Artificial intelligence land biodiversity real-time online monitoring evaluation-risk early warning-risk elimination method and system

Abstract

The invention provides an artificial intelligence land biodiversity real-time on-line monitoring evaluation-risk early warning-risk elimination method and system, and the method comprises the steps: collecting and monitoring biological information, geographic and phenological information, environment quality information and other information, carrying out the data comparison and analysis, carrying out the data evaluation and influence factor analysis, and carrying out the risk diagnosis and early warning. A risk elimination method; the system sequentially comprises a data acquisition and monitoring module, a data analysis and diversity evaluation module, a risk diagnosis and early warning module, a risk elimination module and the like. Scientific basis and technical support are provided for land biodiversity protection, management and decision making, and the land biodiversity protection strength is enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent monitoring of biodiversity, and specifically relates to an artificial intelligence terrestrial biodiversity real-time online monitoring assessment - risk warning - risk elimination method and system. Background Art

[0002] Biodiversity is the cornerstone of the sustainable development of the Earth and human society. Affected by geographical phenology, environmental pollution, human interference, and foreign invasions, the loss of biodiversity and the deterioration of ecosystems are accelerating, and it is urgent to strengthen biodiversity protection.

[0003] At present, biodiversity monitoring is limited to irregular manual surveys of key species in key core areas, which generally have the disadvantages of large manpower input, high cost, long time and poor timeliness. In addition, due to factors such as limited survey time and funds, uneven level of investigators, and imperfect survey technology system, the data obtained are not satisfactory in terms of comprehensiveness, systematicness, timeliness and temporal continuity. It is difficult to synthesize a large database of terrestrial biodiversity and conduct a systematic evaluation, which also affects the effect of horizontal comparison and analysis of data across time and regions, resulting in the inability to issue real-time warnings and take effective biodiversity protection measures.

[0004] Existing biodiversity monitoring technologies provide certain technical support for biodiversity conservation and management, but there are still some drawbacks and challenges. The present invention proposes an artificial intelligence terrestrial biodiversity real-time online monitoring assessment-risk warning-risk elimination method and system, which collects and monitors effective information in real time by deploying data collection and monitoring networks; based on species recognition models and data analysis technology, it can automatically establish a species database, evaluate the biodiversity status in real time, and scientifically determine the threatening factors of each species. With the help of horizontal comparison of data in the database, it can provide early warning information and risk elimination measures in real time, solving a series of drawbacks of existing technologies. Summary of the invention

[0005] In view of the shortcomings of the existing technology, the present invention provides an artificial intelligence terrestrial biodiversity real-time online monitoring and assessment - risk warning - risk elimination method and system, which provides a scientific basis and technical support for terrestrial biodiversity protection, management and decision-making, and strengthens the protection of terrestrial biodiversity.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] An artificial intelligence terrestrial biodiversity real-time online monitoring and assessment-risk warning-risk elimination method, comprising the following steps:

[0008] Build a biological monitoring network to collect and monitor biological information, geographic phenology information, environmental quality information and other information within the preset land area;

[0009] Perform species identification based on the biological information to obtain a species identification result;

[0010] Establishing a biological basic information database for each species based on the species identification results, the biological information, geographic phenological information, and environmental quality information;

[0011] Extracting the target parameters of the geographic phenological information, the environmental quality information and the biological information, and based on the biological basic information database, conducting an assessment of the current status of biodiversity and the regional distribution pattern, and obtaining an assessment result of suitable habitats for species;

[0012] Compare the historical biological information in the biological basic information database horizontally, analyze the correlation between geographic phenology information, environmental quality information, other information and biological information of each species, obtain the threat factors of species, and generate the change trend over the years;

[0013] Based on the assessment results and the historical data of the threatened factors of the species, the main contribution rate of the threatened factors of each species is obtained, and the critical limit of species risk is obtained based on the main contribution rate of the threatened factors of each species;

[0014] Compare the real-time data with the critical risk limits of the species to conduct biodiversity risk diagnosis and obtain risk diagnosis results;

[0015] Provide different levels of early warning for risk diagnosis results based on changes in different geographical phenology, environmental quality impacts, and other information;

[0016] Providing risk elimination measures based on the risk diagnosis results.

[0017] Preferably, the biological information includes biological information of terrestrial higher plants, terrestrial mammals, birds, amphibians, reptiles, insects and fungi within the monitored land area;

[0018] The geographical phenological information includes wind direction, wind speed, temperature, humidity, rainfall, snowfall, latitude and longitude, altitude, aspect and slope information;

[0019] The environmental quality information includes air environment, surface water and sediment, groundwater, soil, noise, and electromagnetic and radiation information;

[0020] The other information includes information on human disturbance, protected lists, protection policies and historical data.

[0021] Preferably, the morphological characteristics, audio-visual characteristics and gene biological spectrum of the organisms in the biological information are extracted and compared with existing data to identify species, obtain a species list and gallery, and classify endemic species, endangered species, new species and alien invasive species;

[0022] Preferably, the method for extracting the morphological features is:

[0023] Crawl the existing species database within the preset land area, clean the crawled data, and convert it into standardized species corpus;

[0024] Entities are extracted and annotated from the species corpus, and a knowledge graph of biological species and related information is constructed in a top-down manner; the nodes in the knowledge graph include species and species morphological characteristics;

[0025] Based on the biological information, obtaining a biological image;

[0026] Extracting color features, texture features, and contour features of the species in the biological image, and concatenating all the features to obtain a manual feature vector;

[0027] Use stacked graph attention layers to build graph convolutional neural networks;

[0028] Using the graph convolutional neural network, the biological image labels are indexed in the knowledge graph to obtain the morphological feature vectors of the corresponding nodes of the corresponding species in the knowledge graph;

[0029] Extracting components of the morphological feature vector and the manual feature vector, and performing cosine similarity calculation on the extracted components to obtain similarity loss;

[0030] Based on the similarity loss, obtaining a morphological similarity feature vector;

[0031] Extracting a semantic representation vector from the biological image by using VIT that introduces a Transformer mechanism;

[0032] The manual feature vector, the morphological similarity feature vector and the semantic representation vector are fused to obtain the final morphological feature of the species.

[0033] Preferably, the method for conducting biodiversity status assessment and regional distribution pattern evaluation to obtain the assessment results is:

[0034] Preset target parameters; the target parameters include species diversity parameters, ecosystem diversity parameters, genetic diversity parameters, species distribution pattern parameters, and threat and protection parameters;

[0035] Based on the biological basic information library, screening and extracting data related to target parameters;

[0036] Conduct statistical and spatial analysis on the data related to the target parameters to assess the status and level of biodiversity and obtain biodiversity assessment results;

[0037] Extract species distribution data and ecosystem type data from data related to target parameters, draw biological maps, analyze biodiversity characteristics and differences in different regions, and obtain species regional spatial distribution patterns and change trends;

[0038] The assessment results are obtained based on the biodiversity assessment results and the regional spatial distribution pattern and change trend of species.

[0039] Preferably, the method for obtaining the species threat factor is:

[0040] Standardize historical biological information to obtain standard background data;

[0041] Conduct time series and spatial distribution change analysis on the standard background data to obtain the spatiotemporal characteristics of species diversity;

[0042] Using principal component analysis, the effects of geographic phenology information, environmental quality information and other information on the spatiotemporal characteristics of species diversity are analyzed to obtain a first analysis result;

[0043] Analyze the correlation between biological information, geographic phenological information, environmental quality information and other information of each species to obtain the second analysis result;

[0044] By combining the first analysis result and the second analysis result, the threat factors of species are obtained and the change trends over the years are generated.

[0045] Preferably, the method for obtaining risk diagnosis results is:

[0046] By comparing and analyzing real-time data and historical data, we can identify species at risk and the main factors that cause abnormalities, and give the main contribution rate of each threat factor based on historical monitoring data to form species risk critical limits. We can compare real-time data with the species risk critical limits to diagnose biodiversity risks and obtain risk diagnosis results. We can also provide different levels of early warnings for different risk diagnosis results based on changes in different geographical phenology, environmental quality impacts, and other information.

[0047] Providing risk elimination measures based on the risk diagnosis results.

[0048] The present invention also provides an artificial intelligence terrestrial biodiversity real-time online monitoring and assessment-risk warning-risk elimination system for implementing the method, comprising:

[0049] Data collection module, used to build a biological monitoring network, collect and monitor biological information, geographic phenology information, environmental quality information and other information within a preset land area;

[0050] An information base construction module is used to perform species identification based on the biological information to obtain species identification results; based on the species identification results, the biological information, geographic phenological information and environmental quality information, establish a biological basic information base for each species;

[0051] A diversity assessment module is used to extract the target parameters of the geographic phenology information, the environmental quality information and the biological information, and to evaluate the current status of biodiversity and the regional distribution pattern based on the biological basic information database to obtain the evaluation results of species suitable habitats;

[0052] An influencing factor analysis module is used to compare the historical biological information in the biological basic information database horizontally, and analyze the correlation between geographical phenological information, environmental quality information, other information and biological information of each species, obtain the threat factors of species, and generate the change trend over the years;

[0053] A risk diagnosis module, for obtaining the main contribution rate of the threatened factors of each species based on the assessment results and the historical data of the threatened factors of the species, and obtaining the species risk critical limit based on the main contribution rate of the threatened factors of each species; comparing the real-time data with the species risk critical limit, performing biodiversity risk diagnosis, and obtaining risk diagnosis results;

[0054] The risk warning module is used to provide different degrees of warning on risk diagnosis results based on changes in different geographical phenology, environmental quality impacts, and other information;

[0055] The risk elimination module is used to provide risk elimination measures based on the risk diagnosis results.

[0056] Compared with the prior art, the beneficial effects of the present invention are as follows: by constructing a biological monitoring network, biological information, geographic phenological information, environmental quality information and other relevant information within a preset land area can be collected and monitored comprehensively and in real time, ensuring the comprehensiveness and timeliness of the data. Species identification based on the collected biological information not only improves the accuracy of species identification, but also establishes a biological basic information database for each species, providing accurate data support for subsequent biodiversity protection and resource utilization. By extracting key parameters and combining with the biological basic information database, the current status of biodiversity and its regional distribution pattern can be scientifically assessed, which helps to discover hot spots and vulnerable areas of biodiversity and provide a basis for formulating targeted protection measures. By horizontally comparing historical biological information and analyzing the correlation between various types of information, the threat factors of species can be accurately identified, and the trend of changes over the years can be generated, which helps to predict possible biodiversity changes in the future and provides the possibility of taking preventive measures in advance. Based on the assessment results and the main contribution rate of the threat factors, the critical limit of species risk is determined, and then the real-time data is compared to perform biodiversity risk diagnosis. This mechanism can timely discover potential risks, and through different degrees of early warning, remind relevant parties to take necessary risk elimination measures to protect biodiversity from further threats. By providing risk elimination measures based on different risk diagnosis results, it provides scientific decision-making support for governments, scientific research institutions and environmental protection organizations, and promotes the realization of biodiversity conservation and sustainable development. The operation of the system and the release of data will also help to enhance the public's understanding and awareness of biodiversity conservation, promote the participation of all sectors of society in biodiversity conservation actions, and create a good social atmosphere. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0058] Figure 1 This is a flow chart of a method for real-time online monitoring and assessment of terrestrial biodiversity - risk warning - risk elimination according to an embodiment of the present invention;

[0059] Figure 2 It is a schematic diagram of the structure of an artificial intelligence terrestrial biodiversity real-time online monitoring and assessment - risk warning - risk elimination system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0062] Embodiment 1

[0063] like Figure 1 As shown, an artificial intelligence terrestrial biodiversity real-time online monitoring and assessment-risk warning-risk elimination method includes the following steps:

[0064] S1: construct a biological monitoring network to collect and monitor biological information, geographical phenological information, environmental quality information and other information within a preset land area; in this embodiment, the biological information includes monitoring biological information of terrestrial higher plants, terrestrial mammals, birds, amphibians, reptiles, insects and fungi within the land area;

[0065] Geographic phenological information includes wind direction, wind speed, temperature, humidity, rainfall, snowfall, latitude and longitude, altitude, aspect and slope information;

[0066] Environmental quality information includes air environment, surface water and sediment, groundwater, soil, noise, and electromagnetic and radiation information;

[0067] Other information includes information on human disturbance, conservation lists, conservation policies, and historical data.

[0068] A further implementation method is to construct a biological monitoring network within a preset land area by using infrared cameras and sensors powered by solar panels to monitor biological information, geographic phenological information, environmental quality information and other information at preset monitoring points, and to construct a biological monitoring network based on wireless communication technology; in this embodiment, a monitoring network is constructed on the basis of protected areas, forests and grasslands, and the country is divided into more than nine million grids. Monitoring tools are deployed in each grid, and ultimately a national 1km*1km monitoring network system is constructed.

[0069] A biological monitoring network is established based on wireless communication technology to transmit data between monitoring points and the central control system; the central control system is used to receive data from each monitoring point in real time and store and analyze the data;

[0070] Based on the central control system, image data processing technology is used to identify species of received biological information; data analysis technology is used to analyze received geographic phenology information, environmental quality information and other information;

[0071] If the central control system cannot identify the species, manual identification will be performed, and the species identification results and data analysis results will be entered into the biological basic information database.

[0072] In this embodiment, the infrared camera, sensor and solar panel are regularly inspected and maintained to ensure their normal operation. Data security measures are strengthened to prevent data leakage or tampering. During the monitoring process, attention should be paid to protecting the privacy of the monitored species and personnel.

[0073] S2: performing species identification based on the biological information to obtain a species identification result;

[0074] Extract the morphological characteristics, audio-visual characteristics and gene biological spectrum of the organisms in the biological information and compare them with existing data to identify species, obtain a species list and gallery, and classify endemic species, endangered species, new species and alien invasive species;

[0075] The gene biological spectrum library is obtained by artificially collecting samples for experiments;

[0076] New species or alien invasive species are manually identified and distinguished, and the identification results are entered into the biological basic information database. The method for extracting morphological characteristics is:

[0077] Crawl the existing species database within the preset land area, clean the crawled data, and convert it into standardized species corpus;

[0078] Entities are extracted and annotated from species corpus, and a knowledge graph of biological species and related information is constructed in a top-down manner; nodes in the knowledge graph include species and species morphological characteristics;

[0079] Based on the biological information, a biological image is obtained;

[0080] Extract the color features, texture features, and contour features of species in biological images, and concatenate all features to obtain manual feature vectors;

[0081] Use stacked graph attention layers to build graph convolutional neural networks;

[0082] Using graph convolutional neural networks, biological image labels are indexed in the knowledge graph to obtain the morphological feature vectors of the corresponding nodes of the corresponding species in the knowledge graph;

[0083] Extract the components of the morphological feature vector and the manual feature vector, and perform cosine similarity calculation on the extracted components to obtain the similarity loss;

[0084] Based on the similarity loss, the morphological similarity feature vector is obtained;

[0085] Using VIT with the Transformer mechanism, the semantic representation vectors in biological images are extracted;

[0086] The manual feature vector, morphological similarity feature vector and semantic representation vector are fused to obtain the final morphological characteristics of the species.

[0087] S3: Establishing a biological basic information database for each species based on the species identification results, the biological information, geographic phenological information, and environmental quality information;

[0088] S4: extracting the target parameters of the geographic phenological information, the environmental quality information and the biological information, and based on the biological basic information database, conducting a biodiversity status assessment and a regional distribution pattern assessment to obtain a species suitable habitat assessment result;

[0089] A further implementation method is to conduct an assessment of the current status of biodiversity and an evaluation of the regional distribution pattern, and the method for obtaining the assessment results is:

[0090] Preset target parameters; target parameters include species diversity parameters (number of species, species richness, endemic species), ecosystem diversity parameters (type, structure, function), genetic diversity parameters (genetic variation, genetic distance, population genetic structure), species distribution pattern parameters (species distribution range, population density, spatial heterogeneity) and threat and protection parameters (threatened species, protected area coverage, human activity impact);

[0091] Based on the biological basic information library, screen and extract data related to target parameters;

[0092] Conduct statistical and spatial analysis on the data related to the target parameters to evaluate the status and level of biodiversity and obtain the results of biodiversity assessment; specifically, use species richness index, diversity index, etc. to quantify the degree and changes of biodiversity. Analyze characteristics such as species composition, ecosystem structure, genetic diversity, reveal the internal mechanism of biodiversity, and use the above analysis results as a basis for biodiversity assessment.

[0093] Extract species distribution data and ecosystem type data from data related to target parameters, draw biological maps, analyze biodiversity characteristics and differences in different regions, and obtain species regional spatial distribution patterns and change trends;

[0094] The assessment results are obtained based on the biodiversity assessment results and the regional spatial distribution patterns and changing trends of species.

[0095] In this embodiment, first, the evaluation indicators of biodiversity, spatial distribution pattern and change trend are determined, and the range method is used to standardize each evaluation indicator; the information entropy of each indicator is calculated, and the indicator weight is defined; a weighted matrix is ​​constructed based on the indicator weight; based on the weighted matrix, the optimal solution and the worst solution are determined, and the Euclidean distance between the optimal solution and the worst solution is calculated; based on the Euclidean distance, a comprehensive evaluation index is calculated, and the comprehensive evaluation index is between 0 and 1, and the larger the value, the better the evaluation effect.

[0096] S5: Horizontally compare the historical biological information in the biological basic information database, and analyze the correlation between geographic phenology information, environmental quality information, other information and biological information of each species, obtain the threat factors of species, and generate the change trend over the years;

[0097] A further implementation method is that the method for obtaining the species threat factor is:

[0098] Standardize historical biological information to obtain standard background data;

[0099] Conduct time series and spatial distribution change analysis on the standard background data to obtain the spatiotemporal characteristics of species diversity;

[0100] Using principal component analysis, the effects of geographic phenology information, environmental quality information and other information on the spatiotemporal characteristics of species diversity are analyzed to obtain a first analysis result;

[0101] Analyze the correlation between biological information, geographic phenological information, environmental quality information and other information of each species to obtain the second analysis result;

[0102] By combining the first analysis result and the second analysis result, the threat factors of species are obtained and the change trends over the years are generated.

[0103] In this embodiment, taking the analysis of environmental quality information and other information as an example, the principal component analysis method is used to reduce the dimension of environmental factors to obtain mutually orthogonal principal components, and the principal components are used as explanatory variables of GAM (generalized additive model, which essentially uses high-order polynomials to simulate the linear and complex nonlinear effects of environmental factors on the distribution of terrestrial species). In order to eliminate the influence of different standard deviations of various environmental factors, the environmental factors are first standardized and then the principal component analysis is performed. The conditional probability formula is used to express the probability of each species appearing under the conditions of the principal components and various environmental factors. Each environmental factor is matched with the spatial position time series. When the probability of species occurrence meets the preset threshold, the environmental factors corresponding to its spatiotemporal characteristics are clustered to obtain the cluster center of the environmental factors. The cluster center is used as the average level of the environmental factor values ​​and is sorted. Human interference and environmental factors are weightedly combined to obtain the factors that threaten species, that is, the weighted values ​​of environmental factors and other information that affect the spatiotemporal distribution of species.

[0104] S6: Based on the assessment results and the historical data of the threatened factors of the species, obtain the main contribution rate of the threatened factors of each species, and obtain the species risk critical limit based on the main contribution rate of the threatened factors of each species;

[0105] S7: Compare the real-time data with the critical risk limit of the species to perform biodiversity risk diagnosis and obtain risk diagnosis results;

[0106] By comparing and analyzing real-time data and historical data, we can identify species at risk and the main factors that cause abnormalities, and give the main contribution rate of each threat factor based on historical monitoring data to form species risk critical limits. We can compare real-time data with the species risk critical limits to diagnose biodiversity risks and obtain risk diagnosis results. We can also provide different levels of early warnings for different risk diagnosis results based on changes in different geographical phenology, environmental quality impacts, and other information.

[0107] Providing risk elimination measures based on the risk diagnosis results.

[0108] S8: Provide different levels of early warning for risk diagnosis results based on changes in different geographical phenology, environmental quality impacts, and other information;

[0109] S9: Provide risk elimination measures based on the risk diagnosis results. Specifically, risk elimination includes proposing targeted response measures and policy recommendations based on the risk diagnosis results, and carrying out risk elimination work, including but not limited to artificial intervention in natural phenology, prevention and control of alien invasive species, removal of environmental pollution sources or cutting off pollution pathways, forcing the cessation of human interference activities, and pushing targeted protection policies to management departments.

[0110] In this embodiment, biodiversity risk diagnosis is carried out by comparative analysis of real-time data and historical data. Once the diagnosis result is abnormal, an early warning is immediately issued to clarify the relevant information of the species at risk and the threat factors that cause the abnormality, including but not limited to natural phenological factors, invasion of alien species, environmental pollution factors, human interference activities, etc., and the main contribution rate of each threat factor is given based on historical monitoring data.

[0111] According to the risk diagnosis results and the analysis of the threat factors of species, targeted response measures or policy recommendations are output, and the execution terminal is started to eliminate the threat factors, including but not limited to artificial intervention in natural phenology, killing invasive alien species, removing environmental pollution sources or cutting off pollution pathways, forcing human interference activities to stop, and pushing targeted protection policies to management departments. When illegal logging or poaching is detected, real-time monitoring is uploaded to the relevant forestry system.

[0112] In summary, the specific steps of the present invention are as follows:

[0113] (1) Collect and monitor biological information, geographic phenology information, environmental quality information and other information, build an information monitoring network, monitor and collect biodiversity background data, geographic phenology data, environmental quality data and other information data within the preset land area;

[0114] (2) Data comparison and analysis, data evaluation and analysis of influencing factors: Generate species lists and image libraries by comparing morphology, audio and video, and genetic biological spectra, establish a species information database, and combine geographical phenology, environmental quality and other information to form a basic information database for each species, conduct assessments of the current status of terrestrial biodiversity and regional distribution patterns; analyze the correlation between species' biological information, geographical phenology, environmental quality information, and other information, scientifically determine the survival status of each species, and identify the main threatening factors.

[0115] (3) Risk diagnosis and early warning: using real-time data and historical data for comparative analysis, identify species at risk and the main factors causing abnormalities, and give the main contribution rate of each threat factor based on historical monitoring data to form species risk critical limits. Compare real-time data with the species risk critical limits to conduct biodiversity risk diagnosis and obtain risk diagnosis results;

[0116] (4) Risk elimination: Based on the risk diagnosis results, put forward targeted response measures and policy recommendations, and carry out risk elimination work.

[0117] This invention provides a scientific basis and technical support for the protection, management and decision-making of terrestrial biodiversity, and provides support for the implementation of the Convention on Biological Diversity and the Kunming-Montreal Global Biodiversity Framework.

[0118] Embodiment 2

[0119] like Figure 2 As shown, the present invention also provides an artificial intelligence terrestrial biodiversity real-time online monitoring and assessment-risk warning-risk elimination system, which is used to implement the method, including:

[0120] Data collection module, used to build a biological monitoring network, collect and monitor biological information, geographic phenology information, environmental quality information and other information within a preset land area;

[0121] An information base construction module is used to perform species identification based on the biological information to obtain species identification results; based on the species identification results, the biological information, geographic phenological information and environmental quality information, establish a biological basic information base for each species;

[0122] A diversity assessment module is used to extract the target parameters of the geographic phenology information, the environmental quality information and the biological information, and to evaluate the current status of biodiversity and the regional distribution pattern based on the biological basic information database to obtain the evaluation results of species suitable habitats;

[0123] An influencing factor analysis module is used to compare the historical biological information in the biological basic information database horizontally, and analyze the correlation between geographical phenological information, environmental quality information, other information and biological information of each species, obtain the threat factors of species, and generate the change trend over the years;

[0124] The risk diagnosis module is used to obtain the main contribution rate of the threatened factors of each species based on the assessment results and the historical data of the threatened factors of the species, and obtain the species risk critical limit based on the main contribution rate of the threatened factors of each species; compare the real-time data with the species risk critical limit, perform biodiversity risk diagnosis, and obtain risk diagnosis results; in this embodiment, the real-time data and historical data are compared and analyzed to clarify the species at risk and the main influencing factors causing the abnormality, and the main contribution rate of each threatening factor is given based on the historical monitoring data to form the species risk critical limit, and the real-time data is compared with the species risk critical limit to perform biodiversity risk diagnosis and obtain risk diagnosis results.

[0125] The risk warning module is used to provide different degrees of warning on risk diagnosis results based on changes in different geographical phenology, environmental quality impacts, and other information;

[0126] The risk elimination module is used to provide risk elimination measures based on the risk diagnosis results. Risk elimination includes proposing targeted response measures and policy recommendations based on the risk diagnosis results, and carrying out risk elimination work, including but not limited to artificial intervention in natural phenology, prevention and control of alien invasive species, removal of environmental pollution sources or cutting off pollution pathways, forcing the cessation of human interference activities, and pushing targeted protection policies to management departments.

[0127] The above embodiments are only descriptions of the preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should fall within the protection scope determined by the claims of the present invention.

Claims

1. An artificial intelligence terrestrial biodiversity real-time online monitoring and assessment - risk warning - risk elimination method, characterized in that: The following steps are involved: Build a biological monitoring network to collect and monitor biological information, geographic phenology information, environmental quality information and other information within the preset land area; Perform species identification based on the biological information to obtain a species identification result; Establishing a biological basic information database for each species based on the species identification results, the biological information, geographic phenological information, and environmental quality information; Extracting the target parameters of the geographic phenological information, the environmental quality information and the biological information, and based on the biological basic information database, conducting an assessment of the current status of biodiversity and the regional distribution pattern, and obtaining an assessment result of suitable habitats for species; Compare the historical biological information in the biological basic information database horizontally, analyze the correlation between geographic phenology information, environmental quality information, other information and biological information of each species, obtain the threat factors of species, and generate the change trend over the years; Based on the assessment results and the historical data of the threatened factors of the species, the main contribution rate of each threatening factor is obtained, and the critical risk limit of the species is obtained based on the main contribution rate of each threatening factor; the real-time data is compared with the critical risk limit of the species to perform biodiversity risk diagnosis and obtain risk diagnosis results; Provide different levels of early warning for risk diagnosis results based on changes in different geographical phenology, environmental quality impacts, and other information; Providing risk elimination measures based on the risk diagnosis results.

2. The artificial intelligence terrestrial biodiversity real-time online monitoring and assessment-risk warning-risk elimination method according to claim 1 is characterized in that: The biological information includes biological information of terrestrial higher plants, terrestrial mammals, birds, amphibians, reptiles, insects and fungi within the monitored land area; The geographical phenological information includes wind direction, wind speed, temperature, humidity, rainfall, snowfall, latitude and longitude, altitude, aspect and slope information; The environmental quality information includes air environment, surface water and sediment, groundwater, soil, noise, and electromagnetic and radiation information; The other information includes information on human disturbance, protected lists, protection policies and historical data.

3. The artificial intelligence terrestrial biodiversity real-time online monitoring and assessment-risk warning-risk elimination method according to claim 2 is characterized in that: The morphological characteristics, audio-visual characteristics and gene biological spectrum of the organisms in the biological information are extracted and compared with the existing data to identify the species, obtain a species list and gallery, and classify endemic species, endangered species, new species and alien invasive species.

4. The artificial intelligence terrestrial biodiversity real-time online monitoring and assessment-risk warning-risk elimination method according to claim 3 is characterized in that: The method for extracting the morphological features is: Crawl the existing species database within the preset land area, clean the crawled data, and convert it into standardized species corpus; Entities are extracted and annotated from the species corpus, and a knowledge graph of biological species and related information is constructed in a top-down manner; the nodes in the knowledge graph include species and species morphological characteristics; Based on the biological information, obtaining a biological image; Extracting color features, texture features, and contour features of the species in the biological image, and concatenating all the features to obtain a manual feature vector; Use stacked graph attention layers to build graph convolutional neural networks; Using the graph convolutional neural network, the biological image labels are indexed in the knowledge graph to obtain the morphological feature vectors of the corresponding nodes of the corresponding species in the knowledge graph; Extracting components of the morphological feature vector and the manual feature vector, and performing cosine similarity calculation on the extracted components to obtain similarity loss; Based on the similarity loss, obtaining a morphological similarity feature vector; Extracting a semantic representation vector from the biological image by using VIT that introduces a Transformer mechanism; The manual feature vector, the morphological similarity feature vector and the semantic representation vector are fused to obtain the final morphological feature of the species.

5. The artificial intelligence terrestrial biodiversity real-time online monitoring and assessment-risk warning-risk elimination method according to claim 3 is characterized in that: The methods for evaluating the current status of biodiversity and regional distribution patterns and obtaining the evaluation results are as follows: Preset target parameters; the target parameters include species diversity parameters, ecosystem diversity parameters, genetic diversity parameters, species distribution pattern parameters, and threat and protection parameters; Based on the biological basic information library, screening and extracting data related to target parameters; Conduct statistical and spatial analysis on the data related to the target parameters to assess the status and level of biodiversity and obtain biodiversity assessment results; Extract species distribution data and ecosystem type data from data related to target parameters, draw biological maps, analyze biodiversity characteristics and differences in different regions, and obtain species spatial distribution patterns and change trends; The assessment results are obtained based on the biodiversity assessment results and the regional spatial distribution pattern and change trend of species.

6. The artificial intelligence terrestrial biodiversity real-time online monitoring and assessment-risk warning-risk elimination method according to claim 1 is characterized in that: The method to obtain the threat factors of species is: Standardize historical biological information to obtain standard background data; Conduct time series and spatial distribution change analysis on the standard background data to obtain the spatiotemporal characteristics of species diversity; Using principal component analysis, the effects of geographic phenology information, environmental quality information and other information on the spatiotemporal characteristics of species diversity are analyzed to obtain a first analysis result; Analyze the correlation between biological information, geographic phenological information, environmental quality information and other information of each species to obtain the second analysis result; By combining the first analysis result and the second analysis result, the threat factors of species are obtained and the change trends over the years are generated.

7. The artificial intelligence terrestrial biodiversity real-time online monitoring and assessment-risk warning-risk elimination method according to claim 1 is characterized in that: The method to obtain risk diagnosis results is: By comparing and analyzing real-time data with historical data, we can identify species at risk and the main factors that cause abnormalities, and give the main contribution rate of each threat factor based on historical monitoring data to form species risk critical limits. We can then compare real-time data with the species risk critical limits to conduct biodiversity risk diagnosis and obtain risk diagnosis results. Provide different levels of early warning for risk diagnosis results based on changes in different geographical phenology, environmental quality impacts, and other information; Providing risk elimination measures based on the risk diagnosis results.

8. An artificial intelligence terrestrial biodiversity real-time online monitoring and assessment-risk warning-risk elimination system, characterized by: The method for implementing any one of claims 1 to 7 comprises: Data collection module, used to build a biological monitoring network, collect and monitor biological information, geographic phenology information, environmental quality information and other information within a preset land area; An information base construction module is used to perform species identification based on the biological information to obtain species identification results; based on the species identification results, the biological information, geographic phenological information and environmental quality information, establish a biological basic information base for each species; A diversity assessment module is used to extract the target parameters of the geographic phenology information, the environmental quality information and the biological information, and to evaluate the current status of biodiversity and the regional distribution pattern based on the biological basic information database to obtain the evaluation results of species suitable habitats; An influencing factor analysis module is used to compare the historical biological information in the biological basic information database horizontally, and analyze the correlation between geographical phenological information, environmental quality information, other information and biological information of each species, obtain the threat factors of species, and generate the change trend over the years; A risk diagnosis module, for obtaining the main contribution rate of the threatened factors of each species based on the assessment results and the historical data of the threatened factors of the species, and obtaining the species risk critical limit based on the main contribution rate of the threatened factors of each species; comparing the real-time data with the species risk critical limit, performing biodiversity risk diagnosis, and obtaining risk diagnosis results; The risk warning module is used to provide different degrees of warning on risk diagnosis results based on changes in different geographical phenology, environmental quality impacts, and other information; The risk elimination module is used to provide risk elimination measures based on the risk diagnosis results.

Images