A method and system for riverway management based on image processing technology

Abstract

The application discloses a kind of based on image processing technique's river course management method and system, it is related to urban river course management technical field, the system is by using GIS, river network distribution and DEM model, management area is divided into several sub-reaches, and video frame, water quality image and pollution characteristic image are collected;Through image processing, the motion trajectory of floating object, distribution characteristics, water color information and abnormal color area are extracted, the pollution change trend and mutation characteristics are analyzed;Through monitoring the diffusion disturbance coefficient of surface garbage, the state of surface garbage is judged in combination with threshold value, and strategy response is triggered;When water quality is abnormal, whether water color is normal is judged through water color difference anomaly coefficient analysis;If there is pollution, pollution gradient change is analyzed along river course countercurrent direction section by section, locks pollution source and generates pollution upstream inference report;The application effectively improves river pollution monitoring precision and response efficiency, helps ecological management and environmental protection.

Description

Technical Field

[0001] This invention belongs to the field of urban river management technology, specifically relating to a method and system for river management based on image processing technology. Background Technology

[0002] In urban river management, the rapid identification and source tracing of water pollutants have always been key technical challenges in water environment management. Currently, common water quality monitoring methods mostly rely on traditional manual sampling, physicochemical analysis, or online sensor equipment for indicator detection. While these methods have a certain degree of accuracy, they suffer from long response delays, limited spatial coverage, and difficulty in real-time tracking of pollution change paths, making it difficult to meet the needs for rapid early warning and dynamic management of complex urban water pollution events.

[0003] Especially in urban tributaries or areas with a high density of discharge outlets where pollutant sources are complex and rapidly changing, relying solely on point-based water quality data cannot fully reconstruct the spatial distribution and evolution of pollution, easily leading to missed or incorrect diagnoses and affecting treatment efficiency. Furthermore, water pollution is often accompanied by visual characteristics such as increased surface debris, abnormal water color variations, and flow disturbances; these pollution manifestations are easily identifiable through images. Therefore, river management methods based on image processing technology can achieve non-contact monitoring and intelligent early warning of pollution events from dimensions such as video images, water surface color, and pollution dynamics, offering significant advantages. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for river management based on image processing technology, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solution: a river management system based on image processing technology, comprising a data acquisition module, an image processing module, a floating debris monitoring and tracking module, a water body monitoring module, and a pollution source investigation module;

[0006] The data acquisition module divides the treatment area into several sub-river sections based on GIS, river network distribution and DEM model, and acquires video frames, water quality images and pollution feature images through fixed high-definition cameras, bank image sampling equipment and UAV multispectral cameras respectively;

[0007] The image processing module is used to extract the motion trajectory and distribution characteristics of floating objects, water color information and abnormal color areas, and pollution change trends and abrupt change characteristics in images over multiple time periods.

[0008] The floating object monitoring and tracking module is used to monitor the density changes and velocity gradient changes of floating object targets in the fused image of the sub-river section in real time, calculate and obtain the surface debris diffusion disturbance coefficient FDI, and compare and analyze it with the first threshold Q1 to determine whether the overall state of surface debris in the sub-river section is stable. If it is unstable, a strategy is given.

[0009] The water monitoring module is used to monitor the water quality of the current sub-river section in real time when it receives the first warning instruction. It calculates the water color difference anomalous coefficient SCP by combining historical water color standard data and compares it with the second threshold Q2 to determine whether the water color of the current sub-river section is normal. If it is abnormal, a strategy is given.

[0010] The pollution source investigation module is used to analyze the river water quality segment by segment along the upstream direction of the river and calculate the pollution gradient change rate ΔSCP when a second early warning instruction is received. j And to find a way to calculate the rate of change of the pollution gradient ΔSCP j The sub-channel region Jcq with the maximum value j They identified the source of the pollution and provided corresponding strategies.

[0011] Preferably, the data acquisition module includes a region division unit and an acquisition unit;

[0012] The regional division unit is used to establish a two-dimensional hydrological spatial coupling model of the river channel based on remote sensing geographic information (GIS) and urban river network distribution data, combined with digital elevation model (DEM), and to divide the treatment area into several sub-river segment regions, labeled as: Jcq1, Jcq2, ..., JcqM; M represents the number of sub-river segment regions;

[0013] The acquisition unit is used to monitor the sub-river section area. It acquires continuous video frame sequences of the river by deploying fixed high-definition water surface monitoring cameras along the river; it acquires high-definition water quality image samples by deploying static image sampling equipment on the bank; and it acquires water surface pollution characteristic images and color characteristic images by using a drone equipped with a multispectral camera.

[0014] Preferably, the image processing module includes an optical flow field processing unit, a water chromatography processing unit, and a pollution time sequence processing unit;

[0015] The optical flow field processing unit is used to extract motion information from a continuous video image sequence of the river channel. It uses the optical flow estimation algorithm PWCNet to obtain the motion trajectory information of floating objects between images; extracts the velocity vector features and aggregation region boundaries of the floating objects to obtain the distribution status and movement trend map of the floating objects in the river channel; and extracts the image features of the density change of the floating objects in the region to obtain the aggregation area of ​​the floating objects in the current frame.

[0016] The water chromatography processing unit is used to extract color information from static water surface images, and uses the HSI color gamut conversion method to obtain the hue, saturation and brightness features in the image; compares the color difference spectrum of water bodies in different areas, extracts abnormal color areas, and generates water surface color feature layer data;

[0017] The pollution time-series processing unit is used to dynamically extract the visual features of pollution from image sequences collected at different time points; analyze the changing trends of floating debris distribution maps, water color maps, and disturbance feature maps in the time dimension; identify abrupt change regions in the pollution change process, and extract key image parameters before and after the change.

[0018] Preferably, the floating object monitoring and tracking module includes a first computing unit and a first analysis unit;

[0019] The first calculation unit is used to monitor in real time the density changes and velocity gradient changes of floating object boundaries in the fused image of the sub-river section. After dimensionless processing, the surface debris diffusion disturbance coefficient FDI is calculated and obtained, as follows:

[0020]

[0021] In the formula, This represents the set of velocity vectors of objects floating in an optical flow field. This represents the variance of the local velocity vector. A represents the change in the boundary density gradient of a floating object per unit time. c A represents the area of ​​the concentrated floating debris region in real time. t represents the average total area of ​​floating objects at time t in the past, and w1, w2 and w3 represent weighting coefficients;

[0022]

[0023] In the formula, N represents the number of pixels in the image, and represents the magnitude of the velocity vector of the floating object at the i-th pixel. This represents the average magnitude of the velocity vector of a floating object;

[0024]

[0025] In the formula, D t D represents the pollution attribute parameter value in the image at time t. t-1 Δt represents the pollution attribute parameter value in the image at time t, which is the time interval between image frames.

[0026] Preferably, the first analysis unit is used to preset a first threshold Q1, and compare the surface debris diffusion disturbance coefficient FDI with the first threshold Q1 to obtain a first evaluation result, including:

[0027] When the surface debris diffusion disturbance coefficient FDI ≤ the first threshold Q1, it indicates that the overall state of surface debris in the current sub-river section is stable, there are no signs of diffusion, and there is no risk of increased debris accumulation. Continuous monitoring is required.

[0028] When the surface debris diffusion disturbance coefficient FDI > the first threshold Q1, it indicates that the overall state of surface debris in the current sub-river section is unstable, with signs of diffusion and a risk of increased debris accumulation. This triggers the first warning instruction and generates the first strategy: activate the floating debris interception equipment along the route and arrange unmanned vessels to prioritize path cleaning; adjust the flow velocity at the upstream drainage gate to alleviate flow velocity disturbance; and conduct water quality testing on the current sub-river section.

[0029] Preferably, the water monitoring module includes a second computing unit and a second analysis unit;

[0030] The second calculation unit is used to monitor the river water quality in the current sub-river section in real time when the first early warning command is received. After combining historical water color standard data and performing dimensionless processing, it calculates and obtains the water color difference anomaly coefficient SCP, as shown in the following formula:

[0031] SCP = a1*(1-uh) 2 +a2*qs+a3*|uI-Iref|+a4*Dspec;

[0032] In the formula, uh represents the mean color of the water body, qs represents the standard deviation of saturation, uI represents the mean brightness, Iref represents the standard brightness reference value, Dspec represents the chromatographic distribution difference value, and a1, a2, a3 and a4 represent weighting coefficients.

[0033] Preferably, the second analysis unit is used to pre-set a second threshold Q2, and compare the water body color difference anomalous coefficient SCP with the second threshold Q2 to obtain a second evaluation result, including:

[0034] When the water color difference anomalous coefficient SCP is less than the second threshold Q2, it indicates that the water color in the current sub-river section is normal and there are no pollution characteristics, and continuous monitoring is required.

[0035] When the water color difference anomaly coefficient SCP is greater than or equal to the second threshold Q2, it indicates that the water color in the current sub-river section is abnormal and has pollution characteristics, triggering the second early warning instruction and generating the second strategy: to activate the upstream investigation mechanism to investigate the source of pollution.

[0036] Preferably, the pollution source investigation module includes a third calculation unit, a fourth calculation unit, and an analysis strategy unit;

[0037] The third calculation unit is used to analyze the river water quality segment by segment along the upstream direction of the river when the second early warning command is received. After dimensionless processing, it calculates and obtains the pollution gradient change rate ΔSCP. j The formula is as follows:

[0038] ΔSCP j =SCP j -SCP j-1 ;

[0039] In the formula, SCP j SCP represents the water color difference anomaly coefficient in the j-th sub-river segment. j-1 This represents the water color difference anomaly coefficient of the (j-1)th sub-channel region, where the (j-1)th sub-channel region is the upstream adjacent sub-channel region of the j-th sub-channel region.

[0040] The fourth calculation unit is used to calculate and obtain the pollution gradient change rate ΔSCP. j In all continuous sub-channel regions, after dimensionless processing, the rate of change of pollution gradient ΔSCP is sought to calculate. j The sub-channel region Jcq with the maximum value j The formula is as follows:

[0041]

[0042] In the formula, ΔSCP j This indicates the rate of change of the pollution gradient.

[0043] Preferably, the analysis strategy unit is used to measure the pollution gradient change rate ΔSCP. j The sub-channel region Jcq with the maximum value j The source of the pollution was identified, and the Jcq section of the river was targeted. j It calls up historical image sequences for comparison, performs retrospective detection, and automatically generates an "Upstream Inference Report on Industrial Pollution Discharge" to notify ecological regulatory units to prioritize its handling.

[0044] Preferably, a river management method based on image processing technology includes the following steps:

[0045] Step 1: Based on GIS, river network distribution and DEM model, the treatment area is divided into several sub-river sections, and video frames, water quality images and pollution feature images are collected by using fixed high-definition cameras, bank image sampling equipment and UAV multispectral cameras respectively;

[0046] Step 2: Extract the motion trajectory and distribution characteristics of floating objects, water color information and abnormal color areas, and pollution change trends and abrupt change characteristics in images over multiple time periods;

[0047] Step 3: Monitor the density changes and velocity gradient changes of floating object boundaries in the fused image of the sub-river section in real time, calculate the surface debris diffusion disturbance coefficient FDI, and compare it with the first threshold Q1 to determine whether the overall state of surface debris in the sub-river section is stable. If it is unstable, a strategy is given.

[0048] Step 4: When the first warning instruction is received, monitor the water quality of the current sub-river section in real time, calculate the water color difference anomalous coefficient SCP by combining historical water color standard data, and compare it with the second threshold Q2 to determine whether the water color of the current sub-river section is normal. If it is abnormal, a strategy is given.

[0049] Step 5: Upon receiving the second early warning instruction, analyze the river water quality segment by segment along the upstream direction of the river, and calculate the pollution gradient change rate ΔSCP. j And to find a way to calculate the rate of change of the pollution gradient ΔSCP j The sub-channel region Jcq with the maximum value j They identified the source of the pollution and provided corresponding strategies.

[0050] This invention provides a method and system for river management based on image processing technology. It has the following beneficial effects:

[0051] (1) This method and system for river management based on image processing technology combines fixed high-definition cameras, static image sampling equipment, and UAV multispectral imaging technology to construct a multi-angle, three-dimensional data acquisition network, enabling high-frequency image acquisition of multiple sub-sections of urban rivers. Compared with traditional manual or sensor-based monitoring methods, this invention has advantages such as non-contact, no blind spots, and wide coverage, enabling full-process recording and real-time early warning of the dynamic evolution of water pollution events, significantly improving monitoring response speed and coverage accuracy.

[0052] (2) This method and system for river management based on image processing technology constructs an image processing module that integrates floating object optical flow tracing, water color difference analysis, and pollution time sequence extraction. This module can efficiently extract key features such as the movement pattern, color anomalies, and temporal abrupt changes of water pollution from visual images, and construct quantitative indicators such as FDI (Fouling Dispersion Disturbance Coefficient) and SCP (Water Color Difference Anomaly Coefficient). This processing method enables quantitative assessment of pollution risk.

[0053] (3) This method and system for river management based on image processing technology constructs a tracking model based on the rate of change of pollution gradient. After an abnormal pollution event is triggered, it can automatically backtrack and analyze image data along the river to calculate the region with the maximum rate of change of pollution gradient, thereby intelligently locating potential pollution sources. Compared with manual inspection, this method has the advantages of high automation, fast positioning speed, and high accuracy, effectively improving the efficiency of pollution tracking and the ability to supervise and deal with pollution.

[0054] (4) This method and system for river management based on image processing technology, in addition to detection, identification, and tracking, further designs multiple response strategies, such as initiating unmanned vessel cleaning path planning, adjusting the flow velocity of upstream sluice gates, and pushing out pollution source tracing reports, to achieve closed-loop control of the entire process from pollution identification to treatment response. The system can also generate image sequence comparison reports for abnormal areas, providing data support for urban water environment management and improving the scientific nature and timeliness of the treatment work. Attached Figure Description

[0055] Figure 1 This is a system block diagram and flowchart of a river management method based on image processing technology according to the present invention.

[0056] Figure 2 This is a schematic diagram illustrating the steps of a river management method based on image processing technology according to the present invention. Detailed Implementation

[0057] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] Example 1

[0059] Please see Figure 1 This invention provides a method and system for river management based on image processing technology, including a data acquisition module, an image processing module, a floating object monitoring and tracking module, a water body monitoring module, and a pollution source investigation module;

[0060] The data acquisition module divides the treatment area into several sub-river sections based on GIS, river network distribution and DEM model, and acquires video frames, water quality images and pollution feature images through fixed high-definition cameras, bank image sampling equipment and UAV multispectral cameras respectively;

[0061] The image processing module is used to extract the motion trajectory and distribution characteristics of floating objects, water color information and abnormal color areas, and pollution change trends and abrupt change characteristics in images over multiple time periods.

[0062] The floating object monitoring and tracking module is used to monitor the density changes and velocity gradient changes of floating object targets in the fused image of the sub-river section in real time, calculate and obtain the surface debris diffusion disturbance coefficient FDI, and compare and analyze it with the first threshold Q1 to determine whether the overall state of surface debris in the sub-river section is stable. If it is unstable, a strategy is given.

[0063] The water monitoring module is used to monitor the water quality of the current sub-river section in real time when it receives the first warning instruction. It calculates the water color difference anomalous coefficient SCP by combining historical water color standard data and compares it with the second threshold Q2 to determine whether the water color of the current sub-river section is normal. If it is abnormal, a strategy is given.

[0064] The pollution source investigation module is used to analyze the river water quality segment by segment along the upstream direction of the river and calculate the pollution gradient change rate ΔSCP when a second early warning instruction is received. j And to find a way to calculate the rate of change of the pollution gradient ΔSCP j The sub-channel region Jcq with the maximum value j They identified the source of the pollution and provided corresponding strategies.

[0065] In this embodiment, by integrating GIS geographic information, river network distribution, and DEM elevation model, and combining multi-source image acquisition and processing methods, it is possible to achieve refined segmented monitoring of urban rivers and accurate identification of pollution status. This significantly improves the real-time nature of pollution event detection and the accuracy of regional positioning, providing a scientific basis and efficient means for subsequent pollution prevention and control and source tracing.

[0066] Example 2

[0067] This embodiment is an explanation based on Embodiment 1. Please refer to it. Figure 1 Specifically, the data acquisition module includes a region division unit and an acquisition unit;

[0068] The regional division unit is used to establish a two-dimensional hydrological spatial coupling model of the river channel based on remote sensing geographic information (GIS) and urban river network distribution data, combined with digital elevation model (DEM), and to divide the treatment area into several sub-river segment regions, labeled as: Jcq1, Jcq2, ..., JcqM; M represents the number of sub-river segment regions;

[0069] The acquisition unit is used to monitor the sub-river section area. It acquires continuous video frame sequences of the river by deploying fixed high-definition water surface monitoring cameras along the river; it acquires high-definition water quality image samples by deploying static image sampling equipment on the bank; and it acquires water surface pollution characteristic images and color characteristic images by using a drone equipped with a multispectral camera.

[0070] In this embodiment, by setting up regional division units and acquisition units in the data acquisition module, a two-dimensional hydrological spatial coupling model of the river channel is constructed using GIS, river network distribution and DEM model, so as to realize accurate segmentation and identification of the river channel. Combined with multi-angle and multi-device image acquisition methods, the spatial coverage of pollution image information and the comprehensiveness of data acquisition are improved, providing high-quality and structured raw data support for subsequent image recognition and pollution analysis.

[0071] Example 3

[0072] This embodiment is an explanation based on Embodiment 2. Please refer to it. Figure 1 Specifically, the image processing module includes an optical flow field processing unit, a water chromatography processing unit, and a pollution time sequence processing unit;

[0073] The optical flow field processing unit is used to extract motion information from a continuous video image sequence of the river channel. It uses the optical flow estimation algorithm PWCNet to obtain the motion trajectory information of floating objects between images; extracts the velocity vector features and aggregation region boundaries of the floating objects to obtain the distribution status and movement trend map of the floating objects in the river channel; and extracts the image features of the density change of the floating objects in the region to obtain the aggregation area of ​​the floating objects in the current frame.

[0074] The water chromatography processing unit is used to extract color information from static water surface images, and uses the HSI color gamut conversion method to obtain the hue, saturation and brightness features in the image; compares the color difference spectrum of water bodies in different areas, extracts abnormal color areas, and generates water surface color feature layer data;

[0075] The pollution time-series processing unit is used to dynamically extract the visual features of pollution from image sequences collected at different time points; analyze the changing trends of floating debris distribution maps, water color maps, and disturbance feature maps in the time dimension; identify abrupt change regions in the pollution change process, and extract key image parameters before and after the change.

[0076] In this embodiment, by arranging an optical flow field processing unit, a water body chromatogram processing unit and a pollution time-series processing unit in the image processing module, multi-dimensional in-depth analysis of river channel pollution images can be implemented. The movement trail and aggregation state of floating objects can be accurately obtained, and abnormal water color areas and dynamic characteristics of pollution evolution over time can also be extracted, which effectively improves the accuracy and timeliness of pollution identification, and provides rich and high-dimensional image feature support for subsequent pollution early warning and traceability analysis.

[0077] Example 4

[0078] This embodiment is an explanation made based on Embodiment 3, please refer to Figure 1 , specifically, the floating object monitoring and tracking module comprises a first calculation unit and a first analysis unit;

[0079] The first calculation unit is configured to monitor the density change and flow velocity gradient change of the target boundary of floating objects in the fused image of the sub-river reach area in real time, and after dimensionless processing, calculate and obtain the water surface garbage diffusion disturbance coefficient FDI, the formula is as follows:

[0080]

[0081] In the formula, represents the velocity vector set of floating objects in the optical flow field, represents the variance of the local velocity vector, represents the change of the density gradient of the floating object boundary per unit time, A c represents the area of the real-time floating object concentration area, A t represents the average total area of floating objects at the t-th time in the past, w1, w2 and w3 represent weight coefficients, 0<w1<1, 0<w2<1, 0<w3<1 and w1+w2+w3=1;

[0082]

[0083] In the formula, N represents the number of pixels in the image, represents the magnitude of the motion velocity vector of the floating object on the i-th pixel, and v represents the average magnitude of the velocity vector of the floating object;

[0084]

[0085] In the formula, D t represents the pollution attribute parameter value in the image at time t, D t-1 represents the pollution attribute parameter value in the image at the time of the previous cycle before time t, and Δt represents the time interval between image frames.

[0086] In this embodiment, by setting up a first calculation unit and a first analysis unit in the floating object monitoring and tracking module, the surface debris diffusion disturbance coefficient FDI is constructed using multiple dynamic image features. Combined with dimensionless processing and weight factor optimization, the dynamic change status of floating debris pollution in the sub-river section can be accurately assessed, and real-time early warning of abnormal diffusion trends of surface debris can be achieved, effectively improving the sensitivity and response efficiency of water quality monitoring.

[0087] Example 5

[0088] This embodiment is an explanation based on Embodiment 4. Please refer to it. Figure 1 Specifically, the first analysis unit is used to pre-set a first threshold Q1 and compare the surface debris diffusion disturbance coefficient FDI with the first threshold Q1 to obtain a first evaluation result, including:

[0089] When the surface debris diffusion disturbance coefficient FDI ≤ the first threshold Q1, it indicates that the overall state of surface debris in the current sub-river section is stable, there are no signs of diffusion, and there is no risk of increased debris accumulation. Continuous monitoring is required.

[0090] When the surface debris diffusion disturbance coefficient FDI > the first threshold Q1, it indicates that the overall state of surface debris in the current sub-river section is unstable, with signs of diffusion and a risk of increased debris accumulation. This triggers the first warning instruction and generates the first strategy: activate the floating debris interception equipment along the route and arrange unmanned vessels to prioritize path cleaning; adjust the flow velocity at the upstream drainage gate to alleviate flow velocity disturbance; and conduct water quality testing on the current sub-river section.

[0091] In this embodiment, the surface debris diffusion disturbance coefficient (FDI) is compared and analyzed in real time using a first threshold Q1 preset by the first analysis unit, which can promptly identify the diffusion trend of surface debris in the sub-river section. When the unstable state of debris is detected, an early warning is automatically triggered and effective strategies are adopted, such as activating floating debris interception equipment and unmanned vessel cleaning, adjusting water flow, and conducting water quality testing, thereby effectively preventing the aggravation of debris accumulation and ensuring the stability of river water quality.

[0092] Example 6

[0093] This embodiment is an explanation based on Embodiment 5. Please refer to it. Figure 1 Specifically, the water monitoring module includes a second computing unit and a second analysis unit;

[0094] The second calculation unit is used to monitor the river water quality in the current sub-river section in real time when the first early warning command is received. After combining historical water color standard data and performing dimensionless processing, it calculates and obtains the water color difference anomaly coefficient SCP, as shown in the following formula:

[0095] SCP = a1*(1-uh)2 +a2*qs+a3*|uI-Iref|+a4*Dspec;

[0096] wherein, uh represents the average value of water body hue, qs represents the standard deviation of saturation, uI represents the average value of brightness, Iref represents the standard brightness reference value, Dspec represents the difference value of chromatographic distribution, a1, a2, a3 and a4 represent weight coefficients, 0<a1<1, 0<a2<1, 0<a3<1, 0<a4<1, and a1+a2+a3+a4=1.

[0097] In this embodiment, the second calculation unit is configured to monitor the water quality condition of the river channel in real time and calculate the water color difference anomaly coefficient SCP. The present invention can effectively identify the abnormal change of water body color and detect the sign of water pollution in time. When the water body color is abnormal, the system can automatically trigger an early warning and perform detailed analysis, which provides an accurate basis for the subsequent pollution source investigation and treatment measures, and improves the sensitivity and accuracy of pollution detection.

[0098] Example 7

[0099] This embodiment is an explanation made based on Embodiment 6, please refer to Figure 1 , specifically, the second analysis unit is configured to preset a second threshold Q2 in advance, and perform comparative analysis on the water color difference anomaly coefficient SCP and the second threshold Q2, and obtaining a second evaluation result comprises:

[0100] when the water color difference anomaly coefficient SCP < the second threshold Q2, it indicates that the water body color in the current sub-river reach area is normal, has no pollution characteristics, and continuous monitoring is performed;

[0101] when the water color difference anomaly coefficient SCP ≥ the second threshold Q2, it indicates that the water body color in the current sub-river reach area is abnormal, has pollution characteristics, triggers a second early warning instruction, and generates a second strategy: starting an upstream investigation mechanism to investigate the pollution source.

[0102] In this embodiment, through the comparative analysis of the water color difference anomaly coefficient SCP and the second threshold Q2 by the second analysis unit, the present invention can accurately identify abnormal water body color, detect pollution signs in time and trigger an early warning. The system can quickly start a pollution source investigation mechanism, help locate the pollution source in time, optimize the pollution treatment process, and ensure the real-time performance and accuracy of water quality monitoring.

[0103] Example 8

[0104] This embodiment is an explanation made based on Embodiment 7, please refer to Figure 1 , specifically, the pollution source investigation module comprises a third calculation unit, a fourth calculation unit and an analysis strategy unit;

[0105] The third calculation unit is used to analyze the river water quality segment by segment along the upstream direction of the river when the second early warning command is received. After dimensionless processing, it calculates and obtains the pollution gradient change rate ΔSCP. j The formula is as follows:

[0106] ΔSCP j =SCP j -SCP j-1 ;

[0107] In the formula, SCP j SCP represents the water color difference anomaly coefficient in the j-th sub-river segment. j-1 This represents the water color difference anomaly coefficient of the (j-1)th sub-channel region, where the (j-1)th sub-channel region is the upstream adjacent sub-channel region of the j-th sub-channel region.

[0108] The fourth calculation unit is used to calculate and obtain the pollution gradient change rate ΔSCP. j In all continuous sub-channel regions, after dimensionless processing, the rate of change of pollution gradient ΔSCP is sought to calculate. j The sub-channel region Jcq with the maximum value j The formula is as follows:

[0109]

[0110] In the formula, ΔSCP j This indicates the rate of change of the pollution gradient.

[0111] In this embodiment, through the combined operation of the third and fourth calculation units of the pollution source investigation module, the present invention can accurately calculate the rate of change of pollution gradient and quickly identify the sub-river area with the largest pollution gradient change by analyzing the river water quality segment by segment. This system can efficiently pinpoint pollution source areas, optimize pollution source investigation and control strategies, and improve the accuracy and response speed of pollution source identification.

[0112] Example 9

[0113] This embodiment is an explanation based on Embodiment 8. Please refer to it. Figure 1 Specifically, the analysis strategy unit is used to calculate the pollution gradient change rate ΔSCP. j The sub-channel region Jcq with the maximum value j The source of the pollution was identified, and the Jcq section of the river was targeted. j It calls up historical image sequences for comparison, performs retrospective detection, and automatically generates an "Upstream Inference Report on Industrial Pollution Discharge" to notify ecological regulatory units to prioritize its handling.

[0114] In this embodiment, by analyzing the automated backtracking detection of the strategy unit and comparing it with historical image sequences, the present invention can accurately locate the source of pollution and automatically generate an "Upstream Inference Report on Industrial Discharge," providing scientific basis for ecological regulatory units to prioritize the treatment of pollution sources. This strategy improves the accuracy and response efficiency of pollution source location, ensuring the timeliness and effectiveness of water pollution control.

[0115] Example 10

[0116] A method for river management based on image processing technology, please refer to... Figure 2 This includes the following steps:

[0117] Step 1: Based on GIS, river network distribution and DEM model, the treatment area is divided into several sub-river sections, and video frames, water quality images and pollution feature images are collected by using fixed high-definition cameras, bank image sampling equipment and UAV multispectral cameras respectively;

[0118] Step 2: Extract the motion trajectory and distribution characteristics of floating objects, water color information and abnormal color areas, and pollution change trends and abrupt change characteristics in images over multiple time periods;

[0119] Step 3: Monitor the density changes and velocity gradient changes of floating object boundaries in the fused image of the sub-river section in real time, calculate the surface debris diffusion disturbance coefficient FDI, and compare it with the first threshold Q1 to determine whether the overall state of surface debris in the sub-river section is stable. If it is unstable, a strategy is given.

[0120] Step 4: When the first warning instruction is received, monitor the water quality of the current sub-river section in real time, calculate the water color difference anomalous coefficient SCP by combining historical water color standard data, and compare it with the second threshold Q2 to determine whether the water color of the current sub-river section is normal. If it is abnormal, a strategy is given.

[0121] Step 5: Upon receiving the second early warning instruction, analyze the river water quality segment by segment along the upstream direction of the river, and calculate the pollution gradient change rate ΔSCP. j And to find a way to calculate the rate of change of the pollution gradient ΔSCP j The sub-channel region Jcq with the maximum value j They identified the source of the pollution and provided corresponding strategies.

[0122] In this embodiment, efficient and accurate identification and response to river water quality and pollution sources can be achieved through five steps of coordinated monitoring and analysis. First, by combining GIS, river network distribution, and DEM models, the treatment area is accurately delineated, ensuring comprehensive monitoring. Second, based on collaborative acquisition from multiple devices and multi-time-period image processing, real-time monitoring of floating debris distribution, water color changes, and pollution mutation characteristics is possible. Third, by calculating and comparing thresholds of the surface debris diffusion disturbance coefficient (FDI) and the water color difference anomalous coefficient (SCP), abnormal conditions of surface debris and water color are effectively identified, triggering timely warnings and generating response strategies. Finally, by analyzing the pollution gradient change rate, the pollution source is automatically located, and an upstream inference report is generated, providing precise treatment basis for ecological regulatory units.

[0123] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value, it is acceptable.

[0124] The above formulas are all derived from software simulation using a large amount of data and are selected to be close to the actual values. The coefficients in the formulas are set by those skilled in the art according to the actual situation. The above description is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any equivalent substitutions or changes made by those skilled in the art within the technical scope disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the protection scope of the present invention.

Claims

1. A river management system based on image processing technology, characterized in that, It includes a data acquisition module, an image processing module, a floating object monitoring and tracking module, a water body monitoring module, and a pollution source investigation module; The data acquisition module divides the treatment area into several sub-river sections based on GIS, river network distribution and DEM model, and acquires video frames, water quality images and pollution feature images through fixed high-definition cameras, bank image sampling equipment and UAV multispectral cameras respectively; The image processing module is used to extract the motion trajectory and distribution characteristics of floating objects, water color information and abnormal color areas, and pollution change trends and abrupt change characteristics in images over multiple time periods. The floating object monitoring and tracking module is used to monitor the density changes and velocity gradient changes of floating object targets in the fused image of the sub-river section in real time, calculate and obtain the surface debris diffusion disturbance coefficient FDI, and compare and analyze it with the first threshold Q1 to determine whether the overall state of surface debris in the sub-river section is stable. If it is unstable, a strategy is given. The floating object monitoring and tracking module includes a first computing unit and a first analysis unit; The first calculation unit is used to monitor in real time the density changes and velocity gradient changes of floating object boundaries in the fused image of the sub-river section. After dimensionless processing, the surface debris diffusion disturbance coefficient FDI is calculated and obtained, as follows: In the formula, This represents the set of velocity vectors of objects floating in an optical flow field. This represents the variance of the local velocity vector. This represents the change in the density gradient at the boundary of the floating object per unit time. This indicates the area of ​​the concentrated floating debris zone in real time. represents the average total area of ​​floating objects at time t in the past, and w1, w2 and w3 represent weighting coefficients; In the formula, N represents the number of pixels in the image, and represents the magnitude of the velocity vector of the floating object at the i-th pixel. This represents the average magnitude of the velocity vector of a floating object; In the formula, This represents the pollution attribute parameter value in the image at time t. This represents the pollution attribute parameter value in the image of the previous cycle at time t. Indicates the time interval between image frames; The water monitoring module is used to monitor the water quality of the current sub-river section in real time when it receives the first warning instruction. It calculates the water color difference anomalous coefficient SCP by combining historical water color standard data and compares it with the second threshold Q2 to determine whether the water color of the current sub-river section is normal. If it is abnormal, a strategy is given. The water monitoring module includes a second calculation unit and a second analysis unit; The second calculation unit is used to monitor the river water quality in the current sub-river section in real time when the first early warning command is received. After combining historical water color standard data and performing dimensionless processing, it calculates and obtains the water color difference anomaly coefficient SCP, as shown in the following formula: In the formula, This indicates the average color tone of the water body. Indicates the standard deviation of saturation. Indicates the average brightness. This represents the standard brightness reference value. These represent the chromatographic distribution difference values, where a1, a2, a3, and a4 represent weighting coefficients. The pollution source investigation module is used to analyze the river water quality segment by segment along the upstream direction of the river and calculate the rate of change of the pollution gradient when a second early warning instruction is received. And to find ways to calculate the rate of change of the pollution gradient. Sub-channel region with maximum value Identify the source of pollution and provide corresponding strategies; The pollution source investigation module includes a third calculation unit, a fourth calculation unit, and an analysis strategy unit; The third calculation unit is used to analyze the river water quality segment by segment along the upstream direction of the river when the second early warning command is received, and calculate the pollution gradient change rate after dimensionless processing. The formula is as follows: In the formula, This represents the water color difference anomaly coefficient in the j-th sub-river segment. Indicates the first The coefficient of water color difference anomaly in the individual river channel area, the first Each sub-channel region is the upstream adjacent sub-channel region of the j-th sub-channel region; The fourth calculation unit is used to obtain the pollution gradient change rate based on the calculation. In all continuous sub-channel regions, after dimensionless processing, the rate of change of pollution gradient is sought to calculate. Sub-channel region with maximum value The formula is as follows: In the formula, This indicates the rate of change of the pollution gradient.

2. The river management system based on image processing technology according to claim 1, characterized in that, The data acquisition module includes a region division unit and an acquisition unit; The regional division unit is used to establish a two-dimensional hydrological spatial coupling model of the river channel based on remote sensing geographic information (GIS) and urban river network distribution data, combined with digital elevation model (DEM), and to divide the treatment area into several sub-river segment regions, labeled as: Jcq1, Jcq2, ..., JcqM; M represents the number of sub-river segment regions; The acquisition unit is used to monitor the sub-river section area. It acquires continuous video frame sequences of the river by deploying fixed high-definition water surface monitoring cameras along the river; it acquires high-definition water quality image samples by deploying static image sampling equipment on the bank; and it acquires water surface pollution characteristic images and color characteristic images by using a drone equipped with a multispectral camera.

3. The river management system based on image processing technology according to claim 2, characterized in that, The image processing module includes an optical flow field processing unit, a water chromatography processing unit, and a pollution time sequence processing unit; The optical flow field processing unit is used to extract motion information from a continuous video image sequence of the river channel. It uses the optical flow estimation algorithm PWCNet to obtain the motion trajectory information of floating objects between images; extracts the velocity vector features and aggregation region boundaries of the floating objects to obtain the distribution status and movement trend map of the floating objects in the river channel; and extracts the image features of the density change of the floating objects in the region to obtain the aggregation area of ​​the floating objects in the current frame. The water chromatography processing unit is used to extract color information from static water surface images and uses the HSI color gamut conversion method to obtain the hue, saturation and brightness features in the image; By comparing the color difference maps of water bodies in different regions, abnormal color areas are extracted, and water surface color feature layer data is generated. The pollution time-series processing unit is used to dynamically extract pollution visual features from image sequences acquired at different time points; Analyze the changing trends of floating debris distribution maps, water color maps, and disturbance feature maps over time; identify abrupt change regions in the pollution process and extract key image parameters before and after the changes.

4. The river management system based on image processing technology according to claim 1, characterized in that, The first analysis unit is used to preset a first threshold Q1, and compare the surface debris diffusion disturbance coefficient FDI with the first threshold Q1 to obtain a first evaluation result, including: When the surface debris diffusion disturbance coefficient FDI ≤ the first threshold Q1, it indicates that the overall state of surface debris in the current sub-river section is stable, there are no signs of diffusion, and there is no risk of increased debris accumulation. Continuous monitoring is required. When the surface debris diffusion disturbance coefficient FDI > the first threshold Q1, it indicates that the overall state of surface debris in the current sub-river section is unstable, with signs of diffusion and a risk of increased debris accumulation. This triggers the first warning instruction and generates the first strategy: activate the floating debris interception equipment along the route and arrange unmanned vessels to prioritize path cleaning; adjust the flow velocity at the upstream drainage gate to alleviate flow velocity disturbance; and conduct water quality testing on the current sub-river section.

5. A river management system based on image processing technology according to claim 1, characterized in that, The second analysis unit is used to pre-set a second threshold Q2 and compare the water body color difference anomalous coefficient SCP with the second threshold Q2 to obtain the second evaluation result, including: When the water color difference anomalous coefficient SCP is less than the second threshold Q2, it indicates that the water color in the current sub-river section is normal and there are no pollution characteristics, and continuous monitoring is required. When the water color difference anomaly coefficient SCP is greater than or equal to the second threshold Q2, it indicates that the water color in the current sub-river section is abnormal and has pollution characteristics, triggering the second early warning instruction and generating the second strategy: to activate the upstream investigation mechanism to investigate the source of pollution.

6. The river management system based on image processing technology according to claim 1, characterized in that, The analysis strategy unit is used to measure the rate of change of the pollution gradient. Sub-channel region with maximum value The source of the pollution was identified, and the sub-river area was targeted. It calls up historical image sequences for comparison, performs retrospective detection, and automatically generates an "Upstream Inference Report on Industrial Pollution Discharge" to notify ecological regulatory units to prioritize its handling.

7. A method for river management based on image processing technology, comprising the river management system based on image processing technology as described in any one of claims 1 to 6, characterized in that, Includes the following steps: Step 1: Based on GIS, river network distribution and DEM model, the treatment area is divided into several sub-river sections, and video frames, water quality images and pollution feature images are collected by using fixed high-definition cameras, bank image sampling equipment and UAV multispectral cameras respectively; Step 2: Extract the motion trajectory and distribution characteristics of floating objects, water color information and abnormal color areas, and pollution change trends and abrupt change characteristics in images over multiple time periods; Step 3: Monitor the density changes and velocity gradient changes of floating object boundaries in the fused image of the sub-river section in real time, calculate the surface debris diffusion disturbance coefficient FDI, and compare it with the first threshold Q1 to determine whether the overall state of surface debris in the sub-river section is stable. If it is unstable, a strategy is given. Step 4: When the first warning instruction is received, monitor the water quality of the current sub-river section in real time, calculate the water color difference anomalous coefficient SCP by combining historical water color standard data, and compare it with the second threshold Q2 to determine whether the water color of the current sub-river section is normal. If it is abnormal, a strategy is given. Step 5: Upon receiving the second early warning instruction, analyze the river water quality segment by segment along the upstream direction of the river channel and calculate the rate of change of the pollution gradient. And to find ways to calculate the rate of change of the pollution gradient. Sub-channel region with maximum value They identified the source of the pollution and provided corresponding strategies.

Images