Evaluation method for resisting alien species invasion in giant panda national park

Abstract

The invention discloses a method and a device for evaluating alien species invasion resistance of a giant panda national park, relates to the technical field of ecological protection, and aims to solve the technical problem that a method for evaluating alien species invasion resistance of the giant panda national park is lacked in the prior art. The method comprises the steps that after target alien species are defended, real-time video data of the giant panda national park are acquired, and attention weight vectors of the target alien species are acquired based on the real-time video data; constructing context features based on analysis of influence factors of historical growth of target alien species; constructing growth state characteristics of the target alien species population, and obtaining historical alien species invasion information and corresponding measure decisions; based on the historical alien species intrusion information and the measure decision, obtaining a historical information empirical feature vector; and based on the growth state features of the target alien species population and the historical information empirical feature vector, evaluating the alien species invasion resisting method of the giant panda national park.

Description

Technical Field

[0001] The present application relates to the field of ecological protection technology, and in particular to an evaluation method and device for the Giant Panda National Park to resist the invasion of alien species. Background Art

[0002] In 2021, my country officially established the Giant Panda National Park. The protection of giant pandas has been transformed from scattered protection in multiple nature reserves to overall protection after the establishment of the national park. The Giant Panda National Park is located in the high mountain canyon area where the Sichuan Basin transitions to the Qinghai-Tibet Plateau. It has complex topography and landforms, rich wildlife resources, and many rare and endemic species. There are 316 species of nationally protected wild animals and plants living in the area, which is known as the "natural gene bank". The Giant Panda National Park spans the three provinces of Sichuan, Shaanxi, and Gansu, with a total area of ​​22,000 square kilometers, of which 87.9% is located in Sichuan Province. While protecting giant pandas, the Giant Panda National Park also has an umbrella effect on other wild animals and plants distributed in the same area, which is of great significance to the protection of regional biodiversity and ecosystems.

[0003] Biological invasion is the second largest cause of biodiversity loss, second only to habitat destruction, and has become a serious threat to global ecological security and human health. China is one of the countries with the richest biodiversity in the world, and is also one of the countries most threatened by biological invasion. Alien target alien species such as Eichhornia crassipes and Eupatorium adenophorum have swept across the southwest, causing huge losses to agriculture, forestry and animal husbandry in the invaded areas.

[0004] As the first national park established in the Sichuan nature reserve system, the Giant Panda National Park is the cornerstone of biodiversity conservation in the giant panda habitat. Evaluating the effectiveness of the Giant Panda National Park in resisting the invasion of alien species, revealing the spatial distribution pattern of alien species in the national park, and clarifying the risk of alien species spread in the national park and its surrounding buffer zone are crucial to proposing a biosafety strategy in nature reserves with national parks as the main body. Summary of the invention

[0005] The present application provides a method, device, equipment and medium for evaluating the Giant Panda National Park's resistance to alien species invasion, aiming to solve the technical problem that the prior art lacks an evaluation method for the Giant Panda National Park's resistance to alien species invasion.

[0006] To solve the above technical problems, the present application embodiment provides: an evaluation method for the Giant Panda National Park to resist the invasion of alien species, comprising the following steps:

[0007] After adopting the target defense measures to defend against the target alien species, obtaining real-time video data of the Giant Panda National Park; extracting the real-time growth characteristics of the target alien species based on the real-time video data of the Giant Panda National Park; and obtaining the attention weight vector of the target alien species based on the real-time growth characteristics of the target alien species;

[0008] Based on the analysis of factors influencing the historical growth of the target alien species, context features are constructed; the context features include target alien species information, ecological environment characteristics, climate conditions, biodiversity data and target resistance measures;

[0009] Based on the attention weight vector of the target alien species and the context feature, construct a multimodal growth feature vector of the target alien species; based on multiple multimodal growth feature vectors of the target alien species, construct a growth state feature of a target alien species population;

[0010] Based on the semantic extraction model, historical alien species invasion information and corresponding Giant Panda National Park measures and decisions are obtained; based on the historical alien species invasion information and the Giant Panda National Park measures and decisions, historical information experience feature vectors are obtained;

[0011] Based on the growth status characteristics of the target alien species population and the historical information empirical feature vector, the growth characteristics of the target alien species at time t+1 are predicted; based on the growth characteristics of the target alien species at time t+1, the method of the Giant Panda National Park to resist the invasion of alien species is evaluated.

[0012] As some optional implementations of the present application, the step of obtaining the attention weight vector of the target alien species based on the real-time growth characteristics of the target alien species includes:

[0013] Obtain historical growth characteristic sequences of target alien species;

[0014] Based on the historical growth feature sequence of the target alien species and the real-time growth feature of the target alien species, an attention weight vector of the target alien species is obtained.

[0015] As some optional implementations of the present application, the step of obtaining the attention weight vector of the target alien species based on the historical growth feature sequence of the target alien species and the real-time growth feature of the target alien species includes:

[0016] By extracting the attention weights of different defense measures and the current growth feature vector of the target alien species from the historical growth feature sequence of the target alien species;

[0017] Based on the current growth feature vector and attention weight of the target alien species, an attention weight vector of the target alien species is obtained.

[0018] As some optional implementations of the present application, the step of constructing a multimodal growth feature vector of the target alien species based on the attention weight vector of the target alien species and the context feature includes:

[0019] The attention weight vector of the target alien species and the context feature are converted into binary code by using one-hot encoding and multi-hot encoding to obtain growth feature vector encoding information of the target alien species;

[0020] The growth feature vector encoding information of the target alien species and the associated historical growth feature vector are input into the trained deep interest network to construct a multimodal growth feature vector of the target alien species.

[0021] As some optional implementation methods of the present application, the steps of obtaining historical alien species invasion information and corresponding Giant Panda National Park measures decisions based on the semantic extraction model; and obtaining historical information empirical feature vectors based on the historical alien species invasion information and the Giant Panda National Park measures decisions include:

[0022] Obtaining a BERT semantic extraction model; wherein the BERT semantic extraction model is obtained based on historical data training, and the historical data includes information on the time of biological invasion, the scope of biological invasion, changes in the degree of biological invasion, and types of measures taken by the Giant Panda National Park;

[0023] Based on the BERT semantic extraction model, key information on historical biological invasions and a set of guiding measures for the Giant Panda National Park were obtained;

[0024] The key information of the historical biological invasion and the set of guiding measures of the Giant Panda National Park are encoded and mapped into a high-dimensional information matrix, and a preliminary biological invasion status description vector is extracted by embedding;

[0025] The preliminary biological invasion status description vector is input into the trained EventKG model to realize the temporal relationship modeling between the measures taken and the changes in the biological invasion status, and then the relationship between the measures taken and the changes in the growth status of the alien species population is represented as a historical information experience vector in a continuous vector space.

[0026] As some optional implementations of the present application, the step of predicting the growth characteristics of the target alien species at time t+1 based on the growth state characteristics of the target alien species population and the historical information experience feature vector includes:

[0027] Inputting the growth status characteristics of the target alien species population into the trained stacked transformers network model to extract the historical experience information extraction network, the context information Ct extraction network and the real-time status feature extraction network;

[0028] The historical experience information extraction network, the context information Ct extraction network and the real-time status feature extraction network are respectively decoded to obtain decoding results; the decoding results are respectively input into the prediction module to predict the growth characteristics of the alien species population at time t+1.

[0029] As some optional implementations of the present application, the step of inputting the decoding results into a prediction module to predict the growth characteristics of the alien species population at time t+1 includes:

[0030] Based on the decoding result, a plurality of candidate state allocation regions are obtained, and a plurality of regions are obtained; wherein each region represents a growth state of the alien species;

[0031] Determine the area where the growth status characteristics of the target alien species population are located, calculate the growth status and confidence generated under different optional measures, define an overall loss function based on the empirical characteristics of historical information; and train the parameters in the loss function to obtain a prediction network model;

[0032] The growth status characteristics of the current target alien species population are input into the prediction network model to obtain the growth status type of the alien species population at time t+1 under different measures; wherein the prediction network model uses stacked transformers as the prediction network architecture.

[0033] On the other hand, the embodiment of the present application further provides: an evaluation device for the Giant Panda National Park to resist the invasion of alien species, comprising:

[0034] The first acquisition module is used to acquire real-time video data of the Giant Panda National Park after adopting the target defense measures to defend against the target alien species; based on the real-time video data of the Giant Panda National Park, extract the real-time growth characteristics of the target alien species; based on the real-time growth characteristics of the target alien species, obtain the attention weight vector of the target alien species;

[0035] The second acquisition module is used to construct context features based on the analysis of factors affecting the historical growth of the target alien species; the context features include target alien species information, ecological environment characteristics, climate conditions, biodiversity data and target resistance measures;

[0036] Constructing a target alien species population growth state module, which is used to construct a target alien species multimodal growth feature vector based on the attention weight vector of the target alien species and the context feature; constructing a target alien species population growth state feature based on multiple target alien species multimodal growth feature vectors;

[0037] A semantic extraction module, for obtaining historical alien species invasion information and corresponding Giant Panda National Park measures decisions based on a semantic extraction model; and obtaining historical information empirical feature vectors based on the historical alien species invasion information and the Giant Panda National Park measures decisions;

[0038] An evaluation module is used to predict the growth characteristics of the target alien species at time t+1 based on the growth status characteristics of the target alien species population and the historical information empirical feature vector; based on the growth characteristics of the target alien species at time t+1, the method of the Giant Panda National Park to resist the invasion of alien species is evaluated.

[0039] On the other hand, an embodiment of the present application also provides: a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-mentioned evaluation method for the Giant Panda National Park to resist the invasion of alien species.

[0040] On the other hand, an embodiment of the present application further provides: a computer-readable storage medium having a computer program stored thereon, and a processor executes the computer program to implement the above-mentioned evaluation method for the Giant Panda National Park to resist the invasion of alien species.

[0041] Compared with the existing technology, the technical solution of this application solves the shortcomings of the existing technology in the assessment method of resisting alien species invasion in the Giant Panda National Park by introducing multiple data sources and analysis methods, which is specifically reflected in the following aspects:

[0042] 1. Fusion of multimodal data

[0043] Current alien species invasion assessment methods usually rely on a single data source, such as ecological surveys or expert assessments, and lack real-time, dynamic information updates. This technical solution achieves all-round monitoring of the status of alien species by collecting real-time video data, combining the growth characteristics of target alien species and multiple information of the ecological environment (such as climate conditions and biodiversity data). This integration of multimodal data improves the accuracy and timeliness of assessment results.

[0044] 2. Application of Attention Weight Vector

[0045] By extracting the attention weight vector of the target alien species, this solution can identify the key factors that affect the growth of alien species. Compared with the traditional method that only relies on statistical models for data analysis, this is more flexible and dynamic, and can reflect the adaptability and changes of alien species under specific environmental conditions in real time. This method enhances the pertinence and scientificity of the assessment.

[0046] 3. Effective use of historical information

[0047] The solution uses a semantic extraction model to obtain historical information on alien species invasion and corresponding management measures, combining historical data with current assessments. The construction of historical information experience feature vectors can provide an important reference for current assessments, thereby improving the referenceability and effectiveness of assessments. By analyzing historical data, the solution can better identify successful management measures and existing problems, providing a scientific basis for future decision-making.

[0048] 4. Improved predictive capabilities

[0049] By constructing the growth status characteristics of the target alien species group and the historical information experience feature vector, the technical solution can predict the growth characteristics of alien species at time t+1. This predictive ability enables the assessment to not only stop at the analysis of the current situation, but also to look forward to future trends. This provides managers with an early warning mechanism on the one hand, and a scientific basis for policy making on the other.

[0050] 5. Objectivity and evaluation standardization

[0051] By constructing a multimodal growth feature vector and combining contextual features, this scheme makes the assessment method more objective and reduces the interference of human factors on the assessment results. At the same time, the standardized method in the assessment process allows data from different times and locations to be compared horizontally, providing a quantitative reference standard for long-term ecological management.

[0052] In summary, the technical solution of this application has greatly improved the evaluation ability of the Giant Panda National Park to resist alien species invasion by integrating advanced technologies such as real-time monitoring, multimodal data analysis, historical data utilization and predictive modeling. It is conducive to improving the effectiveness of evaluating the Giant Panda National Park in resisting biological invasions; predicting the distribution of suitable habitats for invasive species and delineating key prevention and control areas; and providing a basis for formulating the most practical management measures for the prevention and control of alien invasive species in the Giant Panda National Park. This not only solves the problem of the lack of systematicness and objectivity of existing technologies, but also provides a scientific basis and practical reference for the formulation and adjustment of relevant management measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the specific implementation of the present application or the technical solution in the prior art, the following is a brief introduction to the drawings required for the specific implementation or the prior art description. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual scale.

[0054] Figure 1 A distribution map of the survey routes of alien invasions in forest, grassland and wetland ecosystems in Sichuan Province involved in this application;

[0055] Figure 2 A diagram showing the relationship between the key invasive alien species involved in this application and the Giant Panda National Park;

[0056] Figure 3 This is a heat map of the spatial distribution of invasive alien species in the forestry of 21 cities and prefectures in Sichuan Province involved in this application;

[0057] Figure 4 A flowchart of the evaluation method for the Giant Panda National Park to resist the invasion of alien species in this application;

[0058] Figure 5 This is a schematic diagram of the structure of the local attention module in this application;

[0059] Figure 6 This is a schematic diagram of the fusion representation of multimodal features of target alien species under alien species in this application;

[0060] Figure 7 The logic diagram for obtaining and describing the historical information experience feature vector H in this application;

[0061] Figure 8 This is a schematic diagram of the structure of the alien species growth status prediction network in this application;

[0062] Fig. 9 Schematic diagram of the training strategy based on region division and decision-making in this application;

[0063] Fig.10 This is a schematic diagram of the structure of the evaluation device for the Giant Panda National Park to resist the invasion of alien species involved in this application.

[0064] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0066] It should be noted that all directional indications in the embodiments of the present application (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components in a certain specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0067] In this application, unless otherwise clearly specified and limited, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0068] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the meaning of "and / or" appearing in the full text includes three parallel schemes. Taking "A and / or B" as an example, it includes scheme A, or scheme B, or a scheme that satisfies both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0069] The computer device of the hardware operating environment involved in the solution of the present application may include: a processor, such as a central processing unit (CPU), a communication bus, a user interface, a network interface, and a memory. Among them, the communication bus is used to realize the connection and communication between these components. The user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface may also include a standard wired interface and a wireless interface. The network interface may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (WIreless-FIdelity, WI-FI) interface). The memory may be a high-speed random access memory (Random Access Memory, RAM) memory, or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory may also be a storage device independent of the aforementioned processor. The memory as a storage medium may include an operating system, a network communication module, a user interface module, and an electronic program, and may also include a data storage module. The network interface is mainly used for data communication with the network server; the user interface is mainly used for data interaction with the user; the processor and memory in the computer device of this embodiment can be set in the computer device, and the computer device calls the evaluation device for the Giant Panda National Park to resist the invasion of alien species stored in the memory through the processor, and executes the evaluation method for the Giant Panda National Park to resist the invasion of alien species provided in this embodiment.

[0070] It should be noted that the above-mentioned computer device may be an external hardware device that can run independently, or it may be a hardware device that comes with the internal part of the evaluation device for resisting the invasion of alien species in the above-mentioned Giant Panda National Park.

[0071] It should be noted that the data source of the technical solution described in this application is the census of alien invasive species in forest, grassland and wetland ecosystems in Sichuan Province conducted from 2021 to 2023. Figure 1 The following is a distribution map of alien invasion census routes in forest, grassland and wetland ecosystems in Sichuan Province drawn based on survey data, as well as Figure 2 The relationship between the key invasive alien species and the Giant Panda National Park is shown, and Figure 3The heat map of the spatial distribution of invasive alien species in the forestry of 21 cities and prefectures in Sichuan Province is shown. The survey covers all 183 counties (cities and districts) within the administrative scope of Sichuan Province, including protected areas at all levels with clear boundaries as independent survey units. According to the types and distribution points of invasive species, the spatial distribution pattern of key invasive species in the Giant Panda National Park and its surrounding areas is clarified, the number, occurrence area and growth characteristics of invasive species in the Giant Panda National Park are calculated, and the effectiveness of the national park in resisting the invasion of alien species is evaluated.

[0072] Reference Figure 4 Based on the aforementioned hardware environment, this embodiment also provides an evaluation method for the Giant Panda National Park to resist the invasion of alien species, comprising the following steps:

[0073] Step S10: After adopting target defense measures to defend against target alien species, obtain real-time video data of the Giant Panda National Park; based on the real-time video data of the Giant Panda National Park, extract the real-time growth characteristics of the target alien species; based on the real-time growth characteristics of the target alien species, obtain the attention weight vector of the target alien species.

[0074] It should be noted that the real-time video data of the Giant Panda National Park is mainly derived from the network cameras of the Giant Panda National Park. These cameras are installed in key areas of the Giant Panda National Park, such as entrance areas and transit passages, which are the most likely paths for alien species to enter. By monitoring these key entrances, the first appearance of alien species can be captured in time, and measures can be taken quickly to prevent their further spread. Or water sources, which are hotspots of biodiversity and attract various organisms to drink water and live. Monitoring these areas helps to observe the interaction between alien species and local species and assess whether alien species pose a threat to the ecological balance of water sources. Or important ecological areas, such as forests and grasslands with special ecological functions, these areas are often important support points for biodiversity. Installing cameras in these areas can help researchers monitor the potential impact of alien species on local ecosystems. Or areas with intensive human activities, alien species may be introduced through human activities (such as tourism, transportation, etc.). Monitoring these areas can help identify the role of human factors in the spread of alien species. Or areas where alien species are known to be active. For areas where alien species have already appeared, continuous monitoring of their dynamics is necessary in order to evaluate the effectiveness of measures taken and adjust strategies in a timely manner.

[0075] By installing webcams in these key areas, we can maximize monitoring coverage, obtain direct evidence of alien species invasion in a timely manner, and effectively evaluate and adjust protection measures. Such an installation strategy can provide a comprehensive, real-time ecological monitoring system for the Giant Panda National Park, helping to maintain its biodiversity and ecological balance.

[0076] For example, in the San Bernardino Mountains in California, a camera system was installed to monitor the invasion of non-native species, successfully recording the activities of foreign wild boars and domestic cats, which are essential for taking timely management measures. In Serengeti National Park in Africa, by installing cameras around water holes and rivers, researchers were able to observe the impact of foreign species invasion on the activities of local crocodiles and hippos. In Tasmania, Australia, researchers deployed cameras in native forest areas to record the impact of foreign foxes on local wombats and other small mammal populations.

[0077] In practical situations, extracting real-time growth characteristics of alien species can help researchers and managers accurately understand the health status, reproduction and population dynamics of these species. This information is crucial for assessing the specific impact of alien species on native ecosystems. And by real-time monitoring of the growth and spread patterns of alien species, managers can promptly identify potential ecological risks and hot spots, so as to implement early intervention measures, such as adjusting ecological isolation areas or launching emergency population control plans. And detailed growth data supports scientific research, such as ecological adaptability, competitive relationships and predation pressure, which are key scientific issues in understanding and responding to the invasion of alien species.

[0078] When using machine learning and deep learning models to analyze video data, the attention mechanism can help the model more accurately identify and track target alien species, especially in complex natural environments. The attention weight vector can indicate which parts of the image the model should focus on, thereby improving recognition accuracy.

[0079] Attention models can be used to analyze the growth patterns of alien species, such as their activity frequency and habits at specific times and places. This helps to study their ecological habits and environmental adaptability, and provides data support for the development of management strategies. By understanding which areas and time periods are most active for alien species (based on the distribution of attention weights), managers can more effectively allocate monitoring resources and control measures, such as increasing patrol frequency during peak activity periods or optimizing the configuration of monitoring equipment.

[0080] Therefore, it can be seen that the ability of ecological protection areas to monitor and respond to alien species invasions can be greatly enhanced by extracting real-time growth characteristics of alien species through technical means and applying attention weight vectors. This not only helps to protect native biodiversity, but also improves the scientificity and effectiveness of management measures.

[0081] In actual situations, the growth characteristics of the target alien species can quantify the current reproduction degree of the target alien species to a certain extent, so the present application intends to realize the unified characterization of the growth of the target alien species integrating the influencing factors based on the real-time growth characteristics of the target alien species, as shown in step S20.

[0082] Specifically, based on the real-time growth data of the target alien species, the current growth vector of the target alien species is adaptively calculated by extracting the attention weight vectors of the current growth in different periods from its historical growth sequence; redundant information on growth is effectively removed, and the current reproduction status of the target alien species is accurately reflected.

[0083] Specifically, the step of obtaining the attention weight vector of the target alien species based on the real-time growth characteristics of the target alien species includes: obtaining the historical growth feature sequence of the target alien species; and obtaining the attention weight vector of the target alien species based on the historical growth feature sequence of the target alien species and the real-time growth characteristics of the target alien species.

[0084] Among them, the step of obtaining the attention weight vector of the target alien species based on the historical growth feature sequence of the target alien species and the real-time growth features of the target alien species includes: extracting the attention weights of different defense measures and the current growth feature vector of the target alien species from the historical growth feature sequence of the target alien species; and obtaining the attention weight vector of the target alien species based on the current growth feature vector and attention weight of the target alien species.

[0085] In one example, suppose the target species is the American mink, which is found to be a threat to native bird populations in a certain area. The research team wants to use the latest technology to monitor the growth and spread of the mink in order to develop effective control measures. The above steps may include:

[0086] 1) Obtain the historical growth characteristic sequence of the target alien species:

[0087] First, we collected growth data on American minks in the past few years, including weight, body length, reproduction rate, food source, etc. Using this data to build a time series model will help us understand how American minks adapt to new environments and infer their possible growth trends. Based on the historical growth characteristic sequence, we can predict how the population size and distribution of American minks may change in the future.

[0088] 2) Obtain the attention weight vector of the target alien species:

[0089] By setting up wildlife monitoring cameras, we collect the activity and growth data of American minks in real time. We combine historical growth data with real-time observation data, use deep learning models to analyze the importance of current data, and generate attention weight vectors. We identify the places and times where American minks are most active under specific environmental conditions. These are the key areas emphasized by the attention weight vectors.

[0090] 3) Extract the attention weights of different defense measures and current growth features:

[0091] Analyze the effects of different control measures implemented in the past (such as capture, fencing, introduction of ecological competitors, etc.) on the growth characteristics of American mink. Based on the effects of these measures, adjust the model's attention weights to highlight those growth characteristics associated with effective control measures. If it is found that the growth rate of American mink slows down or the population decreases under certain specific measures, then give these characteristics a higher weight in the model. This will help to more accurately predict and implement the most effective control strategies in the future.

[0092] Through the above analysis and application, we can see that combining historical and real-time growth characteristics and optimizing alien species management measures through attention mechanisms can not only improve the accuracy of monitoring and response, but also make ecological management more scientific and systematic.

[0093] Step S20: construct context features based on the analysis of factors affecting the historical growth of the target alien species; the context features include target alien species information, ecological environment characteristics, climate conditions, biodiversity data and target defense measures.

[0094] In the actual technical solution, the characteristics of the target alien species under alien species have the characteristics of multiple types and multiple structures. In order to realize the fusion representation of the multimodal characteristics of the target alien species, the unified feature encoding method is first used to convert the multimodal feature data into binary codes that can be uniformly processed; then a deep interest network based on the attention module is constructed to realize the association between the historical growth feature vector and the real-time growth feature vector of the target alien species.

[0095] The following is a detailed description of the steps of converting multimodal feature data into binary codes that can be uniformly processed by using a unified feature coding method, with reference to an example:

[0096] Suppose the target alien species of the study is an invasive plant with rich multimodal characteristics, Eichhornia crassipes. The growth characteristics of Eichhornia crassipes include visual data (such as images), biological data (such as DNA sequences), and environmental data (such as water quality conditions), which are multi-type and multi-structured data. Eichhornia crassipes is a fast-growing aquatic plant that can quickly cover the entire surface of a water body, hindering water flow, affecting water quality, and posing a huge threat to local ecosystems. Effective management and control of its spread requires a detailed analysis of its growth characteristics.

[0097] Step 211: Collection of feature data

[0098] Visual data: Use drones and underwater cameras to collect images and video data of water hyacinth, recording its growth density, distribution range, etc.

[0099] Biological data: The DNA sequence of water hyacinth was obtained through sample collection and its genetic variation was analyzed.

[0100] Environmental data: Measure environmental parameters such as water temperature, pH, dissolved oxygen, etc. in the growing area.

[0101] Step 212: Feature preprocessing

[0102] Image preprocessing: crop, scale, and standardize the image to make it suitable for subsequent calculation and analysis.

[0103] DNA data processing: converting DNA sequence data into numerical features, such as through certain encoding strategies (such as k-mer counting).

[0104] Environmental data standardization: Normalize environmental parameters to eliminate the impact of dimension.

[0105] Step 213: Multimodal Data Encoding

[0106] Feature vectorization: All pre-processed feature data are converted into vector form. For example, images are extracted through convolutional neural networks (CNNs), DNA data is converted through specific bioinformatics tools, and environmental data is directly input as numerical values.

[0107] Binary encoding: Use an autoencoder or other suitable deep learning model to further compress the feature vector and convert it into binary encoding. The key to this step is to reduce the data dimension and unify the data format from different sources to facilitate unified processing in the deep network.

[0108] Step 214: Comprehensive feature expression

[0109] Unified Processing: The binary-encoded feature data is fed into a deep interest network based on an attention module that identifies and reinforces associations between key features, such as the interaction of historical growth patterns with real-time environmental conditions.

[0110] Through the above steps, the multimodal data of water hyacinth can be effectively converted into a unified internal expression, which can be further analyzed and predicted using a deep learning model. This method not only improves the efficiency of data processing, but also enhances the model's ability to understand and predict the growth dynamics of alien species.

[0111] In some other embodiments, the step of constructing a multimodal growth feature vector of the target alien species based on the attention weight vector of the target alien species and the context feature may further include:

[0112] The attention weight vector of the target alien species and the context features are converted into binary codes using one-hot encoding and multi-hot encoding to obtain the growth feature vector encoding information of the target alien species; the growth feature vector encoding information of the target alien species and the associated historical growth feature vector are input into a trained deep interest network to construct a multimodal growth feature vector of the target alien species.

[0113] Taking the small grass (a common invasive plant) as an example to show how to construct its multimodal growth feature vector through the attention weight vector and context features, the specific implementation steps are as follows:

[0114] Step 221: Collect feature data

[0115] Attention weight vector: This may include weight assessments of the environmental factors that have the greatest impact on P. sphaerocephala, such as sunshine, temperature, moisture, etc. These weights can be obtained through experimental data or field monitoring.

[0116] Contextual features: For example, soil pH, surrounding vegetation type, land use, etc.

[0117] Step 222: Feature preprocessing and encoding

[0118] One-hot encoding: Applicable to categorical features, each category has a single position. For example, land use (farmland, urban, forest land) will be converted into a three-element vector, where only one element is 1 and the rest are 0.

[0119] Multi-hot encoding: Applicable to features that can belong to multiple categories. Suppose that sedge grows in multiple different climate conditions, such as dry season and rainy season, these conditions will be encoded in the same vector.

[0120] Step 223: Convert to binary encoding

[0121] Feature vector construction: Combine the one-hot and multi-hot encoded data to form a complete growth feature vector encoding information. This information will reflect the adaptability and growth status of small grass under different environmental conditions.

[0122] Step 224: Input Deep Interest Network

[0123] Application of feature vectors: The encoded current feature vector is input into a deep interest network together with the historical growth feature vector. This network may be based on deep learning techniques and is specially trained to predict changes in plant growth patterns and potential distributions.

[0124] Deep learning model: The model analyzes the input feature vector, identifies the key factors affecting the growth of P. truncatum, and predicts its future growth trend and potential invasion risk.

[0125] Through the above steps, researchers or managers can obtain a comprehensive view of how P. sphaerocephala interacts with its environment and develop more effective management strategies to control or utilize this exotic plant. This multimodal analysis approach can help to more accurately predict and respond to plant invasion problems, especially in the current context of rapid environmental change.

[0126] Based on the technical solution described above, a table can be designed to show the feature encoding of the target alien species (such as small grass). This table will contain multiple columns, showing different features and their encoding methods. The following is an example table showing how to convert the attention weight vector and context features into one-hot and multi-hot encoding forms.

[0127]

[0128] The step of constructing a deep interest network based on an attention module to realize the association between the historical growth feature vector and the real-time growth feature vector of the target alien species includes:

[0129] Step S221: Figure 5As shown, this step adopts a local attention module (the local attention module refers to the following process, through the outer product of the historical growth feature vector of the target alien species and the current growth feature vector of the target alien species), to adaptively calculate the growth representation of the current target alien species, that is: select any growth feature vector in the current growth of the target alien species, and perform outer product processing with the historical feature growth weight vector of the target alien species, and then connect the two vectors themselves, and finally pass them into the two-layer MLP layer to obtain N real weight values. The calculation process is shown in the following formula:

[0130]

[0131] Among them, (e1, e2, …e H ) is a list of embedding vectors of length H for the growth of target alien species, e j is the jth growth of the target alien species; v u (A) is the weighted vector of the historical growth characteristic vector and the real-time growth characteristic vector of the target alien species finally obtained; v A is the historical growth characteristic vector of the target alien species; w j is a feedforward network a(e j ,v A ) output weight value.

[0132] Step S222: local attention layer real number weight value w j No normalization is performed. Therefore, to a certain extent, Sum(w i ) can be equivalent to the attention weight vector of the target alien species. This value can effectively quantify the current reproduction status of the target alien species. The higher the Sum value, the greater the probability that the target alien species reproduces too quickly.

[0133] Step S30: construct a target alien species multimodal growth feature vector based on the attention weight vector of the target alien species and the context feature; and construct a target alien species population growth state feature based on multiple target alien species multimodal growth feature vectors.

[0134] The multimodal feature fusion representation of the target alien species under the alien species described in this application is as follows Figure 6As shown, it can be seen that the present application first obtains the real-time growth sequence of the target alien species based on the surveillance video of the Giant Panda National Park, and associates it to obtain the historical growth sequence of the target alien species; the real-time growth sequence and the historical growth sequence of the target alien species are feature encoded and input into the embedding layer, so that the embedding layer outputs the historical growth feature vector and the real-time growth feature vector of the target alien species; the historical growth feature vector and the real-time growth feature vector of the target alien species are then input into the attention module to obtain the current reproduction degree of the target alien species; at the same time, in order to improve the prediction accuracy, the present application also feature encodes the contextual features and inputs them into the embedding layer and the connection layer; the contextual features and the attention weight vector of the target alien species are simultaneously input into the connection layer to obtain the multimodal growth characteristics of the target alien species.

[0135] After obtaining the multimodal growth characteristics of each target alien species through the above steps, the growth status characteristics of the target alien species group are constructed based on multiple factors to facilitate the subsequent evaluation of the growth status of alien species, namely:

[0136] Step S40: Based on the semantic extraction model, obtain historical alien species invasion information and corresponding Giant Panda National Park measures and decisions; based on the historical alien species invasion information and the Giant Panda National Park measures and decisions, obtain historical information empirical feature vectors.

[0137] In actual situations, there is a clear correlation between the evolutionary patterns of the various states of alien species in the Giant Panda National Park and the measures taken by the National Park, and relevant empirical information can be effectively obtained from historical biological invasions. Therefore, firstly, the historical biological invasion characteristics and the measures that can be taken by the Giant Panda National Park are obtained through a semantic extraction model; and based on the historical biological invasion characteristics and the measures decisions of the Giant Panda National Park, the growth status characteristics of the target alien species population are obtained, including: obtaining a BERT semantic extraction model; wherein the BERT semantic extraction model is obtained based on historical data training, and the historical data includes the time of biological invasion, the scope of biological invasion, the change information of the degree of biological invasion and the type of measures taken by the Giant Panda National Park; based on the BERT semantic extraction model, the key information of historical biological invasion and the guiding measures set of the Giant Panda National Park are obtained; the key information of historical biological invasion and the guiding measures set of the Giant Panda National Park are encoded and mapped to a high-dimensional information matrix, and a preliminary biological invasion status description vector is extracted by embedding; the preliminary biological invasion status description vector is input into the trained EventKG model to realize the temporal relationship modeling between the taking of measures and the change of biological invasion status, and then the relationship between the taking of measures and the change of growth status of alien species population is represented as a historical information experience vector in a continuous vector space.

[0138] That is Figure 7 As shown, a BERT semantic extraction model is obtained by training the historical data in the Giant Panda National Park database; semantic extraction is performed based on the BERT semantic extraction model, and the information obtained by the semantic extraction is encoded and input into the embedding layer so that it outputs historical biological invasion characteristics; and based on the historical biological invasion characteristics, the event correlation relationship is extracted to classify the status of alien species, and the historical information experience vector is output.

[0139] The extraction of event association relationships refers to judging the relationship between the growth state and defense measures of alien species in the structure of alien species growth state and defense measures based on the historical biological invasion characteristics. For example, when alien species A grows to state i, if measure k is taken, how will the growth state change; if the growth state changes to state j, what defense measures should be taken, such as taking measure I, etc. It should be noted that the Figure 7 It is mainly used to express the structure of measures. The English part has no practical meaning. It is just a naming reference for storing data such as growth status or defense measures. Therefore, I will not go into details here.

[0140] Specifically, it is achieved through the following steps:

[0141] Step S41, using a semantic extraction model based on the BERT model to obtain key information in the historical information: such as the time and location of biological invasion, the type of measures taken by the Giant Panda National Park, changes in the growth status of alien species, etc.;

[0142] Step S42: extract a set of defense measures E for alien species from the relevant data on defense against alien species invasion in the Giant Panda National Park, and form a defense measure set feature vector E = (e1, ..., e n );

[0143] Step S43, encoding and mapping the corresponding historical biological invasion multi-dimensional information obtained into a high-dimensional information matrix, and extracting a preliminary description vector of the alien species growth state by embedding;

[0144] Step S44: The relationship between growth status and defense measures is represented by the EventKG model, so as to realize the temporal relationship modeling between defense measures and changes in growth status of alien species. Among them, eventKG-r represents an invasion event, rdf:type represents the invasion event type, such as event or relation, and the remaining attributes are other relevant descriptions such as the start and end time, location, effect, etc. of the invasion event.

[0145] Step S45, classifying each historical alien species through EventKG clustering according to the different degrees of development status of the alien species;

[0146] Step S46: The relationship between the measures taken and the changes in the growth status of alien species is expressed as a historical information experience vector in a continuous vector space. The vector contains the time sequence and change relationship between the development of the growth status of alien species and the defense measures taken by the Giant Panda National Park, such as Figure 8 As shown. That is:

[0147] In order to effectively utilize the empirical information contained in the historical biological invasion of Giant Panda National Park and the contextual information provided by the dynamic graph state feature set, the following measures are taken: Figure 8 The stacked transformers network shown in the figure realizes the feature extraction fusion of multiple data sources. It consists of multiple transformers, which are the historical experience information extraction network, the context information Ct extraction network and the real-time state feature extraction network. Each module unit uses the result from the previous transformer as the input of its decoder. The decoder part constructs the feature vector selector of the giant panda national park measures. For different candidate measures, multiple context information channels from the encoder are aggregated and independent predictions are made. The measures are orthogonal to each other. Different measures are independently decoded in parallel and input into the prediction module. Finally, the growth status X of alien species under different giant panda national parks is realized. t+1 The prediction is made, and the effectiveness of the defense measures taken is evaluated. That is, step S50:

[0148] Step S50, predicting the growth characteristics of the target alien species at time t+1 based on the growth status characteristics of the target alien species population and the historical information empirical feature vector; and evaluating the method of the Giant Panda National Park to resist the invasion of alien species based on the growth characteristics of the target alien species at time t+1.

[0149] Specifically, the step of predicting the growth characteristics of the target alien species at time t+1 based on the growth status characteristics of the target alien species group and the historical information experience feature vector includes: inputting the growth status characteristics of the target alien species group into a trained stacked transformers network model to extract a historical experience information extraction network, a context information Ct extraction network and a real-time state feature extraction network; decoding the historical experience information extraction network, the context information Ct extraction network and the real-time state feature extraction network respectively to obtain decoding results; and inputting the decoding results into prediction modules respectively to predict the growth characteristics of the alien species group at time t+1.

[0150] Among them, the step of inputting the decoding results into the prediction module respectively to predict the growth characteristics of the alien species group at time t+1 includes: based on the decoding results, obtaining multiple candidate state allocation areas and obtaining multiple areas; wherein each area represents a state of the alien species; determining the area where the growth state characteristics of the target alien species group are located, and calculating the growth state and confidence generated under different optional measures, defining an overall loss function based on historical information experience characteristics; and training the parameters in the loss function to obtain a prediction network model; inputting the current target alien species group growth state characteristics into the prediction network model to obtain the growth state type of the alien species group at time t+1 under different measures; wherein the prediction network model uses stacked transformers as the prediction network architecture.

[0151] That is to say, Fig. 9 As shown, this application predicts the predicted status and predicted probability of alien species in the Giant Panda National Park through the following steps:

[0152] (1) Assign a region to each candidate state, where each region represents a growth state R of the alien species. In the training phase, by selecting the region where the Ground Truth information is located (historical experience features), the trajectories and confidences generated under different optional measures are calculated, and the loss function between the trajectory and the Ground Truth is calculated.

[0153] (2) The historical information empirical feature vector H is used as training data to train the groundtruth. For a given historical dynamic graph real-time growth state feature x t , from the candidate defense measures set feature vector E = (e1,…,e n ) Select the defense measures i , i = 1 ~ n, according to the corresponding defense measures, select the corresponding dynamic graph real-time growth state feature from the historical information experience feature vector H as x gt . Predicted state x i (t+1) and x gt There is a loss function L reg :

[0154]

[0155] Among them, x i (t+1) represents the predicted growth status of the alien species at time t+1 generated under the selection of the i-th measure.

[0156] (3) The relationship between the growth of alien species under different growth states and the defense measures taken by the Giant Panda National Park is defined as a probability distribution. Therefore, the probability prediction of the change in the growth state of alien species under different defense measures of the Giant Panda National Park is equivalent to the difference evaluation between the probability distribution P (prediction) and the probability distribution G (groundtruth).

[0157] (4) For a certain area, corresponding defensive measures E can be taken region =(e1,…,e k ), the MLP network m(·) is applied to generate the corresponding confidence, which represents the probability of taking measures to make the growth state of alien species evolve to this area:

[0158]

[0159] m(·) represents the MLP network, E region =(e1,…,e k ) indicates that defensive measures can be taken, k indicates the number of defensive measures that can be taken, and R indicates the state area.

[0160] (5) Suppose there is a set of optional defense measures in a certain area, E region =(e1,…,e k ), whose probability distribution is:

[0161]

[0162] m(·) represents the MLP network; E region =(e1,…,e k ) indicates that defensive measures can be taken; τ(·) represents the probability distribution.

[0163] (6) The error of the probability prediction of alien species status under different defense measures taken by the Giant Panda National Park can be defined as:

[0164]

[0165] Among them, D(x,x gt ) represents the distance between the predicted state and the actual state in the mapping space, D(x,x gt )=||xx gt ||2;D(x i , x gt ) refers to the distance between the predicted growth state generated under the selection of the i-th measure and the actual growth state in the mapping space.

[0166] Predicted growth status x i (t+1) and x gt The probability distribution loss function L confAs shown below:

[0167]

[0168] Dkl stands for KL divergence, which is a calculation that measures the difference between two probability distributions.

[0169] (7) In summary, during the training process, the overall training loss function based on the historical information experience characteristics can be defined as:

[0170]

[0171] Among them, σ1, σ2 are the loss weights determined by learning and training; σ i Represents (σ1,σ2).

[0172] (8) Based on the training model, stacked transformers is used as the prediction network architecture to output the growth state type and probability of alien species in the next time period under different measures.

[0173] In some embodiments, the method of resisting alien species invasion in the Giant Panda National Park based on the growth characteristics of the target alien species at time t+1 is evaluated, that is, based on the growth characteristics of the target alien species at time t+1, the effectiveness of the method of resisting alien species invasion in the Giant Panda National Park is evaluated, and strategies that need to be adjusted in the future can also be predicted. That is, based on the output of the model, the potential threat of alien species to giant pandas and their ecosystems is evaluated, and suggestions for improving defense measures are put forward, such as adjusting physical barriers, chemical treatments or biological control strategies. Assess the impact of the spread of alien species on local biodiversity, especially the impact on the food chain and habitat of giant pandas. Analyze environmental changes that may be caused by alien species, such as soil nutrient imbalance and water pollution. Quantify the effectiveness of current measures to slow the spread of alien species, and determine which measures are most effective and which need to be improved. Evaluate the cost-effectiveness of different measures and optimize resource allocation.

[0174] Below, taking Water Hyacinthus as an example of an alien species, we explain in detail how to use the method described in this application to assess its invasion in the Giant Panda National Park and evaluate the effectiveness of current defense strategies:

[0175] 1. Data collection and real-time monitoring:

[0176] Surveillance cameras installed in the national park were used to capture real-time images of the water surface, and regular aerial photography was carried out using drones to obtain the coverage and growth rate of water hyacinth.

[0177] Collect data related to water quality and climate change, as these factors may affect the growth rate and distribution of water hyacinth.

[0178] 2. Feature extraction and analysis:

[0179] The growth characteristics of water hyacinth such as coverage area, density, and growth rate are extracted from real-time images.

[0180] Contextual features are constructed by combining water quality parameters (such as pH, dissolved oxygen, etc.) and climate data (such as temperature, precipitation).

[0181] 3. Model building and prediction:

[0182] Train a prediction model to predict the distribution and density of Eichhornia crassipes at a future time point t+1 based on current and historical growth data.

[0183] Use historical invasion information to assess the effectiveness of past measures, such as the efficiency of mechanical removal or chemical treatments.

[0184] 4. Real-time simulation and predictive evaluation:

[0185] Based on the data predicted by the model, an assessment is made of the specific impacts that water hyacinth may have on aquatic ecosystems, especially on photosynthesis and oxygen supply of organisms in the waters, if no further measures are taken.

[0186] To assess the potential impacts of model-predicted water hyacinth spread on fish and other aquatic organisms in the Giant Panda National Park that are part of the giant panda's food chain.

[0187] 5. Adjust and optimize control strategies:

[0188] Based on the model output, if the prediction results show that current measures are not sufficient to control the rapid spread of water hyacinth, it is proposed to increase the frequency of mechanical cleaning or use biological control methods, such as introducing natural enemies.

[0189] Plan to implement new management strategies, such as using specific chemical treatments during certain seasons or in areas where water hyacinth grows rapidly.

[0190] That is, the model can help predict the impact of increased water hyacinth coverage on water oxygen exchange and light, and assess the long-term impact on aquatic plants and animals. And by comparing the model prediction results and actual observation data at different time points, it can quantify the effectiveness of current control measures, determine which methods are most effective, and whether strategies need to be adjusted. And by combining the prediction model with real-time monitoring data, managers can implement an early warning system to respond promptly to abnormal spread of water hyacinth and effectively protect the water quality and ecological security of national parks.

[0191] It can be seen that such a systematic assessment and prediction method can help managers understand and respond to the challenges of alien species invasion more accurately and protect the ecological environment of the Giant Panda National Park from the negative impacts of alien species.

[0192] That is, this application uses video analysis to combine the multimodal data of the Giant Panda National Park and historical biological invasion experience to define and predict the growth status of alien species at various stages. Specifically, it involves taking the growth status area division and decision-making as the core, based on multimodal data and historical biological invasion experience characteristics, and ultimately predicting the status and probability of alien species in the Giant Panda National Park, and evaluating whether the defense measures taken are effective.

[0193] That is, this application proposes a method for defining the status of alien species and predicting the future based on multimodal data and based on historical information experience characteristics, which is adapted to the scenario of the Giant Panda National Park; specifically, the method includes: for the multimodal data of the Giant Panda National Park, a unified expression of the growth characteristics of the target alien species is made, and the multimodal target alien species individual growth characteristic representation based on the deep interest network is studied; for the historical alien species of the Giant Panda National Park, historical information experience characteristics are extracted to study the definition of the status of alien species; based on the multimodal data characteristics and historical information experience characteristics, the prediction of the development status and probability of alien species is studied.

[0194] On the other hand, Fig.10 As shown, in order to solve the above technical problems, the embodiment of the present application further provides: an evaluation device for the Giant Panda National Park to resist the invasion of alien species, comprising:

[0195] The first acquisition module is used to acquire real-time video data of the Giant Panda National Park after adopting the target defense measures to defend against the target alien species; based on the real-time video data of the Giant Panda National Park, extract the real-time growth characteristics of the target alien species; based on the real-time growth characteristics of the target alien species, obtain the attention weight vector of the target alien species;

[0196] The second acquisition module is used to construct context features based on the analysis of factors affecting the historical growth of the target alien species; the context features include target alien species information, ecological environment characteristics, climate conditions, biodiversity data and target resistance measures;

[0197] Constructing a target alien species population growth state module, which is used to construct a target alien species multimodal growth feature vector based on the attention weight vector of the target alien species and the context feature; constructing a target alien species population growth state feature based on multiple target alien species multimodal growth feature vectors;

[0198] A semantic extraction module, for obtaining historical alien species invasion information and corresponding Giant Panda National Park measures decisions based on a semantic extraction model; and obtaining historical information empirical feature vectors based on the historical alien species invasion information and the Giant Panda National Park measures decisions;

[0199] An evaluation module is used to predict the growth characteristics of the target alien species at time t+1 based on the growth status characteristics of the target alien species population and the historical information empirical feature vector; based on the growth characteristics of the target alien species at time t+1, the method of the Giant Panda National Park to resist the invasion of alien species is evaluated.

[0200] Based on the same inventive concept as the aforementioned embodiment, this embodiment provides a computer-readable storage medium, on which a computer program is stored, and a processor executes the computer program to implement the aforementioned method.

[0201] In some embodiments, the computer readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or various devices including one or any combination of the above memories. The computer may be various computing devices including intelligent terminals and servers.

[0202] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for evaluating the Giant Panda National Park's resistance to alien species invasion, characterized in that: The following steps are involved: After adopting the targeted defense measures to defend against the targeted alien species, obtaining real-time video data of the Giant Panda National Park; extracting the real-time growth characteristics of the targeted alien species based on the real-time video data of the Giant Panda National Park; Based on the real-time growth characteristics of the target alien species, obtaining an attention weight vector of the target alien species; Based on the analysis of factors influencing the historical growth of the target alien species, context features are constructed; the context features include target alien species information, ecological environment characteristics, climate conditions, biodiversity data and target resistance measures; Based on the attention weight vector of the target alien species and the context feature, construct a multimodal growth feature vector of the target alien species; based on multiple multimodal growth feature vectors of the target alien species, construct a growth state feature of a target alien species population; Based on the semantic extraction model, historical alien species invasion information and corresponding Giant Panda National Park measures and decisions are obtained; based on the historical alien species invasion information and the Giant Panda National Park measures and decisions, historical information experience feature vectors are obtained; Based on the growth status characteristics of the target alien species population and the historical information empirical feature vector, the growth characteristics of the target alien species at time t+1 are predicted; based on the growth characteristics of the target alien species at time t+1, the method of the Giant Panda National Park to resist the invasion of alien species is evaluated.

2. The method for evaluating the Giant Panda National Park's resistance to alien species invasion according to claim 1, characterized in that: The step of obtaining the attention weight vector of the target alien species based on the real-time growth characteristics of the target alien species comprises: Obtain historical growth characteristic sequences of target alien species; Based on the historical growth feature sequence of the target alien species and the real-time growth feature of the target alien species, an attention weight vector of the target alien species is obtained.

3. The method for evaluating the Giant Panda National Park's resistance to alien species invasion according to claim 2, characterized in that: The step of obtaining the attention weight vector of the target alien species based on the historical growth feature sequence of the target alien species and the real-time growth feature of the target alien species comprises: Extracting the attention weights of different defense measures and the current growth feature vector of the target alien species from the historical growth feature sequence of the target alien species; Based on the current growth feature vector and attention weight of the target alien species, an attention weight vector of the target alien species is obtained.

4. The method for evaluating the Giant Panda National Park against alien species invasion according to claim 1, characterized in that: The step of constructing a multimodal growth feature vector of the target alien species based on the attention weight vector of the target alien species and the context feature comprises: The attention weight vector of the target alien species and the context feature are converted into binary code by using one-hot encoding and multi-hot encoding to obtain growth feature vector encoding information of the target alien species; The growth feature vector encoding information of the target alien species and the associated historical growth feature vector are input into the trained deep interest network to construct a multimodal growth feature vector of the target alien species.

5. The method for evaluating the Giant Panda National Park's resistance to alien species invasion according to claim 1, characterized in that: The steps of obtaining historical alien species invasion information and corresponding Giant Panda National Park measures decisions based on the semantic extraction model; and obtaining historical information experience feature vectors based on the historical alien species invasion information and the Giant Panda National Park measures decisions include: Obtaining a BERT semantic extraction model; wherein the BERT semantic extraction model is obtained based on historical data training, and the historical data includes information on the time of biological invasion, the scope of biological invasion, changes in the degree of biological invasion, and types of measures taken by the Giant Panda National Park; Based on the BERT semantic extraction model, key information on historical biological invasions and a set of guiding measures for the Giant Panda National Park were obtained; The key information of the historical biological invasion and the set of guiding measures of the Giant Panda National Park are encoded and mapped into a high-dimensional information matrix, and a preliminary biological invasion status description vector is extracted by embedding; The preliminary biological invasion status description vector is input into the trained EventKG model to realize the temporal relationship modeling between the measures taken and the changes in the biological invasion status, and then the relationship between the measures taken and the changes in the growth status of the alien species population is represented as a historical information experience vector in a continuous vector space.

6. The method for evaluating the Giant Panda National Park's resistance to alien species invasion according to claim 1, characterized in that: The step of predicting the growth characteristics of the target alien species at time t+1 based on the growth state characteristics of the target alien species population and the historical information empirical feature vector comprises: Inputting the growth status characteristics of the target alien species population into the trained stacked transformers network model to extract the historical experience information extraction network, the context information Ct extraction network and the real-time status feature extraction network; The historical experience information extraction network, the context information Ct extraction network and the real-time status feature extraction network are respectively decoded to obtain decoding results; the decoding results are respectively input into the prediction module to predict the growth characteristics of the alien species population at time t+1.

7. The method for evaluating the Giant Panda National Park's resistance to alien species invasion according to claim 6, characterized in that: The step of inputting the decoding results into the prediction module to predict the growth characteristics of the alien species population at time t+1 includes: Based on the decoding result, a plurality of candidate state allocation regions are obtained, and a plurality of regions are obtained; wherein each region represents a growth state of the alien species; Determine the area where the growth status characteristics of the target alien species population are located, calculate the growth status and confidence generated under different optional measures, define an overall loss function based on the empirical characteristics of historical information; and train the parameters in the loss function to obtain a prediction network model; The growth status characteristics of the current target alien species population are input into the prediction network model to obtain the growth status type of the alien species population at time t+1 under different measures; wherein the prediction network model uses stacked transformers as the prediction network architecture.

8. An evaluation device for the Giant Panda National Park to resist the invasion of alien species, characterized in that: include: The first acquisition module is used to acquire real-time video data of the Giant Panda National Park after adopting targeted defense measures to defend against targeted alien species; Based on the real-time video data of the Giant Panda National Park, extracting the real-time growth characteristics of the target alien species; Based on the real-time growth characteristics of the target alien species, obtaining an attention weight vector of the target alien species; The second acquisition module is used to construct context features based on the analysis of factors affecting the historical growth of the target alien species; the context features include target alien species information, ecological environment characteristics, climate conditions, biodiversity data and target resistance measures; Constructing a target alien species population growth state module, which is used to construct a target alien species multimodal growth feature vector based on the attention weight vector of the target alien species and the context feature; constructing a target alien species population growth state feature based on multiple target alien species multimodal growth feature vectors; A semantic extraction module, for obtaining historical alien species invasion information and corresponding Giant Panda National Park measures decisions based on a semantic extraction model; and obtaining historical information empirical feature vectors based on the historical alien species invasion information and the Giant Panda National Park measures decisions; An evaluation module, used to predict the growth characteristics of the target alien species at time t+1 based on the growth state characteristics of the target alien species population and the historical information empirical feature vector; Based on the growth characteristics of the target alien species at time t+1, the methods used by the Giant Panda National Park to resist the invasion of alien species were evaluated.

9. A computer device, characterized in that: The computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the evaluation method for the Giant Panda National Park to resist the invasion of alien species according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the processor executes the computer program to implement the evaluation method for the Giant Panda National Park to resist the invasion of alien species according to any one of claims 1-7.

Images