CN-121661429-B - Ecological system drawing method and system for natural resource multi-element linkage

CN121661429BCN 121661429 BCN121661429 BCN 121661429BCN-121661429-B

Abstract

The invention relates to the technical field of ecological system drawing, in particular to a natural resource multi-element linkage ecological system drawing method and system; the method comprises the steps of collecting multi-source natural resource investigation monitoring data, conducting semantic cleaning and fusion to construct a natural resource data base plate, mapping the natural resource data base plate into an initial ecosystem classification map based on an artificial intelligent coupling model, continuously monitoring update flow of the multi-source natural resource investigation monitoring data, automatically detecting data change events, calculating change intensity indexes corresponding to the change events based on multi-source change information, and generating a target ecosystem classification map. According to the invention, the multi-source data flow is continuously monitored through the data fingerprint comparison technology, the change event is automatically captured, the intelligent classification and decision are carried out on the detected change, and the automation and the intellectualization of drawing are realized on the premise of ensuring the balance of the data topology and the area through the pattern spot aggregation based on the knowledge graph.

Inventors

GU JUAN
ZHANG YI
WU JUN
YIN SIYANG
DUAN HONGZHI
XIE YANFENG
ZENG JINGWEN
GAO XUZHI
GONG YUN

Assignees

北京市测绘设计研究院

Dates

Publication Date: 20260512
Application Date: 20260206

Claims (9)

1. A natural resource multi-element linkage ecological system drawing method is characterized by comprising the following steps: Collecting multi-source natural resource investigation monitoring data, carrying out semantic cleaning and fusion, and constructing a natural resource data base plate; Mapping the natural resource data base plate into an initial ecosystem classification map based on an artificial intelligence coupling model; continuously monitoring the update flow of the multi-source natural resource investigation monitoring data, and automatically detecting a data change event; classifying the detected data change event into a first type event and a second type event; for the second class of events, calculating a change intensity index corresponding to the change event based on multi-source change information, and comparing and processing the change intensity index with a type stability threshold corresponding to an ecosystem type: If the change intensity index is larger than the type stability threshold, triggering the type update of the initial ecosystem classification map; if the change intensity index is smaller than or equal to the type stability threshold, triggering the pattern spot aggregation and correction flow of the natural resource data base plate to obtain a corrected natural resource data base plate; Updating the initial ecosystem classification map according to the corrected natural resource data base plate to generate a target ecosystem classification map; mapping the natural resource data base plate into an initial ecosystem classification map based on an artificial intelligence coupling model comprises: For each pattern spot object in the natural resource data base plate, extracting and fusing the pattern spot objects from the multi-source attribute data of the pattern spot objects to construct a uniform multi-dimensional feature vector; Inputting the multi-dimensional feature vector into a pre-trained artificial intelligence coupling model, the artificial intelligence coupling model configured to: Weighting each feature in the multi-dimensional feature vector through an internal attention mechanism layer to generate a weighted feature representation; inputting the weighted characteristic representation to a fully-connected classification network, and calculating the probability that the plaque object belongs to each preset ecosystem type; For each plaque object, selecting an ecosystem type with the highest probability value from probabilities output by the fully-connected classification network, taking the selected ecosystem type as a final ecosystem classification result of the plaque object, and taking the highest probability value as the confidence level of the classification result; And generating and outputting the initial ecosystem classification map based on the final ecosystem classification result, the corresponding confidence and the space geometric information of all the map spot objects.
2. A natural resource multi-element linked ecological system mapping method as in claim 1, wherein classifying the detected data change event comprises: If the data change event is associated with a legal administrative approval file, classifying the data change event as a first type event, and directly triggering the type update of the initial ecosystem classification map; and if the data change event is not associated with a legal administrative approval file, classifying the data change event as a second type event.
3. The natural resource multi-element linked ecological system mapping method of claim 1, wherein the process of continuously monitoring the update flow of the multi-source natural resource survey monitoring data comprises: Establishing and maintaining a dynamic data fingerprint for each data source, wherein the data fingerprint is a unique identification code calculated based on data content and is used for representing the complete data state of the data source at a specific time point; periodically recalculating the data fingerprint and comparing the data fingerprint with a fingerprint value stored in the previous time to continuously monitor the update flow of each data source; When the current data fingerprint of any data source is inconsistent with the previous fingerprint, automatically judging that the data source has data change, and extracting and generating a corresponding data change event; wherein the multi-source natural resource survey monitoring data comprises a plurality of data sources.
4. The method for mapping an ecosystem of natural resource multi-element linkage according to claim 1, wherein the process of calculating the change intensity index corresponding to the change event based on the multi-source change information comprises: Calculating an area adjustment coefficient according to the change area; Calculating a time urgency coefficient according to the monitored emergency type; The change intensity index is calculated according to the inherent difficulty of type conversion, the area adjustment coefficient and the time urgency coefficient.
5. The method for mapping an ecosystem of natural resource multi-element linkage according to claim 1, wherein the step of triggering the plaque aggregation and correction process of the natural resource data base plate to obtain the corrected natural resource data base plate comprises the steps of: Extracting each ecosystem change pattern spot which changes in the initial ecosystem classification map based on the change area provided by the second class event; extracting corresponding original image spots from the natural resource data base plate according to each ecological system change image spot to serve as image spots to be processed; Performing graph traversal on each graph to be processed in the natural resource data base plate knowledge graph so as to determine a target merging graph, and performing graph aggregation; Calculating the real-time total area change rate of the target merging pattern spots, and checking the aggregation result according to the comparison result of the standard total area change rate interval and the real-time total area change rate; And synchronously updating the aggregation correction result which passes the inspection to the natural resource data base plate and the natural resource data base plate knowledge graph to generate the corrected natural resource data base plate and the corrected natural resource data base plate knowledge graph.
6. The method for mapping an ecosystem of natural resource multi-element linkage according to claim 5, wherein the process of determining a target merged map spot and performing map spot aggregation comprises: Automatically checking and determining the types of the to-be-processed image spots according to area threshold comparison, average width, local morphology clustering analysis and topology rules to obtain one type of problem image spots, two types of problem image spots, three types of problem image spots and four types of problem image spots; Selecting a pattern spot aggregation mode according to the type and the priority of the pattern spots to be processed; performing pattern spot polymerization according to a corresponding pattern spot polymerization mode; the priority is that a fourth aggregation mode > a third aggregation mode > a second aggregation mode > a first aggregation mode; the pattern spot polymerization modes comprise a first polymerization mode, a second polymerization mode, a third polymerization mode and a fourth polymerization mode; the first aggregation mode is to merge the to-be-processed image spots into a target merged image spot and update attribute information according to the attribute type of the image spot; the second polymerization mode is to distribute the image spots to be processed to the target merging image spots on two adjacent sides; the third polymerization mode is to execute the flow of simplifying the boundary contour; The fourth aggregation is to perform a slot process or an overlap process according to the type of problem.
7. The method for mapping an ecosystem of natural resource multi-element linkage according to claim 5, wherein the process of verifying the aggregate result according to a comparison result of a standard total area change rate interval and a real-time total area change rate comprises: comparing the standard total area change rate interval with the real-time total area change rate: if the real-time total area change rate is within the standard total area change rate interval, judging that the aggregation correction result is obtained through inspection; If the real-time total area change rate is not in the standard total area change rate interval, judging the positive and negative of the real-time total area change rate: If the real-time total area change rate is negative, reselecting the target merging pattern spots for merging by increasing the strong constraint; If the real-time total area change rate is positive, executing a segmentation release strategy to reselect the target merging pattern spots for merging.
8. The natural resource multi-element linkage ecological system mapping method according to claim 5, wherein the process of calculating the real-time total area change rate of the target merged map spot is as follows: determining an inspection area according to the minimum circumscribed range and the buffer zone of each pattern to be processed; Extracting the original total area of each category in the inspection area from the natural resource data base plate before aggregation; Extracting new total areas of all categories in the inspection area from the aggregated corrected natural resource data base plate; And calculating the percentage of the part of the new total area exceeding the original total area to the original total area, and obtaining the real-time total area change rate.
9. An ecological system drawing system for natural resource multi-element linkage, which is applied to the ecological system drawing method for natural resource multi-element linkage according to any one of claims 1 to 8, and is characterized by comprising the following steps: The data fusion processing module is used for collecting multi-source resource investigation and monitoring data, carrying out semantic cleaning and fusion, and constructing a natural resource data base plate; the classification drafting module is connected with the data fusion processing module and is used for mapping the natural resource data base plate into an initial ecosystem classification chart based on an artificial intelligent coupling model; The change detection module is respectively connected with the data fusion processing module and the classification drawing module and is used for continuously monitoring the update flow of the multi-source resource investigation monitoring data and automatically detecting a data change event; The event classification module is respectively connected with the data fusion processing module and the change detection module and is used for classifying the detected data change event into a first type event and a second type event; the response correction module is connected with the event classification module, calculates a change intensity index corresponding to the change event based on multi-source change information for the second type event, and compares and processes the change intensity index with a type stability threshold corresponding to the type of the ecological system: If the change intensity index is larger than the type stability threshold, triggering the type update of the initial ecosystem classification map; if the change intensity index is smaller than or equal to the type stability threshold, triggering the pattern spot aggregation and correction flow of the natural resource data base plate to obtain a corrected natural resource data base plate; And the linkage updating module is respectively connected with the event classifying module, the response correcting module, the data fusion processing module and the classifying drawing module and is used for updating the initial ecosystem classifying diagram according to the corrected natural resource data base plate so as to generate a target ecosystem classifying diagram.

Description

Ecological system drawing method and system for natural resource multi-element linkage Technical Field The invention relates to the technical field of ecological system drawing, in particular to a natural resource multi-element linkage ecological system drawing method and system. Background Along with the increasingly urgent demands on ecological civilization construction and natural resource fine management, how to accurately and dynamically draw an ecosystem distribution map has become a key technical challenge, at present, the field of natural resource management has formed multi-source, multi-element and continuously updated mass data rich minerals, such as data of annual land change investigation, comprehensive monitoring of forest, grass and wet waste, national soil space planning, plowing quality monitoring and the like, which theoretically provide possibility for high-precision and current ecosystem drawing, the existing ecosystem drawing method generally relies on a single or small number of data sources, adopts a static classification model to classify the data at a certain time point at one time to generate a static ecosystem classification map, and for data updating, the prior art often adopts a mode of periodically re-running the whole classification flow or manually or semi-automatically updates local changes through simple rules, and meanwhile, on the data processing level, a set of drawing 'noise' such as small map spots, long narrow map spots and the like generated by factor acquisition, updating or model errors lacks a set of drawing comprehensive and correction mechanism coupled with logic depth of the ecosystem classification. CN119313026a discloses a method, device, equipment and drawing system for ecological division of a plateau traffic corridor, in which meteorological station data, digital elevation model data and land utilization type data in a research area are obtained based on the vector range of the ecological division research area of the plateau traffic corridor. And (3) obtaining the surface relief degree and the ecosystem type of the research area based on the data, providing a missing air temperature data interpolation method of the meteorological site, and interpolating the missing air temperature. Based on the interpolated air temperature data, the ground surface relief and the ecosystem type, three zone indexes of the plateau climate zone type, the plateau landform type and the road domain ecology type of the research area are determined and manufactured by combining a preset mapping relation, and the three indexes are subjected to space coupling analysis to obtain a coupling result. The method comprises the steps of carrying out multi-index-plaque clustering analysis based on a coupling result to obtain a clustering result, carrying out ecological division on a research area based on the clustering result, and therefore, the conventional ecological system drawing is used for massive and multi-source data flows, automatic detection of light-weight and high-performance changes is lacking, instant discovery and response of change events are difficult to realize, update monitoring efficiency is low, and classification precision of an ecological system classification diagram is low. Disclosure of Invention Therefore, the invention provides a natural resource multi-element linkage ecological system drawing method and system, which are used for solving the problems of low classification precision of an ecological system classification map caused by low update monitoring efficiency due to lack of automatic detection of lightweight and high-performance change facing mass and multi-source data flows in the prior art. In order to achieve the above object, in one aspect, the present invention provides a method for mapping an ecosystem by natural resource multi-element linkage, including: Collecting multi-source natural resource investigation monitoring data, carrying out semantic cleaning and fusion, and constructing a natural resource data base plate; Mapping the natural resource data base plate into an initial ecosystem classification map based on an artificial intelligence coupling model; continuously monitoring the update flow of the multi-source natural resource investigation monitoring data, and automatically detecting a data change event; classifying the detected data change event into a first type event and a second type event; for the second class of events, calculating a change intensity index corresponding to the change event based on multi-source change information, and comparing and processing the change intensity index with a type stability threshold corresponding to an ecosystem type: If the change intensity index is larger than the type stability threshold, triggering the type update of the initial ecosystem classification map; if the change intensity index is smaller than or equal to the type stability threshold, triggering the pattern spot aggregation and correct