A method for monitoring submerged vegetation cover and rapidly identifying suitable restoration areas

Abstract

This invention proposes a method for monitoring submerged vegetation coverage and rapidly identifying suitable restoration areas, relating to the field of aquatic ecological protection and restoration. The method includes: acquiring underwater images to form an underwater image dataset; classifying the underwater images using an image classification model to identify the submerged vegetation characteristics, coverage information, and sediment type information at each monitoring point; using a spatial interpolation algorithm, processing the submerged vegetation characteristics, coverage information, sediment type information, and water depth data at each monitoring point to generate a spatial distribution map of water depth, sediment type, submerged vegetation type, and coverage in the target water area; and using the spatial distribution map, evaluating it through a comprehensive ecological suitability assessment method to determine the submerged vegetation restoration type, suitable restoration areas, and corresponding restoration priorities for the target water area. This invention can rapidly monitor the characteristics of submerged plants and sediment in aquatic bodies, accurately identify suitable areas for submerged plant restoration, and rationally arrange the restoration priority sequence.

Description

Technical Field

[0001] This invention relates to the field of water ecological protection and restoration, and in particular to a method for monitoring the coverage of submerged vegetation and rapidly identifying suitable restoration areas. Background Technology

[0002] Submerged vegetation, as aquatic plant communities that grow on the bottom of water bodies with roots anchored to the sediment and stems and leaves completely or partially submerged in water, is an indispensable component of aquatic ecosystems. By constructing complex underwater structures, they provide vital habitats, breeding grounds, and shelters for various aquatic animals, serving as spawning substrates for fish and attachment interfaces for plankton. Simultaneously, submerged vegetation plays crucial ecological functions in improving water quality (such as absorbing nutrients and increasing water transparency), inhibiting the excessive proliferation of harmful algae (such as algal blooms), stabilizing sediment, and reducing resuspension, thus being essential for maintaining aquatic biodiversity and the health of the ecosystem.

[0003] However, with the intensification of human activities, many lakes and other water bodies are facing severe eutrophication problems, leading to the rapid degradation and even extinction of submerged vegetation. To address this challenge, large-scale artificial restoration projects for submerged vegetation have been widely carried out both domestically and internationally. The core objective is to enhance the self-purification capacity of water bodies by rebuilding submerged plant communities, effectively controlling algal blooms, thereby improving overall water quality and enhancing the species diversity and structural stability of aquatic ecosystems. In traditional preliminary work for vegetation restoration, the investigation and assessment of the current status of submerged vegetation and sediment conditions mainly rely on manual methods. For example, vegetation monitoring often uses visual estimation or relies on tools such as grass clips for sampling and analysis to determine coverage and species composition; sediment surveys typically use sediment sampling buckets or columnar sediment samples to obtain samples, which are then analyzed in the laboratory for their physicochemical properties such as water content and particle size distribution, as well as pollutant load, to determine whether the target area meets the basic conditions for vegetation restoration. These traditional methods have significant limitations: they not only struggle to achieve comprehensive coverage of the entire water area, but also consume substantial human and material resources, resulting in low operational efficiency. Furthermore, the sampling process may disturb or even damage fragile submerged plants and their substrate habitats. More importantly, traditional methods often fail to adequately consider the spatial complexity and heterogeneity of water depth, existing vegetation patch distribution characteristics, and sediment conditions, as well as their interrelationships. This often leads to a lack of sufficient scientific basis and precise spatial positioning in the selection of restoration areas.

[0004] Chinese invention patent application CN202110973730.X discloses a rapid monitoring system and control method for submerged plant communities. This technical solution uses shipborne equipment to collect real-time data on water transparency, nutrients, water depth, and images of submerged plants. After data processing, the data is analyzed, and a control device, in conjunction with hydraulic facilities, automatically adjusts water depth or nutrient concentration, aiming to achieve dynamic monitoring and control of the existing submerged plant growth environment. This system effectively improves the level of automated management of the growth status of existing submerged plant communities. However, it primarily focuses on maintaining and optimizing the growth conditions of existing vegetation areas, and does not provide specific technical means for systematically identifying and assessing areas with restoration potential in broader, potentially vegetation-deficient waters. How to effectively conduct large-scale suitability assessments and site selection in the early stages of vegetation restoration remains a technical problem that needs further resolution. Summary of the Invention

[0005] In view of this, the present invention provides a method for monitoring the coverage of submerged vegetation and rapidly identifying suitable restoration areas, which can quickly monitor the characteristics of submerged plants and bottom sediments in water bodies, accurately identify suitable restoration areas for submerged plants, and rationally arrange the priority sequence of restoration.

[0006] The technical solution of this invention is implemented as follows:

[0007] This invention provides a method for monitoring submerged vegetation cover and rapidly identifying suitable restoration areas, comprising:

[0008] S1. Using underwater image acquisition equipment, underwater image sequences are acquired in the target water area according to a preset route, and the corresponding geographic coordinates and water depth data of the underwater images are recorded simultaneously to form an underwater image dataset with geographic coordinates and water depth information.

[0009] S2. The underwater image dataset is processed and classified using an image classification model to identify the submerged vegetation type, coverage information and bottom sediment type information at each monitoring point; and the classification results with low confidence are manually verified by composite confidence, and the model parameters of the image classification model are optimized by using the verification results as feedback.

[0010] S3. Using a spatial interpolation algorithm, the submerged vegetation type, coverage information, bottom sediment type information, and water depth data at each monitoring point are used to perform spatial interpolation processing to generate a spatial distribution map of the target water area.

[0011] S4. Using the spatial distribution map of the target water area and the preset ecological evaluation factors, a quantitative assessment is conducted through the comprehensive ecological suitability evaluation method to determine the types of submerged vegetation to be restored, suitable restoration areas, and corresponding restoration priorities in the target water area.

[0012] Preferably, in step S1, the underwater image acquisition device includes a high-definition underwater camera, a depth sounder, and a GPS locator mounted on a cruising vessel; the preset route is a spiral or zigzag route; the underwater image acquisition device moves along the route and acquires data.

[0013] Preferably, in step S2, the image classification model is an adaptive multi-threshold underwater image enhancement and classification ensemble model, which sequentially performs a series of image processing operations on the underwater image, including image enhancement preprocessing, multi-scale boundary enhancement processing, and adaptive classification processing, to obtain preliminary classification labels for submerged vegetation and bottom sediment at each monitoring point.

[0014] Preferably, the composite confidence score is used to evaluate the initial classification label, and its calculation method is as follows:

[0015] Conf(p)=γ1·P max (p)+γ2·(1-Entropy(p))+γ3·S spatial (p)

[0016] Where Conf(p) is the classification confidence score of point p; P max (p) represents the highest category probability value in the classification results corresponding to point p; Entropy(p) represents the normalized probability distribution entropy of the classification results corresponding to point p; S spatial (p) represents the spatial consistency score between the classification result of point p and the classification result of its neighborhood; γ1, γ2, γ3 are preset weight coefficients, satisfying γ1+γ2+γ3=1;

[0017] If the Conf(p) of point p is lower than the preset confidence threshold Conf threshold If so, the preliminary classification label of that location will be marked as a label to be manually verified.

[0018] Preferably, in step S3, the spatial interpolation algorithm integrates the hydrodynamically modified spatial correlation structure model and ecological threshold constraints to perform spatial interpolation processing on the submerged vegetation type, coverage information, sediment type information and water depth data of each monitoring point, and combines feature boundary-preserving interpolation adjustment to generate a spatial distribution map of the target water area, including a spatial distribution map of submerged vegetation coverage, a spatial distribution map of sediment type and a spatial distribution map of water depth.

[0019] Preferably, the spatial correlation structure model with hydrodynamic correction uses a preset hydrodynamic correction function. After correcting the semivariance function γ0(h), the relationship is obtained as follows:

[0020]

[0021] In the formula, γ(h) is the corrected semivariance function; h is the distance vector between two points in space; θ represents the dominant flow direction vector; θ is the flow intensity parameter.

[0022] Preferably, the spatial distribution map generated in step S3 further includes: a spatial distribution map of vegetation species composition and a distribution map of sediment health status.

[0023] Preferably, in step S4, the comprehensive ecological suitability evaluation method is to calculate the integrated ecological suitability evaluation index IECSI using a fusion optimization method. The calculation formula for IECSI is:

[0024]

[0025] Wherein, IECSI(p) is the comprehensive ecological suitability index of point p within the target water area; Let be the standardized score of the i-th core constraint factor at point p; The preset weight index of the i-th core constraint factor satisfies K represents the number of core limiting factors; C neighbor (p) represents the neighboring vegetation connectivity index at point p; α is the neighboring gain adjustment coefficient. The standardized score of the j-th auxiliary support factor at point p; Let the preset weight coefficients of the j-th auxiliary support factor satisfy the following conditions: M represents the number of auxiliary support factors; Ψ(p) represents the temporal stability correction coefficient at point p.

[0026] Preferably, the core limiting factors include historical vegetation distribution, water depth suitability, and sediment health status; the auxiliary supporting factors include sediment particle size suitability and sediment moisture content suitability.

[0027] Preferably, the method further includes:

[0028] Based on the comprehensive evaluation index value of ecological suitability at each location, and according to the preset index threshold range, the target water area is divided into priority restoration area, secondary restoration area, observation area and unsuitable area, and outputs the submerged vegetation restoration types and suitable restoration area classification map of the target water area.

[0029] In areas where submerged vegetation restoration measures have been implemented, steps S1, S2, and S3 are repeated periodically to obtain information on the type, coverage, and sediment type of submerged vegetation in the restored waters, as well as corresponding spatial distribution maps. The monitoring results are compared and analyzed with the preset restoration targets, and subsequent restoration strategies or monitoring plans are adjusted based on the analysis results.

[0030] The present invention has the following advantages over the prior art:

[0031] (1) This invention can efficiently and accurately acquire multi-source spatial data such as submerged vegetation, bottom sediment characteristics and water depth in the target water area, and conduct comprehensive evaluation and zoning based on ecological principles. It provides strong technical support for the scientific site selection and priority determination of restoration areas in water ecological restoration projects, and improves the scientific nature and decision-making efficiency of submerged vegetation restoration work.

[0032] (2) This invention uses an underwater image acquisition device equipped with a high-definition underwater camera, depth sounder and GPS positioning device, and performs automated and synchronized data acquisition along a preset route. Compared with traditional manual sampling and fixed-point monitoring methods, it greatly improves the efficiency of data acquisition and the integrity of spatial coverage. At the same time, the acquired vegetation images, water depth and geographic coordinate data have accurate spatiotemporal correspondence, and effectively avoids human disturbance to underwater bottom sediment and vegetation habitat.

[0033] (3) The present invention adopts an adaptive multi-threshold underwater image enhancement and classification integration model, combined with composite confidence assessment and necessary manual verification feedback optimization mechanism, which can automatically and accurately identify the submerged vegetation type, coverage and bottom sediment type of the collected underwater image data, effectively reducing the subjectivity and cumbersomeness of traditional manual interpretation and improving the objectivity and reliability of the identification results.

[0034] (4) By using spatial interpolation algorithms, this invention transforms discrete monitoring point data into continuous spatial distribution maps. Furthermore, by combining ecological suitability comprehensive evaluation methods, it can systematically reveal the spatial heterogeneity of key environmental factors for submerged vegetation growth and their coupling relationship with vegetation distribution. This allows for the scientific and quantitative identification of areas with different restoration potentials, providing an intuitive spatial decision-making basis for the formulation of submerged vegetation restoration plans.

[0035] (5) This invention can not only quickly identify suitable restoration areas for submerged vegetation, but also classify the restoration priority of target water areas by establishing a submerged vegetation restoration type and suitable restoration area classification system, and support periodic dynamic evaluation and feedback optimization of the effects of the implemented restoration measures. Thus, a closed-loop technical process from early site selection, mid-term planning to late management and adaptive adjustment is formed, which helps to improve the long-term effectiveness and sustainability of submerged vegetation restoration projects. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart of the method of the present invention;

[0038] Figure 2 This is a diagram illustrating the technical implementation of the present invention;

[0039] Figure 3 This is a schematic diagram illustrating the unsupervised classification and identification of submerged vegetation according to an embodiment of the present invention.

[0040] Figure 4 This is a schematic diagram of supervised classification and identification of sediment samples according to another embodiment of the present invention. Detailed Implementation

[0041] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0042] like Figure 1 As shown, this invention provides a method for monitoring submerged vegetation cover and rapidly identifying suitable restoration areas, comprising:

[0043] S1. Using underwater image acquisition equipment, underwater image sequences are acquired in the target water area according to a preset route, and the corresponding geographic coordinates and water depth data of the underwater images are recorded simultaneously to form an underwater image dataset with geographic coordinates and water depth information.

[0044] S2. The underwater image dataset is processed and classified using an image classification model to identify the submerged vegetation type, coverage information and bottom sediment type information at each monitoring point; and the classification results with low confidence are manually verified by composite confidence, and the model parameters of the image classification model are optimized by using the verification results as feedback.

[0045] S3. Using a spatial interpolation algorithm, the submerged vegetation type, coverage information, bottom sediment type information, and water depth data at each monitoring point are used to perform spatial interpolation processing to generate a spatial distribution map of the target water area.

[0046] S4. Using the spatial distribution map of the target water area and the preset ecological evaluation factors, a quantitative assessment is conducted through the comprehensive ecological suitability evaluation method to determine the types of submerged vegetation to be restored, suitable restoration areas, and corresponding restoration priorities in the target water area.

[0047] like Figure 2As shown, this invention proposes a method for monitoring submerged vegetation coverage and rapidly identifying suitable restoration areas. By integrating underwater data acquisition, intelligent image processing, and spatial ecological analysis technologies, a complete technology chain of "data acquisition - intelligent classification - manual verification - spatial modeling - ecological evaluation" is constructed. Specifically, this method first utilizes a cruising vessel, such as an unmanned surface vessel (USV), equipped with a high-definition underwater camera, depth sounder, and GPS positioning system to collect underwater image sequences and water depth data with geographic coordinates along a preset route. Then, an adaptive multi-threshold underwater image enhancement and classification ensemble model is used to process and classify the collected underwater images, identifying the submerged vegetation coverage and sediment type at each monitoring point. Low-confidence points are then manually verified using a composite confidence level, forming a feedback optimization closed loop. Next, a spatial interpolation algorithm is used to integrate a hydrodynamically corrected spatial correlation structure model and ecological threshold constraints, transforming discrete monitoring data into a continuous spatial distribution map. Finally, based on the generated spatial distribution map, a fusion-optimized Integrated Ecological Suitability Index (IECSI) is used. This index comprehensively considers core limiting factors (such as historical vegetation distribution, water depth suitability, and sediment health status) and auxiliary supporting factors (such as sediment particle size, water content, water quality, and light conditions) to scientifically quantify and evaluate the ecological suitability of each point. This determines the types of submerged vegetation to be restored, suitable restoration areas, and their restoration priorities, and supports dynamic evaluation and feedback optimization of restoration effects. This invention effectively improves the efficiency of submerged vegetation monitoring and the scientific nature of restoration zone site selection through this data-driven, multi-factor fusion technical method.

[0048] Specifically, in one embodiment of the present invention, step S1 includes:

[0049] The underwater image acquisition equipment is a cruising vessel, such as an unmanned surface vessel, equipped with a high-definition underwater camera, a depth sounder, and a GPS locator. The high-definition underwater camera is mounted vertically downwards on the bottom of the hull to acquire underwater images; the depth sounder obtains the water depth information corresponding to the image location; and the GPS locator records the precise geographic coordinates of the sampling points. These three devices work synchronously to ensure that each underwater image has corresponding geographic location and water depth data.

[0050] The preset route is a spiral or zigzag route:

[0051] Spiral navigation path: Suitable for large open water areas, such as the central area of ​​a lake. This path starts from the center point of the target water area and expands outward in a spiral shape. The spacing between the paths can be adjusted according to the water transparency and the target monitoring accuracy.

[0052] Z-shaped route: suitable for narrow waterways or bays. This route consists of parallel segments connected by short distances to form a continuous "Z" shaped path.

[0053] The underwater image acquisition equipment moves along a preset route and collects data, forming an underwater image dataset. In this embodiment, after acquiring the underwater images, preliminary data preprocessing can be performed, including optical correction, scale calibration, and coordinate association. A unified point attribute table is established, including:

[0054] Serial Number (ID): Unique identification code; Spatial Coordinates (X,Y): Planar coordinates or latitude and longitude; Water Depth (D): Water depth at the sampling point, in meters; Image File Association (Image_ID): Corresponding underwater image file number; Other Environmental Parameters: Water quality parameters, sediment type, etc.

[0055] Specifically, in one embodiment of the present invention, in step S2, the image classification model is an adaptive multi-threshold underwater image enhancement and classification integration model, which sequentially performs a series of image processing operations on the underwater image, including image enhancement preprocessing, multi-scale boundary enhancement processing and adaptive classification processing, to obtain preliminary classification labels for submerged vegetation and bottom sediment at each monitoring point.

[0056] S2.1 Underwater Image Enhancement Preprocessing

[0057] To address imaging challenges under varying water depths and qualities, an adaptive multi-threshold underwater image enhancement algorithm is designed:

[0058] I enhanced (x,y)=I original (x,y)·e α(d)·(1-β·T(x,y))

[0059] Among them, I enhanced (x,y) represents the pixel values ​​of the enhanced image; I original (x,y) represents the original image pixel value; α(d) is the compensation coefficient related to water depth d, α(d)=k1·ln(d+1), k1 is the adjustment parameter (1.0-2.0); β is the light attenuation adjustment parameter (0.3-0.7), which is automatically adjusted according to the turbidity of the water body; T(x,y) is the transmittance estimate, which is calculated by the dark channel prior method.

[0060] S2.2 Multi-scale Boundary Enhancement Processing

[0061] To address the issue of blurred boundaries between vegetation and sediment, multi-scale boundary enhancement techniques were applied:

[0062]

[0063] in, This is the result of multi-scale edge enhancement; For different scales (σ) i Gaussian kernel; w i The weighting coefficients for each scale satisfy the following conditions: K is the number of scales, usually 3-5; * is the convolution operator.

[0064] S2.3 Adaptive Classification Strategy Selection

[0065] Automatically select the best classification strategy based on image characteristics:

[0066]

[0067] Where E(I) is the image entropy value, which measures information complexity; V(I) is the image variance, which measures contrast level; M(I) is the prior model matching score; τ1, τ2, τ3 are adaptive thresholds, which are dynamically adjusted according to historical data.

[0068] S2.4 Unsupervised Classification Method (Submerged Vegetation Identification)

[0069] When choosing unsupervised classification, an improved K-means or ISODATA algorithm with multi-feature fusion is used:

[0070] Extract color feature vectors: hue, saturation, and lightness in the HSV space. Extract texture feature vectors: contrast, correlation, energy, and homogeneity of the gray-level co-occurrence matrix. Perform feature vector normalization and fusion. Execute an improved clustering algorithm to automatically determine the optimal number of categories. Map the clustering results to vegetation categories using expert rules or sample library matching.

[0071] S2.5 Surveillance Classification Method (Sediment Health Status Assessment)

[0072] When choosing supervised classification, an ensemble learning framework is used:

[0073]

[0074] Wherein, P(y i |x) represents the category y to which sample x belongs. i The combined probability; P j (y i |x) represents the probability given by the j-th classifier; ω j Let the weights of the j-th classifier satisfy the following condition: M represents the number of classifiers (3-5).

[0075] The classifier weights are automatically adjusted through periodic evaluations:

[0076]

[0077] Among them Acc j Let be the accuracy of the j-th classifier on the manually verified sample set.

[0078] It should be noted that when performing image classification, it is not necessary to choose only one classification method. The classification strategy described in this embodiment can refer to selecting a classification method at different stages. For example, color extraction can be used first to identify whether there are submerged plants. If so, unsupervised classification can be used to identify the type of submerged plant. After subsequent manual verification, supervised classification can be used to refine the classification results and update the attribute table. Alternatively, a supervised classification algorithm can be used, which requires pre-training with training samples before identifying the classification results.

[0079] S2.6 Composite Confidence Level

[0080] To reduce the workload of manual review, a composite confidence level is designed:

[0081] Conf(p)=γ1·P max (p)+γ2·(1-Entropy(p))+γ3·S spatial (p)

[0082] Where Conf(p) is the classification confidence score of point p; P max (p) represents the highest category probability value in the classification results corresponding to point p; Entropy(p) represents the normalized probability distribution entropy of the classification results corresponding to point p; S spatial (p) represents the spatial consistency score between the classification result of point p and the classification result of its neighborhood; γ1, γ2, γ3 are preset weight coefficients, satisfying γ1+γ2+γ3=1;

[0083] in, N(p) is the set of neighborhood points of point p; w(p,q) is the weight between point p and neighborhood point q, calculated based on distance; δ(y(p),y(q)) is the Kronecker delta function, which is 1 when the classification results of p and q are the same, and 0 otherwise.

[0084] If the Conf(p) of point p is lower than the preset confidence threshold Conf threshold If so, the preliminary classification label of that location will be marked as a label to be manually verified.

[0085] For locations with an automatic classification confidence level below the threshold, manual verification is performed: the original underwater image is extracted, and the automatic classification results are displayed; experts interpret and confirm the actual situation of the location; the location attribute table is updated, and the classification results are corrected; the manual verification results are fed back to the classification algorithm to adjust the model parameters.

[0086] After the above processing, the final output includes the submerged vegetation type, coverage information, and sediment type information for each monitoring point, including: submerged vegetation coverage, submerged vegetation type, sediment type, sediment health status, and confidence score for each classification result.

[0087] Specifically, in one embodiment of the present invention, in step S3, the spatial interpolation algorithm integrates the hydrodynamically modified spatial correlation structure model and ecological threshold constraints to perform spatial interpolation processing on the submerged vegetation type, coverage information, bottom sediment type information and water depth data of each monitoring point, and combines feature boundary-preserving interpolation adjustment to generate a spatial distribution map of the target water area, including a spatial distribution map of submerged vegetation coverage, a spatial distribution map of bottom sediment type and a spatial distribution map of water depth.

[0088] In this embodiment, based on the updated attribute table, the spatial interpolation algorithm ECAI is used to generate various distribution maps:

[0089] S3.1 Hydrodynamic-Assisted Spatial Relational Structure Modeling

[0090] Traditional spatial interpolation methods typically assume that spatial correlation is only affected by distance, neglecting the influence of water flow, a crucial hydrodynamic factor, on the spatial distribution of aquatic ecosystems. In the ECAI algorithm, a hydrodynamic factor is introduced to correct the semivariogram function:

[0091]

[0092] Where γ(h) is the corrected semivariance function; γ0(h) is the original semivariance function (spherical model, exponential model, etc.); This is a hydrodynamic correction function; θ represents the dominant flow direction vector; θ is the flow intensity parameter.

[0093] The hydrodynamic correction function is defined as:

[0094]

[0095] in, Let h be the distance vector and the direction of water flow. The included angle; δ is the directional intensity parameter (0.2-0.8); |h| is the magnitude of the distance vector.

[0096] The preset hydrodynamic correction function has the following characteristics: when the distance vector is in the same direction as the water flow (the angle is 0°), The correction effect is strongest when the distance vector is perpendicular to the water flow direction (the angle is 90°). No directional correction; when the distance vector is opposite to the direction of water flow (angle is 180°), The correlation is reduced; the correction effect decays exponentially with increasing distance, reflecting the distance decay characteristics of the influence of water flow.

[0097] S3.2 Ecological Threshold Constrained Kriging Interpolation

[0098] Traditional Kriging interpolation can produce ecologically inappropriate results. To address this issue, the ECAI algorithm introduces ecological threshold constraints, integrating ecological knowledge into the Kriging interpolation process.

[0099]

[0100] Where the weight λ i The following ecological constraints must be met:

[0101]

[0102] Among them, [a k ,b k [For region R] k The ecological threshold range; R k This refers to sub-regions with similar ecological conditions.

[0103] This constraint ensures that the interpolation results do not produce ecologically unreasonable extreme values, such as impossible vegetation cover or sediment distribution that does not conform to physical laws. For example, for submerged vegetation cover, its value range is guaranteed to be between 0-100%; for specific areas, an upper limit for vegetation cover is set according to environmental conditions (such as water depth and light); for sediment type, the interpolation results are ensured to conform to the natural transition law of sediment distribution.

[0104] S3.3 Multivariate Environmental Factor Information Entropy Discrimination Zoning

[0105] Spatial information entropy is used to determine and partition areas based on environmental factors such as water depth and substrate type.

[0106]

[0107] Where E(R) is the normalized information entropy of region R; p i Let be the normalized probability density of environmental factor i within the region; m is the number of environmental factors considered.

[0108] When E(R)>E threshold At the same time, the region is further subdivided to ensure the ecological homogeneity of the sub-regions.

[0109] Specifically, environmental factors may include water depth gradient, sediment type, shoreline distance, and water quality parameters.

[0110] S3.4 Feature Boundary Preserving Interpolation

[0111] Ecosystems often exhibit distinct boundary features, such as the boundaries between different substrate types and zones of rapid water depth changes. To preserve these ecological boundary features, the ECAI algorithm applies boundary-preserving interpolation adjustment:

[0112] ZBP (s)=Z K (s)·(1-B(s))+Z E (s)·B(s)

[0113] Among them, Z BP (s) represents the boundary-preserving interpolation result; Z K (s) represents the Kriging interpolation result; Z E B(s) represents the local interpolation result of the boundary region; B(s) represents the boundary detection function (0-1).

[0114] The boundary detection function is defined as follows:

[0115]

[0116] Where d(s,E) is the distance from point s to the nearest boundary E; σ B The boundary effect attenuation parameter; r B The radius is affected by the boundary.

[0117] In summary, the complete implementation process of the ECAI algorithm in this embodiment is as follows: 1) Organize the point attribute table obtained in step S2, including the submerged vegetation type, coverage information, bottom sediment type information, and water depth data of each monitoring point; obtain water dynamic parameters, such as the dominant flow direction vector and flow intensity parameters; initialize various interpolation parameters, such as directional intensity parameters and boundary influence parameters; 2) Calculate the spatial distribution characteristics of each environmental factor; apply the multivariate environmental factor information entropy discrimination method to perform spatial partitioning; and determine the ecological threshold constraint interval for each sub-region [a k ,b k 3) Identify the boundary locations of each environmental factor, construct the boundary detection function B(s), and design a local interpolation strategy for the boundary region; 4) Fit the original semivariance function γ0(h) based on the sample point data, and apply the hydrodynamic correction function. Calculate the corrected semivariance function γ(h). 5) For each predicted point s0, determine its sub-region R. k By applying ecological threshold constraints, the optimal weight λ is solved. i Calculate the Kriging interpolation results. 6) For points near the boundary, calculate the boundary detection function value B(s) and the local interpolation result Z in the boundary region. E (s), fused to obtain the final boundary-preserving interpolation result Z BP (s); 7) Generate a spatial distribution map of the target water area, including: spatial distribution map of submerged vegetation coverage, spatial distribution map of sediment type, spatial distribution map of water depth, spatial distribution map of vegetation species composition, and spatial distribution map of sediment health status.

[0118] Specifically, in one embodiment of the present invention, step S4 determines the types of submerged vegetation to be restored in the target water area, the suitable restoration area and the restoration priority based on the spatial distribution map obtained in the previous steps and the preset ecological evaluation factors through a comprehensive evaluation method.

[0119] Based on the generated spatial distribution map, this embodiment constructs an optimized Integrated Ecological Suitability Index (IECSI) to identify suitable restoration areas:

[0120]

[0121] Wherein, IECSI(p) is the comprehensive ecological suitability index of point p within the target water area; Let be the standardized score of the i-th core constraint factor at point p; The preset weight index of the i-th core constraint factor satisfies K represents the number of core limiting factors; C neighbor (p) represents the neighboring vegetation connectivity index at point p; α is the neighboring gain adjustment coefficient. The standardized score of the j-th auxiliary support factor at point p; Let the preset weight coefficients of the j-th auxiliary support factor satisfy the following conditions: M represents the number of auxiliary support factors; Ψ(p) represents the temporal stability correction coefficient at point p.

[0122] The core limiting factors include historical vegetation distribution, water depth suitability, and sediment health, and their specific evaluation methods are as follows:

[0123] Historical vegetation distribution evaluation function:

[0124]

[0125] Water depth suitability evaluation function:

[0126]

[0127] Where, μ D The optimal water depth center value; σ D The standard deviation parameter is suitable for a range of water depths; (D(p)) is an interval indicator function, when D(p)∈[D min D max The value is 1 if the condition is met, otherwise it is 0.

[0128] Sediment health status evaluation function:

[0129]

[0130] Among them, C j(p) represents the concentration of the j-th pollutant at point p; C j,ref is the reference concentration threshold for the j-th pollutant; M is the number of pollutant types considered.

[0131] Method for calculating vegetation connectivity index:

[0132]

[0133] Where N(p) is the search neighborhood of point p (e.g., a 500-meter radius circle); V(q) is the vegetation cover of point q (0-1); d(p,q) is the distance between points p and q; σ d This is the parameter for spatial effect attenuation.

[0134] Supporting factors include sediment particle size suitability and sediment moisture content suitability.

[0135] Sediment particle size suitability evaluation function:

[0136]

[0137] Where G(p) is the particle size of the sediment at point p; G opt,min G opt,max The optimal particle size range for the target vegetation; G max This represents the maximum permissible particle size.

[0138] Sediment moisture content suitability evaluation function:

[0139]

[0140] Where W(p) is the sediment moisture content at point p; W opt The optimal water content for the target vegetation; W max This represents the maximum permissible moisture content.

[0141] Considering the seasonality and interannual variability of hydrological conditions:

[0142]

[0143] Among them, CV T (p) represents the time variation coefficient of the water level at point p; ΔW max (p) represents the maximum water level fluctuation; W avg (p) represents the average water level; β and γ are the adjustment parameters (0-0.5).

[0144] Based on the calculated IECSI index value, step S4 finally outputs the submerged vegetation restoration species, suitable restoration areas, and restoration priorities for the target water area.

[0145] Specifically, based on the preset index threshold range, the target water areas are divided into the following categories:

[0146] Priority recovery zone: IECSI(p) ≥ IECSI high Suboptimal recovery zone: IECSI medium ≤IECSI(p) <IECSI high ; Area to be observed: IECSI low ≤IECSI(p) <IECSI medium Unsuitable area: IECSI(p) <IECSI low .

[0147] Among them IECSI low IECSI medium and IECSI high It is a predefined threshold.

[0148] Considering spatial continuity, adjacent high-priority areas are merged to form continuous recovery patches. A suitable recovery area map and a recovery priority zoning map are then generated.

[0149] Specifically, in one embodiment of the present invention, in areas where submerged vegetation restoration measures have been implemented, a monitoring and feedback mechanism needs to be established: fixed monitoring points are set up in the restoration area, and steps S1, S2, and S3 are periodically repeated to obtain: the type of submerged vegetation, coverage information, sediment type information, and corresponding spatial distribution map of the restored water area. The monitoring results are compared and analyzed with the preset restoration targets: vegetation coverage growth rate, changes in species composition, and evolution of spatial distribution patterns. Subsequent restoration strategies or monitoring plans are adjusted based on the analysis results: for areas with low restoration efficiency, the restoration plan is adjusted; for areas with high efficiency, it is maintained; and based on the monitoring data, model parameters and evaluation systems are optimized.

[0150] The following two examples briefly illustrate the implementation process of the present invention.

[0151] Example 1

[0152] A series of underwater photographs of a lake were taken using an underwater camera. Simple color extraction from these photographs allowed for the identification of submerged vegetation at specific locations. Supervised classification identified the types of submerged vegetation. Through color extraction and unsupervised classification of multiple sets of underwater photographs, attribute tables for different locations were obtained. These attribute tables included the presence or absence of submerged vegetation and the type of submerged vegetation. Spatial interpolation of the attribute tables yielded a preliminary spatial distribution map of whether and what type of submerged vegetation was present at different locations. Combined with visual underwater photographs, photographs of different submerged vegetation were selected for further supervised classification, refining the classification results and updating the attribute tables. Spatial interpolation then revealed the specific underwater characteristics of the water body. The identification results in this embodiment are as follows: Figure 3 As shown, Figure 3 The image on the left shows submerged vegetation. Figure 3The image on the right shows submerged plants identified through supervised classification. This embodiment demonstrates the adaptive selection of unsupervised classification processing in step S2 of the present invention.

[0153] Example 2

[0154] Training samples were prepared for the supervised classification algorithm. Sediment samples with different submerged vegetation and varying pollutant concentrations were collected. In the laboratory, spectral analysis and image acquisition were performed on samples with different pollutant concentrations to extract feature parameters, including sediment color, texture, and infrared spectral characteristics.

[0155] A series of underwater photographs of sediment were obtained through unmanned surface vessel (USV) navigation. The images and features extracted in the previous step were used as training samples. The series of underwater photographs were then classified using both the training samples and supervised classification. After classification, a monitoring point attribute table containing this information was generated, recording the relevant attribute data for each monitoring point. Spatial interpolation was then performed to obtain a visual spatial distribution map of sediment pollution levels. Based on the pollution level combined with water depth distribution and vegetation distribution characteristics, a priority time sequence scheme for submerged vegetation restoration was formulated. The sample identification results in this embodiment are as follows: Figure 4 As shown, Figure 4 The first row represents the sediment sample, and the second row represents the degree of contamination identified through monitoring and classification. Figure 4 From left to right, the images represent sediment with low pollution content (left), sediment with medium pollution content (middle), and sediment with heavy pollution content (right). This embodiment illustrates a scenario where the supervised classification model is applied in step S2 of this invention.

[0156] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for submerged vegetation coverage monitoring and rapid identification of suitable restoration areas, characterized in that, The method comprises the following steps: S1, using an underwater image acquisition device, collecting underwater image sequences according to a preset flight path in a target water area, and synchronously recording the geographical coordinates and water depth data corresponding to the underwater images to form an underwater image dataset with geographical coordinates and water depth information; S2, using an image classification model to process and classify the underwater image dataset, identifying the submerged vegetation type, coverage information and sediment type information of each monitoring point; and through the composite confidence, manually checking the classification results with low confidence, and using the checking results to optimize the model parameters of the image classification model; S3, using a spatial interpolation algorithm, using the submerged vegetation type, coverage information, sediment type information and water depth data of each monitoring point, performing spatial interpolation processing to generate a spatial distribution map of the target water area; S4, using the spatial distribution map of the target water area, combining the preset ecological evaluation factors, and quantitatively evaluating by an ecological suitability comprehensive evaluation method to determine the submerged vegetation restoration type, suitable restoration area and corresponding restoration priority of the target water area; In step S3, the spatial interpolation algorithm integrates the spatial correlation structure model corrected by hydrodynamics and the ecological threshold constraint condition to perform spatial interpolation processing on the submerged vegetation type, coverage information, sediment type information and water depth data of each monitoring point, and combines feature boundary preservation interpolation adjustment to generate a spatial distribution map of the target water area, including a submerged vegetation type distribution map, a coverage spatial distribution map, a sediment type spatial distribution map and a water depth spatial distribution map. In step S4, the ecological suitability comprehensive evaluation method is to calculate by using an optimized ecological suitability comprehensive evaluation index IECSI, and the calculation formula of IECSI is: in, Points within the target water area The comprehensive ecological suitability index; For point First Standardized scores of the core limiting factors; For the first The preset weight index of each core limiting factor satisfies The number of core limiting factors; For point The connectivity index of adjacent vegetation at the location; This is the neighbor gain adjustment coefficient; For point First Standardized scores of each auxiliary support factor; For the first The preset weight coefficients of each auxiliary support factor satisfy... ; The number of auxiliary support factors; For point The timing stability correction coefficient at the location.

2. The method for submerged vegetation coverage monitoring and suitable recovery zone rapid identification according to claim 1, characterized in that, In step S1, the underwater image acquisition device comprises a high-definition underwater camera, a depth sounder and a GPS positioning instrument installed on a cruising boat; the preset flight path is a spiral or zigzag flight path; the underwater image acquisition device moves along the flight path and collects data.

3. The method for submerged vegetation coverage monitoring and suitable recovery zone rapid identification according to claim 1, characterized in that, In step S2, the image classification model is a self-adaptive multi-threshold underwater image enhancement and classification integrated model, which sequentially performs a series of image processing operations including image enhancement preprocessing, multi-scale boundary enhancement processing and self-adaptive classification processing on the underwater image to obtain the preliminary classification labels of the submerged vegetation and the preliminary classification labels of the sediment of each monitoring point.

4. The method for monitoring submerged vegetation coverage and quickly identifying suitable recovery areas according to claim 3, characterized in that, The composite confidence evaluates the preliminary classification labels, and the calculation method is: wherein, is a classification confidence score of the point position ; is a highest class probability value in the classification result of the point position ; is a normalized probability distribution entropy of the classification result of the point position ; is a spatial consistency score of the classification result of the point position and its neighborhood classification result; is a preset weight coefficient, satisfying ; If the location of Below the preset confidence threshold If so, the preliminary classification label of that location will be marked as a label to be manually verified.

5. The method for submerged vegetation coverage monitoring and suitable recovery zone rapid identification according to claim 1, characterized in that, The spatially dependent structure model with hydrodynamic correction is to use a preset hydrodynamic correction function The modified semi-variance function The relationship is obtained as follows: wherein is the modified semi-variance function; is the distance vector between two points in space; is the dominant flow direction vector; is the flow intensity parameter.

6. The method for submerged vegetation coverage monitoring and suitable recovery zone rapid identification according to claim 1, characterized in that, The spatial distribution map generated in step S3 further comprises a vegetation species composition spatial distribution map and a sediment health state distribution map.

7. The method for submerged vegetation coverage monitoring and suitable recovery zone rapid identification according to claim 1, characterized in that, The core limiting factors include historical vegetation distribution, water depth suitability and sediment health state; the auxiliary support factors include sediment particle size suitability and sediment moisture content suitability.

8. The method for submerged vegetation coverage monitoring and suitable recovery zone rapid identification according to claim 1, characterized in that, The method further comprises: According to the ecological suitability comprehensive evaluation index value of each point, the target water area is divided into a priority restoration area, a sub-optimal restoration area, an observation area and an unsuitable area according to a preset index threshold interval, and the submerged vegetation restoration type and the suitable restoration area classification map of the target water area are output. In the area where the submerged vegetation restoration measures have been implemented, the steps S1, S2 and S3 are periodically repeated to obtain the submerged vegetation type, coverage information, sediment type information and corresponding spatial distribution map of the water area after restoration, the monitoring results are compared and analyzed with the preset restoration target, and the subsequent restoration strategy or monitoring scheme is adjusted according to the analysis results.

Images