A method and system for evaluating base improvement construction

Abstract

The present invention relates to the field of data processing technology, and in particular to a method and system for evaluating construction of base improvement. The method comprises the following steps: collecting bottom mud images of water areas by image acquisition equipment to obtain low mud image data of water areas, and extracting organic mud features and inorganic mud features from the low mud image data of water areas to obtain organic mud feature data and inorganic mud feature data; performing bottom mud component correlation processing according to the organic mud feature data and inorganic mud feature data to obtain bottom mud component correlation data; predicting water quality change trends according to the bottom mud component correlation data to obtain water quality change trend data; performing ecosystem simulation and emulation according to the water quality change trend data to obtain ecosystem stability index data to perform base improvement construction evaluation operations. The present invention accurately evaluates the quality and improvement needs of the base by analyzing the composition and characteristics of the bottom mud.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for evaluating substrate improvement construction. Background Art

[0002] In the field of aquatic environmental management and protection, the quality of sediment plays a vital role in the stability of aquatic ecosystems and the maintenance of water quality. The organic and inorganic components in sediment have an important impact on processes such as eutrophication, biological growth, and pollutant transport in water bodies. In order to avoid secondary pollution of sediment, provide a good growth environment for aquatic plants and animals, and ensure the stability of the aquatic ecosystem, by analyzing the impact of sediment on water quality, studying substrate disinfection and substrate trace element ratio measures to create lake habitats and improve the substrate environment. Therefore, accurate analysis and evaluation of sediment composition is of great significance for the effective management of aquatic resources and improvement of water quality environment.

[0003] Traditional methods for analyzing sediment composition usually rely on field sampling and chemical analysis. This method has problems such as insufficient sampling, long data acquisition cycle, and high cost, which limits the monitoring and assessment of large areas of water. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a substrate improvement construction evaluation method and system to solve at least one of the above technical problems.

[0005] The present application provides a method for evaluating substrate improvement construction, the method comprising:

[0006] S1, collecting bottom mud images of the water area by an image acquisition device to obtain low mud image data of the water area, and extracting organic bottom mud features and inorganic bottom mud features from the low mud image data of the water area to obtain organic bottom mud feature data and inorganic bottom mud feature data;

[0007] S2, performing sediment component correlation processing according to the organic sediment characteristic data and the inorganic sediment characteristic data to obtain sediment component correlation data;

[0008] S3, predicting the water quality change trend according to the sediment component correlation data to obtain water quality change trend data;

[0009] S4. Perform ecosystem simulation and emulation based on the water quality change trend data to obtain ecosystem stability index data for base improvement construction assessment operations.

[0010] In the present invention, by collecting and extracting features from the sediment images of the water area, the composition and characteristics of the sediment can be deeply understood, so as to accurately evaluate the quality of the substrate and the need for improvement. The sediment component association processing can deeply analyze the correlation between the organic and inorganic components in the sediment, which helps to identify the source of the sediment, the degree of pollution and the impact on the water quality. Through the prediction of water quality change trends, the changing trend of water quality after sediment improvement can be predicted, including the improvement or decline of water quality, to provide guidance and early warning for improvement construction. Ecosystem simulation and emulation can simulate various dynamic changes of aquatic ecosystems, obtain ecosystem stability index data, and more accurately evaluate the impact of improvement construction on the stability of aquatic ecosystems. Based on the ecosystem stability index data, the substrate improvement construction can be scientifically guided to ensure that the impact of the construction process on the aquatic ecosystem is minimized and the ecological environment quality of the aquatic area is maximized.

[0011] Optionally, the present invention further provides a substrate improvement construction evaluation system for executing the substrate improvement construction evaluation method as described above, wherein the substrate improvement construction evaluation system comprises:

[0012] A water area low mud feature extraction module is used to collect water area bottom mud images through an image acquisition device to obtain water area low mud image data, and to perform organic bottom mud feature extraction and inorganic bottom mud feature extraction on the water area low mud image data to obtain organic bottom mud feature data and inorganic bottom mud feature data;

[0013] A sediment component correlation processing module is used to perform sediment component correlation processing according to the organic sediment characteristic data and the inorganic sediment characteristic data to obtain sediment component correlation data;

[0014] A water quality change trend prediction module is used to predict the water quality change trend according to the sediment component correlation data to obtain water quality change trend data;

[0015] The ecosystem simulation and emulation module is used to simulate and emulate the ecosystem based on the water quality change trend data to obtain ecosystem stability index data for base improvement construction assessment operations.

[0016] The purpose of the present invention is to extract organic and inorganic sediment features from water area sediment image data, so as to understand the composition and characteristics of water area sediment, help identify organic and inorganic components in sediment, and provide basic data for subsequent sediment component association processing. Sediment component association processing can reveal the association rules between various organic and inorganic components in sediment. By mining these association rules, the complexity and change trend of sediment components can be better understood. Based on sediment component association data, water quality change trend prediction can be carried out, and the change trend of water quality in water area can be effectively predicted, including the change direction, rate and influencing factors of water quality parameters, etc., providing a scientific basis for water area ecological environment management and protection, and showing the change and development of water area ecosystem. Ecosystem simulation and simulation are carried out using water quality change trend data to simulate the dynamic change process of water area ecosystem under different water quality conditions. By evaluating the stability index of the ecosystem, the impact of water quality changes on the ecosystem is reflected, and a reference is provided for the evaluation of base improvement construction. Ecosystem stability index data provides scientific guidance for base improvement construction. By comprehensively considering water quality change trends and ecosystem stability, targeted base improvement plans can be formulated to minimize the impact on the ecological environment, ensure the sustainability and ecological safety of construction, and contribute to protecting the natural environment and achieving harmonious coexistence between man and nature. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting implementations made with reference to the following drawings:

[0018] Figure 1 A flow chart showing the steps of a method for evaluating substrate improvement construction according to an embodiment is shown;

[0019] Figure 2 A flowchart showing a method for collecting sediment images in a water area according to an embodiment of the present invention is provided;

[0020] Figure 3 A flow chart showing the steps of a water area boundary extraction method according to an embodiment is shown;

[0021] Figure 4 A flowchart of the steps of a multi-dimensional water boundary data fusion method according to an embodiment is shown;

[0022] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0023] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0024] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0025] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0026] See also Figures 1 to 4 The present application provides a method for evaluating substrate improvement construction, the method comprising:

[0027] S1, collecting bottom mud images of the water area by an image acquisition device to obtain low mud image data of the water area, and extracting organic bottom mud features and inorganic bottom mud features from the low mud image data of the water area to obtain organic bottom mud feature data and inorganic bottom mud feature data;

[0028] Specifically, an underwater camera or a submersible is used to collect sediment images in the target waters, and the collected sediment image data is processed and analyzed to extract the characteristics of organic sediment and inorganic sediment, such as color, texture, particle size, etc., to obtain organic sediment characteristic data and inorganic sediment characteristic data.

[0029] S2, performing sediment component correlation processing according to the organic sediment characteristic data and the inorganic sediment characteristic data to obtain sediment component correlation data;

[0030] Specifically, the characteristic data of organic sediment and inorganic sediment are subjected to association analysis, such as using statistical methods or machine learning algorithms to mine the correlation and regularity between them, and obtain sediment component association data, including information such as the degree of association between them, frequency distribution, etc. The correlation coefficient between the features is calculated to understand the linear correlation between them, and the correlation coefficient includes the Pearson correlation coefficient, the Spearman correlation coefficient, etc. For variance analysis, it is possible to analyze whether there is a significant difference in the variance between the features to determine their contribution to the sediment components. It can also be carried out in the form of model analysis, such as model construction through existing historical data to establish a sediment component association analysis model, and the organic sediment characteristic data and the inorganic sediment characteristic data are subjected to sediment component association processing through the sediment component association analysis model to obtain sediment component association data. The construction steps of the model construction are specifically: obtaining historical organic sediment characteristic data, historical inorganic sediment characteristic data, and sediment component association relationship data; generating a decision tree according to the historical organic sediment characteristic data, the historical inorganic sediment characteristic data, and the sediment component association relationship data to obtain a sediment component association analysis model.

[0031] S3, predicting the water quality change trend according to the sediment component correlation data to obtain water quality change trend data;

[0032] Specifically, based on the existing sediment component correlation data, a water quality change trend prediction model is established. For example, time series analysis, regression analysis or neural network methods are used for prediction to obtain trend data of water quality changes in the future, including information such as the direction of change of water quality parameters and trend amplitude.

[0033] Specifically, in another embodiment, the water quality index is used to comprehensively evaluate the pollution degree and change trend of the water body. The water quality index is calculated based on different water quality parameters and weights, and is a single index or a combination of multiple indexes. The monitoring data is analyzed based on the change trend of the water quality index to predict the future change trend of water quality. For example, according to the importance and influence of the water quality parameters, the corresponding weight is determined for each parameter. For example, dissolved oxygen has a greater impact on the ecological health of the water body and is given a higher weight; while total phosphorus mainly affects the eutrophication degree of the water body and is given a lower weight. The water quality index is calculated based on the selected water quality parameters and the corresponding weights. A simple weighted summation method can be used, or a more realistic calculation formula can be used based on the concentration of different parameters and water quality standards. The calculation result of the water quality index will reflect the comprehensive pollution degree of the water body.

[0034] have water quality parameters, denoted as X1, X2, X3, ..., X n , their concentrations are C1, C2, C3, …, C nThe corresponding water quality standard values ​​are S1, S2, S3, ..., S n . The water quality parameter X i and the corresponding water quality standard value S i The difference between them is taken as the predictor variable, namely Δ i =C i -S i , indicating the parameter X i According to the interaction between parameters, a multivariate linear regression model is constructed to predict the water quality index. The form of the model is as follows: ; … is the weighted coefficient corresponding to the deviation degree of different water quality parameters, is the initial value of the water quality index.

[0035] Specifically, in another embodiment, the water quality change trend is modeled and predicted using the statistical characteristics of time series data, such as mean, variance, trend, periodicity, etc. Traditional time series analysis methods, such as moving average, exponential smoothing, ARIMA model, etc., are used to analyze the historical change patterns of water quality data and predict future water quality change trends.

[0036] Specifically, in another embodiment, a water quality model is established to simulate the pollution diffusion and change process of the water body. The water quality model can simulate the transmission, transformation and diffusion of pollutants in the water body based on physical principles or empirical formulas, combined with hydrological, hydrodynamic and other parameters. By simulating the impact of different pollution sources and changes in hydrological conditions, the trend of water quality changes in the future can be predicted. The parameters of the empirical formula are generated by consulting or querying the hydrological data or information stored in the database for parameter extraction.

[0037] S4. Perform ecosystem simulation and emulation based on the water quality change trend data to obtain ecosystem stability index data for base improvement construction assessment operations.

[0038] Specifically, using water quality change trend data as input parameters, combined with existing ecosystem models and simulation software, ecosystem simulation and simulation analysis are carried out. By simulating the impact of water quality changes on the ecosystem, the impact of different base improvement schemes on ecosystem stability is evaluated, and finally ecosystem stability index data is obtained, providing a scientific basis for base improvement construction evaluation.

[0039] Specifically, according to the data of ecological characteristics, topography, environmental parameters and other data of the water area, an ecosystem model is established, including the description of ecological processes such as aquatic vegetation distribution, aquatic animal community structure, and nutrient circulation. The water quality change trend data is integrated with the ecosystem model as the input parameters of the simulation, including the concentration change of water quality parameters, the change trend of water quality index and other information. In the simulation process, different substrate improvement schemes are simulated according to the water quality change trend data, including measures such as changing the bottom of the water body, vegetation planting, and bioremediation. Through simulation, the impact of different substrate improvement schemes on the ecosystem is evaluated, including changes in the degree of water quality improvement, biodiversity, and changes in ecosystem structure and function. According to the simulation results, the stability indicators of the ecosystem are calculated, such as the stable state, tolerance, and recovery capacity of the ecosystem. These indicators can reflect the response capacity and stability of the ecosystem to external interference. The obtained ecosystem stability index data are compared and analyzed with the effects of different substrate improvement schemes to provide a scientific basis for the evaluation of substrate improvement construction.

[0040] In the present invention, by collecting and extracting features from the sediment images of the water area, the composition and characteristics of the sediment can be deeply understood, so as to accurately evaluate the quality of the substrate and the need for improvement. The sediment component association processing can deeply analyze the correlation between the organic and inorganic components in the sediment, which helps to identify the source of the sediment, the degree of pollution and the impact on the water quality. Through the prediction of water quality change trends, the changing trend of water quality after sediment improvement can be predicted, including the improvement or decline of water quality, to provide guidance and early warning for improvement construction. Ecosystem simulation and emulation can simulate various dynamic changes of aquatic ecosystems, obtain ecosystem stability index data, and more accurately evaluate the impact of improvement construction on the stability of aquatic ecosystems. Based on the ecosystem stability index data, the substrate improvement construction can be scientifically guided to ensure that the impact of the construction process on the aquatic ecosystem is minimized and the ecological environment quality of the aquatic area is maximized.

[0041] Optionally, the collecting of bottom mud images of the water area by an image collection device to obtain low mud image data of the water area includes:

[0042] S11, collecting water area images through GIS equipment to obtain water area image data;

[0043] Specifically, aerial photography drones or satellite remote sensing technology are used to collect high-resolution images of the target water area to obtain image data of the water area.

[0044] S12, identifying the main body of the water area according to the water area image data to obtain the main body data of the water area;

[0045] Specifically, the image of the water area is analyzed and identified by using image processing and computer vision technology, and the main targets or features in the water area, such as water bodies, land, buildings, etc., are extracted to obtain the main data of the water area. The water area image is preprocessed, including denoising, grayscale, image enhancement and other operations. Image processing technology and feature extraction algorithm are used to extract representative features from the preprocessed image, including color, texture, shape, etc. Different areas or targets in the image are segmented, and targets such as water bodies, land, buildings, etc. are distinguished. The segmentation methods include threshold segmentation, edge detection, region growing, etc. The segmented image areas are identified and classified by using machine learning algorithms, deep learning models or traditional image classification algorithms (generally preset neural network models), and they are classified into different main categories such as water bodies, land, buildings, etc. The main data of the water area after identification and classification is output in a structured form, such as pixel coordinates, bounding box information, category labels, etc., for subsequent processing and analysis.

[0046] The specific steps of recognition and classification are as follows: extract features from the segmented image area, including color histogram, texture features (such as LBP, GLCM), shape features (such as edge detection results), corner features, etc. Represent the extracted features in vector form, and generally concatenate the features of each image area into a feature vector. Preprocess the extracted features, including normalization, standardization, or other data conversion methods. Select a specific machine learning classifier, such as support vector machine (SVM), K-nearest neighbor (K-NN), decision tree, etc. Use a labeled training data set to train the selected classifier. During the training process, the classifier will learn how to correctly classify the image area into main categories such as water, land, and buildings based on the provided image features. Use the validation data set to evaluate the trained classifier to evaluate its classification performance. Evaluation indicators include accuracy, recall, F1 score, etc.

[0047] The specific process of K-NN is as follows: the input training data set contains image regions and feature vectors that have been labeled with categories. During the training process, the distance or similarity between each sample needs to be calculated. After the training is completed, the K-NN algorithm will obtain a model for classification. The specific training process is as follows: select the K value (number of neighbors) and determine the distance measurement method (such as Euclidean distance, Manhattan distance, etc.). For each training sample, calculate the distance or similarity between it and all other samples, and calculate it using the selected distance measurement method. For each sample, according to the calculated distance or similarity, select the K samples closest to it as its neighbors. For each sample to be classified, vote according to the categories of its K nearest neighbors, select the category with the most occurrences among the K neighbors as the category of the sample, and use the category obtained by voting as the predicted category of the sample. The performance of the algorithm can be optimized by adjusting hyperparameters such as the K value and the distance measurement method.

[0048] S13, extracting water area boundaries from the water area image data according to the water area main body data to obtain water area boundary data;

[0049] Specifically, based on the main data of the water area, the boundary information of the water area is extracted from the water area image using image processing and edge detection algorithms to obtain the water area boundary data. For example, the water area image is preprocessed, including denoising, grayscale, image enhancement and other operations. Edge detection algorithms such as Sobel, Canny, Prewitt, etc. are used to detect the edge of the preprocessed image, and the areas with large grayscale changes in the image are identified, which will generate a series of edge points. The detected edge points are connected to connect adjacent edge points into line segments or curves, such as Hough Transform. The connected edges are screened, and the line segments or curves related to the water boundary are retained. The screening is based on the length, direction, shape and other features to exclude the interference of non-water boundary. The boundary of the water area is determined based on the screened edge segments or curves, and further processing such as closed curves and filled areas is performed to obtain complete water boundary information.

[0050] S14, performing topographic analysis and environmental parameter extraction according to the water area boundary data to obtain topographic data and environmental parameter data;

[0051] Specifically, geographic information system (GIS) tools are used to conduct topographic analysis of water boundary data, including the extraction of parameters such as water depth and terrain elevation. At the same time, environmental parameters such as water quality, temperature, salinity and other data are extracted based on the water area location and topographic characteristics.

[0052] S15. Determine the deployment position according to the water area boundary data, the topographic data and the environmental parameter data to obtain the deployment position data, and collect the bottom mud image of the water area through the image acquisition device to obtain the low mud image data of the water area.

[0053] Specifically, the optimal deployment location of the image acquisition equipment is determined based on the water area boundary, topography and environmental parameter data to ensure comprehensive and accurate collection of water area bottom mud image data, thereby obtaining water area low-mud image data. The equipment includes underwater cameras, submersibles, etc.

[0054] The present invention collects water area image data through GIS equipment, and can accurately obtain image information of the water area, including the scope, shape and surrounding environment of the water area. By performing water area subject identification and water area boundary extraction on the water area image data, the boundary of the water area can be accurately identified, and the specific scope and position of the water area can be determined. According to the water area boundary data, topographic and geomorphic analysis and environmental parameter extraction can be performed, and the topographic and geomorphic characteristics of the water area and the surrounding environmental parameters can be reasonably analyzed, providing a basis for the subsequent deployment position determination. In combination with the water area boundary data, topographic and geomorphic data and environmental parameter data, the best deployment position can be scientifically determined to ensure that the image acquisition device can effectively cover the water area and obtain accurate sediment image data. By analyzing the water area information and environmental conditions and determining the appropriate deployment position, the data acquisition efficiency and accuracy of the image acquisition device can be improved, and high-quality data support can be provided for the subsequent sediment feature extraction.

[0055] Optionally, extracting water area boundaries from the water area image data according to the water area main body data to obtain water area boundary data includes:

[0056] S131, extracting edge water area boundaries from the water area image data according to the water area main body data to obtain first water area boundary data;

[0057] Specifically, the edge detection algorithm is used to extract the edge part, that is, the boundary information of the water area, from the water area image to obtain the first water area boundary data. For example, the water area image is preprocessed, including denoising, grayscale, image enhancement and other operations. Edge detection algorithms such as Sobel, Canny, Prewitt, etc. are used to perform edge detection on the preprocessed image to identify the area with large grayscale changes in the image, which usually generates a series of edge points. The detected edge points are connected to connect adjacent edge points into line segments or curves, such as Hough Transform. The connected edges are screened, and the line segments or curves related to the water area boundary are retained. The screening is performed based on the length, direction, shape and other features to exclude the interference of non-water area boundaries. The boundary of the water area is determined based on the screened edge segments or curves, and further processing such as closing curves and filling areas is performed to obtain complete water area boundary information.

[0058] More importantly, edge point extraction is performed on the water area image data according to the water area main body data to obtain water area edge point data; edge connection is performed on the water area edge point data to obtain water area edge connection data; similarity screening is performed on the water area edge connection data to obtain first water area boundary data; wherein the similarity screening is performed by calculating the water area similarity calculation formula to obtain the water area similarity index for screening, and the water area similarity calculation formula is specifically:

[0059] ;

[0060] is the water area similarity index, is the sequence item data corresponding to the edge connection data of the water area, is the quantity data corresponding to the edge connection data of the water area, is similarity weighted data, is the similarity index of the water area edge connection data and the water area main data. , calculate its relationship with the main data of the water area Similarity index This similarity index can be defined based on specific features and measurement methods, such as the distance or similarity measurement between feature vectors. Then, each similarity index is multiplied by its corresponding weighting coefficient. , and sum all items to get the total similarity index . Each water area edge connection data There is a corresponding similarity weight coefficient These weights are determined based on their importance in the similarity calculation. For example, if certain edge connection data are more critical in identifying water boundaries, they can be given higher weights. Once the overall similarity index is calculated , set a threshold to filter out the water area edge connection data that meets the requirements. For example, if the similarity between the boundary connection data and the preset main data is required to reach a certain level, then set a corresponding threshold. All boundary connection data with a similarity index higher than the threshold will be retained to form the first water area boundary data.

[0061] S132, performing hydrodynamic simulation processing according to the main body data of the water area to obtain water area hydrodynamic simulation data;

[0062] Specifically, based on the main data of the water area, numerical simulation methods (such as computational fluid dynamics models) are used to simulate and calculate the hydrodynamic characteristics of the water flow, waves, etc. in the water area to obtain the hydrodynamic simulation data of the water area. For example, the water flow in the water area is simulated by hydrodynamic simulation software (such as MIKE model, Delft3D, etc.), including parameters such as water flow speed, water flow direction, and water depth. These simulation data are obtained through numerical calculations and reflect the flow of water bodies at different locations.

[0063] S133, extracting a water area boundary from the water area image data according to the water area hydrodynamic simulation data to obtain second water area boundary data;

[0064] Specifically, the hydrodynamic simulation data is used to determine the areas with higher flow velocity in the water area, which are located near the boundary of the water body. The hydrodynamic simulation data can be combined with the water area image data using image processing technology to extract the areas with higher water velocity or shallower water depth, thereby determining the boundary position of the water area, thereby obtaining the water area boundary information based on the hydrodynamic simulation data. These boundary data are a series of coordinate points or boundary line segments, representing the boundary contour of the water area, and obtaining the second water area boundary data.

[0065] A hydrodynamic simulation of a lake was performed to obtain water flow velocity data at different locations in the lake. The boundary position of the lake was determined based on the change in water flow velocity. Generally speaking, the water flow velocity at the boundary of the lake will be larger because the lake is in contact with the external environment and the water flow velocity is greatly affected by external factors. Using this feature and combining image processing technology, the areas with larger water flow velocity are extracted from the water area image, and then the boundary position of the lake is determined based on the distribution of these areas. Finally, these boundary positions are sorted into boundary segments or coordinate points to obtain the second water area boundary data.

[0066] The specific steps of extracting the area with higher water velocity from the water area image are as follows: aligning the hydrodynamic simulation data with the water area image to ensure that the two correspond accurately in spatial position. Preset a threshold value, and mark the pixel points in the hydrodynamic simulation data where the water velocity exceeds the threshold as the area with higher water velocity. Starting from a seed point in the water area image, according to certain growth criteria (such as the similarity of water velocity of adjacent pixels), the region is gradually grown until the condition of stopping growth is met to form an area with higher water velocity. Use edge detection algorithms (such as Canny operator, Sobel operator, etc.) to detect the edges with higher water velocity in the water area image, so as to obtain the area with higher water velocity. Post-process the extracted area with higher water velocity, including removing discontinuous areas, filling holes, smoothing boundaries, etc., to extract the area with higher water velocity.

[0067] S134, performing boundary comparison on the first water area boundary data and the second water area boundary data to obtain boundary comparison data;

[0068] Specifically, the first water area boundary data and the second water area boundary data are compared to find out the differences and overlaps between them and obtain boundary comparison data.

[0069] S135, when it is determined that the boundary comparison data is boundary overlap data, boundary data fusion is performed on the boundary comparison data corresponding to the boundary overlap data to obtain water area boundary data;

[0070] Specifically, if it is found that there are overlapping parts in the boundary comparison data, that is, boundary overlapping data, then these data are fused to obtain the final water area boundary data.

[0071] S136. When it is determined that the boundary comparison data is boundary overlap data, the first water area boundary data corresponding to the boundary overlap data is corrected according to the second water area boundary data corresponding to the boundary overlap data to obtain the water area boundary data; or, historical water area boundary data is obtained, and boundary correction is performed on the second water area boundary data corresponding to the boundary overlap data and the first water area boundary data corresponding to the boundary overlap data according to the historical water area boundary data to obtain the water area boundary data.

[0072] Specifically, the boundary is corrected based on the boundary overlap data and combined with the historical water boundary data to obtain more accurate water boundary data. For example, for the overlapping part, the boundary is corrected in combination with the historical water boundary data. Some techniques, such as linear interpolation and curve fitting, are used to adjust the position and shape of the boundary to make it more consistent with the historical data.

[0073] The overlapping part is corrected according to the second water boundary data, such as comparing the two sets of data and adjusting the first water boundary data to make it consistent or close to the second water boundary data. For example, the overlapping part of the first water boundary data is adjusted to be consistent with the second water boundary data by using interpolation or curve fitting technology, which is completed by mathematical modeling or geographic information system software. Alternatively, according to the shape, direction and length of the second water boundary data, the first water boundary data is partially adjusted to make it closer to the second water boundary data.

[0074] In the present invention, the boundary of the water area is extracted by the main data of the water area, and the boundary information of the water area can be accurately obtained to ensure the accuracy and reliability of subsequent processing. The water boundary is extracted by using hydrodynamic simulation data, which can take into account the dynamics and changes of water flow, so as to more accurately depict the water boundary. By processing the boundary overlapping data, the boundary information of different data sources can be fused to improve the integrity and accuracy of the boundary data. By correcting the boundary data, the water boundary can be further improved and corrected according to the evolution process and historical data of the water boundary to ensure that it is more in line with the actual situation. Comprehensively considering multiple data sources and historical evolution of the water boundary, the credibility and reliability of the water boundary data can be improved, and more reliable data support can be provided for subsequent water management and planning. Through boundary comparison, fusion and correction, the boundary error and uncertainty between different data sources can be reduced, the consistency and stability of the water boundary data can be improved, and more reliable basic data can be provided for related decision-making and application. Traditional methods often only use one data source or method for boundary extraction, and cannot fully obtain the boundary information of the water area. For example, the satellite image-based method is affected by weather conditions, cloud cover, etc. and cannot obtain clear boundary information, while the hydrological data-based method ignores the impact of important factors such as topography on the boundary. The present invention uses the main data of the water area and the hydrodynamic simulation data combined with the image data to extract the boundary, making full use of the information of multiple data sources to improve the accuracy and completeness of the boundary extraction.

[0075] Optionally, performing boundary data fusion on the boundary comparison data corresponding to the boundary overlap data to obtain water area boundary data includes:

[0076] S1351, obtaining historical boundary data;

[0077] Specifically, previous water boundary data is obtained through historical records, satellite images or other reliable data sources as a reference.

[0078] S1352, performing boundary fusion according to the historical boundary data and the boundary comparison data corresponding to the boundary overlap data to obtain first water area boundary data;

[0079] Specifically, the boundary comparison data corresponding to the historical boundary data and the boundary overlap data are combined to merge and correct the boundary to obtain the first water area boundary data. For example, for the boundary overlap part, the average value or weighted average value method is used to merge the two boundaries to balance the contribution of the historical boundary data and the boundary overlap data.

[0080] S1353, performing sediment particle movement analysis according to the water area hydrodynamic simulation data to obtain sediment particle movement data;

[0081] Specifically, the movement of sediment particles in the water area is analyzed and simulated by numerical simulation methods or experimental tests using hydrodynamic simulation data to obtain sediment particle movement data. Obtain hydrodynamic simulation data of the water area, including information such as flow velocity, flow direction, and turbulence intensity. Prepare characteristic data of sediment particles, such as particle size, density, and shape, which can be obtained through established literature or preset databases. Based on the hydrodynamic simulation data and the characteristics of sediment particles, a numerical simulation model for the movement of sediment particles is established, using fluid mechanics models (such as the Reynolds average Navier-Stokes equation) and particle dynamics models (such as the Diskrete Element Method). The hydrodynamic simulation data is used as input, combined with the established numerical simulation model, to simulate the movement of sediment particles in the water area. In the simulation, the impact, friction, suspension, and other effects of water flow on sediment particles, as well as the interaction between particles, are considered. Analyze the simulation results to obtain data such as the movement trajectory, velocity, and force of sediment particles. Output the movement data of sediment particles, such as the movement path diagram of particles and the velocity distribution diagram.

[0082] S1354, performing boundary data fusion on the boundary comparison data corresponding to the boundary overlap data according to the bottom mud particle movement data to obtain second water area boundary data;

[0083] Specifically, the boundary data is evolved according to the movement data of the sediment particles to obtain the boundary evolution data, and the boundary data is weighted and fused according to the boundary evolution data and the boundary comparison data to obtain the second water area boundary data. The evolution of the water area boundary is analyzed using the sediment particle movement data. The movement of the sediment particles is affected by factors such as water flow and bottom quality, which leads to changes in the water area boundary. By analyzing the position and movement direction of the sediment particles at different times, the evolution of the water area boundary over time is inferred. The weight of the boundary data is determined according to the boundary evolution data and the boundary comparison data. The boundary evolution data reflects the changing trend of the water area boundary over time, while the boundary comparison data provides comparison information with historical data. The credibility and importance of the boundary data can be determined more accurately by weighted fusion of these two aspects of information. According to the boundary evolution data and the boundary comparison data, a weight is assigned to each boundary data point. The weight can be determined according to factors such as the stability and change trend of the data. For example, a higher weight can be assigned to an area with obvious boundary evolution. Then, the weight is applied to the boundary comparison data, and the boundary data of different time periods or different data sources are weighted averaged or weighted fused according to the weight to obtain the final second water area boundary data.

[0084] S1355. Perform water area boundary evolution processing according to the first water area boundary data and the second water area boundary data to obtain water area boundary data.

[0085] Specifically, the first water area boundary data and the second water area boundary data are combined to perform boundary evolution simulation or analysis to determine the final form and characteristics of the water area boundary and obtain water area boundary data. The first water area boundary data and the second water area boundary data are aligned and matched, and the two sets of data are spatially corresponded to find boundary data points in the same area. The evolution of the water area boundary is processed using a simulation method or an analysis method, such as by hydrodynamic simulation, weight distribution of the first water area boundary data and the second water area boundary data or boundary evolution simulation calculation is performed, and the calculation process uses a mathematical model based on the hydrodynamic equation to perform evolution simulation. According to the results of the boundary evolution simulation or analysis, the water area boundary data is updated, including adjusting the position of the boundary line and correcting the coordinates of the boundary points to obtain the water area boundary data.

[0086] The present invention combines historical boundary data and hydrodynamic simulation data, can make full use of past water area information and current simulation results, and provide multi-faceted information support for the fusion of water area boundary data. Comprehensively considering historical data and simulation results can reduce the uncertainty of data sources and improve the credibility and reliability of water area boundary data. Through the evolution analysis of water area boundary data, the dynamic change process of water area boundary can be understood, which helps to understand the evolution law and environmental change trend of water area, and provides a reference basis for water area management and protection. The analysis of sediment particle movement data can help understand the distribution and movement law of water area sediment, and combined with boundary data fusion, it can better understand the geomorphic characteristics and sediment distribution of water area. Through the evolution analysis of water area boundary data, the boundary data can have better spatiotemporal consistency, conform to the actual situation, and provide a more reliable data basis for water area management and planning. Conventional methods for processing boundary overlapping data often simply superimpose data or select boundaries by simple rules, lack of in-depth analysis and processing of data, and easily lead to inaccuracy and instability of extraction results. The present invention makes full use of the influence of hydrodynamic simulation data on the boundary, and further improves the accuracy of boundary extraction by analyzing the influence of hydrodynamic simulation data on the movement of sediment particles.

[0087] Optionally, the step of extracting organic sediment features and inorganic sediment features from the water area low-mud image data to obtain organic sediment feature data and inorganic sediment feature data includes:

[0088] Performing color feature extraction and texture feature extraction on the water area low-mud image data to obtain color feature data and texture feature data;

[0089] Specifically, image processing technology is used to perform color analysis and texture extraction on low-mud images of water areas to extract color features and texture feature data, such as extracting color features such as color histograms and color matrices, and extracting texture feature data using filters, texture algorithms and other methods.

[0090] Extracting sediment particle morphology from the low-mud image data of the water area to obtain sediment particle morphology data;

[0091] Specifically, the sediment particles in the image are separated from the background through techniques such as image segmentation and morphological processing, and then the morphological characteristics of the sediment particles, such as particle size, shape, distribution density, etc., are extracted to obtain the sediment particle morphological data.

[0092] The sediment characteristic type is determined based on the sediment particle morphology data, the color characteristic data and the texture characteristic data according to the water area hydrodynamic simulation data, so as to obtain organic sediment characteristic data and inorganic sediment characteristic data.

[0093] Specifically, combined with the hydrodynamic simulation data, the machine learning algorithm or expert experience rules are used to analyze and synthesize the sediment particle morphological data, color feature data and texture feature data to determine the characteristic types of organic sediment and inorganic sediment. For example, organic sediment has specific color and texture characteristics, while inorganic sediment has different morphological characteristics and color characteristics. Such as morphological data: size and shape of sediment particles; color feature data: color information of sediment particles; texture feature data: texture information of sediment particles; hydrodynamic simulation data: water flow speed, flow direction, etc. Through the preset machine learning model, the organic sediment feature data and the inorganic sediment feature data are obtained by screening. The preset machine learning model includes support vector machine (SVM), random forest, deep neural network, etc., and the preset machine learning model is obtained by training the model with the labeled data set. Alternatively, the rule base is constructed using expert experience rules, and the rules for distinguishing organic sediment and inorganic sediment are formulated according to the features such as the morphology, color and texture of the sediment particles. For example, when the flow rate is relatively fast and shallow, the color characteristics will be distributed in light blue or light green, the bottom mud will be mostly floating, and the bottom mud texture will show a diffusion from the slow flow end to the fast flow end.

[0094] The present invention can obtain the characteristic information of the water area sediment, including the characteristics of color distribution, texture structure and sediment particle morphology, etc., to reflect the deep structural data of the water area sediment, so as to provide accurate and reliable data support for the feature division according to the depth data of the water area sediment. Combined with the low mud image data of the water area and the hydrodynamic simulation data, the multi-source data information can be fully utilized to improve the grasp and accuracy of the sediment characteristics. Determined by the sediment feature type, the sediment feature data can be divided into organic sediment feature data and inorganic sediment feature data, and the sediment components can be further distinguished, providing different types of data support for subsequent water quality change trend prediction and ecosystem simulation. The acquisition of organic sediment feature data and inorganic sediment feature data is helpful to accurately evaluate the sediment situation of the water area, and then predict the water quality change trend and the stability of the simulated ecosystem, and provide accurate data support for the evaluation of substrate improvement construction. Combined with the analysis of sediment feature data, the management and protection of the water area can be scientifically guided, and a scientific basis is provided for formulating effective protection measures and management strategies.

[0095] Optionally, determining the sediment feature type of the sediment particle morphology data, the color feature data and the texture feature data according to the water area hydrodynamic simulation data to obtain organic sediment feature data and inorganic sediment feature data includes:

[0096] Performing sediment particle characteristic type analysis on the sediment particle morphology data according to the water area hydrodynamic simulation data to obtain sediment particle characteristic type data;

[0097] Specifically, the hydrodynamic simulation data is used to analyze the movement of the water flow on the sediment particles, and then the characteristic types of the sediment particles, such as the settling velocity and suspension concentration of the particles, are determined to obtain the characteristic type data of the sediment particles. For example, the hydrodynamic simulation data of the water area, including information such as water flow velocity, flow direction, and flow velocity gradient, is obtained through numerical simulation methods (such as computational fluid dynamics simulation) or experimental tests. The hydrodynamic simulation data is used to analyze the movement of the water flow on the sediment particles, including parameters such as the settling velocity, suspension concentration, and transport distance of the sediment particles in the water flow. By analyzing the flow velocity and flow direction of the water flow, as well as the density, particle size and other characteristics of the sediment particles, the behavior of the sediment particles in the water flow is inferred. Based on the movement of the sediment particles and the hydrodynamic simulation data, the characteristic types of the sediment particles are determined. For example, according to parameters such as the settling velocity and suspension concentration of the sediment particles, the sediment particles are divided into different types, such as sedimentary sediment and suspended sediment.

[0098] Performing color feature data analysis and texture feature analysis according to the water area hydrodynamic simulation data to obtain color feature analysis data and texture feature analysis data;

[0099] Specifically, based on the hydrodynamic simulation results, the color distribution and texture feature changes of the sediment particles in the water area are analyzed, such as the influence of water flow velocity and direction on color and texture, to obtain color feature analysis data and texture feature analysis data.

[0100] The color feature data and the texture feature data are screened according to the color feature analysis data and the texture feature analysis data to obtain primary organic sediment feature data and primary inorganic sediment feature data;

[0101] Specifically, according to the results of color feature analysis and texture feature analysis, important features related to hydrodynamic simulation, such as color and texture features related to changes in water flow velocity, are screened out to obtain primary organic sediment feature data and primary inorganic sediment feature data. The color feature data and texture feature data are analyzed to understand the changing trend of color changes and texture features of sediment particles in hydrodynamic simulation. By analyzing the correlation between the hydrodynamic simulation data and these feature data, it is determined which color and texture features are related to changes in water flow velocity. According to the results of color feature analysis and texture feature analysis, important features related to hydrodynamic simulation are selected. Features include color brightness, saturation, and texture uniformity, directionality, etc. The screened features will become candidate sets of primary organic sediment feature data and primary inorganic sediment feature data. According to the screened feature set, corresponding feature values ​​are extracted from the original color feature data and texture feature data, and these feature values ​​will constitute primary organic sediment feature data and primary inorganic sediment feature data.

[0102] Sediment characteristics are generated for the sediment particle characteristic type data, the primary organic sediment characteristic data and the primary inorganic sediment characteristic data to obtain organic sediment characteristic data and inorganic sediment characteristic data.

[0103] Specifically, comprehensive analysis and processing are carried out in combination with the sediment particle characteristic types and primary characteristic data to generate complete organic sediment characteristic data and inorganic sediment characteristic data for subsequent analysis and evaluation.

[0104] In the present invention, combined with hydrodynamic simulation data, the morphology, color characteristics and texture characteristics of the sediment particles are analyzed, and multiple characteristics of the sediment can be considered to improve the comprehensiveness and accuracy of the sediment characteristics. The type of sediment characteristics can be determined using hydrodynamic simulation data, and the scientificity and accuracy of the analysis and judgment of the sediment characteristics can be improved based on scientific simulation data. The characteristic data of the sediment particles are screened according to the color characteristic analysis data and the texture characteristic analysis data, and the characteristics of organic sediment and inorganic sediment can be preliminarily determined. By comprehensively considering the hydrodynamic simulation data and the sediment characteristic data, the credibility and accuracy of the sediment characteristic data can be improved. The acquisition of organic sediment characteristic data and inorganic sediment characteristic data can provide important data support for the evaluation of substrate improvement construction and guide the implementation and evaluation of substrate improvement work.

[0105] Optionally, performing sediment component association processing according to the organic sediment characteristic data and the inorganic sediment characteristic data to obtain sediment component association data includes:

[0106] Performing association rule mining based on the organic sediment characteristic data and the inorganic sediment characteristic data to obtain association rule data;

[0107] Specifically, data mining technology is used to analyze the organic sediment characteristic data and the inorganic sediment characteristic data to find out the association rules between them. For example, through the association analysis algorithm (such as the Apriori algorithm), the frequent item sets and association rules between the organic sediment characteristic data and the inorganic sediment characteristic data are mined to find the correlation and regularity between them.

[0108] A sediment component association network is constructed according to the association rule data to obtain sediment component association data.

[0109] Specifically, based on the association rule data obtained by mining, a sediment component association network is constructed, in which nodes represent sediment feature data and edges represent the association relationship between them. For example, a graph theory method is used to construct an association network to more intuitively display the association relationship and complexity between sediment components.

[0110] In the present invention, by mining the association rules of organic sediment characteristic data and inorganic sediment characteristic data, the association rules and potential relationships between sediment components can be revealed, which is helpful to understand the mutual influence and action mechanism between sediment components. Based on the association rule data mined, a sediment component association network is constructed, which can intuitively show the relationship network between sediment components, and is helpful to comprehensively analyze the complex characteristics and mutual relationships of sediment. The acquisition of sediment component association data can provide a basis for the prediction of water quality change trends. The composition and characteristics of sediment are closely related to water quality. By analyzing the association relationship of sediment components, the change trend of water quality can be more accurately predicted. Sediment component association data provides important input parameters for ecosystem simulation and simulation. Understanding the association relationship between sediment components can better simulate the dynamic changes of ecosystems in water areas, and provide a scientific basis for ecological environment protection and management. The acquisition of sediment component association data helps to evaluate the effect and impact of substrate improvement construction. By analyzing the association relationship of sediment components, the impact of substrate improvement on the ecological environment of water areas can be more comprehensively evaluated, providing a basis for the formulation and adjustment of construction plans.

[0111] Optionally, the predicting of water quality change trend according to the sediment component correlation data to obtain water quality change trend data includes:

[0112] Using a preset water quality change trend prediction model to predict the water quality change trend of the sediment component correlation data, to obtain water quality change trend data;

[0113] The steps for constructing the preset water quality change trend prediction model are specifically as follows:

[0114] Obtain historical sediment composition association data and corresponding water quality label data;

[0115] Specifically, the sediment component-related data during the historical period are collected, including organic sediment characteristic data and inorganic sediment characteristic data, and the corresponding water quality label data are recorded, such as the pH value, dissolved oxygen content, turbidity, etc. of the water body.

[0116] Performing time series processing on the historical sediment component associated data to obtain sediment component associated time series data;

[0117] Specifically, according to the concept of time series, the historical sediment component correlation data are processed into time series, and the sediment component correlation data are arranged in chronological order to form a time series data set.

[0118] Performing convolution calculation on the sediment component-related time series data to obtain sediment component-related convolution layer data;

[0119] Specifically, the preset convolution layer data is used to perform convolution calculations on the time series data, extract the spatial features and time features in the data, and generate sediment component-associated convolution layer data. Alternatively, water quality convolution layer data is generated based on the water quality label data, and convolution calculations are performed on the sediment component-associated time series data based on the water quality convolution layer data to obtain sediment component-associated convolution layer data; for example, if the water quality label data reflects that the water quality is relatively clean, a larger convolution layer (size is 5X5, offset is 2, no padding) is generated for processing to capture a wider range of features; if the water quality label data reflects that the water quality is relatively turbid, a smaller convolution layer (size is 3X3, offset is 1, padding) is generated for processing to focus on local features;

[0120] Performing self-attention calculation on the sediment component associated convolutional layer data to obtain sediment component associated self-attention data;

[0121] Specifically, the sediment component-associated convolutional layer data is passed as input to the self-attention layer (including the calculation of query, key and value, and the weighted summation of the calculated attention weight and value). The self-attention layer calculates the attention weight between each data point and other data points. According to the calculated attention weight, the sediment component-associated convolutional layer data is weighted and the weighted summation result is obtained. The weighted summation result is the sediment component-associated self-attention data, which highlights the important features and associations in the data.

[0122] Performing pooling layer processing on the sediment component associated self-attention data to obtain sediment component associated pooling layer data;

[0123] Specifically, pooling operations (maximum pooling layer, average pooling layer, and minimum pooling layer) are used to perform dimensionality reduction processing on the sediment component associated self-attention data to reduce the data dimension and extract key features, thereby generating sediment component associated pooling layer data.

[0124] Performing full connection calculation on the sediment component associated pooling layer data to obtain the sediment component associated fully connected layer data;

[0125] Specifically, the connection weights and biases of the fully connected layer neurons are used to perform linear and nonlinear transformations on the sediment component-associated pooling layer data to generate sediment component-associated fully connected layer data.

[0126] A model is constructed for the sediment component associated fully connected layer data and the water quality label data to obtain a water quality change trend prediction model.

[0127] Specifically, the above steps (convolution calculation, self-attention calculation, pooling layer processing and full connection calculation) are repeated to iteratively train the model on the fully connected layer data associated with the sediment components and the water quality label data, and the model is optimized through model optimization functions (such as stochastic gradient descent (SGD), Adam, etc.) to establish a water quality change trend prediction model.

[0128] In the present invention, by using a preset water quality change trend prediction model to predict the sediment component associated data, the calculation steps such as convolution, self-attention, pooling and full connection in the historical data and model construction process can be used to consider the correlation between different sediment components and the impact on water quality changes, thereby improving the accuracy of water quality change prediction. By performing time series processing and convolution calculation on the historical sediment component associated data, the time series characteristics of the sediment components can be comprehensively considered, and the change trend and periodic change between the sediment components can be captured, so as to more accurately predict the change trend of water quality. The application of the self-attention calculation step can effectively extract the key information and mutual influence degree in the sediment component associated data, which is helpful to identify the sediment components that have the greatest impact on water quality changes, thereby optimizing the modeling process of the water quality change trend prediction model. By performing convolution, self-attention and pooling on the sediment component associated data, the fully connected layer data obtained can effectively express the correlation between the sediment components, making the model construction process more interpretable, and helping to understand the influence mechanism of the sediment components on water quality changes. The construction of the water quality change trend prediction model can provide a scientific basis for water quality management and protection. Accurately predicting the changing trends of water quality will help to take timely measures to regulate and improve water quality, protect the aquatic ecological environment, and maintain human health and quality of life.

[0129] Optionally, performing ecosystem simulation and emulation according to the water quality change trend data to obtain ecosystem stability index data includes:

[0130] Acquiring ecosystem basic element data, and constructing an ecosystem model based on the ecosystem basic element data to obtain ecosystem model data;

[0131] Specifically, collect data on basic elements of the ecosystem, including vegetation distribution, soil type, hydrological conditions, etc., and then build an ecosystem model based on these data. You can use methods such as ecosystem dynamic models and ecosystem process models to describe the interactions and dynamic changes within the ecosystem, so as to obtain ecosystem model data.

[0132] Dynamically simulating the ecosystem model data using the water quality change trend data to obtain ecosystem model dynamic simulation data;

[0133] Specifically, the water quality change trend data is used as input parameters and combined with the ecosystem model data, and the ecosystem is dynamically simulated using mathematical models and computer simulation technology. During the simulation process, the impact of water quality changes on various elements of the ecosystem is considered, and the interaction and dynamic changes of various biological and environmental elements within the ecosystem are simulated to obtain dynamic simulation data of the ecosystem under different water quality conditions.

[0134] The stability index of the ecosystem model dynamic simulation data is calculated to obtain ecosystem stability index data.

[0135] Specifically, according to ecological theory and indicator system, stability analysis is conducted on the dynamic simulation data of the ecosystem model, and various stability indicators of the ecosystem are calculated, such as biodiversity index, ecosystem tolerance index, energy flow stability index, etc. The indicators can reflect the structure, function and stability level of the ecosystem, thereby evaluating the stability performance of the ecosystem under different water quality conditions.

[0136] The ecosystem model in the present invention can be dynamically simulated according to the water quality change trend data, simulating the evolution process of the ecosystem under different water quality conditions, which is helpful to understand the response of the ecosystem to water quality changes, including changes in the composition of biological communities, adjustments in ecological niches, changes in ecosystem structure, etc. By calculating the stability index of the dynamic simulation data of the ecosystem model, the degree of influence of water quality changes on the stability of the ecosystem can be evaluated, which helps to determine the impact of water quality changes on the functions of the ecosystem, including the tolerance and anti-interference ability of the ecosystem, and thus provide a scientific basis for the management and protection of the ecosystem. Ecosystem stability index data can provide support for ecological environment management and protection decisions. By analyzing the stability change trend of the ecosystem, the vulnerability and risk of the ecosystem can be identified in time, and corresponding protection strategies and measures can be formulated to promote the recovery and healthy development of the ecosystem. Combining the ecosystem simulation and simulation results and stability index data, the ecosystem management strategy can be optimized. According to the simulation results, targeted control measures are formulated to strengthen ecological environment monitoring and management, reduce the adverse effects of water quality changes on the ecosystem, and protect and improve the ecological environment of the water area.

[0137] Optionally, the present invention further provides a substrate improvement construction evaluation system for executing the substrate improvement construction evaluation method as described above, wherein the substrate improvement construction evaluation system comprises:

[0138] A water area low mud feature extraction module is used to collect water area bottom mud images through an image acquisition device to obtain water area low mud image data, and to perform organic bottom mud feature extraction and inorganic bottom mud feature extraction on the water area low mud image data to obtain organic bottom mud feature data and inorganic bottom mud feature data;

[0139] A sediment component correlation processing module is used to perform sediment component correlation processing according to the organic sediment characteristic data and the inorganic sediment characteristic data to obtain sediment component correlation data;

[0140] A water quality change trend prediction module is used to predict the water quality change trend according to the sediment component correlation data to obtain water quality change trend data;

[0141] The ecosystem simulation and emulation module is used to simulate and emulate the ecosystem based on the water quality change trend data to obtain ecosystem stability index data for base improvement construction assessment operations.

[0142] The purpose of the present invention is to extract organic and inorganic sediment features from water area sediment image data, so as to understand the composition and characteristics of water area sediment, help identify organic and inorganic components in sediment, and provide basic data for subsequent sediment component association processing. Sediment component association processing can reveal the association rules between various organic and inorganic components in sediment. By mining these association rules, the complexity and change trend of sediment components can be better understood. Based on sediment component association data, water quality change trend prediction can be carried out, and the change trend of water quality in water area can be effectively predicted, including the change direction, rate and influencing factors of water quality parameters, etc., providing a scientific basis for water area ecological environment management and protection, and showing the change and development of water area ecosystem. Ecosystem simulation and simulation are carried out using water quality change trend data to simulate the dynamic change process of water area ecosystem under different water quality conditions. By evaluating the stability index of the ecosystem, the impact of water quality changes on the ecosystem is reflected, and a reference is provided for the evaluation of base improvement construction. Ecosystem stability index data provides scientific guidance for base improvement construction. By comprehensively considering water quality change trends and ecosystem stability, targeted base improvement plans can be formulated to minimize the impact on the ecological environment, ensure the sustainability and ecological safety of construction, and contribute to protecting the natural environment and achieving harmonious coexistence between man and nature.

[0143] Therefore, from any point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the attached application documents rather than the above description, and it is intended that all changes falling within the meaning and scope of equivalent elements of the application documents are included in the present invention.

[0144] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for evaluating substrate improvement construction, characterized in that: The method comprises: S1, collecting water area sediment images by image acquisition equipment to obtain water area sediment image data, and extracting organic sediment features and inorganic sediment features from the water area sediment image data to obtain organic sediment feature data and inorganic sediment feature data; S2, performing sediment component correlation processing according to the organic sediment characteristic data and the inorganic sediment characteristic data to obtain sediment component correlation data; S3, predicting the water quality change trend according to the sediment component correlation data to obtain water quality change trend data; S4. Perform ecosystem simulation and emulation based on the water quality change trend data to obtain ecosystem stability index data for base improvement construction assessment operations.

2. The method according to claim 1, characterized in that: The method of collecting the bottom mud image of the water area by the image collection device to obtain the bottom mud image data of the water area includes: Collect water area images through GIS equipment to obtain water area image data; Performing water area subject recognition based on the water area image data to obtain water area subject data; Extracting water area boundaries from the water area image data according to the water area main body data to obtain water area boundary data; Performing topographic analysis and environmental parameter extraction according to the water area boundary data to obtain topographic data and environmental parameter data; The deployment position is determined according to the water area boundary data, the topographic data and the environmental parameter data to obtain the deployment position data, so as to acquire the water area bottom mud image through the image acquisition device to obtain the water area bottom mud image data.

3. The method according to claim 2, characterized in that The step of extracting the water area boundary from the water area image data according to the water area main body data to obtain the water area boundary data comprises: Extracting the edge water area boundary of the water area image data according to the water area main body data to obtain first water area boundary data; Performing hydrodynamic simulation processing according to the main data of the water area to obtain water area hydrodynamic simulation data; Extracting a water area boundary from the water area image data according to the water area hydrodynamic simulation data to obtain second water area boundary data; Performing boundary comparison on the first water area boundary data and the second water area boundary data to obtain boundary comparison data; When it is determined that the boundary comparison data is boundary overlap data, boundary data fusion is performed on the boundary comparison data corresponding to the boundary overlap data to obtain water area boundary data; When it is determined that the boundary comparison data is boundary overlap data, the first water area boundary data corresponding to the boundary overlap data is corrected according to the second water area boundary data corresponding to the boundary overlap data to obtain the water area boundary data; or, historical water area boundary data is obtained, and boundary correction is performed on the second water area boundary data corresponding to the boundary overlap data and the first water area boundary data corresponding to the boundary overlap data according to the historical water area boundary data to obtain the water area boundary data.

4. The method according to claim 3, characterized in that: The performing boundary data fusion on the boundary comparison data corresponding to the boundary overlap data to obtain water area boundary data includes: Obtain historical boundary data; Perform boundary fusion according to the historical boundary data and the boundary comparison data corresponding to the boundary overlap data to obtain first water area boundary data; Performing sediment particle movement analysis based on the water area hydrodynamic simulation data to obtain sediment particle movement data; Perform boundary data fusion on the boundary comparison data corresponding to the boundary overlap data according to the bottom mud particle movement data to obtain second water area boundary data; A water area boundary evolution process is performed according to the first water area boundary data and the second water area boundary data to obtain water area boundary data.

5. The method according to claim 3, characterized in that: The step of extracting organic sediment features and inorganic sediment features from the water area sediment image data to obtain organic sediment feature data and inorganic sediment feature data includes: Performing color feature extraction and texture feature extraction on the water area bottom mud image data to obtain color feature data and texture feature data; Extracting sediment particle morphology from the sediment image data of the water area to obtain sediment particle morphology data; The sediment characteristic type is determined based on the sediment particle morphology data, the color characteristic data and the texture characteristic data according to the water area hydrodynamic simulation data, so as to obtain organic sediment characteristic data and inorganic sediment characteristic data.

6. The method according to claim 5, characterized in that The determining of the sediment feature type of the sediment particle morphology data, the color feature data and the texture feature data according to the water area hydrodynamic simulation data to obtain organic sediment feature data and inorganic sediment feature data includes: Performing sediment particle characteristic type analysis on the sediment particle morphology data according to the water area hydrodynamic simulation data to obtain sediment particle characteristic type data; Performing color feature data analysis and texture feature analysis according to the water area hydrodynamic simulation data to obtain color feature analysis data and texture feature analysis data; The color feature data and the texture feature data are screened according to the color feature analysis data and the texture feature analysis data to obtain primary organic sediment feature data and primary inorganic sediment feature data; Sediment characteristics are generated for the sediment particle characteristic type data, the primary organic sediment characteristic data and the primary inorganic sediment characteristic data to obtain organic sediment characteristic data and inorganic sediment characteristic data.

7. The method according to claim 1, characterized in that The step of performing sediment component correlation processing according to the organic sediment characteristic data and the inorganic sediment characteristic data to obtain sediment component correlation data includes: Performing association rule mining based on the organic sediment characteristic data and the inorganic sediment characteristic data to obtain association rule data; A sediment component association network is constructed according to the association rule data to obtain sediment component association data.

8. The method according to claim 1, characterized in that The water quality change trend prediction is performed according to the sediment component correlation data to obtain water quality change trend data, including: Using a preset water quality change trend prediction model to predict the water quality change trend of the sediment component correlation data, to obtain water quality change trend data; The steps for constructing the preset water quality change trend prediction model are specifically as follows: Obtain historical sediment composition association data and corresponding water quality label data; Performing time series processing on the historical sediment component associated data to obtain sediment component associated time series data; Performing convolution calculation on the sediment component-related time series data to obtain sediment component-related convolution layer data; Performing self-attention calculation on the sediment component associated convolutional layer data to obtain sediment component associated self-attention data; Performing pooling layer processing on the sediment component associated self-attention data to obtain sediment component associated pooling layer data; Performing full connection calculation on the sediment component associated pooling layer data to obtain the sediment component associated fully connected layer data; A model is constructed for the sediment component associated fully connected layer data and the water quality label data to obtain a water quality change trend prediction model.

9. The method according to claim 1, characterized in that: The ecosystem simulation and emulation are performed according to the water quality change trend data to obtain ecosystem stability index data, including: Acquiring ecosystem basic element data, and constructing an ecosystem model based on the ecosystem basic element data to obtain ecosystem model data; Dynamically simulating the ecosystem model data using the water quality change trend data to obtain ecosystem model dynamic simulation data; The stability index of the ecosystem model dynamic simulation data is calculated to obtain ecosystem stability index data.

10. A base improvement construction evaluation system, characterized in that: Used to execute the substrate improvement construction evaluation method according to claim 1, the substrate improvement construction evaluation system comprises: A water area sediment feature extraction module is used to collect water area sediment images through an image acquisition device to obtain water area sediment image data, and to perform organic sediment feature extraction and inorganic sediment feature extraction on the water area sediment image data to obtain organic sediment feature data and inorganic sediment feature data; A sediment component correlation processing module is used to perform sediment component correlation processing according to the organic sediment characteristic data and the inorganic sediment characteristic data to obtain sediment component correlation data; A water quality change trend prediction module is used to predict the water quality change trend according to the sediment component correlation data to obtain water quality change trend data; The ecosystem simulation and emulation module is used to simulate and emulate the ecosystem based on the water quality change trend data to obtain ecosystem stability index data for base improvement construction assessment operations.

Images