CN-121996973-A - Blue space advanced and advanced diagnosis method, system, computer equipment and medium

CN121996973ACN 121996973 ACN121996973 ACN 121996973ACN-121996973-A

Abstract

The invention provides a blue space advanced and retreated diagnosis method, a system, computer equipment and a medium, which belong to the field of ecological environment monitoring and comprise the steps of obtaining remote sensing images, climate data and human activity data of a blue space to be diagnosed, extracting water body images of a plurality of time points of the blue space to be diagnosed from the remote sensing images, respectively calculating multiple types of landscape morphology indexes of each grid in the water body images, clustering the water body images based on the multiple types of landscape morphology indexes of each grid to obtain a blue space combination mode corresponding to each grid, inputting the climate data, the human activity data and the landscape morphology indexes into a random forest model to obtain contribution degree, screening and obtaining water body change driving factors from the climate data, the human activity data and the landscape morphology indexes based on the contribution degree, and determining advanced and retreated diagnosis results of the blue space to be diagnosed. The blue space combination mode constructed by the method can reflect structural differences of the blue space, and ensures the accuracy of degradation diagnosis of the blue space.

Inventors

LI GUO
HU WENMIN
LIU LI
LIU MAOQING
CAO JIE

Assignees

南宁学院

Dates

Publication Date: 20260508
Application Date: 20260129

Claims (10)

1. A blue space advanced/retreated diagnosis method, comprising: acquiring remote sensing images, climate data and human activity data of a blue space to be diagnosed, and extracting water body images at a plurality of time points from a time sequence of the remote sensing images; The method comprises the steps of dividing a water body image into a plurality of grids, respectively calculating a multi-type landscape morphology index of each grid in the water body image, clustering the water body image based on the multi-type landscape morphology index of each grid, and taking the clustering type of each grid as a blue space combination mode corresponding to the grid; Inputting the climate data, the human activity data, the landscape morphology indexes and the area time series data corresponding to the blue space combination mode into a pre-trained random forest model to obtain the contribution degree of the climate data, the human activity data and the landscape morphology indexes to the area time series data, wherein the area time series data is determined based on the areas of the blue space combination modes in the water body images at different time points; And determining the advanced and retreated diagnosis result of the blue space to be diagnosed according to the change trend of the area time series data and the relation between the area time series data and a critical threshold, wherein the critical threshold is determined according to the relation between the water body change driving factor and the area time series data.
2. The method of diagnosing blue space advanced and retreated as set forth in claim 1, wherein extracting the water body image at a plurality of time points from the time series of the remote sensing images comprises: and carrying out water body identification pixel by pixel in the time sequence of the remote sensing image based on MNDWI indexes to obtain water body images at different time points.
3. The method of claim 1, wherein the landscape morphology index comprises plaque density, shannon diversity index, fractal dimension index, maximum plaque index, landscape shape index, plaque association index, spreading index, and edge density.
4. The method for diagnosing blue space advanced and advanced as claimed in claim 1, wherein clustering the water body image based on the multi-class landscape morphology index of each grid, taking the clustering class of each grid as the blue space combination mode corresponding to the grid comprises: k-means clustering is carried out on each grid in the water body image according to Euclidean distances of the multiple types of landscape morphology indexes of each grid, and a blue space combination mode corresponding to each grid is obtained, wherein grid dimensions of the K-means clusters are determined by variation fluctuation of statistical features relative to dimensions of the landscape morphology indexes, and the number of clusters of the K-means clusters is determined by differences among the grids through an elbow method.
5. The method for diagnosing blue space advanced and advanced as claimed in claim 1, wherein the grid includes a deviation grid, the difference between the probability of the water body type of the deviation grid and the blue space combination pattern is higher than a preset value, and the method further comprises, after taking the clustering category of each grid as the blue space combination pattern corresponding to the grid: Determining the water body type probability of each grid by a pre-trained XGBoost model according to the multi-type landscape morphological indexes of each grid; Analyzing contribution degree of the landscape morphology indexes of multiple types to the water body type probability through SHAP; and re-dividing the deviation grids according to the contribution degree of the multi-type landscape morphology indexes to the water body type probability to obtain corrected blue space combination modes corresponding to each grid.
6. The method for diagnosing blue space advanced and retreated according to claim 1, further comprising, after the step of obtaining the water body change driving factor: constructing a partial dependency graph according to the relation between the water body change driving factor and the area time sequence data; An inflection point is identified from the partial dependency graph by a second order difference method based on curvature change, from which a critical threshold is determined.
7. The blue space advanced and retreating diagnostic method of claim 1, wherein the blue space combined mode includes a low-break-up-high-connectivity mode and a high-break-up-low-connectivity mode, and the change trend of the area time series data includes the change trend of the areas of the low-break-up-high-connectivity mode and the high-break-low-connectivity mode.
8. A blue space advanced and advanced diagnosis system, comprising: The image acquisition module is used for acquiring remote sensing images, climate data and human activity data of a blue space to be diagnosed, and extracting water body images of a plurality of time points from a time sequence of the remote sensing images; The system comprises a water body clustering module, a blue space combination mode and a blue space combination mode, wherein the water body clustering module is used for dividing the water body image into a plurality of grids and respectively calculating a plurality of types of landscape morphology indexes of each grid in the water body image; The contribution degree calculation module is used for inputting the climate data, the human activity data, the landscape morphology indexes and the area time series data corresponding to the blue space combination mode into a pre-trained random forest model to obtain the contribution degree of the climate data, the human activity data and the landscape morphology indexes to the area time series data, wherein the area time series data is determined based on the areas of the blue space combination modes in the water body images at different time points; And the degradation diagnosis module is used for screening and obtaining a water body change driving factor from the climate data, the human activity data and the landscape morphology index based on the contribution degree, and determining a degradation diagnosis result of the blue space to be diagnosed according to the change trend of the area time sequence data and the relation between the area time sequence data and a critical threshold value, wherein the critical threshold value is determined based on the relation between the water body change driving factor and the area time sequence data.
9. A computer device comprising a memory, a processor and a computer program stored on the memory, characterized in that the processor executes the computer program to implement the steps of a blue space advanced diagnosis method according to any one of claims 1 to 7.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that the computer program, when loaded by a processor, is capable of executing the steps of a blue space advanced/retreated diagnosis method according to any one of claims 1 to 7.

Description

Blue space advanced and advanced diagnosis method, system, computer equipment and medium Technical Field The invention belongs to the field of ecological environment monitoring, and particularly relates to a blue space advanced and retreated diagnosis method, a blue space advanced and retreated diagnosis system, computer equipment and medium. Background The blue space refers to a natural or semi-natural ecological system taking water bodies as cores, such as rivers, lakes, wetlands and the like, and is used as a key carrier of water circulation and an important habitat of biodiversity, and the morphological structure and dynamic change of the blue space are directly related to the sustainability of water resources, the safety of water quality and the ecological toughness of areas. However, under the dual stress of climate change and human activity, the morphological evolution of the blue space is increasingly complex, the driving mechanism of the advancing and retreating process is not clear, and the dynamic rule is revealed through accurate diagnosis so as to support the formulation of the water resource adaptability management strategy. The existing degradation of the blue space depends on diagnosis by combining remote sensing data with morphological index analysis, such as calculating landscape pattern index by combining remote sensing data, so as to quantify the morphology of the blue space. However, the water body has highly similar spectrum and texture characteristics, the method for quantifying the morphology of the blue space based on a single water body index is difficult to classify the water body coating types, and the responses of different water body coating types to the influences of climate, human activities and the like are obviously different, so that the advanced and retreated diagnosis of the blue space is limited. Disclosure of Invention In order to solve the existing problems, the invention provides a blue space advanced and retreated diagnosis method, a system, computer equipment and a medium. In order to achieve the above object, the present invention provides the following technical solutions: A blue space advanced and advanced diagnosis method, comprising: acquiring remote sensing images, climate data and human activity data of a blue space to be diagnosed, and extracting water body images at a plurality of time points from a time sequence of the remote sensing images; dividing the water body image into a plurality of grids, and respectively calculating a multi-type landscape morphology index of each grid in the water body image; Inputting climate data, human activity data, landscape morphology indexes and area time series data corresponding to the blue space combination mode into a pre-trained random forest model to obtain contribution degrees of the climate data, the human activity data and the landscape morphology indexes to the area time series data, wherein the area time series data is determined based on areas of the blue space combination modes in the water body images at different time points; And determining an advancing and retreating diagnosis result of the blue space to be diagnosed according to the change trend of the area time series data and the relation between the area time series data and a critical threshold, wherein the critical threshold is determined based on the relation between the water body change driving factor and the area time series data. The invention also provides a blue space advanced and advanced diagnosis system, which comprises: The image acquisition module is used for acquiring remote sensing images, climate data and human activity data of a blue space to be diagnosed, and extracting water body images of a plurality of time points from a time sequence of the remote sensing images; the system comprises a water body clustering module, a blue space combination mode and a blue space combination mode, wherein the water body clustering module is used for dividing a water body image into a plurality of grids and respectively calculating a multi-type landscape morphology index of each grid in the water body image; The contribution degree calculation module is used for inputting the climate data, the human activity data, the landscape morphology indexes and the area time series data corresponding to the blue space combination mode into a pre-trained random forest model to obtain the contribution degree of the climate data, the human activity data and the landscape morphology indexes to the area time series data, wherein the area time series data is determined based on the areas of the blue space combination modes in the water body images at different time points; The degradation diagnosis module is used for screening and obtaining a water body change driving factor from climate data, human activity data and landscape morphology indexes based on the contribution degree, and determining a degradation diagnosis result of a blue space to be diagnosed accordin