Artificial Intelligence-Based Method and System for Monitoring the Behavior of Wild Giant Pandas

Abstract

The present invention provides an artificial-intelligence-based method and system for monitoring the behaviors of wild giant pandas, which relates to the technical field of animal behavior monitoring and includes: obtaining the original data set of the target giant panda; performing feature extraction and processing according to video information and environmental information, by extracting the action, body posture and micro-expression information of the target giant panda, and analyzing the feature change trends in different time periods and different scenarios in combination with the environmental conditions to obtain a feature set; performing numerical processing according to the feature set to obtain a numerical representation of a virtual individual; performing modeling processing according to the numerical representation of the virtual individual to construct a locomotor function model of the giant panda; performing behavioral trajectory deduction according to the locomotor function model of the giant panda and geographical location information to obtain a behavioral trajectory deduction result; and generating a behavioral monitoring report of the target giant panda based on the behavioral trajectory deduction result. The present invention can analyze the behavior patterns of giant pandas more comprehensively and accurately, and improve the monitoring efficiency and protection effect.

Description

Technical Field

[0001] The present invention relates to the technical field of animal behavior monitoring. Specifically, it relates to a method and system for monitoring wild giant panda behavior based on artificial intelligence. Background Art

[0002] Currently, with the increasing demands for wildlife protection and ecological research, the technology for monitoring wild giant panda behavior has gradually become one of the key technologies in biological research, ecological protection, and behavioral analysis. The existing methods for monitoring giant panda behavior mainly rely on manual observation and video surveillance, where observers record the animal's activity trajectories and action characteristics. However, these methods have many problems: First, manual observation is inefficient and easily affected by the subjective factors of observers; Second, although video surveillance can provide data, its processing relies on manual analysis and often fails to effectively capture and predict the dynamic behavior patterns of giant pandas in real time. In addition, the existing technology has deficiencies in processing data across monitoring points. Especially under the influence of time asynchrony and environmental interference, the data collected from multiple monitoring points lacks effective synchronization, resulting in inaccurate and inconsistent analysis results. These existing technologies cannot comprehensively and systematically analyze complex factors such as the spatio-temporal changes and emotional fluctuations of giant panda behavior, so there are limitations in behavior prediction and dynamic monitoring. The existing technology has not effectively combined the movement, health, and emotional states of animals for unified modeling and mostly relies on manual analysis, lacking automated and accurate behavior prediction capabilities, resulting in the inability to accurately reflect the dynamic behavior and health status of giant pandas in real time.

[0003] Based on the above disadvantages of the existing technology, there is an urgent need for a method and system for monitoring wild giant panda behavior based on artificial intelligence. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for monitoring wild giant panda behavior based on artificial intelligence to improve the above problems. To achieve the above purpose, the technical solutions adopted by the present invention are as follows:

[0005] In a first aspect, the present application provides a method for monitoring wild giant panda behavior based on artificial intelligence, including:

[0006] Obtain the original data set of the target giant panda, where the original data set is the video information, geographical location information of the same wild giant panda collected from multiple monitoring locations, and the environmental information of the monitoring locations;

[0007] Perform feature extraction processing based on the video information and the environmental information. By extracting the action, body posture, and micro-expression information of the target giant panda, and combining the environmental conditions to analyze the feature change trends in different time periods and different scenarios, a feature set is obtained;

[0008] Perform numerical processing according to the described feature set, and obtain the numerical representation of the virtual individual by quantifying the action features, body posture features, and micro-expression features into the activity intensity index, body posture balance index, and emotional fluctuation index respectively;

[0009] Perform modeling processing according to the numerical representation of the virtual individual, and construct the giant panda motor function model by simulating the skeletal movement posture and combining the influence of health and emotional state on individual behavior selection;

[0010] Deduce the behavior trajectory according to the giant panda motor function model and the geographical location information, and use the time series model and the Markov decision process to simulate the dynamic behavior selection process of the giant panda in different scenarios to obtain the behavior trajectory deduction result;

[0011] Generate a behavior monitoring report for the target giant panda based on the behavior trajectory deduction result.

[0012] In a second aspect, the present application also provides an artificial intelligence-based wild giant panda behavior monitoring system, including:

[0013] An acquisition module for acquiring the original data set of the target giant panda, where the original data set is the video information, geographical location information of the same wild giant panda collected from multiple monitoring locations, and the environmental information of the monitoring locations;

[0014] An extraction module for performing feature extraction processing according to the video information and the environmental information, and obtaining a feature set by extracting the action, body posture, and micro-expression information of the target giant panda and analyzing the feature change trend in different time periods and different scenarios in combination with the environmental conditions;

[0015] A conversion module for performing numerical processing according to the feature set, and obtaining the numerical representation of the virtual individual by quantifying the action features, body posture features, and micro-expression features into the activity intensity index, body posture balance index, and emotional fluctuation index respectively;

[0016] A modeling module for performing modeling processing according to the numerical representation of the virtual individual, and constructing the giant panda motor function model by simulating the skeletal movement posture and combining the influence of health and emotional state on individual behavior selection;

[0017] A deduction module for deducing the behavior trajectory according to the giant panda motor function model and the geographical location information, and using the time series model and the Markov decision process to simulate the dynamic behavior selection process of the giant panda in different scenarios to obtain the behavior trajectory deduction result;

[0018] An output module for generating a behavior monitoring report for the target giant panda based on the behavior trajectory deduction result.

[0019] The beneficial effects of the present invention are:

[0020] The present invention obtains multi-source data, including surveillance videos, geographical locations, and environmental information, extracts features from the data to obtain a feature set of the actions, postures, and micro-expressions of giant pandas, numerically represents the action features and emotional fluctuations of giant pandas through virtual individual modeling, establishes a motor function model and a decision-making model, and realizes the deduction of behavioral trajectories, which can analyze the behavior patterns of giant pandas more comprehensively and accurately, improve the monitoring efficiency and protection effect, and effectively solve the problems of cross-monitoring point data synchronization and environmental interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0022] Figure 1 It is a schematic flowchart of a method for monitoring the behavior of wild giant pandas based on artificial intelligence according to an embodiment of the present invention;

[0023] Figure 2 It is a schematic structural diagram of a system for monitoring the behavior of wild giant pandas based on artificial intelligence according to an embodiment of the present invention;

[0024] Figure 3 It is a schematic structural diagram of a device for monitoring the behavior of wild giant pandas based on artificial intelligence according to an embodiment of the present invention.

[0025] Reference signs in the figure: 800, a device for monitoring the behavior of wild giant pandas based on artificial intelligence; 801, a processor; 802, a memory; 803, a multimedia component; 804, an I / O interface; 805, a communication component; 901, an acquisition module; 902, an extraction module; 903, a conversion module; 904, a modeling module; 905, a deduction module; 906, an output module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein can be arranged and designed in a variety of different configurations. Therefore, the detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0027] It should be noted that like reference numerals and letters denote like items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive distinction and cannot be construed as indicating or implying relative importance.

[0028] Embodiment 1:

[0029] This embodiment provides a method for monitoring the behavior of wild giant pandas based on artificial intelligence.

[0030] See Figure 1 , which shows that this method includes steps S100 to S500.

[0031] Step S100: Obtain the original data set of the target giant panda. The original data set is the video information, geographical location information of the same wild giant panda collected from multiple monitoring locations, and the environmental information of the monitoring locations.

[0032] It should be explained that the video information provides dynamic behavior data of the giant panda at different time points, including key features such as its activity range, behavior patterns, and posture changes; the geographical location information accurately records the spatial distribution and activity trajectories of the giant panda in the ecological environment through GPS or other positioning technologies; the environmental information includes environmental variables such as temperature, humidity, air pressure, and vegetation density.

[0033] Step S200: Perform feature extraction processing based on the video information and environmental information. By extracting the action, body posture, and micro-expression information of the target giant panda, and analyzing the feature change trends in different time periods and different scenarios in combination with the environmental conditions, a feature set is obtained.

[0034] It is understandable that action feature extraction analyzes the dynamic behaviors of giant pandas in videos, usually including their motion states, behavior types (such as foraging, resting, moving, socializing, etc.), and activity intensities, etc. Posture feature extraction focuses on the postures and morphological changes of giant pandas, especially the transitions between postures such as walking, standing, and crouching. Micro-expression feature extraction focuses on the subtle changes in the facial expressions of giant pandas, especially capturing details such as the subtle movements of the ears and changes in eye expressions through facial recognition technology. These subtle facial changes can reflect the emotional states of giant pandas, such as anxiety, relaxation, or pleasure, etc., which in turn affect their subsequent behavior choices. Combining environmental factors (such as temperature, humidity, or food resource conditions), these micro-expression features provide a background of deep emotional fluctuations for analyzing the behaviors of giant pandas.

[0035] Step S300: Perform numerical processing according to the feature set. By quantifying the action features, posture features, and micro-expression features into an activity intensity index, a posture balance index, and an emotional fluctuation index respectively, a numerical representation of the virtual individual is obtained;

[0036] Specifically, first, the quantification of action features is achieved by converting the activity intensity, frequency, and duration of giant pandas into an activity intensity index. This quantification process needs to combine the motion data in video analysis, including the moving speed, activity range, and motion patterns of giant pandas within a specific time period. The quantification of posture features focuses on the body states of giant pandas, especially their posture balance. By extracting the key points of the skeleton, posture changes, and the situation of the body touching the ground in the video, a posture balance index is calculated. This index can not only quantify the physical health of giant pandas but also reflect their adaptation to balance and stability in different environments. The quantification of micro-expression features obtains an emotional fluctuation index through the analysis of facial expressions and behavior changes. For example, by using a facial expression analysis model to extract the micro-expression changes of giant pandas and combining the influence of environmental factors on emotions, an emotional fluctuation index is obtained. This index can reflect the psychological states of giant pandas, such as emotional fluctuations of anxiety, relaxation, or pleasure, which further affect their behavior choices and activity patterns.

[0037] Step S400: Perform modeling processing according to the numerical representation of the virtual individual. By simulating the skeletal motion postures and combining the influence of health and emotional states on individual behavior choices, a locomotor function model of giant pandas is constructed;

[0038] It should be noted that in this step, by integrating the simulation of skeletal motion and the analysis of the influence of health and emotional states, not only can the motion trajectories of giant pandas be accurately captured, but also the physiological and emotional factors behind their behavior choices can be deeply revealed. This enables the model to more realistically reflect the behavior performances of giant pandas in complex environments and provides a basis for subsequent behavior trajectory deduction and prediction.

[0039] Step S500: Conduct behavioral trajectory deduction based on the giant panda's motor function model and geographical location information, and use the time series model and Markov decision process to simulate the dynamic behavior selection process of the giant panda in different scenarios, obtaining the behavioral trajectory deduction result;

[0040] It can be understood that, first of all, in this step, the time series model is used to capture the changes in the giant panda's behavior patterns at different times and environmental conditions. Through the training of the time series model, it is possible to predict its behavioral trends in a specific future time period based on historical data, making the behavioral trajectory deduction have stronger time series correlation. Secondly, the Markov decision process is used to simulate the decision-making process of the giant panda in a complex environment. Specifically, the Markov model defines the relationships between states, actions, and rewards to simulate how the giant panda makes optimal behavior choices based on the current state. Based on this model, the transition probabilities between different behaviors of the giant panda can be described, and its behavioral path can be further simulated.

[0041] Step S600: Generate a behavior monitoring report for the target giant panda based on the behavioral trajectory deduction result.

[0042] It should be noted that through in-depth analysis of the giant panda's behavioral trajectory, this report can provide the activity patterns, behavioral change trends of the giant panda within a specific time period, and its behavioral adaptability in different environments, providing scientific and systematic behavior prediction information for reserve managers or researchers.

[0043] Furthermore, step S200 includes steps S210 to S240.

[0044] Step S210: Perform action feature extraction processing based on the video information. By using the sparse coding algorithm to sparsely decompose the action features in the video, extract the key frames and action patterns of the giant panda's movement trajectory to construct an action feature dictionary, obtaining an action feature set;

[0045] It can be understood that the application of the sparse coding algorithm is to represent the action features of the giant panda in the video as a set of sparse basis vectors. These basis vectors can capture the key information in different behavior patterns of the giant panda, such as walking, standing, crawling, eating, etc. The advantage of sparse coding is that it can effectively reduce data redundancy, represent complex action sequences as a small number of basis vectors and corresponding weights, thereby extracting the key features of the giant panda's behavior. Sparse coding can not only extract the representative actions in the giant panda's movement trajectory but also identify the most representative key frames during the movement process. These key frames reflect the important turning points in the giant panda's movement process, such as the transition from rest to movement or the switching of movement modes. The finally constructed action feature dictionary can not only represent the basic actions of the giant panda but also reflect its behavioral changes in different environments, with high generalization.

[0046] Step S220: Perform body feature extraction and processing based on the video information. By applying the non - negative matrix factorization algorithm, dimensionality reduction and feature decomposition are carried out on the body information in the video, and the body changes in different scenarios are extracted to obtain a body feature set.

[0047] It can be understood that the non - negative matrix factorization algorithm can decompose high - dimensional body information into several meaningful and non - negative basis vectors, and represent the body changes of the giant panda in different scenarios by combining these basis vectors. Body changes include actions such as standing, walking, crawling, and sitting or lying down. These changes directly affect the behavior performance and health status of the giant panda. Through the non - negative matrix factorization algorithm, the body features of the giant panda can be efficiently extracted from video data, reducing noise and redundant information, thereby improving the analysis accuracy.

[0048] Step S230: Perform micro - expression feature extraction and processing based on the video information. By using the local binary pattern algorithm, local features of ear jitter and eye movement changes in the facial area of the giant panda are extracted, and the emotional state is analyzed by combining the preset typical facial structures and expression patterns to obtain a micro - expression feature set.

[0049] Specifically, the local binary pattern algorithm is used to analyze the micro - expressions such as ear jitter and eye movement changes in the facial area of the giant panda. These subtle changes reflect the emotional state of the giant panda. By extracting the local features of the facial area, the local binary pattern algorithm can capture detailed change patterns, such as micro - expressions of emotions like anxiety, pleasure, and tension. These micro - expression changes are not easily noticeable, but have a profound impact on the behavior choices of the giant panda. For example, when the giant panda is in an anxious state, it may choose to leave certain areas, while when it is in a pleasant mood, it may stay in a certain place for a longer time.

[0050] Step S240: Perform feature integration processing based on the action feature set, body feature set, and micro - expression feature set. By combining action, body, and micro - expression features with environmental information, analyze the impact of environmental factors on the behavior of the giant panda, and through feature weighting and normalization processing, obtain a comprehensive feature set.

[0051] It should be noted that by performing weighting and normalization processing on these features, it is ensured that the contribution of each feature in data integration is reasonably balanced, avoiding the bias impact of a certain feature on the model. The finally obtained comprehensive feature set can comprehensively reflect the behavior, health, and emotional state of the giant panda.

[0052] Furthermore, step S300 includes steps S310 to S340.

[0053] Step S310: Perform behavior time series alignment processing based on the action features in the feature set. Use the dynamic time warping algorithm to perform non-linear alignment on the unequal-length action sequences across monitoring points, and introduce the environmental light intensity as a time stretching penalty factor to calculate the activity intensity index.

[0054] It can be understood that the purpose of behavior time series alignment processing is to solve the problem of data time asynchrony across monitoring points, especially when the action sequences collected at different monitoring locations have inconsistent lengths and unsynchronized timestamps. To achieve this goal, the dynamic time warping algorithm is used to perform non-linear alignment on the unequal-length action sequences across monitoring points. The dynamic time warping algorithm can deform and align the action sequences on the time axis to align similar action patterns in the time series. To further improve the accuracy of time series alignment, the environmental light intensity is introduced as a time stretching penalty factor, which can reasonably adjust the time axis according to the light change, considering that the light intensity may affect the execution rate of the giant panda's behavior (for example, in a low-light environment, the giant panda's activity speed may be slower). After this processing, the obtained activity intensity index can effectively reflect the activity level of the giant panda under different environmental conditions.

[0055] Step S320: Perform body posture space embedding processing based on the body posture features in the feature set. Use the isometric feature mapping algorithm to map the skeleton key points extracted from multi-view videos to a low-dimensional manifold space, and calculate the covariance matrix of the principal curvature change rate and the ground contact pressure to obtain the body posture balance index.

[0056] It should be noted that the body posture space embedding processing focuses on how to extract accurate body posture balance information from the giant panda's skeleton key point data extracted from multi-view videos. By using the isometric feature mapping algorithm, the high-dimensional skeleton data is mapped to a low-dimensional manifold space, thus simplifying the representation of body posture data and retaining important body posture information. This method can capture the movement balance and stability of the giant panda by calculating the relative position changes of each key point of the skeleton. Further, by calculating the covariance matrix of the principal curvature change rate of the skeleton and the ground contact pressure, the balance of the giant panda's body posture can be quantitatively evaluated, and the body posture balance index is generated. This index can reflect the coordination of the giant panda during movement, especially its movement performance in complex terrains or dynamic environments. The calculation formula of the body posture balance index is as follows:

[0057] ;

[0058] where, is the body posture balance index, which measures the balance of the giant panda under different environmental and behavioral conditions; represents the serial number of the moment; represents the partial derivative; is a time function of the principal curvature, representing the change in the skeleton curvature of the giant panda at moment; is a time function of the ground contact pressure, representing the contact pressure between the skeleton and the ground at moment; is the principal curvature of the th sample point, representing the change in the skeleton posture; is the ground contact pressure of the th sample point; is the total number of sample points; and are time intervals, used to consider the dynamic effect of the body posture change over a period of time; is a weight parameter; is a regularization parameter; and are respectively the posture change matrix and the contact pressure matrix of the giant panda skeleton, and are and 's Frobenius norm, representing the overall change intensity; represents the regularization parameter; is the covariance matrix, used to represent the interaction between the principal curvature and the contact pressure; represents the determinant of the matrix.

[0059] Step S330: Perform emotion coupling analysis and processing according to the micro-expression features in the feature set, apply fuzzy cognitive maps to model the non-linear relationship between micro-expression electromyogram signals, environmental temperature and humidity, and feeding behavior, and determine the phase transition threshold of the emotional state by calculating the Lyapunov exponent spectrum of the attractor domain to obtain the emotional fluctuation index;

[0060] Specifically, using fuzzy cognitive maps can model the relationship between the electromyogram signals of micro-expressions, environmental changes (such as temperature and humidity), and the behavior of giant pandas (such as feeding behavior). By introducing the analysis framework of fuzzy cognitive maps, complex non-linear connections can be established between emotions, the environment, and behavior, thus more accurately quantifying the impact of emotional fluctuations on the behavior selection of giant pandas. By calculating the Lyapunov exponent spectrum of the attractor domain, the phase transition threshold of the emotional state can be determined, predicting the critical point of the emotional change of giant pandas, and then obtaining the emotional fluctuation index.

[0061] Step S340: Perform multi-modal fusion processing according to the activity intensity index, body posture balance index, and emotional fluctuation index, conduct a Pareto front search for all indices, and generate a numerical representation of the virtual individual by defining a fitness function that includes an energy consumption function and a social contact probability.

[0062] It should be noted that the Pareto front search can optimize multiple objectives, ensuring to find the optimal balance among different indices and avoiding any one feature from having an excessive impact on the final result. Meanwhile, the defined fitness function includes the energy consumption function and the social contact probability, which helps to optimize the behavior model of virtual individuals, enabling them to consider the interaction frequency of giant pandas in social behaviors while minimizing energy consumption. The fitness function is expressed as:

[0063] ;

[0064] where F is the fitness function, d represents the objective function for the behavior optimization of giant pandas; I activity (t) is the activity intensity index, representing the activity intensity of the giant panda at time t; I emotion (t) is the emotional fluctuation index, representing the emotional fluctuation intensity of the giant panda at time t; S contact (t) is the social contact probability, representing the contact frequency between the giant panda and other individuals; E consumption (t) is the energy consumption function, representing the energy consumption of the giant panda at time t; are all weight coefficients; is the time decay factor, used to control the change trend of energy consumption with the growth of time; T represents the time range; e is the base of the natural logarithm.

[0065] Furthermore, step S400 includes steps S410 to S440.

[0066] Step S410: Conduct skeletal dynamics modeling processing based on the numerical representation of virtual individuals, describe the joint motion constraints using differential-algebraic equations, and obtain a rigid-flexible coupled skeletal model;

[0067] It can be understood that by considering factors such as the binding force, motion range, and moment of inertia between joints, differential-algebraic equations can accurately simulate the dynamic behavior of giant panda skeletons. Differential-algebraic equations can not only describe the relative motion of each skeletal part of the giant panda but also capture the complex interaction relationships between bones in actual movements, such as moving, standing, walking, jumping, etc. By constructing a rigid-flexible coupled skeletal model, the rigidity (bone stiffness) and flexibility (flexibility of joints and soft tissues) of bones can be considered simultaneously, making the simulation results more in line with the real situation.

[0068] Step S420: Conduct health state integration processing based on the rigid-flexible coupled skeletal model, apply the stochastic optimal control theory to establish a metabolic energy - motion efficiency conversion model, and describe the probability density evolution of health indicators through the Fokker - Planck equation to obtain a set of physiological constraint motion strategies;

[0069] It should be noted that the stochastic optimal control theory aims to describe the conversion relationship between energy consumption and movement efficiency during the movement of giant pandas. Through the stochastic optimal control theory, the model can take into account the energy optimization process of giant pandas when performing movement tasks, that is, completing necessary behaviors without consuming too much energy. At the same time, the Fokker-Planck equation is used to describe the probability density evolution of health indicators, considering the physiological changes of giant pandas in different health states, which in turn affect their movement efficiency and behavior selection.

[0070] Step S430: Perform emotion-behavior coupling processing according to the physiological constraint movement strategy set, and obtain an emotion-enhanced decision tree by modeling the influence intensity of emotion fluctuations on path selection.

[0071] It can be understood that in this step, by modeling the influence intensity of emotion fluctuations on the path selection of giant pandas, it is possible to predict how emotion changes affect the behavior decision-making of giant pandas in different environments. For example, when a giant panda is in an anxious state, its behavior path may tend to a safe area, while when it is relaxed or happy, its movement path may be wider and more exploratory. This process models the relationship between emotion fluctuations and behavior selection by establishing an emotion-enhanced decision tree. The emotion-enhanced decision tree can generate a possible behavior selection tree under the influence of the emotion fluctuations of giant pandas, so as to take into account the influence of emotion states in the simulation and provide a more detailed decision-making basis for behavior deduction.

[0072] Step S440: Perform multi-scale fusion processing according to the emotion-enhanced decision tree to obtain a movement function model of giant pandas.

[0073] The core of this step is to fuse decision-making information at different levels, including behavior patterns, health states, and emotion fluctuations, to form a multi-dimensional movement decision-making framework. It comprehensively reflects the behavior patterns of the target giant pandas in different environments and emotion states, providing accurate model support for subsequent behavior deduction and monitoring.

[0074] Furthermore, step S500 includes steps S510 to S540.

[0075] Step S510: Perform terrain manifold modeling processing according to geographical location information, use the diffusion mapping algorithm to fuse slope, bamboo density, and stream distribution data, and generate a terrain reachability map by calculating the terrain similarity matrix and combining joint torque constraints.

[0076] In this step, the scatter mapping algorithm not only considers the complexity of the terrain in which the giant panda is located, but also evaluates the accessibility of different terrain areas by calculating the terrain similarity matrix. In addition, combined with joint torque constraints, the algorithm can take into account the movement limitations of its bones and muscles when simulating the giant panda's behavior, thereby more realistically reflecting the giant panda's range of activities. The resulting terrain accessibility map provides an environment-based feasibility framework for the giant panda's behavior deduction, which can accurately represent the potential paths of the giant panda's movement in complex terrain.

[0077] Step S520: Discretize the spatiotemporal state according to the terrain accessibility map, apply the non-uniform Poisson process to establish a behavior event flow model, and use the Hawkes process to model the self-motivation effect of food discovery events on the movement path to obtain a potential behavior transfer network;

[0078] Specifically, first, a behavioral event flow model is established using the non-uniform Poisson process, which can capture the flow characteristics of giant pandas' behavioral events under different spatiotemporal states. The non-uniform Poisson process can flexibly model the occurrence of events at irregular time intervals, and is particularly suitable for describing the behavioral dynamics of giant pandas in complex environments. In addition, the use of the Hawkes process to model the self-motivational effect of food discovery events on the movement path of giant pandas can accurately describe how giant pandas' behaviors change rapidly after discovering food and stimulate a series of subsequent behavioral choices. Through the modeling of these spatiotemporal dynamics, the generated potential behavioral transfer network can depict the possible behavioral paths of giant pandas in different environments and reveal how their behavioral choices are affected by environmental factors and emotional changes.

[0079] Step S530: Perform multimodal decision modeling processing according to the potential behavior transfer network, generate a neural-behavioral coupling strategy set by jointly modeling visual attention focus, hormone level and exercise energy consumption, and learning the mapping relationship from environmental features to action strategies;

[0080] In this process, the neural-behavioral coupling strategy set used can jointly analyze the giant panda's sensory input, internal physiological state and behavioral output. By modeling the focus of visual attention, it is possible to simulate how giant pandas choose to observe, focus on and respond to certain visual stimuli in a specific environment; changes in hormone levels (such as physiological changes such as appetite and anxiety) affect the giant panda's behavioral choices; and exercise energy consumption is directly related to the giant panda's activity and behavior duration. By learning the mapping relationship from environmental features to action strategies, the giant panda's behavioral strategy can be adaptively adjusted according to environmental changes, thereby generating behavioral predictions that are more in line with actual situations. The final generated neural-behavioral coupling strategy set provides multi-dimensional support for behavioral prediction and decision-making, and can take into account the giant panda's complex physiological, psychological and environmental interaction factors.

[0081] Step S540: Perform dynamic trajectory synthesis processing according to the neural-behavior coupling strategy set, apply differential game theory to establish an individual-environment game model and solve for the safe reachable set, and obtain the behavioral trajectory deduction result.

[0082] It can be understood that this process provides an optimization framework for the panda's behavior path selection. Especially when environmental changes or emotional fluctuations affect behavior decisions, the game model can simulate how the panda selects the optimal movement path while ensuring safety. The generated behavioral trajectory deduction result not only accurately predicts the panda's action trajectory but also takes into account the dynamic changes under different environments and emotional states, improving the accuracy and practicality of the prediction.

[0083] Furthermore, step S600 includes steps S610 to S640.

[0084] Step S610: Perform spatio-temporal semantic segmentation processing based on the behavioral trajectory deduction result. By combining the spectral clustering algorithm with terrain data, divide the continuous trajectory into behavioral units that conform to the activity rhythm characteristics of the panda, and obtain the set of behavioral semantic segments;

[0085] It can be understood that spectral clustering is a technique for clustering analysis by constructing the spectrum of a graph, which can effectively process non-linear and complex spatio-temporal data. In this step, the spectral clustering algorithm will combine the panda's behavioral trajectory data and terrain data, and identify continuous behavioral patterns in similar environments or states by calculating the similarity between trajectory data points. Through spectral analysis of the spatio-temporal distribution of the panda's behavioral trajectory, the trajectory can be divided into several units with similar behavioral patterns, and these units reflect a series of behavioral activities of the panda at different time periods. Secondly, by combining the activity rhythm characteristics of the panda, that is, the changes in its behavioral patterns at different times (such as diurnal activity rhythm, seasonal behavioral changes, etc.), more representative behavioral units can be divided more accurately. For example, the foraging behavior of the panda is usually closely related to certain specific environmental conditions (such as bamboo growth areas or near water sources). The spectral clustering algorithm can segment the trajectory based on these environmental characteristics and identify different behavioral stages of the panda in a specific environment. In this way, each segment in the trajectory can reflect the panda's behavioral pattern in a specific time and space dimension. Through spatio-temporal semantic segmentation in this step, complex behavioral trajectory data can be transformed into behavioral units with clear meanings, providing more structured data support for subsequent behavioral analysis and prediction.

[0086] Step S620: Perform pattern association mining processing based on the set of behavioral semantic segments. Apply the tensor decomposition algorithm to fuse the three-dimensional features of time, space, and environment, and extract the common patterns across behavioral units through non-negative constrained tensor decomposition to obtain the behavioral pattern feature tensor;

[0087] It should be noted that, first of all, the tensor decomposition algorithm is used for pattern mining of the behavior semantic fragment set. Tensor decomposition is a high-dimensional data decomposition method that can extract relevant information from multi-dimensional data (such as time, space, and environment), revealing the potential correlations between different dimensions. In this step, each behavior unit in the behavior semantic fragment set is regarded as an element of a tensor, and each dimension of the tensor represents the time, space, and environmental characteristics of the behavior respectively. Through non-negative constrained tensor decomposition, common patterns across behavior units can be effectively extracted from these high-dimensional data, avoiding the influence of negative values in the data and maintaining the positive characteristics in practical applications.

[0088] Through this processing, the tensor decomposition algorithm can not only reveal the correlations between different behavior units, but also identify the behavior patterns of giant pandas under time, space, and environmental conditions. For example, the behavior patterns of giant pandas may vary significantly in different geographical regions, but their basic activity patterns (such as foraging, resting, exploring, etc.) will show certain commonalities in the time series. Tensor decomposition can extract these common patterns and generate a behavior pattern feature tensor, thus providing a more refined pattern representation for subsequent behavior analysis and prediction.

[0089] Step S630: Perform risk situation assessment processing based on the behavior pattern feature tensor, use a fuzzy cognitive map to model the correlation between human activity interference, the probability of natural enemy appearance, and behavior abnormality, and solve the risk propagation path through a particle swarm optimization algorithm to obtain an ecological risk heat map;

[0090] Specifically, a fuzzy cognitive map is a knowledge representation method based on a graph structure that describes the causal relationships between variables through nodes and edges. In this step, the fuzzy cognitive map can relate external environmental factors (such as human activity interference with the habitat, the frequency of natural enemy appearance, etc.) to the behavior abnormality of giant pandas (such as evasive behavior, aggressive behavior, or changes in the activity range). By establishing this association model, the potential threats of different environmental factors to the behavior of giant pandas can be quantified, and the possible reactions of giant pandas when encountering threats can be predicted. Next, the particle swarm optimization algorithm is used to solve the risk propagation path. Particle swarm optimization is an optimization algorithm based on swarm intelligence that can simulate the process of particles searching for the optimal solution in the search space. In this step, the particle swarm optimization algorithm can efficiently solve the propagation path of ecological risks, and find the path of risk diffusion and potential high-risk areas by simulating how different environmental risk factors (such as the influence of natural enemy appearance and human activities) affect the behavior of giant pandas.

[0091] Finally, using the ecological risk heat map generated by the above method, by combining the risk propagation path with the behavioral pattern characteristics of the giant panda, a graphical output showing the risk concentration area is formed. This heat map can intuitively display the ecological risk distribution of the giant panda under different environmental conditions.

[0092] Step S640: Perform multi-modal report generation processing based on the ecological risk heat map. Adopt the template adaptation technology based on the knowledge graph, construct a structured narrative logic through semantic role annotation and event extraction, and generate a visual behavior monitoring report in combination with 3D terrain rendering.

[0093] First, the template adaptation technology based on the knowledge graph is used to generate the framework and structure of the report. The knowledge graph is a graphical structure that represents entities and their relationships through nodes and edges, and is used to represent domain knowledge. During the report generation process, the knowledge graph can help organize various information structurally, such as the behavioral patterns of giant pandas, environmental conditions, ecological risk assessment results, etc. Through the template adaptation technology, the content and display form of the report can be dynamically adjusted according to different monitoring requirements and data types, ensuring the personalization and pertinence of the report content. For example, in different risk assessment scenarios, the report template can display the dynamic changes in the behavior of giant pandas as needed, or highlight the high-risk points in specific areas.

[0094] Next, semantic role annotation and event extraction technologies are applied to extract specific risk information from the ecological risk heat map and transform it into a structured narrative. This process uses automated technologies to identify the core elements in the event, such as human activities, natural enemy threats, abnormal behaviors, etc., and annotate the relationships and roles of these elements in the entire ecosystem. Through semantic role annotation, the semantic relationships between various entities and behaviors can be clearly identified, such as "human activities lead to changes in the habitat of giant pandas", and corresponding descriptions can be made; while event extraction is used to extract specific events from the ecological risk data, such as "frequent activities of natural enemies lead to changes in the behavior pattern of giant pandas", forming a complete structured narrative logic to ensure that the content of the report is not only accurate but also easy to understand.

[0095] Finally, 3D terrain rendering is used to visualize geographical data to more intuitively display the behavior monitoring results of giant pandas and the spatial distribution of ecological risks. Through the 3D terrain rendering technology, the report can dynamically display the behavior trajectories of giant pandas, high-risk areas, and the specific distribution of ecological threats in maps or charts. Combining environmental data, the 3D view can not only display the habitat areas of giant pandas, but also highlight the factors affecting their activities, such as the distribution of food resources, the distribution of natural enemies, etc.

[0096] Example 2:

[0097] As Figure 2As shown in the figure, this embodiment provides an artificial intelligence-based wild giant panda behavior monitoring system, which includes:

[0098] An acquisition module 901, configured to acquire an original data set of a target giant panda. The original data set is video information, geographical location information of the same wild giant panda collected from multiple monitoring locations, and environmental information of the monitoring locations;

[0099] An extraction module 902, configured to perform feature extraction processing based on the video information and environmental information. By extracting the action, body posture, and micro-expression information of the target giant panda, and combining the environmental conditions to analyze the feature change trends in different time periods and different scenarios, a feature set is obtained;

[0100] A conversion module 903, configured to perform numerical processing based on the feature set. By quantifying the action features, body posture features, and micro-expression features into an activity intensity index, a body posture balance index, and an emotional fluctuation index respectively, a virtual individual numerical representation is obtained;

[0101] A modeling module 904, configured to perform modeling processing based on the virtual individual numerical representation. By simulating the skeletal movement posture and combining the influence of health and emotional states on individual behavior selection, a giant panda motor function model is constructed;

[0102] A deduction module 905, configured to perform behavior trajectory deduction based on the giant panda motor function model and geographical location information. Using a time series model and a Markov decision process to simulate the dynamic behavior selection process of the giant panda in different scenarios, a behavior trajectory deduction result is obtained;

[0103] An output module 906, which generates a behavior monitoring report of the target giant panda based on the behavior trajectory deduction result.

[0104] In some embodiments disclosed in this application, the extraction module 902 includes:

[0105] A first extraction unit, configured to perform action feature extraction processing based on the video information. By using a sparse coding algorithm to perform sparse decomposition on the action features in the video, the key frames and action patterns of the giant panda's movement trajectory are extracted to construct an action feature dictionary, and an action feature set is obtained;

[0106] A second extraction unit, configured to perform body posture feature extraction processing based on the video information. By applying a non-negative matrix factorization algorithm to perform dimensionality reduction and feature decomposition on the body posture information in the video, the body posture changes in different scenarios are extracted, and a body posture feature set is obtained;

[0107] A third extraction unit, configured to perform micro-expression feature extraction processing based on video information, perform local feature extraction on the ear jitter and eye movement of the giant panda's facial area through the local binary pattern algorithm, analyze the emotional state by combining the preset typical facial structures and expression patterns, and obtain a micro-expression feature set;

[0108] A fourth extraction unit, configured to perform feature integration processing based on the action feature set, body posture feature set, and micro-expression feature set. By combining the action, body posture, and micro-expression features with environmental information, analyze the impact of environmental factors on the behavior of the giant panda, and obtain a comprehensive feature set through feature weighting and normalization processing.

[0109] In some embodiments disclosed in the present application, the conversion module 903 includes:

[0110] A first conversion unit, configured to perform behavior time series alignment processing based on the action features in the feature set. Non-linearly align the unequal-length action sequences across monitoring points through the dynamic time warping algorithm, and introduce the environmental light intensity as a time stretching penalty factor to calculate the activity intensity index;

[0111] A second conversion unit, configured to perform body posture space embedding processing based on the body posture features in the feature set. Map the skeleton key points extracted from multi-view videos to a low-dimensional manifold space using the isometric feature mapping algorithm, and obtain the body posture balance index by calculating the covariance matrix of the principal curvature change rate and the ground contact pressure;

[0112] A third conversion unit, configured to perform emotion coupling analysis processing based on the micro-expression features in the feature set. Apply a fuzzy cognitive map to model the non-linear relationship between the micro-expression electromyogram signal, environmental temperature and humidity, and feeding behavior, and determine the phase transition threshold of the emotional state by calculating the Lyapunov exponent spectrum of the attractor domain to obtain the emotion fluctuation index;

[0113] A fourth conversion unit, configured to perform multi-modal fusion processing based on the activity intensity index, body posture balance index, and emotion fluctuation index. Perform a Pareto front search on all indices, and generate a virtual individual numerical representation by defining a fitness function that includes an energy consumption function and a social contact probability.

[0114] In some embodiments disclosed in the present application, the modeling module 904 includes:

[0115] A first modeling unit, configured to perform rigid-flexible coupling bone model processing based on the virtual individual numerical representation, and describe the joint motion constraints using differential algebraic equations to obtain a rigid-flexible coupling bone model;

[0116] A second modeling unit, configured to perform health status integration processing based on the rigid-flexible coupled bone model, establish a metabolic energy - motion efficiency conversion model by applying stochastic optimal control theory, describe the probability density evolution of health indicators through the Fokker - Planck equation, and obtain a set of physiological constraint motion strategies;

[0117] A third modeling unit, configured to perform emotion - behavior coupling processing based on the set of physiological constraint motion strategies, and obtain an emotion - enhanced decision tree by modeling the influence intensity of emotional fluctuations on path selection;

[0118] A fourth modeling unit, configured to perform multi - scale fusion processing based on the emotion - enhanced decision tree to obtain a giant panda locomotor function model.

[0119] In some embodiments disclosed in the present application, the deduction module 905 includes:

[0120] A first deduction unit, configured to perform terrain manifold modeling processing based on geographical location information, fuse slope, bamboo density, and stream distribution data using the diffusion mapping algorithm, and generate a terrain accessibility map by calculating a terrain similarity matrix and combining joint torque constraints;

[0121] A second deduction unit, configured to perform spatio - temporal state discretization processing based on the terrain accessibility map, establish a behavioral event flow model using a non - homogeneous Poisson process, and model the self - exciting effect of food discovery events on the movement path through the Hawkes process to obtain a potential behavior transition network;

[0122] A third deduction unit, configured to perform multi - modal decision - making modeling processing based on the potential behavior transition network, jointly model the visual attention focus, hormone level, and motion energy consumption, and learn the mapping relationship from environmental features to action strategies to generate a set of neural - behavior coupling strategies;

[0123] A fourth deduction unit, configured to perform dynamic trajectory synthesis processing based on the set of neural - behavior coupling strategies, establish an individual - environment game model using differential game theory and solve for the safe reachable set to obtain a behavioral trajectory deduction result.

[0124] Embodiment 3:

[0125] Corresponding to the above - mentioned method embodiment, in this embodiment, there is also provided a field giant panda behavior monitoring device based on artificial intelligence. The field giant panda behavior monitoring device described below and the field giant panda behavior monitoring method described above can be correspondingly referred to each other.

[0126] Figure 3 It is a block diagram of a field giant panda behavior monitoring device 800 shown according to an exemplary embodiment. As Figure 3As shown, the wild giant panda behavior monitoring device 800 based on artificial intelligence may include: a processor 801 and a memory 802. The wild giant panda behavior monitoring device 800 based on artificial intelligence may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0127] Among them, the processor 801 is used to control the overall operation of the wild giant panda behavior monitoring device 800 based on artificial intelligence to complete all or part of the steps in the above-mentioned wild giant panda behavior monitoring method based on artificial intelligence. The memory 802 is used to store various types of data to support the operation of the wild giant panda behavior monitoring device 800 based on artificial intelligence. These data may include, for example, instructions for any application or method operating on the wild giant panda behavior monitoring device 800 based on artificial intelligence, as well as application-related data, such as contact data, sent and received messages, pictures, audio, video, and so on. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 803 may include a screen and an audio component. Among them, the screen may be a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone, and the microphone is used to receive external audio signals. The received audio signals may be further stored in the memory 802 or sent through the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules. The above-mentioned other interface modules may be a keyboard, a mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the wild giant panda behavior monitoring device 800 based on artificial intelligence and other devices. Wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G or 4G, or a combination of one or more of them. Therefore, the corresponding communication component 805 may include: a Wi-Fi module, a Bluetooth module, and an NFC module.

[0128] In an exemplary embodiment, an artificial intelligence-based wild giant panda behavior monitoring device 800 can be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components, and is used to execute the above-mentioned artificial intelligence-based wild giant panda behavior monitoring method.

[0129] In another exemplary embodiment, a computer-readable storage medium including program instructions is further provided. When the program instructions are executed by a processor, the steps of the above-mentioned artificial intelligence-based wild giant panda behavior monitoring method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 802 including program instructions, and the above program instructions can be executed by the processor 801 of an artificial intelligence-based wild giant panda behavior monitoring device 800 to complete the above-mentioned artificial intelligence-based wild giant panda behavior monitoring method.

[0130] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A method for monitoring the behavior of wild giant pandas based on artificial intelligence, characterized in that: include: Acquire an original data set of a target giant panda, wherein the original data set is video information, geographic location information, and environmental information of the same wild giant panda collected from multiple monitoring locations; Performing feature extraction processing according to the video information and the environmental information, extracting the action, posture and micro-expression information of the target giant panda, and analyzing the feature change trends in different time periods and different scenes in combination with the environmental conditions, to obtain a feature set; Numerical processing is performed according to the feature set, and the action features, posture features, and micro-expression features are quantified into an activity intensity index, a posture balance index, and an emotion fluctuation index, respectively, to obtain a virtual individual numerical representation; Modeling is performed according to the virtual individual numerical representation, and a giant panda movement function model is constructed by simulating the skeletal movement posture and combining the influence of health and emotional state on individual behavior selection; Deducing the behavior trajectory according to the giant panda's motor function model and the geographic location information, simulating the dynamic behavior selection process of the giant panda in different scenarios using a time series model and a Markov decision process, and obtaining the behavior trajectory deduction result; A behavior monitoring report of the target giant panda is generated based on the behavior trajectory deduction results.

2. The method for monitoring wild giant panda behavior based on artificial intelligence according to claim 1, characterized in that: Perform feature extraction processing based on the video information and the environmental information, extract the action, posture and micro-expression information of the target giant panda, and analyze the feature change trends in different time periods and different scenes in combination with the environmental conditions to obtain a feature set, including: Performing motion feature extraction processing according to the video information, performing sparse decomposition on the motion features in the video by using a sparse coding algorithm, extracting key frames and motion patterns of the giant panda's motion trajectory to construct a motion feature dictionary, and obtaining a motion feature set; Performing posture feature extraction processing according to the video information, performing dimension reduction and feature decomposition on the posture information in the video by applying a non-negative matrix decomposition algorithm, extracting its posture changes in different scenes, and obtaining a posture feature set; Performing micro-expression feature extraction processing based on the video information, extracting local features of ear shaking and eye expression changes in the facial area of ​​the giant panda using a local binary pattern algorithm, analyzing the emotional state in combination with a preset typical facial structure and expression pattern, and obtaining a micro-expression feature set; Feature integration processing is performed based on the action feature set, the posture feature set and the micro-expression feature set. The influence of environmental factors on the behavior of giant pandas is analyzed by combining the action, posture and micro-expression features with environmental information. A comprehensive feature set is obtained through feature weighting and normalization processing.

3. The method for monitoring wild giant panda behavior based on artificial intelligence according to claim 1, characterized in that: According to the feature set, numerical processing is performed, and the action features, posture features and micro-expression features are quantified into activity intensity index, posture balance index and emotion fluctuation index respectively, so as to obtain the virtual individual numerical representation, including: Performing behavior timing alignment processing according to the action features in the feature set, performing nonlinear alignment on action sequences of unequal lengths across monitoring points through a dynamic time warping algorithm, and introducing ambient light intensity as a time expansion penalty factor to calculate an activity intensity index; Performing posture space embedding processing according to the posture features in the feature set, using an isometric feature mapping algorithm to map the skeleton key points extracted from the multi-view video to a low-dimensional manifold space, and obtaining a posture balance index by calculating the covariance matrix of the principal curvature change rate and the ground contact pressure; Emotional coupling analysis is performed based on the micro-expression features in the feature set, the nonlinear relationship between micro-expression electromyographic signals, environmental temperature and humidity, and eating behavior is modeled using fuzzy cognitive graphs, the phase transition threshold of the emotional state is determined by calculating the Lyapunov exponent spectrum of the attractor domain, and the emotional fluctuation index is obtained; Multimodal fusion processing is performed according to the activity intensity index, the body balance index and the mood fluctuation index, a Pareto frontier search is performed on all indexes, and a virtual individual numerical representation is generated by defining a fitness function including an energy consumption function and a social contact probability.

4. The method for monitoring wild giant panda behavior based on artificial intelligence according to claim 1, characterized in that: Modeling is performed based on the virtual individual numerical representation, and by simulating the skeletal movement posture and combining the influence of health and emotional state on individual behavior selection, a giant panda movement function model is constructed, including: Performing skeletal dynamics modeling processing according to the virtual individual numerical representation, using differential algebraic equations to describe joint motion constraints, and obtaining a rigid-flexible coupling skeletal model; According to the rigid-flexible coupled skeleton model, health status integration processing is performed, and a metabolic energy-exercise efficiency conversion model is established by applying stochastic optimal control theory. The probability density evolution of health indicators is described by the Focke-Planck equation to obtain a set of physiological constraint exercise strategies; Performing emotion-behavior coupling processing according to the physiological constraint movement strategy set, and obtaining an emotion enhancement decision tree by modeling the influence intensity of emotion fluctuations on path selection; Multi-scale fusion processing is performed according to the emotion enhancement decision tree to obtain a giant panda motor function model.

5. The method for monitoring wild giant panda behavior based on artificial intelligence according to claim 1, characterized in that: The behavior trajectory is deduced according to the giant panda's motor function model and the geographic location information, and the dynamic behavior selection process of the giant panda in different scenarios is simulated using a time series model and a Markov decision process to obtain the behavior trajectory deduction results, including: Performing terrain manifold modeling according to the geographic location information, using a diffusion mapping algorithm to fuse slope, bamboo density and stream distribution data, and generating a terrain accessibility map by calculating a terrain similarity matrix and combining joint torque constraints; According to the terrain accessibility map, the spatiotemporal state is discretized, a behavior event flow model is established by applying a non-uniform Poisson process, and the self-motivation effect of food discovery events on the movement path is modeled by a Hawkes process to obtain a potential behavior transfer network; Performing multimodal decision modeling processing according to the potential behavior transfer network, generating a neural-behavioral coupling strategy set by jointly modeling visual attention focus, hormone level and exercise energy consumption, and learning the mapping relationship from environmental features to action strategies; Dynamic trajectory synthesis processing is performed according to the neural-behavioral coupling strategy set, and the individual-environment game model is established by applying differential game theory and solving the safe reachable set to obtain the behavior trajectory deduction result.

6. A wild giant panda behavior monitoring system based on artificial intelligence, characterized in that: include: An acquisition module is used to acquire an original data set of a target giant panda, wherein the original data set is video information, geographic location information, and environmental information of the same wild giant panda collected from multiple monitoring locations; An extraction module is used to perform feature extraction processing according to the video information and the environmental information, extract the action, posture and micro-expression information of the target giant panda, and analyze the feature change trends in different time periods and different scenes in combination with the environmental conditions to obtain a feature set; A conversion module, used for performing numerical processing according to the feature set, and obtaining a virtual individual numerical representation by quantifying the action features, body features and micro-expression features into an activity intensity index, a body balance index and an emotion fluctuation index respectively; A modeling module is used to perform modeling processing according to the virtual individual numerical representation, and construct a giant panda movement function model by simulating the skeletal movement posture and combining the influence of health and emotional state on individual behavior selection; A deduction module is used to deduce the behavior trajectory according to the giant panda's motor function model and the geographic location information, and to simulate the dynamic behavior selection process of the giant panda in different scenarios by using a time series model and a Markov decision process to obtain a behavior trajectory deduction result; The output module generates a behavior monitoring report of the target giant panda based on the behavior trajectory deduction results.

7. The wild giant panda behavior monitoring system based on artificial intelligence according to claim 6 is characterized in that: The extraction module comprises: A first extraction unit is used to perform motion feature extraction processing according to the video information, and sparsely decompose the motion features in the video by using a sparse coding algorithm, extract key frames and motion patterns of the giant panda's motion trajectory to construct a motion feature dictionary, and obtain a motion feature set; A second extraction unit is used to perform posture feature extraction processing according to the video information, and perform dimension reduction and feature decomposition on the posture information in the video by applying a non-negative matrix decomposition algorithm, extract the posture changes in different scenes, and obtain a posture feature set; A third extraction unit is used to perform micro-expression feature extraction processing based on the video information, extract local features of ear shaking and eye expression changes in the giant panda's facial area through a local binary pattern algorithm, analyze the emotional state in combination with a preset typical facial structure and expression pattern, and obtain a micro-expression feature set; The fourth extraction unit is used to perform feature integration processing based on the action feature set, the posture feature set and the micro-expression feature set, analyze the impact of environmental factors on the behavior of giant pandas by combining the action, posture and micro-expression features with environmental information, and obtain a comprehensive feature set through feature weighting and normalization processing.

8. The wild giant panda behavior monitoring system based on artificial intelligence according to claim 6 is characterized in that: The conversion module comprises: A first conversion unit is used to perform behavior timing alignment processing according to the action features in the feature set, perform nonlinear alignment on action sequences of unequal lengths across monitoring points through a dynamic time warping algorithm, and introduce ambient light intensity as a time expansion penalty factor to calculate an activity intensity index; A second conversion unit is used to perform posture space embedding processing according to the posture features in the feature set, map the skeleton key points extracted from the multi-view video to the low-dimensional manifold space by using an isometric feature mapping algorithm, and obtain a posture balance index by calculating the covariance matrix of the principal curvature change rate and the ground contact pressure; A third conversion unit is used to perform emotion coupling analysis and processing according to the micro-expression features in the feature set, use fuzzy cognitive graphs to model the nonlinear relationship between micro-expression electromyographic signals, environmental temperature and humidity, and eating behavior, determine the phase change threshold of the emotional state by calculating the Lyapunov exponent spectrum of the attractor domain, and obtain the emotion fluctuation index; The fourth conversion unit is used to perform multimodal fusion processing according to the activity intensity index, the body balance index and the mood fluctuation index, perform Pareto frontier search on all indexes, and generate a virtual individual numerical representation by defining a fitness function including an energy consumption function and a social contact probability.

9. The wild giant panda behavior monitoring system based on artificial intelligence according to claim 6 is characterized in that: The modeling module includes: A first modeling unit is used to perform skeletal dynamics modeling processing according to the virtual individual numerical representation, using differential algebraic equations to describe joint motion constraints to obtain a rigid-flexible coupling skeletal model; The second modeling unit is used to perform health status integration processing according to the rigid-flexible coupled skeleton model, establish a metabolic energy-exercise efficiency conversion model by applying the stochastic optimal control theory, describe the probability density evolution of health indicators by the Focke-Planck equation, and obtain a set of physiological constraint exercise strategies; A third modeling unit is used to perform emotion-behavior coupling processing according to the physiological constraint movement strategy set, and obtain an emotion enhancement decision tree by modeling the influence intensity of emotion fluctuations on path selection; The fourth modeling unit is used to perform multi-scale fusion processing according to the emotion enhancement decision tree to obtain a giant panda motor function model.

10. The wild giant panda behavior monitoring system based on artificial intelligence according to claim 6 is characterized in that: The deduction module includes: A first deduction unit is used to perform terrain manifold modeling according to the geographical location information, fuse the slope, bamboo density and stream distribution data using a diffusion mapping algorithm, and generate a terrain accessibility map by calculating a terrain similarity matrix and combining joint torque constraints; The second deduction unit is used to discretize the spatiotemporal state according to the terrain accessibility map, establish a behavior event flow model by applying a non-uniform Poisson process, and model the self-motivation effect of food discovery events on the movement path by using a Hawkes process to obtain a potential behavior transfer network; A third deduction unit is used to perform multimodal decision modeling processing according to the potential behavior transfer network, and generate a neural-behavioral coupling strategy set by jointly modeling visual attention focus, hormone level and exercise energy consumption, and learning the mapping relationship from environmental features to action strategies; The fourth deduction unit is used to perform dynamic trajectory synthesis processing according to the neural-behavioral coupling strategy set, apply differential game theory to establish an individual-environment game model and solve the safe reachable set to obtain the behavior trajectory deduction result.

Images