment result; S5. According to the smear judgment result obtained in step S4, water pollution is judged and pollution is predicted according to the pollutant coverage area to obtain the identification result of water pollutants; Step S3 specifically includes the following steps: Taking the moving speed of the center of gravity of the pollutant obtained in step S2 as the constraint condition, a two-dimensional flow field equation group is constructed and expressed by the following formula: in, is the water density; is the velocity of the water in the x direction; v is the velocity of the water in the y direction; is the viscosity coefficient of water; P is the micro-element pressure of fluid; is the volume force in the x direction; is the volume force in the y direction; The coordinates are Velocity vector of position; is the velocity of the water flow in the x direction at the center of gravity of the pollutant; is the velocity of the water flow in the y direction at the center of gravity of the pollutant; To solve the two-dimensional flow field equations , v as the basis, the calculation result matrix of the flow field in the x direction is U, and the calculation result matrix of the flow field in the y direction is V; The result matrix is ​​expanded and modularized, and the expansion matrix module process is expressed by the following formula: in, is the front of the matrix U OK; is the front of the matrix V OK; is the matrix U OK; is the matrix V OK; is the first row of the matrix U; is the first row of the matrix V; is the Mth row of matrix U; is the Mth row of matrix V; The signal covariance matrices in the x-direction and y-direction are calculated respectively, and the signal covariance matrices in the x-direction and y-direction are expressed by the following formulas: Among them, R is the covariance matrix; H is the conjugate transpose operator of the matrix; E is the mathematical expectation operator; Design the expansion matrix, expressed as follows: Among them, matrix K is the expansion matrix; Insert the extended matrix into the two-dimensional flow field equations to obtain the expanded water surface flow field velocity matrix, which is expressed by the following formula: Among them, the matrix KK is the velocity matrix of the water surface flow field after interpolation and expansion; The coordinates of the expanded elements of the velocity matrix.

2. The surface water body trailing judgment method based on flow field matrix expansion according to claim 1 is characterized in that: In step S1, the water body image data is captured at the sewage outlet using an industrial camera and divided into grids.

3. The surface water body trailing judgment method based on flow field matrix expansion according to claim 1 is characterized in that: Step S2 includes the following steps: Performing grayscale transformation on the water body image data obtained in step S1; Based on the second-order partial derivatives of the grayscale transformation results in the x-direction and the y-direction, the edge curve of the pollutant on the water surface is calculated; Design pixel variance correction factor based on water image saturation, and design pixel exposure correction factor based on ambient brightness; Based on the pixel variance correction factor and the pixel exposure correction coefficient, the edge curve of the pollutant on the water surface is corrected to obtain the corrected edge curve; According to the corrected edge curve, the moving speed of the center of gravity of the pollutant is calculated.

4. The surface water body trailing judgment method based on flow field matrix expansion according to claim 3 is characterized in that: Step S2 specifically includes the following steps: The water body image data obtained in step S1 is subjected to grayscale transformation and is calculated using the following formula: in, is the blue component of the pixel with the horizontal coordinate x and the vertical coordinate y in the Nth image of the water body image data; The green component of the pixel point with the horizontal coordinate x and the vertical coordinate y in the water body image data; The red component of the pixel point with the horizontal coordinate x and the vertical coordinate y in the water body image data; is the result of grayscale transformation of the Nth image in the water body image data; Based on the second-order partial derivatives of the grayscale transformation results in the x and y directions, the edge curve of the pollutant on the water surface is calculated, and the pollutant contour line is obtained by solving the following formula: in, is the sum of the second-order partial derivatives of the image in the x-direction and the y-direction; is the image gray value; The pixel variance correction factor is designed based on the saturation of the water image and calculated using the following formula: in, is the saturation of the Nth image in the water body image data; Design the pixel exposure correction factor based on the ambient brightness and calculate it using the following formula: in, is the ambient brightness parameter; Based on the pixel variance correction factor and the pixel exposure correction coefficient, the edge curve of the pollutant on the water surface is corrected to obtain the corrected edge curve, which is expressed by the following formula: in, is the corrected edge curve; is the pixel variance correction factor; is the pixel exposure correction factor; According to the corrected edge curve, the moving speed of the center of gravity of the pollutant is calculated by the following formula: in, is the velocity of the water flow in the x direction at the center of gravity of the pollutant; is the velocity of the water flow in the y direction at the center of gravity of the pollutant; a is the corrected edge curve The projection line segment in the x direction; b is the corrected edge curve The projection line segment in the y direction; t is the time difference of the camera taking pictures when sampling the water body image data.

5. The surface water body trailing judgment method based on flow field matrix expansion according to claim 4 is characterized in that: Step S3 includes the following steps: Using the moving speed of the center of gravity of the pollutant obtained in step S2 as a constraint condition, construct a two-dimensional flow field equation group; Solve the two-dimensional flow field equations and obtain the result matrix; Expand and modularize the result matrix; Calculate the signal covariance matrix in the x-direction and y-direction respectively; Based on the signal covariance matrix in the x-direction and the y-direction, an expansion matrix is ​​designed; Insert the extended matrix into the two-dimensional flow field equations to obtain the expanded water surface flow field velocity matrix.

6. The surface water body trailing judgment method based on flow field matrix expansion according to claim 5 is characterized in that: Step S4 specifically includes the following steps: Based on the expanded water surface flow field velocity matrix obtained in step S3, select the coordinates of any point on the edge of the pollutant The following formula is used to express the judgment based on several images before and after the target image in the time axis within t seconds: Among them, ZZ is the coordinate The position is the state of a circle whose center is taken as the radius approaching infinitesimal; For coordinates The instantaneous velocity of the position; It is the position state of the pollutant after it moves, which is the sum of its initial position and velocity vector; Based on the judgment basis, the water pollutants are judged by dragging, which is expressed by the following formula: in, The left limit of the initial state of the coordinate position is 0, and the right limit is 1; ST is the result of the smear judgment; is the state without drag; T is the state of drag; when the pollutant moves, the left and right limits of the position are If the initial states of the coordinate positions are the same, it is determined that no smearing occurs; otherwise, it is determined that a smearing occurs.

7. A surface water pollution identification method comprising the surface water smear judgment method based on flow field matrix expansion as claimed in any one of claims 1 to 6, characterized in that: The following steps are involved: Performing a smear judgment according to the surface water body smear judgment method based on flow field matrix expansion to obtain a smear judgment result; According to the obtained afterimage judgment result, if it is determined that there is no afterimage in the water body image data, the water body pollution is judged and the pollution situation is estimated according to the pollutant coverage area to obtain the identification result of the water body pollutant; otherwise, the water body pollution identification is not performed on the image.

Images