Living environment adaptability evaluation method based on wild animal behavior pattern

Abstract

The invention discloses a habitat adaptability evaluation method based on a wild animal behavior pattern, and the method comprises the steps: S1, arranging a wild animal monitoring station, and obtaining an animal behavior pattern feature data set of a wild animal; s2, acquiring a habitat environment feature data set of wild animals; s3, constructing an ecological niche behavior-environment hypervolume model, identifying environment dimensions having influences on animal behavior modes, and screening out influence factors; s4, taking the screened influence factors as nodes, and constructing a behavior-environment dynamic complex network; and S5, analyzing an influence path and influence intensity of habitat change on an animal behavior mode through a behavior-environment dynamic complex network, dynamically updating probability distribution of habitat adaptability, outputting an adaptability score, and dividing an adaptation grade area according to the score. According to the invention, the dynamic monitoring of the habitat adaptability of wild animals is realized, and the technical problem that the habitat change influence is difficult to dynamically monitor is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of wild animal adaptability monitoring, and in particular to a habitat adaptability evaluation method based on wild animal behavior patterns. Background Art

[0002] Accurately assessing wildlife adaptability to their habitats is crucial in ecological conservation and wildlife research. The survival and reproduction of wildlife are closely linked to their habitats, and even small changes in their habitats can have profound impacts on their behavior and well-being. With global climate change and the continued expansion of human activities, wildlife habitats face unprecedented challenges, such as habitat loss and fragmentation, and intensified competition for resources. Therefore, a precise and comprehensive evaluation method is needed to dynamically monitor and quantify the degree of adaptation of wildlife to different habitat conditions, enabling the timely implementation of targeted conservation measures.

[0003] Traditional wildlife habitat adaptability assessment techniques rely primarily on field observations and simple data recording. Researchers directly observe animal behavior over extended periods in the wild, recording information such as activity frequency, foraging range, and reproductive behavior. This approach also incorporates descriptions of basic environmental factors, such as temperature and vegetation type, to assess habitat adaptability. This method offers the advantages of intuitiveness, authenticity, and the ability to obtain firsthand data, making it invaluable for preliminary research on specific regions and species.

[0004] However, its shortcomings are also significant. On the one hand, field observations are limited by human, material, and time resources, resulting in small sample sizes that make it difficult to fully reflect the behavior and adaptation of wild animals in a wider range of areas and under different environmental conditions. On the other hand, the data obtained through this method is relatively single-source, lacking systematicity and comprehensiveness, making it impossible to deeply analyze the complex interrelationships between environmental factors and animal behavior patterns. Furthermore, traditional methods are mostly qualitative descriptions, making it difficult to accurately quantify habitat adaptability. This leads to highly subjective assessment results, and differences in evaluation criteria between researchers, making research results less comparable.

[0005] With the advancement of science and technology, existing technologies have made some progress in assessing wildlife habitat adaptability. Some studies have begun to utilize sensor technologies, such as GPS trackers and temperature sensors, to automatically monitor wildlife behavior and environmental parameters. This significantly improves the efficiency and accuracy of data collection compared to traditional methods. Geographic Information Systems (GIS) technology has also been widely used to analyze the spatial characteristics of habitats, providing a visual representation of their distribution and changes. In terms of data analysis, statistical analysis methods and simple models are being used to explore the relationship between environmental factors and animal behavior. For example, correlation analysis can be used to determine the extent to which certain environmental variables influence specific animal behaviors. However, existing technologies still have shortcomings. While sensors and GIS technologies can capture large amounts of data, they lack sufficient capabilities for in-depth data mining and integrated analysis. Currently used models are mostly simplistic and fail to fully capture the complex, nonlinear relationships between the environment and animal behavior, making it difficult to accurately assess animal habitat adaptability under the combined influence of multiple environmental factors. Furthermore, existing technologies lack dynamic update mechanisms, preventing them from promptly reflecting the real-time impact of habitat changes on animal behavior patterns, thus failing to meet the timeliness requirements of ecological conservation efforts.

[0006] Existing technologies have made significant progress in assessing wildlife habitat adaptability. The use of sensors and GIS technology has improved the efficiency of data collection and spatial analysis capabilities, and statistical analysis methods can also reveal the relationship between the environment and animal behavior to a certain extent. However, their limitations cannot be ignored. From a data processing perspective, the ability to process and analyze massive amounts of data needs to be improved, and data integration is not perfect, resulting in an inability to fully realize the value of the data. In terms of model construction, simple models cannot simulate complex ecosystems, and the comprehensive consideration of environmental factors is insufficient, affecting the accuracy of the evaluation results. In terms of timeliness, the lack of a dynamic update mechanism causes the evaluation results to lag behind actual changes in the habitat, failing to provide timely and effective support for ecological protection decisions.

[0007] Therefore, both traditional and existing technologies are difficult to fully meet the habitat adaptability evaluation requirements for dynamic wildlife updates. Summary of the Invention

[0008] Based on the above, this application discloses a habitat adaptability evaluation method based on wild animal behavior patterns; including:

[0009] S1. Set up wildlife monitoring stations to obtain behavioral data of wildlife in their natural habitats, including visual behavioral images, thermal infrared activity information, and acoustic communication signal data. Extract key features from each type of data to form a standardized animal behavior pattern feature dataset.

[0010] S2. Based on the time of animal appearance observed at wildlife monitoring stations, collect environmental parameters at the corresponding time to form a standardized habitat environmental characteristic dataset;

[0011] S3. Mapping animal behavior pattern characteristic data and habitat environment characteristic data into high-dimensional space, constructing an ecological niche behavior-environment hypervolume model, identifying environmental dimensions that affect animal behavior patterns, and screening out influencing factors;

[0012] S4. Use the selected influencing factors as nodes and the correlation strength between the influencing factors and animal behavior patterns as edge weights to construct a behavior-environment dynamic complex network;

[0013] S5. Analyze the impact path and intensity of habitat changes on animal behavior patterns through a behavior-environment dynamic complex network. Use Bayesian reasoning to dynamically update the probability distribution of habitat adaptability, quantify the degree of adaptation of wild animals in their current habitat, output adaptability scores, and divide adaptation level areas according to the scores.

[0014] Preferably, S1 extracts key features from the visual behavior images, thermal infrared activity information and sound communication signal data of wild animals. The key features include: animal type, movement posture, and activity trajectory characteristics in the visual behavior images; animal body temperature regulation status, metabolic level, and activity pattern characteristics in the animal thermal infrared activity information; time domain characteristics and frequency domain characteristics of animal sounds in the sound communication signals. The key features are preprocessed to form a standardized animal behavior pattern feature data set.

[0015] Preferably, the environmental parameters in S2 include temperature, humidity, light intensity, soil type, soil pH, vegetation type, vegetation coverage, slope, slope direction, airflow, wind speed, air quality, water environment quality and activity time data, which are formed into a habitat environment characteristic data set after preprocessing.

[0016] Preferably, the preprocessing is noise reduction, normalization and fusion mechanism; noise reduction is performed by acquiring feature data, and the formula is: ,in is the original eigenvalue, is the neighborhood eigenvalue, is the number of neighbors, and the feature data is mapped to the interval [0, 1] by normalization. The processed feature data is fused, and the formula is: ,in is the fusion function, After normalization, feature data, is the total number of feature data, For the The weight coefficient of each feature data.

[0017] Preferably, the animal behavior pattern feature data set is obtained in S3 and habitat environmental characteristics dataset Through the mapping function Get a dimension element , the dimension elements The niche behavior-environment hypervolume model is formed by combining the time and space sequences. The formula is: ,in , and are the corresponding weight coefficients, and The corresponding and feature data, and are the total number of corresponding feature data respectively.

[0018] Preferably, in S3, the environmental dimensions that have an impact on animal behavior patterns are identified through the niche behavior-environment hypervolume model, and the influencing factors are screened out, specifically:

[0019] Each dimension in the niche behavior-environment hypervolume model corresponds to an environmental variable, and the impact index is calculated ,when Greater than a pre-set threshold When the corresponding environmental variables have an impact on the animal behavior pattern, the influencing factors are screened out; the impact index The formula is: ,in, is the number of environmental samples, For the The change in niche behavior-environmental hypervolume in each environmental sample, To change the The first environmental dimension variable The variation of niche behavior-environmental hypervolume in environmental samples.

[0020] Preferably, in said S4, a behavior-environment dynamic complex network is constructed to obtain the degree of influence of influencing factors on animal behavior patterns, specifically:

[0021] The impact factor As a network node, according to the impact factor Add edges to the interaction relationship between them and determine the weight of the edges , the formula is: ,in Impact Factor and The node edge weights, and is the adjacent node influence factor, Impact Factor and After merging, the mutual information with the animal behavior pattern, and Impact Factor and The mutual information with animal behavior patterns is: ,in Represents a dataset of animal behavior pattern features, Impact Factor The value of and animal behavior pattern eigenvalues The joint probability of and They are and The marginal probability of the calculated mutual information value , through the impact factor The weight of the edge between the animal behavior pattern node After constructing the behavior-environment dynamic complex network, the centrality index of the impact factor is calculated using degree centrality , the formula is: , Representation node and The weight of the edge between is the total number of edges, and the degree of influence of the influencing factors on the animal behavior pattern is obtained.

[0022] Preferably, the influence path of habitat change on animal behavior pattern is analyzed in S5 through behavior-environment dynamic complex network, specifically: obtaining the influencing factors affected by habitat change, taking the influencing factors as the starting nodes, searching all paths from the starting nodes to the nodes representing animal behavior patterns, and for the edges on the paths, weights Represents the correlation strength between the influencing factor and the animal behavior pattern, and calculates the influence strength of each path on the animal behavior pattern , the formula is: ,in is the number of path edges, For the The weight of the edge, For the path The importance weight of each node is calculated through the centrality index Determine by comparing the impact strength of different paths , determine the impact path of habitat changes on animal behavior patterns, and obtain the overall impact intensity through the obtained impact path.

[0023] Preferably, the overall impact intensity is obtained through the impact path, specifically:

[0024] Obtain the impact pathways of habitat changes on animal behavior patterns and the synergistic effects between these pathways, and calculate the overall impact intensity , the formula is: ,in is the total number of paths from the starting node to the animal behavior pattern node, is the synergistic effect coefficient, For path and path The correlation coefficient between the impact intensities of the two groups is used to obtain the overall impact intensity.

[0025] Preferably, in S5, Bayesian reasoning is used to dynamically update the probability distribution of habitat adaptability, quantify the degree of adaptation of wild animals in the current habitat, output an adaptability score, and divide the adaptation level areas according to the score, specifically:

[0026] Obtain the adaptation status of wild animals in their current habitats, update the probability distribution of habitat adaptability through the impact path and overall impact intensity of animal behavior patterns, and use Bayesian inference to update the posterior probability of habitat adaptability , the formula is: ,in In the state of adaptation The observed impact value The likelihood probability, In the state of adaptation The observed impact value Likelihood probability, influence value The impact path and overall impact intensity of habitat change on animal behavior patterns, and To adapt to the state and The prior probability of To adapt to the total number of states, and They are the corresponding adjustment factors, and the posterior probability obtained by Calculating fitness scores , the formula is: ,in For each adaptation state The corresponding score value is based on the adaptability score Divide the area into adaptive levels.

[0027] Compared with the prior art, the technical solution of this application has the following technical effects:

[0028] Through multi-source data fusion and complex model construction, this invention achieves a comprehensive and accurate evaluation of the adaptability of wild animals to their habitats. It not only collects visual behavioral images, thermal infrared activity information and sound communication signal data of wild animals, but also synchronously collects rich environmental parameters at the corresponding moment, covering temperature, humidity, light intensity and other aspects, and obtains all-round data from the animal's own behavioral characteristics to the surrounding environmental factors.

[0029] This aspect can capture the impact of habitat changes on influencing factors in real time through the behavior-environment dynamic complex network. Once the habitat changes, the affected influencing factors are used as the starting node to quickly search for all paths representing the animal behavior pattern nodes, calculate the impact intensity of each path and the overall impact intensity, and input the real-time impact information into the Bayesian inference model as new evidence to dynamically update the probability distribution of habitat adaptability. This dynamic update mechanism enables the evaluation results to keep up with the actual changes in the habitat and promptly reflect the degree of adaptation of wild animals at different times.

[0030] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application so that it can be implemented in accordance with the contents of the specification, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following is a detailed description of the preferred embodiment of the present application in conjunction with the accompanying drawings.

[0031] Based on the detailed description of the specific embodiments of the present application in conjunction with the accompanying drawings below, those skilled in the art will become more aware of the above and other objects, advantages and features of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without inventive work. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0033] Figure 1 This is a flow chart of the habitat adaptability evaluation method based on wild animal behavior patterns of the present invention;

[0034] Figure 2 To select a topographic map for evaluating the habitat adaptability of sika deer;

[0035] Figure 3 Map of monitoring sites on the terrain for habitat adaptability assessment of sika deer;

[0036] Figure 4 It is a structural flow chart of the impact path and intensity of sika deer;

[0037] Figure 5 This is the habitat adaptability evaluation area map for sika deer. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. In the following description, specific details such as specific configurations and components are provided only to help fully understand the embodiments of the present application. Therefore, it should be clear to those skilled in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. In addition, for clarity and brevity, the description of known functions and structures has been omitted in the embodiments.

[0039] It should be understood that references throughout this specification to "one embodiment" or "this embodiment" mean that a particular feature, structure, or characteristic associated with the embodiment is included in at least one embodiment of the present application. Therefore, the appearance of "one embodiment" or "this embodiment" throughout this specification does not necessarily refer to the same embodiment. Furthermore, these particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0040] In addition, the present application may repeat reference numerals and / or letters in different examples. This repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and / or settings discussed.

[0041] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist at the same time. The term " / and" in this article describes another type of association object relationship, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " in this article generally indicates that the previous and subsequent associated objects are in an "or" relationship.

[0042] The term "at least one" in this article is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, at least one of A and B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0043] It should also be noted that, in this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include," "comprises," or any other variations thereof are intended to cover non-exclusive inclusion.

[0044] Example 1

[0045] This embodiment mainly describes a habitat adaptability evaluation method based on wild animal behavior patterns, such as Figure 1 Shown, including:

[0046] S1. Set up wildlife monitoring stations to obtain behavioral data of wildlife in their natural habitats, including visual behavioral images, thermal infrared activity information, and acoustic communication signal data. Extract key features from each type of data to form a standardized animal behavior pattern feature dataset.

[0047] S2. Based on the time of animal appearance observed at wildlife monitoring stations, collect environmental parameters at the corresponding time to form a standardized habitat environmental characteristic dataset;

[0048] S3. Mapping animal behavior pattern characteristic data and habitat environment characteristic data into high-dimensional space, constructing an ecological niche behavior-environment hypervolume model, identifying environmental dimensions that affect animal behavior patterns, and screening out influencing factors;

[0049] S4. Use the selected influencing factors as nodes and the correlation strength between the influencing factors and animal behavior patterns as edge weights to construct a behavior-environment dynamic complex network;

[0050] S5. Analyze the impact path and intensity of habitat changes on animal behavior patterns through a behavior-environment dynamic complex network. Use Bayesian reasoning to dynamically update the probability distribution of habitat adaptability, quantify the degree of adaptation of wild animals in their current habitat, output adaptability scores, and divide adaptation level areas according to the scores.

[0051] Furthermore, S1 extracts key features from the visual behavior images, thermal infrared activity information and sound communication signal data of wild animals. The key features include: animal type, movement posture, and activity trajectory characteristics in visual behavior images; animal body temperature regulation status, metabolic level, and activity pattern characteristics in animal thermal infrared activity information; time domain characteristics and frequency domain characteristics of animal sounds in sound communication signals. The key features are preprocessed to form a standardized animal behavior pattern feature data set.

[0052] Furthermore, the environmental parameters in S2 include temperature, humidity, light intensity, soil type, soil pH, vegetation type, vegetation coverage, slope, slope direction, airflow, wind speed, air quality, water environment quality and activity time data, which are pre-processed to form a habitat environment characteristic data set.

[0053] Furthermore, the preprocessing is noise reduction, normalization and fusion mechanism; noise reduction is performed by acquiring feature data, and the formula is: ,in is the original eigenvalue, is the neighborhood eigenvalue, is the number of neighbors, and the feature data is mapped to the interval [0, 1] by normalization. The processed feature data is fused, and the formula is: ,in is the fusion function, After normalization, feature data, is the total number of feature data, For the The weight coefficient of each feature data.

[0054] Furthermore, the animal behavior pattern feature dataset is obtained in S3 and habitat environmental characteristics dataset Through the mapping function Get a dimension element , the dimension elements The niche behavior-environment hypervolume model is formed by combining the time and space sequences. The formula is: ,in , and are the corresponding weight coefficients, and The corresponding and feature data, and are the total number of corresponding feature data respectively.

[0055] Furthermore, in S3, the environmental dimensions that influence animal behavior patterns are identified through the niche behavior-environment hypervolume model, and the influencing factors are screened out, specifically:

[0056] Each dimension in the niche behavior-environment hypervolume model corresponds to an environmental variable, and the impact index is calculated ,when Greater than a pre-set threshold When the corresponding environmental variables have an impact on the animal behavior pattern, the influencing factors are screened out; the impact index The formula is: ,in, is the number of environmental samples, For the The change in niche behavior-environmental hypervolume in each environmental sample, To change the The first environmental dimension variable The variation of niche behavior-environmental hypervolume in environmental samples.

[0057] Furthermore, in S4, a behavior-environment dynamic complex network is constructed to obtain the degree of influence of influencing factors on animal behavior patterns, specifically:

[0058] The impact factor As a network node, according to the impact factor Add edges to the interaction relationship between them and determine the weight of the edges , the formula is: ,in Impact Factor and The node edge weights, and is the adjacent node influence factor, Impact Factor and After merging, the mutual information with the animal behavior pattern, and Impact Factor and The mutual information with animal behavior patterns is: ,in Represents a dataset of animal behavior pattern features, Impact Factor The value of and animal behavior pattern eigenvalues The joint probability of and They are and The marginal probability of the calculated mutual information value , through the impact factor The weight of the edge between the animal behavior pattern node After constructing the behavior-environment dynamic complex network, the centrality index of the impact factor is calculated using degree centrality , the formula is: , Representation node and The weight of the edge between is the total number of edges, and the degree of influence of the influencing factors on the animal behavior pattern is obtained.

[0059] Furthermore, the influence path of habitat change on animal behavior pattern is analyzed in S5 through behavior-environment dynamic complex network, specifically: the influencing factors affected by habitat change are obtained, the influencing factors are used as the starting nodes, and all paths from the starting nodes to the nodes representing animal behavior patterns are searched. For the edges on the paths, the weights Represents the correlation strength between the influencing factor and the animal behavior pattern, and calculates the influence strength of each path on the animal behavior pattern , the formula is: ,in is the number of path edges, For the The weight of the edge, For the path The importance weight of each node is calculated through the centrality index Determine by comparing the impact strength of different paths , determine the impact path of habitat changes on animal behavior patterns, and obtain the overall impact intensity through the obtained impact path.

[0060] Furthermore, the overall impact intensity is obtained through the impact path, specifically:

[0061] Obtain the impact pathways of habitat changes on animal behavior patterns and the synergistic effects between these pathways, and calculate the overall impact intensity , the formula is: ,in is the total number of paths from the starting node to the animal behavior pattern node, is the synergistic effect coefficient, For path and path The correlation coefficient between the impact intensities of the two groups is used to obtain the overall impact intensity.

[0062] Furthermore, in S5, Bayesian reasoning is used to dynamically update the probability distribution of habitat adaptability, quantify the degree of adaptation of wild animals in the current habitat, output an adaptability score, and divide the adaptation level areas according to the score, specifically:

[0063] Obtain the adaptation status of wild animals in their current habitats, update the probability distribution of habitat adaptability through the impact path and overall impact intensity of animal behavior patterns, and use Bayesian inference to update the posterior probability of habitat adaptability , the formula is: ,in In the state of adaptation The observed impact value The likelihood probability, In the state of adaptation The observed impact value Likelihood probability, influence value The impact path and overall impact intensity of habitat change on animal behavior patterns, and To adapt to the state and The prior probability of To adapt to the total number of states, and They are the corresponding adjustment factors, and the posterior probability obtained by Calculating fitness scores , the formula is: ,in For each adaptation state The corresponding score value is based on the adaptability score Divide the area into adaptive levels.

[0064] This implementation describes in detail the construction of a niche-behavior-environment hypervolume model, which maps characteristic data of animal behavior patterns and habitat environmental characteristics into a high-dimensional space. This model meticulously depicts the complex interrelationships between animals and their environments. By calculating impact indices, it identifies the environmental dimensions and factors that influence animal behavior patterns. These influencing factors are then used to construct a dynamic behavior-environment complex network, enabling in-depth analysis of the pathways and intensity of the impact of habitat changes on animal behavior patterns. Finally, Bayesian inference is used to dynamically update the probability distribution of habitat adaptability, quantify the degree of adaptability, output a score, and delineate adaptation levels.

[0065] Based on Example 2, this example describes in detail the specific implementation methods of the present application, specifically:

[0066] Sika deer, a rare even-toed ungulate species of the Cervidae family, occupy a crucial position in ecosystems. Research on their habitat adaptability is crucial for species conservation and ecosystem management. This study, conducted in a specific sika deer habitat in East my country, comprehensively assessed their habitat adaptability, aiming to provide a scientific basis for their conservation.

[0067] like Figure 2As shown in the figure, this study focuses on an area in East my country, covering an area of approximately 452.16 square kilometers. The terrain is undulating, including mountains, hills and mountain basins. The climate is subtropical monsoon with distinct four seasons and abundant annual precipitation, providing a suitable living environment for many organisms. The forest vegetation types in the area are diverse, mainly evergreen broad-leaved forests, deciduous broad-leaved forests and mixed coniferous and broad-leaved forests. The rich plant resources provide a potential food source for sika deer, and there are many streams running through the area with sufficient water resources, making it one of the important habitats of sika deer. In the early stage of the study, the research team went deep into the area and, by reviewing the sika deer monitoring records over the past 10 years, analyzing the historical data of infrared cameras and visiting local forest rangers and residents, preliminarily determined that the area where sika deer frequently move is concentrated in the mountainous forest area at an altitude of 300-800 meters. The area has lush vegetation, a wide variety of plant species, and is close to water sources, which can meet the daily drinking water needs of sika deer.

[0068] Based on the areas where sika deer frequently move, and taking into account local terrain, vegetation distribution, water source location and other environmental factors, multiple monitoring stations were scientifically deployed in the area. Figure 3 As shown, monitoring stations should be set up near valleys, river junctions, and other locations where sika deer frequently occur, to obtain more comprehensive data. At the same time, monitoring stations should be distributed across different altitudes, slope aspects, and vegetation types to cover the various microhabitats inhabited by sika deer.

[0069] High-definition infrared cameras, thermal infrared sensors, and professional microphone arrays are installed at each monitoring station. The high-definition infrared cameras use thermal sensing technology to automatically trigger filming when sika deer pass by, recording their daily behaviors such as foraging, resting, courting, and fighting. The thermal infrared sensors monitor the deer's body temperature regulation and activity patterns in real time. Based on changes in thermal infrared data, the intensity of the deer's activity and whether they are in special periods such as breeding can be determined. The microphone array is used to capture the deer's calls and the sounds produced when they interact with other animals, analyzing their vocal communication patterns.

[0070] Environmental parameters are collected simultaneously with sika deer sightings. High-precision temperature and humidity sensors measure temperature and humidity, and light sensors capture light intensity data. Soil samples are collected monthly and analyzed in the laboratory for soil type and pH. Field surveys combined with satellite remote sensing imagery determine vegetation type and coverage, and precisely measure slope and aspect. A small weather station monitors airflow and wind speed, while air quality monitors measure air quality and water quality testing equipment analyzes water quality. Furthermore, the time of each sika deer sighting is accurately recorded.

[0071] During the three-month data collection process, the temperature (°C), humidity (%) and light intensity (lx) data of monitoring stations Y01 and Y05 are selected for display, as shown in the following table:

[0072] Monitoring site number month Temperature (℃) humidity(%) Light intensity (lx) Y01 January 10.5 72.6 150.3 Y01 February 18.2 68.4 350.7 Y01 March 25.5 80.1 500.5 Y05 January 9.8 75.0 130.8 Y05 February 17.5 70.3 320.4 Y05 March 24.8 82.0 480.6

[0073] Obtain the soil type, soil pH, vegetation type, and vegetation coverage of monitoring stations Y01 and Y05, as shown in the following table:

[0074] Monitoring site number month Soil type Soil pH Vegetation type Vegetation coverage (%) Y01 January red soil 5.8 evergreen broad-leaved forest 70.2 Y01 February red soil 5.9 evergreen broad-leaved forest 75.0 Y01 March red soil 6.0 evergreen broad-leaved forest 78.5 Y05 January red soil 5.6 Mixed coniferous and broad-leaved forest 65.3 Y05 February red soil 5.7 Mixed coniferous and broad-leaved forest 70.0 Y05 March red soil 5.8 Mixed coniferous and broad-leaved forest 73.5

[0075] Obtain the slope, slope direction, airflow, wind speed, air quality index, water environment quality index, and sika deer activity time of monitoring stations numbered Y01 and Y05, as shown in the following table:

[0076] Monitoring site number month Slope (°) Slope Airflow (m / s) Wind speed (m / s) Air Quality Index Water Environment Quality Index Activity time (hours) Y01 January 20.4 southeast 0.62 1.32 58.0 82.2 3.5 Y01 February 21.7 southeast 0.86 1.56 53.0 85.3 4.7 Y01 March 21.9 southeast 1.03 1.84 48.0 88.1 4.5 Y05 January 23.1 southeast 0.55 1.24 60.0 80.6 3.8 Y05 February 23.6 southeast 0.76 1.43 55.0 83.3 3.5 Y05 March 24.8 southeast 0.94 1.68 50.0 86.4 4.1

[0077] The collected raw data are strictly preprocessed, and the preprocessed animal behavior pattern characteristic data are integrated with the habitat environment characteristic data to construct the niche behavior-environment hypervolume model, such as Figure 4 As shown in the figure, during the construction process, an in-depth analysis of the relationship between various environmental factors and sika deer behavior was conducted. Correlation analysis revealed that vegetation cover is highly correlated with sika deer foraging behavior and is therefore given a higher weight in the model. However, some secondary air quality indicators have relatively little impact on sika deer survival and are therefore given lower weights.

[0078] Using a well-established niche-behavior-environment hypervolume model, a detailed analysis was conducted on the impact of various environmental dimensions on sika deer behavior patterns. After multiple rounds of calculations and comparisons, environmental factors with significant impacts on sika deer behavior were identified. The results showed that temperature, humidity, vegetation cover, food abundance (associated with vegetation type and soil nutrients), and distance to water sources significantly influenced sika deer's activity patterns, foraging behavior, and reproductive behavior. When temperatures dropped below 5°C or rose above 30°C, sika deer's activity time decreased significantly. A decrease in vegetation cover of more than 20% significantly increased their foraging difficulty and expanded their range by approximately 30%.

[0079] The screened influencing factors are used as network nodes, and the edge weights are determined according to the correlation strength between the influencing factors and the behavior patterns of sika deer. A behavior-environment dynamic complex network is constructed. The complex network analysis technology is used to comprehensively consider the interactions between the influencing factors. Precipitation changes will affect soil moisture and vegetation growth, thereby affecting the distribution and quantity of food, and ultimately affecting the foraging behavior of sika deer. This indirect influence relationship is reflected in the network through corresponding edges and weights. By calculating the centrality index of the network nodes, the degree of influence of each influencing factor on the behavior pattern of sika deer is evaluated, and the key influencing factors are determined.

[0080] Using a complex behavior-environment dynamic network, we delve deeper into the pathways and magnitudes of the impact of habitat change on sika deer behavior patterns. When one or more environmental factors change, we use network analysis algorithms to track the propagation paths of these changes within the network and calculate the magnitude of each path's impact on deer behavior. For example, when reduced precipitation in a region limits vegetation growth, thereby reducing food availability, network analysis reveals that this change will have a significant impact along the pathway of precipitation-vegetation growth-food availability-sika deer foraging behavior. By comprehensively analyzing all impact pathways, we can determine the overall magnitude of the impact of habitat change on deer behavior.

[0081] Using the Bayesian inference method, combined with the impact path and intensity information obtained from the previous analysis, the adaptability probability distribution of sika deer in the current habitat is dynamically updated. The prior probability is determined based on the historical data and expert experience of sika deer in different adaptation states. The newly acquired habitat change and behavior pattern data are used as evidence to continuously adjust the posterior probability. After multiple iterative calculations, the adaptability score of sika deer in the current habitat is obtained, and the adaptability score of sika deer in the monitoring area in a certain period of time is obtained; based on the obtained adaptability score, the sika deer habitat is divided into four adaptation level areas: green represents high adaptation areas. In these areas, the temperature and humidity are suitable, the vegetation coverage is high, the food resources are abundant, the water source is sufficient and close, the survival needs of sika deer can be fully met, the sika deer are active, the behavior is normal, and the reproduction success rate is high; orange is the adaptation area. The environmental conditions in the adaptation area can basically meet the survival of sika deer, and the vegetation coverage in some areas is high. The temperature and humidity fluctuate in certain seasons, and the water source is a little far away, but the sika deer can still adapt well. The activity and reproduction of the sika deer are relatively stable, but will be affected to a certain extent by environmental changes; orange represents the marginal area, where some large carnivores exist and the water source is relatively close. The sika deer have a smaller activity range, shorter activity time, and less reproduction in the marginal area; blue represents the danger zone. The environmental conditions in the danger zone are characterized by habitat fragmentation, extreme food shortage, and the coexistence of multiple large carnivores. The survival of the sika deer faces a great threat, and their numbers are scarce, or even no sika deer are active anymore.

[0082] By analyzing and counting the adaptability scores of the entire study area, a regional distribution map of adaptability levels was drawn, such as Figure 4 As shown in the figure, the adaptation status of sika deer in different areas is intuitively displayed. It can be seen from the figure that the high adaptation area and the adaptation area are mainly concentrated in the core area with gentle slopes and a single river; while the marginal area and the danger area are mainly distributed in areas with steep slopes and many rivers.

[0083] This example describes in detail the systematic data collection, analysis, and model construction of a sika deer habitat in North my country, comprehensively assessing the habitat adaptability of sika deer in that area. By constructing a niche behavior-environment hypervolume model, screening influencing factors, and constructing a behavior-environment dynamic complex network, the impact path and intensity of habitat changes on sika deer behavior patterns were deeply analyzed, and the degree of sika deer adaptation in their current habitat was quantified using Bayesian inference methods. The adaptation level areas divided according to the adaptability scores clearly present the adaptation status of sika deer in different areas, providing an important basis for formulating scientific and reasonable conservation strategies.

[0084] The above are only preferred embodiments of the present invention, which do not limit the scope of protection of the present invention. For those skilled in the art, the present invention can be modified and varied in various ways. Any changes, modifications, replacements, integrations and parameter changes to these embodiments through conventional substitutions or that can achieve the same functions without departing from the principles and spirit of the present invention fall within the scope of protection of the present invention.

Claims

1. A habitat adaptability evaluation method based on wild animal behavior patterns, characterized by: include: S1. Set up wildlife monitoring stations to obtain behavioral data of wildlife in their natural habitats, including visual behavioral images, thermal infrared activity information, and acoustic communication signal data. Extract key features from each type of data to form a standardized animal behavior pattern feature dataset. S2. Based on the time of animal appearance observed at wildlife monitoring stations, collect environmental parameters at the corresponding time to form a standardized habitat environmental characteristic dataset; S3. Mapping animal behavior pattern characteristic data and habitat environment characteristic data into high-dimensional space, constructing an ecological niche behavior-environment hypervolume model, identifying environmental dimensions that affect animal behavior patterns, and screening out influencing factors; S4. Use the selected influencing factors as nodes and the correlation strength between the influencing factors and animal behavior patterns as edge weights to construct a behavior-environment dynamic complex network; S5. Analyze the impact path and intensity of habitat changes on animal behavior patterns through a behavior-environment dynamic complex network. Use Bayesian reasoning to dynamically update the probability distribution of habitat adaptability, quantify the degree of adaptation of wild animals in their current habitat, output adaptability scores, and divide adaptation level areas according to the scores.

2. The habitat adaptability evaluation method based on wild animal behavior patterns according to claim 1, characterized in that: The S1 extracts key features from the visual behavior images, thermal infrared activity information and sound communication signal data of wild animals. The key features include: animal type, movement posture, and activity trajectory characteristics in the visual behavior images; animal body temperature regulation status, metabolic level, and activity pattern characteristics in the animal thermal infrared activity information; and time domain characteristics and frequency domain characteristics of animal sounds in the sound communication signals. The key features are pre-processed to form a standardized animal behavior pattern feature data set.

3. The habitat adaptability evaluation method based on wild animal behavior patterns according to claim 2 is characterized in that: The environmental parameters in S2 include temperature, humidity, light intensity, soil type, soil pH, vegetation type, vegetation coverage, slope, slope direction, airflow, wind speed, air quality, water environment quality and activity time data, which are preprocessed to form a habitat environment characteristic data set.

4. The habitat adaptability evaluation method based on wild animal behavior patterns according to claim 2 or 3, characterized in that: The preprocessing includes noise reduction, normalization and fusion mechanisms; noise reduction is performed by acquiring feature data, and the formula is: ,in is the original eigenvalue, is the neighborhood eigenvalue, is the number of neighbors, and the feature data is mapped to the interval [0, 1] by normalization. The processed feature data is fused, and the formula is: ,in is the fusion function, After normalization, feature data, is the total number of feature data, For the The weight coefficient of each feature data.

5. The habitat adaptability evaluation method based on wild animal behavior patterns according to claim 1, characterized in that: The animal behavior pattern feature dataset is obtained in S3 and habitat environmental characteristics dataset Through the mapping function Get a dimension element , the dimension elements The niche behavior-environment hypervolume model is formed by combining the time and space sequences. The formula is: ,in , and are the corresponding weight coefficients, and The corresponding and feature data, and are the total number of corresponding feature data respectively.

6. The habitat adaptability evaluation method based on wild animal behavior patterns according to claim 1 or 5, characterized in that: In S3, the niche behavior-environment hypervolume model is used to identify environmental dimensions that influence animal behavior patterns and screen out influencing factors, specifically: Each dimension in the niche behavior-environment hypervolume model corresponds to an environmental variable, and the impact index is calculated ,when Greater than a pre-set threshold When the corresponding environmental variables have an impact on the animal behavior pattern, the influencing factors are screened out; the impact index The formula is: ,in, is the number of environmental samples, For the The change in niche behavior-environmental hypervolume in each environmental sample, To change the The first environmental dimension variable The variation of niche behavior-environmental hypervolume in environmental samples.

7. The habitat adaptability evaluation method based on wild animal behavior patterns according to claim 1, characterized in that: In S4, a behavior-environment dynamic complex network is constructed to obtain the degree of influence of influencing factors on animal behavior patterns, specifically: The impact factor As a network node, according to the impact factor Add edges to the interaction relationship between them and determine the weight of the edges , the formula is: ,in Impact Factor and The node edge weights, and is the adjacent node influence factor, Impact Factor and After merging, the mutual information with the animal behavior pattern, and Impact Factor and The mutual information with animal behavior patterns is: ,in Represents a dataset of animal behavior pattern features, Impact Factor The value of and animal behavior pattern eigenvalues The joint probability of and They are and The marginal probability of the calculated mutual information value , through the impact factor The weight of the edge between the animal behavior pattern node After constructing the behavior-environment dynamic complex network, the centrality index of the impact factor is calculated using degree centrality , the formula is: , Representation node and The weight of the edge between is the total number of edges, and the degree of influence of the influencing factors on the animal behavior pattern is obtained.

8. The method for evaluating habitat adaptability based on wild animal behavior patterns according to claim 1 or 7, characterized in that: In the S5, the influence path of habitat change on animal behavior pattern is analyzed through the behavior-environment dynamic complex network. Specifically, the influencing factors affected by habitat change are obtained, and the influencing factors are used as the starting nodes to search for all paths from the starting nodes to the nodes representing animal behavior patterns. For the edges on the paths, the weights Represents the correlation strength between the influencing factor and the animal behavior pattern, and calculates the influence strength of each path on the animal behavior pattern , the formula is: ,in is the number of path edges, For the The weight of the edge, For the path The importance weight of each node is calculated through the centrality index Determine by comparing the impact strength of different paths , determine the impact path of habitat changes on animal behavior patterns, and obtain the overall impact intensity through the obtained impact path.

9. The method for evaluating habitat adaptability based on wild animal behavior patterns according to claim 8, characterized in that: The overall impact intensity is obtained through the impact path, specifically: Obtain the impact pathways of habitat changes on animal behavior patterns and the synergistic effects between these pathways, and calculate the overall impact intensity , the formula is: ,in is the total number of paths from the starting node to the animal behavior pattern node, is the synergistic effect coefficient, For path and path The correlation coefficient between the impact intensities of the two groups is used to obtain the overall impact intensity.

10. The habitat adaptability evaluation method based on wild animal behavior patterns according to claim 1 or 9, characterized in that: In S5, Bayesian reasoning is used to dynamically update the probability distribution of habitat adaptability, quantify the degree of adaptation of wild animals in the current habitat, output the adaptability score, and divide the adaptation level areas according to the score, specifically: Obtain the adaptation status of wild animals in their current habitats, update the probability distribution of habitat adaptability through the impact path and overall impact intensity of animal behavior patterns, and use Bayesian inference to update the posterior probability of habitat adaptability , the formula is: ,in In the state of adaptation The observed impact value The likelihood probability, In the state of adaptation The observed impact value Likelihood probability, influence value The impact path and overall impact intensity of habitat change on animal behavior patterns, and To adapt to the state and The prior probability of To adapt to the total number of states, and They are the corresponding adjustment factors, and the posterior probability obtained by Calculating fitness scores , the formula is: ,in For each adaptation state The corresponding score value is based on the adaptability score Divide the area into adaptive levels.

Images