Geographic information evaluation method and system based on intelligent algorithm

Abstract

The invention discloses a geographic information evaluation method and system based on an intelligent algorithm, and relates to the technical field of geographic information evaluation. According to the geographic information assessment method based on the intelligent algorithm, geographic information data of a plurality of districts of a to-be-assessed set ecological region are obtained, and feature extraction is performed to obtain a wetland structure assessment index set of each district of the to-be-assessed set ecological region; comprising a landform structure ecological suitability index, a water structure ecological balance index and a vegetation structure health index, and comprehensively analyzing the assessment index set of each area of the to-be-assessed set ecological area to obtain a wetland ecological structure health index of the to-be-assessed set ecological area. According to the method, the preset ecological protection measures are taken for the to-be-evaluated set ecological region through the wetland ecological structure health index, so that the suitability and stability characteristics of the wetland ecological system on the natural structure level can be comprehensively reflected, and the authenticity and accuracy of the evaluation result are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of geographic information evaluation, and specifically to a geographic information evaluation method and system based on intelligent algorithms. Background Art

[0002] Geographic information, as an important data carrier reflecting natural elements and human activities on the earth's surface, is widely used in multiple fields such as urban planning and ecological protection. In the field of ecological protection, geographic information is widely used in tasks such as ecological system structure monitoring and ecological degradation identification. Especially in typical ecologically sensitive areas such as wetlands, forests, and grasslands, the spatial structure characteristics, environmental state distribution, and dynamic change trajectories provided by geographic information have become the core basis for supporting ecological protection planning and regulation. However, most traditional ecological geographic information evaluation methods rely on expert rules, empirical weights, and static index superposition models. However, in complex ecological systems with high multi-factor coupling, multi-scale heterogeneity, and significant dynamic evolution, there are significant deficiencies in both evaluation accuracy and response timeliness, and it is difficult to reveal the non-linear relationships and their change trends among the internal structures of the ecological system. For example, in a wetland ecological system, its health status is not only jointly affected by factors such as the stability of the geomorphic structure, the coherence of the hydrological process, and the continuity of vegetation distribution, but also has high spatial heterogeneity and dynamic evolution characteristics. If only a single data dimension or a static model is used for analysis, it is easy to lead to evaluation distortion and response lag.

[0003] The prior art, such as a patent application with the publication number CN118364413A, discloses an ecological environment monitoring method and system based on geographic information data, including: obtaining the humidity data time series of each air humidity sensor every day, clustering all air humidity sensors, and obtaining the difference between all air humidity sensors in each cluster and other air humidity sensors according to the number of air humidity sensors, each mode, and the time length spanned by each mode; obtaining the abnormality of each air humidity sensor according to the difference between each air humidity sensor and other air humidity sensors and all humidity data within each humidity sequence segment, and screening out several abnormal sensors from all air humidity sensors based on this. Thus, the present invention improves the accuracy of ecological environment monitoring by clustering sensors with the same air humidity change into one category and then calculating the air humidity change and the similarity of all sensors in the cluster within the region.

[0004] Based on the above solutions, it is found that the limitations of the existing technologies at least include the following problems. The existing technologies have a single monitoring object and a narrow evaluation dimension, thus lacking the interaction mechanism among multiple factors within the ecosystem. As a result, it is difficult to accurately restore the overall structural state of the ecological region. In the actual ecosystem, the change of the ecological health state is not driven by a single factor, but by the coupling and mutual regulation among multiple dimensions such as the geomorphic structure, hydrological process, and vegetation distribution. The existing methods ignore the synergistic relationship among these dimensions in terms of spatial structure and response law, and thus lead to weak response ability and insufficient recognition accuracy when facing phenomena with obvious spatial heterogeneity such as geomorphic fracture, hydrological fragmentation, and vegetation patch degradation, and then reduce the accuracy of the evaluation results. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technologies, the present invention provides a geographic information evaluation method and system based on intelligent algorithms, which solves the problems that the existing technologies ignore the coupling of multiple ecological factors and are difficult to accurately identify the structural heterogeneity and the overall ecological health state.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A geographic information evaluation method based on intelligent algorithms includes the following steps: obtaining geographic information data of several areas of a to-be-evaluated set ecological region, where the geographic information data includes wetland geomorphic image data, wetland hydrological geographic data, and wetland vegetation geographic data; respectively performing feature extraction on the geographic information data of each area of the to-be-evaluated set ecological region to obtain a wetland structure evaluation index set for each area of the to-be-evaluated set ecological region, including a geomorphic structure ecological suitability index, a water structure ecological stability index, and a vegetation structure health index; comprehensively analyzing the evaluation index set of each area of the to-be-evaluated set ecological region to obtain a wetland ecological structure health index of the to-be-evaluated set ecological region; and taking preset ecological protection measures for the to-be-evaluated set ecological region based on the wetland ecological structure health index.

[0007] Further, the specific formula for calculating the wetland ecological structure health index of the to-be-evaluated set ecological region is as follows: where KsT is the wetland ecological structure health index of the to-be-evaluated set ecological region, and DsT i is the geomorphic structure ecological suitability index of the i-th area of the to-be-evaluated set ecological region, α1 is the geomorphic adjustment coefficient stored in the database, and SgW i is the water structure ecological stability index of the i-th area of the to-be-evaluated set ecological region, α2 is the water structure adjustment coefficient stored in the database, and ZbG i$H_{vi}$ is the vegetation structure health index of the $i$-th area of the ecological region to be evaluated, $\alpha_3$ is the vegetation adjustment coefficient stored in the database, $\alpha_4$ is the collaborative penalty adjustment coefficient stored in the database, $\alpha_5$ is the vegetation imbalance penalty adjustment coefficient stored in the database, $\zeta$ is the smoothing coefficient stored in the database, $\alpha_6$ is the comprehensive adjustment coefficient stored in the database, $i = 1, 2, 3, \ldots, i_0$, and $i_0$ is the number of areas.

[0008] Further, the wet topographic image data is specifically the pixel value and two-dimensional coordinates of each pixel point in the wet topographic image, and the specific steps to obtain the geomorphic structure ecological suitability index of each area of the ecological region to be evaluated are as follows: Input the wet topographic image data of each area of the ecological region to be evaluated into a pre-trained geomorphic recognition model for comprehensive analysis to obtain a geomorphic evaluation index set of each area of the ecological region to be evaluated, including a geomorphic stability index, a topographic fracture interference index, and a boundary continuity index; Conduct comprehensive analysis on the geomorphic evaluation index set of each area of the ecological region to be evaluated to obtain the geomorphic structure ecological suitability index of each area of the ecological region to be evaluated.

[0009] Further, the geomorphic recognition model is specifically a deep convolutional neural network. The deep convolutional neural network includes an input layer, several convolutional layers, an edge enhancement layer, an upsampling structure layer, a spatial statistics layer, and an output layer. The specific steps to obtain the geomorphic evaluation index set of each area of the ecological region to be evaluated are as follows: In the input layer of the deep convolutional neural network, receive the wet topographic image data of each area of the ecological region to be evaluated and perform preprocessing; In the convolutional layer of the deep convolutional neural network, extract image features from the preprocessed wet topographic image data of each area of the ecological region to be evaluated to obtain a multi-channel geomorphic feature map of each area of the ecological region to be evaluated; In the edge enhancement layer of the deep convolutional neural network, perform edge structure strengthening processing on the multi-channel geomorphic feature map of each area of the ecological region to be evaluated to obtain an enhanced edge feature map set of each area of the ecological region to be evaluated; In the upsampling structure layer of the deep convolutional neural network, perform spatial restoration processing on the enhanced edge feature map set of each area of the ecological region to be evaluated to obtain a structure-restored edge image set of each area of the evaluated ecological region; In the spatial statistics layer of the deep convolutional neural network, perform regional structure statistics processing on the structure-restored edge image set of each area of the evaluated ecological region to obtain a geomorphic feature vector of each area of the evaluated ecological region; In the output layer of the deep convolutional neural network, perform feature fusion processing on the geomorphic feature vector of each area of the evaluated ecological region to obtain the geomorphic stability index, the topographic fracture interference index, and the boundary continuity index of each area of the ecological region to be evaluated, that is, the geomorphic evaluation index set.

[0010] Further, the specific formula for calculating the ecological suitability index of the geomorphic structure of each area in the to-be-evaluated set ecological region is as follows: Among them, DsT i is the ecological suitability index of the geomorphic structure of the i-th area in the to-be-evaluated set ecological region, DmW i is the geomorphic stability index of the i-th area in the to-be-evaluated set ecological region, φ1 is the geomorphic stability coefficient stored in the database, η1 is the geomorphic stability adjustment coefficient stored in the database, BxL i is the boundary continuity index of the i-th area in the to-be-evaluated set ecological region, φ2 is the boundary continuity coefficient stored in the database, η2 is the boundary continuity adjustment coefficient stored in the database, DgR i is the terrain fracture interference index of the i-th area in the to-be-evaluated set ecological region, φ3 is the fracture interference coefficient stored in the database, η3 is the fracture interference adjustment coefficient stored in the database, χ is the suppression adjustment coefficient stored in the database, η4 is the interaction adjustment coefficient stored in the database, and φ1 + φ2 + φ3 = 1, i = 1, 2, 3, …, i0, where i0 is the number of areas.

[0011] Further, the wetland hydrogeographic data includes water depth value, water level fluctuation index, disturbance difference index, water level gradient index, water accumulation connectivity index, and edge water level expansion gradient index. The specific steps for obtaining the water structure ecological stability index of each area in the to-be-evaluated set ecological region are as follows: comprehensively analyze the wetland hydrogeographic data of each area in the to-be-evaluated set ecological region respectively to obtain the hydrogeological evaluation index set of each area in the to-be-evaluated set ecological region, including hydrogeological spatial storage stability index, water level terrain disturbance index, and hydrogeomorphic recharge recovery index; and comprehensively analyze the hydrogeological evaluation index set of each area in the to-be-evaluated set ecological region to obtain the water structure ecological stability index of each area in the to-be-evaluated set ecological region.

[0012] Further, the specific steps for obtaining the hydrogeological evaluation index set of each area in the to-be-evaluated set ecological region are as follows: obtain the water level stability rate value of each area in the to-be-evaluated set ecological region, and comprehensively analyze it in combination with the water depth value and water level fluctuation index to obtain the hydrogeological spatial storage stability index of each area in the to-be-evaluated set ecological region; obtain the break point water level difference value of each area in the to-be-evaluated set ecological region, and comprehensively analyze it in combination with the disturbance difference index and water level gradient index to obtain the water level terrain disturbance index of each area in the to-be-evaluated set ecological region; obtain the environmental correction factor of each area in the to-be-evaluated set ecological region, and comprehensively analyze it in combination with the water accumulation connectivity index and edge water level expansion gradient index to obtain the hydrogeomorphic recharge recovery index of each area in the to-be-evaluated set ecological region.

[0013] Furthermore, the wetland vegetation geographic data includes a vegetation patch concentration index, a root zone water response spatial deviation index, and a water content distribution index. The specific steps to obtain the vegetation structure health index for each area of the to-be-evaluated set ecological region are as follows: Standardize the vegetation patch concentration index, root zone water response spatial deviation index, and water content distribution index for each area of the to-be-evaluated set ecological region; and comprehensively analyze the standardized vegetation patch concentration index, root zone water response spatial deviation index, and water content distribution index for each area of the to-be-evaluated set ecological region to obtain the vegetation structure health index for each area of the to-be-evaluated set ecological region.

[0014] Furthermore, the specific steps to take preset ecological protection measures for the to-be-evaluated set ecological region based on the wetland ecological structure health index are as follows: Judge and analyze the wetland ecological structure health index of the to-be-evaluated set ecological region with the preset wetland ecological structure health index threshold; If the wetland ecological structure health index of the to-be-evaluated set ecological region is lower than or equal to the preset wetland ecological structure health index threshold, then take the first ecological protection measure; If the wetland ecological structure health index of the to-be-evaluated set ecological region is higher than the preset wetland ecological structure health index threshold, then take the second ecological protection measure.

[0015] A geographic information evaluation system based on an intelligent algorithm includes: a data acquisition module for acquiring geographic information data of several areas of the to-be-evaluated set ecological region, where the geographic information data includes wetland topography image data, wetland hydrological geographic data, and wetland vegetation geographic data; a structure feature extraction module for respectively extracting features from the geographic information data of each area of the to-be-evaluated set ecological region to obtain a wetland structure evaluation index set for each area of the to-be-evaluated set ecological region, including a topographic structure ecological suitability index, a water structure ecological balance index, and a vegetation structure health index; a comprehensive evaluation module for comprehensively analyzing the evaluation index set of each area of the to-be-evaluated set ecological region to obtain the wetland ecological structure health index of the to-be-evaluated set ecological region; and an evaluation feedback module for taking preset ecological protection measures for the to-be-evaluated set ecological region based on the wetland ecological structure health index.

[0016] The present invention has the following beneficial effects:

[0017] (1) The geographical information assessment method based on intelligent algorithms constructs a multi-dimensional wetland structure assessment system. Starting from three core dimensions of geomorphology, hydrology, and vegetation, it comprehensively extracts and analyzes various ecological structure characteristics based on relevant geographical information data with the help of intelligent algorithms, and on this basis, integrates to form a wetland ecological structure health index. Thus, it can comprehensively reflect the suitability and stability characteristics of the wetland ecosystem at the natural structure level and provide more scientific and accurate support for protection and intervention. For example, in a certain assessment area, if the vegetation structure health index is relatively high, but the water structure ecological balance index is relatively low, the system will automatically identify through the coupling weight that there is an imbalance phenomenon in the area where the vegetation is well represented but the hydrological structure is fragmented, so as to avoid the risk of overall misjudgment caused by the goodness of a certain dimension, and then effectively improve the authenticity and accuracy of the assessment results.

[0018] (2) The geographical information assessment method based on intelligent algorithms realizes a closed-loop collaborative mechanism from macro comprehensive assessment to ecological intervention decision-making by constructing an overall wetland ecological structure health index. After integrating multi-source structure indexes such as geomorphology, hydrology, and vegetation, the system forms a unified assessment result with spatial ecological meaning, and accordingly makes an overall judgment on the health level of the entire set ecological area. According to the comparison between the assessment result and the preset threshold, the system will automatically trigger corresponding levels of ecological management strategies, thus avoiding one-sided interpretation of indicators, and then strengthening the ability of structural coupling response analysis at the regional scale, and then improving the overall adaptability and response scientificity of ecological protection measures.

[0019] (3) The geographical information assessment method based on intelligent algorithms introduces a deep convolutional neural network architecture in the process of extracting geomorphological structure characteristics, and conducts multi-stage structure modeling for the edge structure and spatial distribution characteristics of wetland geomorphological images, thus improving the recognition sensitivity and expression accuracy of the model for ecological disturbance characteristics. For example, by constructing a multi-layer neural network and combining structural designs such as dilated convolution, direction-sensitive convolution, and skip connection, it effectively captures key features in high-interference areas such as sudden slope changes, fracture structures, and boundary coherence in wetland images, and quantitatively extracts them to generate a geomorphological assessment index set, thereby strengthening the recognition ability of complex ecological structures in the image, and then ensuring the generalization and adaptation ability of the model in different wetland areas.

[0020] (4) The geographic information evaluation system based on intelligent algorithms realizes the high integration of ecological evaluation tasks and the intelligent closed-loop of the processing flow by constructing a full-process collaborative architecture. The system can not only be compatible with multiple types of geographic information data sources, but also accurately identify and structurally model different ecological factors through the structural feature extraction module, so as to quickly generate a wetland ecological structure health index with regional representativeness in the comprehensive evaluation module, thereby significantly improving the processing efficiency and result consistency of executing large-scale ecological evaluation tasks in complex environments. The evaluation results output by the system can directly drive the intelligent response mechanism, thus having strong deployment flexibility and dynamic response capabilities, and then strengthening the response closed-loop and scenario integration capabilities of the ecological governance system.

[0021] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flowchart of a geographic information evaluation method based on intelligent algorithms of the present invention.

[0023] Figure 2 It is a flowchart of the specific steps for obtaining the geomorphic structure ecological suitability index of each area of the to-be-evaluated set ecological area in a geographic information evaluation method based on intelligent algorithms of the present invention.

[0024] Figure 3 It is a sequence diagram of geomorphic stability index areas in a geographic information evaluation method based on intelligent algorithms of the present invention.

[0025] Figure 4 It is a sequence diagram of boundary continuity index areas in a geographic information evaluation method based on intelligent algorithms of the present invention.

[0026] Figure 5 It is a sequence diagram of terrain fracture interference index areas in a geographic information evaluation method based on intelligent algorithms of the present invention.

[0027] Figure 6 It is a block diagram of a geographic information evaluation system based on intelligent algorithms of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] Please refer to Figure 1, an embodiment of the present invention provides a technical solution: a geographical information evaluation method based on intelligent algorithms, including the following steps: obtaining geographical information data of several areas of a to-be-evaluated set ecological area (such as a wetland reserve), where the geographical information data includes wetland topographic image data, wetland hydrogeographical data, and wetland vegetation geographical data; respectively performing feature extraction on the geographical information data of each area of the to-be-evaluated set ecological area to obtain a wetland structure evaluation index set for each area of the to-be-evaluated set ecological area, including a topographic structure ecological suitability index, a water structure ecological balance index, and a vegetation structure health index; comprehensively analyzing the evaluation index sets of each area of the to-be-evaluated set ecological area to obtain a wetland ecological structure health index of the to-be-evaluated set ecological area; and taking preset ecological protection measures for the to-be-evaluated set ecological area based on the wetland ecological structure health index.

[0029] The specific formula for calculating the wetland ecological structure health index of the to-be-evaluated set ecological area is as follows: where KsT is the wetland ecological structure health index of the to-be-evaluated set ecological area, and DsT i is the topographic structure ecological suitability index of the i-th area of the to-be-evaluated set ecological area, α1 is the topographic adjustment coefficient stored in the database, and SgW i is the water structure ecological balance index of the i-th area of the to-be-evaluated set ecological area, α2 is the water structure adjustment coefficient stored in the database, and ZbG i is the vegetation structure health index of the i-th area of the to-be-evaluated set ecological area, α3 is the vegetation adjustment coefficient stored in the database, α4 is the collaborative penalty adjustment coefficient stored in the database, α5 is the vegetation imbalance penalty adjustment coefficient stored in the database, ζ is the smoothing coefficient stored in the database, and its value is 0.01 in this embodiment, α6 is the comprehensive adjustment coefficient stored in the database, i = 1, 2, 3,..., i0, and i0 is the number of areas.

[0030] It should be explained that in the formula this term serves as the denominator penalty term of the structural index to control the output of the wetland ecological structure health index and avoid being too high or too low. The penalty will only weaken when all aspects of the structure are uniform and coordinated.

[0031] α1, α2, α3, α4, α5, α6 can be obtained through the following steps: Using historical data, combined with the ecological suitability index of geomorphic structure, the ecological stability index of water structure, and the health index of vegetation structure, conduct statistical regression analysis to quantify the specific impacts of various factors on the health index of wetland ecological structure, so as to fit the initial weight values. Secondly, adopt the sensitivity analysis method to adjust the value range of each coefficient and observe its impact on the evaluation results of wetland ecological structure health to ensure the stability and rationality of the model. Based on the characteristics of ecological regions and actual situations, correct and optimize the preliminarily fitted coefficients, and finally determine the coefficient values applicable to specific ecological regions.

[0032] Specifically, as Figure 2 shown, the wetland geomorphic image data is specifically the pixel value and two-dimensional coordinates of each pixel point in the wetland geomorphic image, and the specific steps to obtain the ecological suitability index of the geomorphic structure of each area in the to-be-evaluated set ecological region are as follows: Input the wetland geomorphic image data of each area in the to-be-evaluated set ecological region into the pre-trained geomorphic recognition model for comprehensive analysis to obtain the geomorphic evaluation index set of each area in the to-be-evaluated set ecological region, including the geomorphic stability index (representing the degree of terrain undulation balance and disturbance distribution state of the wetland area), the terrain fracture interference index (representing the degree of geomorphic cutting caused by natural fragmentation or human construction in the wetland area), and the boundary continuity index (representing the natural closure and coherence of the wetland area in space); conduct comprehensive analysis on the geomorphic evaluation index set of each area in the to-be-evaluated set ecological region to obtain the ecological suitability index of the geomorphic structure of each area in the to-be-evaluated set ecological region.

[0033] The specific formula for calculating the ecological suitability index of the geomorphic structure of each area in the to-be-evaluated set ecological region is as follows: Among them, DsT i is the ecological suitability index of the geomorphic structure of the i-th area in the to-be-evaluated set ecological region, DmW i is the geomorphic stability index of the i-th area in the to-be-evaluated set ecological region, φ1 is the geomorphic stability coefficient stored in the database, η1 is the geomorphic stability adjustment coefficient stored in the database, BxL i is the boundary continuity index of the i-th area in the to-be-evaluated set ecological region, φ2 is the boundary continuity coefficient stored in the database, η2 is the boundary continuity adjustment coefficient stored in the database, DgR iLet \( \varphi_{3i} \) be the terrain fracture interference index of the \( i \)-th area of the ecological region to be evaluated, \( \varphi_3 \) be the fracture interference coefficient stored in the database, \( \eta_3 \) be the fracture interference adjustment coefficient stored in the database, \( \chi \) be the suppression adjustment coefficient stored in the database (to avoid a zero denominator), and in this implementation example, it takes the value of \( 0.1 \), \( \eta_4 \) be the interaction adjustment coefficient stored in the database, and \( \varphi_1+\varphi_2+\varphi_3 = 1 \), \( i = 1,2,3,\cdots,i_0 \), where \( i_0 \) is the number of areas.

[0034] It should be noted that in the formula This term adopts a main structure enhancement term and a secondary interference compensation term, so that the geomorphic stability index and the boundary continuity index are enhanced through a logarithmic function combination, the terrain fracture interference index is suppressed in a reciprocal compression form, and an independent weight coefficient is introduced to control its comprehensive contribution degree, so as to realize the ecological suitability index of the geomorphic structure with ecological logic.

[0035] \( \varphi_1 \), \( \varphi_2 \), \( \varphi_3 \) can be obtained through the following steps: Read the geomorphic stability index, terrain fracture interference index, and boundary continuity index of each area of the ecological region to be evaluated, and perform a mean analysis to obtain the mean values of the geomorphic stability index, terrain fracture interference index, and boundary continuity index of the ecological region to be evaluated, and perform a summation analysis to obtain the ecological suitability sum value, and perform a ratio analysis of the mean values of the geomorphic stability index and terrain fracture interference index of the ecological region to be evaluated with the ecological suitability sum value, and use the ratio analysis results as the corresponding coefficients.

[0036] \( \eta_1 \), \( \eta_2 \), \( \eta_3 \), \( \eta_4 \) can be obtained through the following steps: Based on historical data, determine the initial influence weights of each variable (geomorphic stability index, terrain fracture interference index, boundary continuity index) on the ecological suitability index of the geomorphic structure through statistical regression analysis. Then, use the sensitivity analysis method to adjust the value range of the coefficients to evaluate the stability and applicability of these parameters to the formula output. Next, further fit the weights through model optimization (such as machine learning algorithms or multi-objective optimization) to ensure that the formula can accurately reflect the ecological suitability of the actual geomorphic structure.

[0037] The specific implementation example of calculating the ecological suitability index of the geomorphic structure of each area of the ecological region to be evaluated is as follows. Now, randomly select the geomorphic stability index, boundary continuity index, and terrain fracture interference index of 5 areas of the ecological region to be evaluated, as shown in Table 1 and Figures 3 - 5 shown below:

[0038] Table 1 Data example of the geomorphic evaluation index of the ecological region to be evaluated

[0039] Geomorphic stability index Boundary continuity index Topographic fracture interference index The first area 0.783 0.719 0.134 The second area 0.672 0.615 0.317 The third area 0.694 0.715 0.331 The fourth area 0.813 0.594 0.418 The fifth area 0.724 0.748 0.295

[0040] The geomorphic stability coefficient φ1 stored in the database is approximately: 0.394;

[0041] The boundary continuity coefficient φ2 stored in the database is approximately: 0.259;

[0042] The fracture interference coefficient φ3 stored in the database is approximately: 0.347;

[0043] The geomorphic stability adjustment coefficient η1 stored in the database is approximately: 0.427;

[0044] The boundary continuity adjustment coefficient η2 stored in the database is approximately: 0.315;

[0045] The fracture interference adjustment coefficient η3 stored in the database is approximately: 0.281;

[0046] The interaction adjustment coefficient η4 stored in the database is approximately: 0.693;

[0047] The suppression adjustment coefficient stored in the database is 0.1;

[0048] Substitute the data in Table 1 and the above coefficients and adjustment coefficients into the specific formula for calculating the geomorphic structure ecological suitability index of each area of the to-be-evaluated set ecological region, and obtain:

[0049] The geomorphic structure ecological suitability index of the first area of the to-be-evaluated set ecological region = (ln(1 + 0.394×0.783 0.427 + 0.259×0.719 0.315 )) + 0.347 / (0.1 + 0.134 0.281 )) / (1 + exp(-0.693×0.134×√|0.783 - 0.719|)) ≈ 0.496;

[0050] The geomorphic structure ecological suitability index of the second area of the to-be-evaluated set ecological region = (ln(1 + 0.394×0.672 0.427 + 0.259×0.615 0.315 )) + 0.347 / (0.1 + 0.317 0.281 )) / (1 + exp(-0.693×0.317×√|0.672 - 0.615|)) ≈ 0.438;

[0051] The geomorphic structure ecological suitability index of the third area of the to-be-evaluated set ecological region = (ln(1 + 0.394×0.694 0.427 + 0.259×0.715 0.315 )) + 0.347 / (0.1 + 0.331 0.281)) / (1 + exp(-0.693×0.331×√|0.694 - 0.715|)) ≈ 0.437;

[0052] The ecological suitability index of the geomorphic structure of the fourth area of the ecological area to be evaluated and set = (ln(1 + 0.394×0.813 0.427 + 0.259×0.594 0.315 ) + 0.347 / (0.1 + 0.418 0.281 )) / (1 + exp(-0.693×0.418×√|0.813 - 0.594|)) ≈ 0.457;

[0053] The ecological suitability index of the geomorphic structure of the fifth area of the ecological area to be evaluated and set = (ln(1 + 0.394×0.724 0.427 + 0.259×0.748 0.315 ) + 0.347 / (0.1 + 0.295 0.281 )) / (1 + exp(-0.693×0.295×√|0.724 - 0.748|)) ≈ 0.454.

[0054] In this implementation plan, through depth image analysis, local undulations, fault zones, and boundary closed structures existing in the wetland terrain are automatically identified, covering three ecologically significant structural dimensions of stability, disturbance, and continuity, thereby avoiding the defects of relying on a single indicator or surface characteristics. Secondly, by using the geomorphic stability index, terrain fracture interference index, and boundary continuity index as the core input parameters, the comprehensive state of the ecological area in terms of structural dimensions such as undulation balance, fragmentation interference risk, and spatial closure can be comprehensively reflected, thereby effectively improving the physical basis and interpretability of ecological suitability assessment. Finally, by introducing interaction term coefficients, the modeling of the interaction effects between geomorphic factors is realized, and typical ecological conflict situations such as stable areas with severe fragmentation and coherent boundaries but violent undulations are identified, thereby improving the adaptability and discrimination ability of the model to complex geomorphic patterns, and converting multiple geomorphic factors into a unified structural ecological suitability index, which can be directly used as the quantitative basis for subsequent spatial decisions such as ecological replanting, hydrological regulation, and boundary restoration, thereby improving the practicality and feasibility of the intelligent assessment system.

[0055] Specifically, the landform recognition model is specifically a deep convolutional neural network. The deep convolutional neural network includes an input layer, several convolutional layers, an edge enhancement layer, an upsampling structure layer, a spatial statistics layer, and an output layer. The specific steps to obtain the landform evaluation index set for each area of the to-be-evaluated set ecological area are as follows: In the input layer of the deep convolutional neural network, receive the wet landform image data of each area of the to-be-evaluated set ecological area (i.e., the pixel values and two-dimensional coordinates of each pixel point in the wet landform image), and perform preprocessing; in the convolutional layer of the deep convolutional neural network, perform image feature extraction on the preprocessed wet landform image data of each area of the to-be-evaluated set ecological area (extract the local slope pattern and texture features of the wet landform image through multi-scale convolution operations, and obtain the response information to different scale perturbation structures. In this process, the texture differences of high-frequency undulations and slope mutations in the local area will be clearly identified. Secondly, enhance the feature response of the mutation areas in the image, such as fractures and edge mutations, through the ReLU activation function, and improve the stability of the feature map distribution and the layer-to-layer gradient propagation efficiency through batch normalization operations. Subsequently, compress the spatial size through max-pooling operations and retain the strongest structural feature responses, effectively extracting the slope difference concentration points and perturbation extreme points in the high-impact areas), to obtain the multi-channel landform feature maps of each area of the to-be-evaluated set ecological area (the set of spatial feature expressions generated after the convolutional layer processing, and each channel represents the response intensity of a landform feature type, such as slope undulation texture, edge intensity distribution, patch continuity, local elevation difference standard deviation, structural balance, slope perturbation density, etc.); in the edge enhancement layer of the deep convolutional neural network, perform edge structure enhancement processing on the multi-channel landform feature maps of each area of the to-be-evaluated set ecological area (introduce direction-sensitive convolution to extract the lateral, longitudinal, and oblique landform splitting features, and construct the terrain fracture structure diagram within the area; use dilated convolution to expand the receptive field and identify potential coherent boundary lines with larger spans; fuse multi-scale edge responses to enhance the perception of different scale boundary manifestation forms; and use the activation function to strengthen the response value distribution of the edge mutation areas, so that high-perturbation areas such as fracture zones and boundary incisions form high-response clusters in the feature map), to obtain the enhanced edge feature map sets of each area of the to-be-evaluated set ecological area (including but not limited to edge response intensity, edge direction distribution, edge pixel density, edge closed-loop degree, edge line connectivity, etc.); in the upsampling structure layer of the deep convolutional neural network, perform spatial restoration processing on the enhanced edge feature map sets of each area of the to-be-evaluated set ecological area (use transposed convolution or bilinear interpolation method to restore the size of the enhanced edge feature map, and restore it from the low-resolution state to the same as the original wet landform Figure OneSecondly, using the skip connection mechanism, the upsampled deep edge structure map is channel-concatenated with the shallow high-resolution edge detail features extracted from the previous layer of the convolutional network to fuse semantic information and spatial location information. Subsequently, the concatenated result is processed by multiple convolutional fusion layers to integrate the structure, eliminate invalid interference responses, and strengthen the real boundary structure, ensuring clear edge morphology and complete structural contours, to obtain a set of structure-restored edge images for each area of the evaluated and set ecological region (where each channel represents boundary types in different directions and scales). In the spatial statistics layer of the deep convolutional neural network, regional structure statistics are performed on the set of structure-restored edge images for each area of the evaluated and set ecological region (for each semantic channel of the set of structure-restored edge images, such as the slope response channel, fracture structure channel, boundary continuity channel, etc., processing methods such as sliding window analysis, pixel direction analysis, connected graph modeling, and contour tracing are respectively used to extract structural statistical features with clear spatial physical meanings; each statistical feature is derived from the corresponding image channel response behavior and obtains quantitative values through local distribution features, global distribution trends, or high-order image structure analysis, and then composes a geomorphic feature vector), to obtain a geomorphic feature vector for each area of the evaluated and set ecological region, which includes but is not limited to the extraction of the standard deviation of local elevation difference (dividing the image into multiple blocks, calculating the standard deviation of each edge intensity map block and averaging), slope disturbance density (counting slope mutation patches, which can be detected by Laplace or second derivative), standard deviation of edge response intensity (standard deviation of all pixel responses), slope direction gradient index (calculating the gradient direction for each pixel and measuring with variance), fracture line density (extracting line segments in the edge map + unit area statistics), mutation point density (detecting edge mutation points using the Laplacian / DoG method), fracture direction distribution entropy (performing distribution entropy analysis on all line segment directions); boundary closure (counting the number of closed / non-closed contours after contour extraction), boundary connectivity (proportion of the maximum connected boundary length), and edge break point rate (dividing the number of break points in the pixel chain / contour map by the total number of nodes); in the output layer of the deep convolutional neural network, feature fusion is performed on the geomorphic feature vector for each area of the evaluated and set ecological region (normalizing the structural statistical features in each dimension, enhancing the response weights of key disturbance features through a non-linear mapping function, and respectively constructing an exponential generation model. For the geomorphic stability index, statistical features closely related to terrain stability are selected from the geomorphic feature vector, including the standard deviation of local elevation difference, slope disturbance density, standard deviation of edge response intensity, and slope direction gradient index. The above features are respectively normalized, and an exponential function mapping is introduced for the normalized disturbance features to enhance the negative impact of high-disturbance regions, that is, the stronger the disturbance, the faster the index value drops. Then, the above mapped features are weighted and combined according to empirical weights or learned weights to form a fusion value, and finally the output value is compressed by the sigmoid function to ensure that the geomorphic stability index is between 0 and 1;For the topographic fracture interference index, extract the fracture line density, mutation point density, fracture direction distribution entropy, and standardize all fracture-related features. Then, logically fuse multiple fracture-related indicators, such as weighted summation or cascade judgment, to construct the interference degree level. Finally, compress the fusion result through the sigmoid function to ensure that the topographic fracture interference index is between 0 and 1. For the boundary continuity index, extract the boundary closure degree, boundary connectivity, and edge break point rate, and perform normalization processing respectively. Then, fuse each index based on Gaussian distribution or empirical weighting. Finally, compress the fusion result through the sigmoid function to ensure that the boundary continuity index is between 0 and 1, obtaining the geomorphic stability index, topographic fracture interference index, and boundary continuity index of each area of the to-be-evaluated set ecological area, that is, the geomorphic evaluation index set.

[0056] Among them, the input layer is used to receive and preprocess the wet landform image data, and complete the size specification and pixel normalization.

[0057] The convolutional layer is used to extract basic geomorphic features such as slope, texture, and edge, and enhance the response of the mutation area.

[0058] The edge enhancement layer is used to strengthen the structural information such as fracture zones and boundary lines, and highlight the splitting direction and edge continuity.

[0059] The upsampling structure layer is used to restore the spatial resolution of the feature map, fuse shallow details and deep semantics, and generate a structure-restored edge image.

[0060] The spatial statistics layer is used to statistically extract structural indicators with physical significance from the restored map and construct a geomorphic feature vector.

[0061] The output layer is used to fuse the feature vector and generate a normalized geomorphic stability index, fracture interference index, and boundary continuity index.

[0062] And the pre-training process of the deep convolutional neural network is as follows:

[0063] Obtain the wetland appearance dataset, which contains typical wetland geomorphic image samples. The image samples cover multiple types of geomorphic structure areas, including stable terrain areas, fracture interference areas, and boundary fragmentation areas, and divide the wetland appearance dataset into a training set and a validation set.

[0064] Initialize the deep convolutional neural network, that is, use the He initialization method for all convolutional layer weights to adapt to the ReLU activation function used in the network and improve the feature convergence efficiency in the initial stage of training. Then, initialize the bias terms, that is, initialize all bias terms to zero to ensure that the network calculation does not deviate in the initial state, and select the activation function. For example, use the ReLU activation function for all intermediate layers, and use the transposed convolution method to initialize the parameters of the upsampling layer. Finally, initialize the loss function, that is, construct a multi-task loss function that includes the structure contour reconstruction loss and the structure category prediction loss, which are used to jointly optimize the feature map quality and the structure semantic recognition ability.

[0065] Train based on the training set, set the number of training loops (such as 100 times), and the processing steps for each training loop are, in turn, input the training images (randomly extract several wetland landform image samples from the training set, including stable regions, fracture regions, and boundary splitting regions, and input their corresponding structure label maps), forward propagation processing (gradually extract features and generate a set of landform evaluation indices through multiple architecture layers, etc.), loss calculation (compare the output set of landform evaluation indices with the true set of landform evaluation indices and calculate the loss, such as cross-entropy or structural similarity loss), and backpropagation to update the parameters (use the Adam optimizer, backpropagate the gradient based on the loss function, and update the parameters of each layer).

[0066] After each training loop ends, conduct an evaluation and analysis based on the validation set, that is, forward evaluation processing (input the images in the validation set into the trained model, output the generated set of landform evaluation indices, and calculate the mean square error with the manual evaluation result, that is, calculate the loss value), and calculate the loss function value and accuracy of the model on the validation set to evaluate the performance of the model. Based on the evaluation result, if the loss on the validation set does not decrease or the accuracy does not increase, adjust the hyperparameters of the network (such as the learning rate, network structure, etc.), and use the early stopping (EarlyStopping) technique. If the loss on the validation set does not improve significantly within several training cycles, stop the training to prevent overfitting.

[0067] When the training is completed and the loss and accuracy on the validation set reach the expected standards, the training process ends, and a trained network model is obtained.

[0068] In this implementation plan, by adopting a deep convolutional neural network as the landform recognition model, the structure extraction accuracy of ecological image data and the scientific construction of evaluation indexes are significantly enhanced. Secondly, through multi-layer convolution processing, the model effectively captures multi-scale textures and slope changes in wet landform images, thus having a strong response ability to landform undulation differences and boundary fragmentation structures. Especially in key areas such as fracture lines and slope mutation points, it shows higher resolution and perceptual sensitivity. Through the collaborative processing of the edge enhancement layer and the spatial statistics layer, the model can extract structural indexes with physical significance from the images, providing solid data support for the subsequent construction of landform stability indexes, fracture interference indexes, and boundary continuity indexes. Finally, after multiple training processes, the evaluation results output by the model are more in line with the actual state of the landform, thereby significantly improving the intelligent level of landform structure evaluation.

[0069] Specifically, wetland hydrogeographic data includes water depth values, water level fluctuation indexes, disturbance difference indexes, water level gradient indexes, water accumulation connectivity indexes, and edge water level expansion gradient indexes. The specific steps to obtain the water structure ecological stability index for each area of the to-be-evaluated set ecological area are as follows: comprehensively analyze the wetland hydrogeographic data for each area of the to-be-evaluated set ecological area respectively to obtain the hydrogeographic evaluation index set for each area of the to-be-evaluated set ecological area, including the hydrogeographic space storage stability index (measuring the stability of the water storage capacity of the water body in this area), the water level terrain disturbance index (measuring the spatial structural water level mutation trend in this area), and the hydrogeomorphic replenishment and recovery index (measuring the ability of the water body in this area to complete hydrogeomorphic replenishment under drought or low water level conditions); and comprehensively analyze the hydrogeographic evaluation index set for each area of the to-be-evaluated set ecological area to obtain the water structure ecological stability index for each area of the to-be-evaluated set ecological area.

[0070] Among them, the water depth value can be obtained by collecting water depths at several representative spatial position points (such as the central area, boundary area, low-lying core area, inflow point, outflow point adjacent points, etc.) pre-laid in this area (which can be through an ultrasonic water depth sensor), and performing mean processing. The obtained result is the water depth value.

[0071] The water level fluctuation index is the degree of water level fluctuation in this area. It can be obtained by collecting water depths at several representative spatial position points pre-laid in this area (which can be through an ultrasonic water depth sensor), and performing standard deviation processing. The obtained result is the water level fluctuation index.

[0072] The disturbance difference index measures the spatial concentration of local disturbances. It can be obtained by collecting water depths at several representative spatial location points pre - arranged in the area (and uploading the results to the database), and obtaining the corresponding three - dimensional position coordinates (which can be obtained by an RTK - GPS locator and then uploaded to the database). The water depth values of each representative spatial location point are sorted in descending order, and the three - dimensional position coordinates of each group of adjacent representative spatial location points after the descending order are comprehensively analyzed (that is, calculating the distance between adjacent representative spatial location points based on the Euclidean distance formula and performing mean processing). The resulting value is the disturbance difference index.

[0073] The water level gradient index is the spatial gradient intensity of the water level fluctuation in the area. It can be obtained by collecting water depths at several representative spatial location points pre - arranged in the area, obtaining the corresponding three - dimensional position coordinates, statistically analyzing to obtain the spatial location point with the maximum water depth and the spatial location point with the minimum water depth, and performing a difference process (that is, the maximum water depth value - the minimum water depth value). Then, analyzing the three - dimensional position coordinates of these two spatial location points to obtain the distance value between them, and based on the difference process result / distance value, the resulting value is the water level gradient index.

[0074] The ponding connectivity index measures the degree of connectivity of the ponding area in the spatial structure of the area. It can be obtained by collecting water depths at several representative spatial location points pre - arranged in the area, obtaining the corresponding three - dimensional position coordinates, and judging the connectivity for any two spatial location points, that is, the absolute value of the difference between the water depth values of the two spatial location points is less than or equal to the set difference threshold (5 cm), and the distance between the two spatial location points is less than or equal to the set distance threshold (10 m), then the two points are regarded as connected points. Count the number of connected points, and perform a ratio process with the total number of spatial location points. The resulting value is the ponding connectivity index.

[0075] The edge water level expansion gradient index evaluates the degree of diffusion of the ponding at the boundary of the area to the edge area. It can be obtained by, within the boundary range of the area, obtaining the water depth values of each point in the boundary area in real - time through multiple preset water level collection points, and obtaining the corresponding three - dimensional position coordinates. The water depth values and three - dimensional position coordinates of each water level collection point are comprehensively analyzed respectively (that is, the absolute value of the difference between the water depth values of two adjacent water level collection points, calculating the distance between two adjacent water level collection points based on the Euclidean distance formula), obtaining several groups of water depth differences and water level distance values between adjacent water level collection points, and performing a ratio analysis (water depth difference / water level distance value). Based on the ratio analysis results, perform mean processing. The resulting value is the edge water level expansion gradient index.

[0076] The specific steps to obtain the hydrological evaluation index set for each area of the to-be-evaluated set ecological region are as follows: Obtain the water level stability rate value for each area of the to-be-evaluated set ecological region, and conduct comprehensive analysis in combination with the water depth value and the water level fluctuation index (that is, first perform normalization processing, and then perform weighted processing based on the normalization processing result) to obtain the hydrological spatial storage stability index for each area of the to-be-evaluated set ecological region; Obtain the break point water level difference value for each area of the to-be-evaluated set ecological region, and conduct comprehensive analysis in combination with the disturbance difference index and the water level gradient index (the comprehensive analysis logic is the same as that of the hydrological spatial storage stability index) to obtain the water level topographic disturbance index for each area of the to-be-evaluated set ecological region; Obtain the environmental correction factor for each area of the to-be-evaluated set ecological region, and conduct comprehensive analysis in combination with the water accumulation connectivity index and the edge water level expansion gradient index (the comprehensive analysis logic is the same as that of the hydrological spatial storage stability index) to obtain the hydrological geomorphic replenishment and restoration index for each area of the to-be-evaluated set ecological region.

[0077] Among them, the water level stability rate value is the degree of water level stability. It can be obtained by collecting water depths at several representative spatial position points pre-laid in this area (which can be done through ultrasonic water depth sensors), and then conducting ratio analysis (i.e., water depth value / water depth reference value) with the corresponding water depth reference value (which can be obtained by taking the mean of the historical water depth values of several times at each spatial position point). Based on the ratio analysis result, weighted processing is performed, and the resulting value is the water level stability rate value.

[0078] The break point water level difference value measures the spatial difference degree of the water level in the wetland boundary area. It can be obtained by arranging multiple water level collection points within the boundary range of this area to obtain the water level values of each point in the boundary area in real time, and then extracting the extreme difference value of the water level in this area, that is, the difference between the maximum water level value and the minimum water level value.

[0079] The environmental correction factor is affected by the meteorological environment (such as wind speed, air temperature, etc.) of this area. It can be obtained by obtaining the wind speed value (obtained by a wind speed sensor), the air temperature value (obtained by a temperature sensor), the humidity value (obtained by a humidity sensor), the wind speed reference value, the air temperature reference value, the humidity reference value, and then conducting ratio processing respectively (such as the absolute value of the difference between the wind speed value and the wind speed reference value / the wind speed reference value), and based on the ratio processing result, weighted processing is performed. The resulting value is the environmental correction factor. The wind speed reference value can be obtained by taking the mean of the historical wind speed values of several times, and the resulting value is the wind speed reference value. The acquisition logics of the air temperature reference value and the humidity reference value are the same as that of the wind speed reference value.

[0080] And the specific formula for calculating the water structure ecological stability index for each area of the to-be-evaluated set ecological region is as follows: Among them, SgW iThe water structure ecological stability index of the \(i\)th area of the ecological region to be evaluated, \(TwK\) i The hydrological spatial storage stability index of the \(i\)th area of the ecological region to be evaluated, \(\mu1\) is the storage stability adjustment coefficient stored in the database, \(ShF\) i The hydrogeomorphic recharge and restoration index of the \(i\)th area of the ecological region to be evaluated, \(\mu2\) is the recharge and restoration adjustment coefficient stored in the database, \(SwD\) i The water level and terrain disturbance index of the \(i\)th area of the ecological region to be evaluated, \(\mu3\) is the water level and terrain disturbance adjustment coefficient stored in the database, \(\mu4\) is the superposition adjustment coefficient stored in the database, \(i = 1, 2, 3, \ldots, i0\), and \(i0\) is the number of areas.

[0081] It should be explained that in the formula This term is used to adjust the superposition effect of the hydrological spatial storage stability index and the hydrogeomorphic recharge and restoration index, avoiding the water structure ecological stability index being too high or too low.

[0082] \(\mu1\), \(\mu2\), \(\mu3\), \(\mu4\) can be obtained through the following steps: Based on historical data, determine the initial influence weights of each variable (hydrological spatial storage stability index, water level and terrain disturbance index, hydrogeomorphic recharge and restoration index) on the water structure ecological stability index through statistical regression analysis. Then, use the sensitivity analysis method to adjust the value range of the coefficients to evaluate the stability and applicability of these parameters to the formula output. Next, further fit the weights through model optimization (such as machine learning algorithms) to ensure that the formula can accurately reflect the actual stability state of the water structure ecology.

[0083] In this implementation plan, by introducing multi-dimensional hydrogeographical parameters with spatial physical significance, a water structure ecological information acquisition system is constructed, and on this basis, three core structural indices are designed and formed, thus avoiding the limitation of relying only on single static data in wetland hydrological assessment, and realizing the multi-angle capture and modeling of the dynamics, spatial structure and functional resilience of hydrological elements. Secondly, in the fusion calculation process of the water structure ecological stability index, a normalization, distribution difference evaluation and adjustment coefficient coupling mechanism is adopted, which can effectively control the mutual influence between various hydrological factors, thus avoiding misjudgment of structural imbalance caused by single-factor anomalies. Finally, through the weight adjustment and model fitting mechanism, the adaptability of each parameter in different ecological regions is ensured, thereby improving the universality and stability of index evaluation in complex wetland systems, and providing a key basic support for realizing the scientific assessment, dynamic monitoring and protection intervention strategies of ecological hydrological structures.

[0084] Specifically, the wetland vegetation geographical data includes the vegetation patch concentration index, the spatial deviation index of root zone water response, and the water content distribution index. The specific steps to obtain the vegetation structure health index for each area of the to-be-evaluated set ecological region are as follows: Standardize the vegetation patch concentration index, the spatial deviation index of root zone water response, and the water content distribution index for each area of the to-be-evaluated set ecological region; and comprehensively analyze (i.e., weighted processing) the standardized vegetation patch concentration index, the spatial deviation index of root zone water response, and the water content distribution index for each area of the to-be-evaluated set ecological region to obtain the vegetation structure health index for each area of the to-be-evaluated set ecological region.

[0085] Among them, the vegetation patch concentration index measures the spatial aggregation degree of plants in this area. It can be obtained by acquiring the area values of several patch areas in this area (a vegetation distribution area with obvious spatial boundaries composed of continuously distributed or spatially adjacent homogeneous plant communities within a certain geographical area, that is, an area where plants grow significantly while there are no or very sparse plants around) (acquire the boundary inflection point coordinates of each patch area based on an RTK-GPS locator, calculate based on spatial geometric formulas, and then upload the results to the database), the three-dimensional position coordinates (i.e., the three-dimensional position coordinates of the midpoint of the patch, which can be obtained by an RTK-GPS locator and then upload the results to the database), and perform mean processing and statistical processing respectively to obtain the mean patch area and the maximum patch area, and perform ratio processing (maximum patch area / mean patch area of the patch area), then perform distance analysis on the three-dimensional position coordinates of each patch area to obtain the patch distance values of several groups of adjacent patch areas, perform standard deviation processing, and perform weighted processing based on the ratio processing results and the standard deviation processing results. The obtained result is the vegetation patch concentration index.

[0086] The spatial deviation index of root zone water response measures the balance of water supply in the root zone of plants in this area. It can be obtained by acquiring the root zone water values (which can be obtained by soil moisture sensors) and the corresponding three-dimensional position coordinates of multiple representative vegetation positions preset in this area (such as the centers and edges of different vegetation patches), performing difference analysis on the root zone water values of each group of adjacent vegetation positions (i.e., the absolute value of the difference between the root zone water values of adjacent vegetation positions) to obtain the water difference of each group of adjacent vegetation positions, calculating the vegetation position distance of each group of adjacent vegetation positions, performing standard deviation processing on the water difference and vegetation position distance of each group of adjacent vegetation positions, and performing weighted processing based on the standard deviation processing results. The obtained result is the spatial deviation index of root zone water response.

[0087] The water content distribution index is used to evaluate the spatial fluctuations of the healthy water status of vegetation. It can be obtained by acquiring the leaf water content of multiple preset representative vegetation positions in the area (that is, randomly measuring the water content of one leaf of the vegetation with a leaf water content meter and uploading it to the database), as well as the corresponding three-dimensional position coordinates. Perform a difference analysis on the leaf water content of each group of adjacent vegetation positions to obtain the difference in leaf water content between each group of adjacent vegetation positions, calculate the distance between each group of adjacent vegetation positions, and perform standard deviation processing on the difference in leaf water content and distance between each group of adjacent vegetation positions. Based on the results of the standard deviation processing, perform weighted processing, and the obtained result is the water content distribution index.

[0088] In this implementation plan, by constructing a triple structural index system, the spatial, structural, and dynamic precise evaluation of the healthy state of vegetation in the wetland ecosystem is realized. Secondly, the spatial distribution characteristics of vegetation, the underground water response ability, and the above-ground physiological state are comprehensively considered to reflect the true state of wetland vegetation in terms of structural integrity and functional adaptability from multiple dimensions. Finally, through spatial positioning and standard deviation analysis, each index is ensured to have refined spatial representativeness and data stability, so that the finally formed vegetation structure health index not only has high interpretability, but also can provide accurate basis for wetland replanting and restoration, hydrological regulation, and ecological monitoring strategy formulation, thereby improving the effectiveness and scientific nature of ecological management.

[0089] Specifically, the specific steps for taking preset ecological protection measures for the to-be-evaluated set ecological area based on the wetland ecological structure health index are as follows: Judge and analyze the wetland ecological structure health index of the to-be-evaluated set ecological area with the preset wetland ecological structure health index threshold; if the wetland ecological structure health index of the to-be-evaluated set ecological area is lower than or equal to the preset wetland ecological structure health index threshold, then take the first ecological protection measure, which is specifically a structural repair-level ecological intervention, that is, artificial vegetation reinforcement (implementing artificial replanting and guiding the restoration of wetland native populations in areas with sparse vegetation), hydrological regulation (setting ecological water replenishment or adjusting flood diversion and regulation measures in areas with large hydrological disturbances), and geomorphic fracture repair (carrying out micro-geomorphic reconstruction and boundary connectivity repair in areas with obvious geomorphic structure fractures); if the wetland ecological structure health index of the to-be-evaluated set ecological area is higher than the preset wetland ecological structure health index threshold, then take the second ecological protection measure, which is specifically an ecological stability maintenance plan, that is, through monitoring, isolation, and ecological diversity maintenance measures (such as continuing to implement the zoned ecological monitoring mechanism, strengthening the dynamic monitoring of key structural parameters, setting buffer forest belts and herbaceous filter zones in the edge areas, and promoting the protection of inter-species diversity), to ensure its long-term structural stability and functional coherence.

[0090] In this implementation, by introducing a hierarchical response mechanism centered around the wetland ecological structure health index, the intelligent triggering and dynamic hierarchical management of ecological protection work have been achieved. Secondly, when the health index is in the low-value range, the system will identify it as a high-priority governance object and actively enable structure repair measures. Starting from the three-dimensional structure of vegetation, hydrology, and landform, it targets the promotion of ecological reinforcement and function reconstruction, thereby shortening the response time from assessment to governance and enabling the intervention plan to be intelligently selected based on the actual structural state of the ecosystem, thus improving the accuracy, timeliness, and operability of governance measures. Finally, parametric adjustment is carried out according to the wetland area to be applicable to multi-level ecological protection projects from regional monitoring to special repair, providing strong technical support for the long-term stable operation of the wetland system.

[0091] Please refer to Figure 6 , an embodiment of the present invention provides a technical solution: a geographic information assessment system based on an intelligent algorithm, including: a data acquisition module for acquiring geographic information data of several areas in a to-be-assessed set ecological area, where the geographic information data includes wetland landform image data, wetland hydrogeographic data, and wetland vegetation geographic data; a structural feature extraction module for respectively extracting features from the geographic information data of each area in the to-be-assessed set ecological area to obtain a wetland structure assessment index set for each area in the to-be-assessed set ecological area, including a landform structure ecological suitability index, a water structure ecological balance index, and a vegetation structure health index; a comprehensive assessment module for comprehensively analyzing the assessment index sets of each area in the to-be-assessed set ecological area to obtain the wetland ecological structure health index of the to-be-assessed set ecological area; and an assessment feedback module for taking preset ecological protection measures for the to-be-assessed set ecological area based on the wetland ecological structure health index.

[0092] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they know the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0093] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims

1. A geographic information evaluation method based on intelligent algorithms, characterized in that, It includes the following steps: Obtain the geographic information data of several areas in the to-be-evaluated set ecological area, where the geographic information data includes wetland topographic image data, wetland hydrological geographic data, and wetland vegetation geographic data; Respectively perform feature extraction on the geographic information data of each area in the to-be-evaluated set ecological area to obtain the wetland structure evaluation index set of each area in the to-be-evaluated set ecological area, including the geomorphic structure ecological suitability index, the water structure ecological balance index, and the vegetation structure health index; Comprehensively analyze the evaluation index set of each area in the to-be-evaluated set ecological area to obtain the wetland ecological structure health index of the to-be-evaluated set ecological area; Based on the wetland ecological structure health index, take preset ecological protection measures for the to-be-evaluated set ecological area.

2. The geographic information evaluation method based on an intelligent algorithm according to claim 1, wherein The specific formula for calculating the wetland ecological structure health index of the to-be-evaluated set ecological area is as follows: Among them, KsT is the wetland ecological structure health index of the to-be-evaluated set ecological region, and DsT i , SgW i , ZbG i are respectively the geomorphic structure ecological suitability index, water structure ecological stability index, and vegetation structure health index of the i-th area of the to-be-evaluated set ecological region. α1, α2, α3, α4, α5, α6 are respectively the geomorphic adjustment coefficient, water structure adjustment coefficient, vegetation adjustment coefficient, collaborative penalty adjustment coefficient, vegetation imbalance penalty adjustment coefficient, and comprehensive adjustment coefficient stored in the database. ζ is the smoothing coefficient stored in the database. i = 1, 2, 3,..., i0, where i0 is the number of areas.

3. The geographic information evaluation method based on an intelligent algorithm according to claim 1, wherein The wetland topographic image data is specifically the pixel value and two-dimensional coordinates of each pixel point in the wetland topographic image, and the specific steps for obtaining the geomorphic structure ecological suitability index of each area in the to-be-evaluated set ecological area are as follows: Input the wetland topographic image data of each area in the to-be-evaluated set ecological area into a pre-trained geomorphic recognition model for comprehensive analysis to obtain the geomorphic evaluation index set of each area in the to-be-evaluated set ecological area, including the geomorphic stability index, the topographic fracture interference index, and the boundary continuity index; Comprehensively analyze the geomorphic evaluation index set of each area in the to-be-evaluated set ecological area to obtain the geomorphic structure ecological suitability index of each area in the to-be-evaluated set ecological area.

4. The geographic information evaluation method based on an intelligent algorithm according to claim 3, characterized in that The geomorphic recognition model is specifically a deep convolutional neural network. The deep convolutional neural network includes an input layer, several convolutional layers, an edge enhancement layer, an upsampling structure layer, a spatial statistics layer, and an output layer. The specific steps for obtaining the geomorphic evaluation index set of each area in the to-be-evaluated set ecological area are as follows: In the input layer of the deep convolutional neural network, receive the wetland topographic image data of each area in the to-be-evaluated set ecological area and perform preprocessing; In the convolutional layer of the deep convolutional neural network, perform image feature extraction on the preprocessed wetland topographic image data of each area in the to-be-evaluated set ecological area to obtain the multi-channel geomorphic feature map of each area in the to-be-evaluated set ecological area; In the edge enhancement layer of the deep convolutional neural network, perform edge structure strengthening processing on the multi-channel geomorphic feature map of each area in the to-be-evaluated set ecological area to obtain the enhanced edge feature map set of each area in the to-be-evaluated set ecological area; In the upsampling structure layer of the deep convolutional neural network, perform spatial restoration processing on the enhanced edge feature map set of each area in the to-be-evaluated set ecological area to obtain the structure restored edge image set of each area in the evaluated set ecological area; In the spatial statistics layer of the deep convolutional neural network, perform regional structure statistics processing on the structure restored edge image set of each area in the evaluated set ecological area to obtain the geomorphic feature vector of each area in the evaluated set ecological area; In the output layer of the deep convolutional neural network, feature fusion processing is performed on the geomorphic feature vectors of each area of the evaluated ecological region to obtain the geomorphic stability index, terrain fracture interference index, and boundary continuity index of each area of the evaluated ecological region, that is, the geomorphic evaluation index set.

5. The geographic information evaluation method based on intelligent algorithm according to claim 3, characterized in that The specific formula for calculating the geomorphic structure ecological suitability index of each area of the evaluated ecological region is as follows: Among them, DsT i , DmW i , BxL i , DgR i are respectively the geomorphic structure ecological suitability index, geomorphic stability index, boundary continuity index, and topographic fracture interference index of the i-th area of the ecological area to be evaluated. φ1, φ2, and φ3 are respectively the geomorphic stability coefficient, boundary continuity coefficient, and fracture interference coefficient stored in the database. η1, η2, η3, and η4 are respectively the geomorphic stability adjustment coefficient, boundary continuity adjustment coefficient, fracture interference adjustment coefficient, and interaction adjustment coefficient stored in the database. χ is the inhibition adjustment coefficient stored in the database, and φ1 + φ2 + φ3 = 1, i = 1, 2, 3, …, i0, where i0 is the number of areas.

6. The geographic information evaluation method based on an intelligent algorithm according to claim 1, characterized in that The wetland hydrogeographic data includes water depth value, water level fluctuation index, disturbance difference index, water level gradient index, water accumulation connectivity index, and edge water level expansion gradient index. The specific steps for obtaining the water structure ecological stability index of each area of the evaluated ecological region are as follows: Comprehensive analysis is respectively performed on the wetland hydrogeographic data of each area of the evaluated ecological region to obtain the hydrogeological evaluation index set of each area of the evaluated ecological region, including hydrogeological spatial storage stability index, water level terrain disturbance index, and hydrogeomorphic recharge recovery index; And comprehensive analysis is performed on the hydrogeological evaluation index set of each area of the evaluated ecological region to obtain the water structure ecological stability index of each area of the evaluated ecological region.

7. The geographic information evaluation method based on an intelligent algorithm according to claim 6, wherein The specific steps for obtaining the hydrogeological evaluation index set of each area of the evaluated ecological region are as follows: Obtain the water level stability rate value of each area of the evaluated ecological region, and perform comprehensive analysis in combination with the water depth value and water level fluctuation index to obtain the hydrogeological spatial storage stability index of each area of the evaluated ecological region; Obtain the break point water level difference value of each area of the evaluated ecological region, and perform comprehensive analysis in combination with the disturbance difference index and water level gradient index to obtain the water level terrain disturbance index of each area of the evaluated ecological region; Obtain the environmental correction factor of each area of the evaluated ecological region, and perform comprehensive analysis in combination with the water accumulation connectivity index and edge water level expansion gradient index to obtain the hydrogeomorphic recharge recovery index of each area of the evaluated ecological region.

8. The geographic information evaluation method based on an intelligent algorithm according to claim 1, characterized in that The wetland vegetation geographic data includes vegetation patch concentration index, root zone water response spatial deviation index, and water content distribution index. The specific steps for obtaining the vegetation structure health index of each area of the evaluated ecological region are as follows: Perform standardization processing on the vegetation patch concentration index, root zone water response spatial deviation index, and water content distribution index of each area of the evaluated ecological region; And perform comprehensive analysis on the vegetation patch concentration index, root zone water response spatial deviation index, and water content distribution index of each area of the evaluated ecological region after standardization processing to obtain the vegetation structure health index of each area of the evaluated ecological region.

9. The geographic information evaluation method based on an intelligent algorithm according to claim 1, wherein The specific steps for taking the preset ecological protection measures for the evaluated ecological region based on the wetland ecological structure health index are as follows: Perform judgment analysis on the wetland ecological structure health index of the evaluated ecological region and the preset wetland ecological structure health index threshold; If the wetland ecological structure health index of the evaluated ecological region is lower than or equal to the preset wetland ecological structure health index threshold, then take the first ecological protection measure; If the wetland ecological structure health index of the to-be-evaluated set ecological region is higher than the preset wetland ecological structure health index threshold, then the second ecological protection measure is taken.

10. A geographic information evaluation system based on an intelligent algorithm, which applies the geographic information evaluation method based on an intelligent algorithm according to any one of claims 1-9, characterized in that Including: A data acquisition module, configured to acquire geographical information data of several areas of the to-be-evaluated set ecological region, where the geographical information data includes wetland topography image data, wetland hydrographic geographical data, and wetland vegetation geographical data; A structural feature extraction module, configured to respectively perform feature extraction on the geographical information data of each area of the to-be-evaluated set ecological region, to obtain a wetland structure evaluation index set of each area of the to-be-evaluated set ecological region, including a landform structure ecological suitability index, a water structure ecological balance index, and a vegetation structure health index; A comprehensive evaluation module, configured to comprehensively analyze the evaluation index set of each area of the to-be-evaluated set ecological region, to obtain the wetland ecological structure health index of the to-be-evaluated set ecological region; An evaluation feedback module, configured to take a preset ecological protection measure for the to-be-evaluated set ecological region based on the wetland ecological structure health index.

Images