Bird flight path intelligent tracking and monitoring method and system based on big data analysis

Abstract

The invention discloses a bird flight path intelligent tracking and monitoring method and system based on big data analysis. The method comprises the following steps: S1, generating an original bird observation data set; s2, outputting a standardized bird observation data set; s3, calling a multi-target tracking algorithm based on the standardized bird observation data set to obtain a bird trajectory fragment data set; s4, constructing a lion group optimization search model, and inputting the bird trajectory fragment data set into the lion group optimization search model to form a candidate trajectory data set; s5, inputting the candidate trajectory data set into a space-time Transform model configured with rotation position embedding, and generating a continuous flight path data set; and S6, outputting a cross-region bird migration monitoring result set. According to the method, bird migration continuous tracking requirements under different time periods, multiple geographical zones and heterogeneous equipment acquisition conditions can be covered, application scenes of ecological protection, migration research and airspace management and control are effectively supported, and the method has remarkable engineering popularization and ecological protection values.

Description

Technical Field

[0001] The present invention relates to the field of bird monitoring technology, and in particular to a method and system for intelligently tracking and monitoring bird flight paths based on big data analysis. Background Art

[0002] With the continuous advancement of ecological monitoring technology, cross-regional continuous tracking of bird migration routes has become an important supporting means in the fields of global biodiversity conservation, climate change research and aviation safety management. At present, bird migration route monitoring mainly relies on the deployment of satellite transmitters, ground radar stations, video surveillance equipment and acoustic sensors at key nodes. By collecting and splicing the bird's activity trajectories, the migration dynamics can be recorded.

[0003] With the application of big data and artificial intelligence technologies, researchers have tried to use recurrent neural networks and self-attention Transformer sequence modeling algorithms to align bird trajectories and interpolate missing segments. However, existing models generally use absolute position encoding methods, which are difficult to handle migration trajectories with long spans, sparse nodes, and drastic changes in geographical location. This can easily lead to a decrease in long-distance dependent capture capabilities, large errors in missing segment interpolation, and unstable identity retention. At the same time, existing data fusion methods often ignore ecological constraints in bird migration, such as wind direction, terrain, and physical and biological boundaries of species differences, and cannot achieve intelligent optimization and reliable evaluation of the trajectory splicing process. Furthermore, existing technologies have not yet achieved linked normalization processing of long-sequence, multi-target, cross-regional, and heterogeneous observation data, making it difficult to output high-confidence, traceable continuous migration path results.

[0004] In summary, there is an urgent need to develop a new cross-regional continuous tracking method for bird flight paths that can effectively solve the above problems. Summary of the Invention

[0005] One purpose of the present invention is to propose a method and system for intelligent tracking and monitoring of bird flight paths based on big data analysis. The present invention can cover the needs of continuous tracking of bird migration under different time periods, across multiple geographical zones, and under heterogeneous equipment collection conditions, effectively supporting application scenarios of ecological protection, migration research, and airspace control, and has significant engineering promotion and ecological protection value.

[0006] According to an embodiment of the present invention, a method for intelligent tracking and monitoring of bird flight paths based on big data analysis includes the following steps:

[0007] S1. Deploy bird monitoring equipment in multiple geographical areas including plateaus, mountains, wetlands, and coasts to collect multimodal bird observation data including images, videos, and sounds in real time and generate original bird observation datasets.

[0008] S2. Perform data preprocessing on the original bird observation dataset. Target detection and recognition algorithms are used to locate and identify bird targets in images and videos. Audio recognition models are used to identify the bird species corresponding to the sound signals. Time and space coordinates are unified based on GPS tags. Low-confidence or duplicate observation records are removed, and outliers are cleaned to output a standardized bird observation dataset.

[0009] S3. Based on a standardized bird observation dataset, a multi-target tracking algorithm is used to establish the temporal trajectory of the same bird in consecutive video frames. The image, video, and sound recognition results are integrated to construct a set of trajectory segments, resulting in a bird trajectory segment dataset.

[0010] S4. Build a lion population optimization search model. Input the bird trajectory segment dataset into the lion population optimization search model. Perform multi-target coarse cascade through global search of male lions, local screening of female lions, and diffusion exploration of cubs. Output a set of candidate trajectories and a confidence matrix to form a candidate trajectory dataset.

[0011] S5. Input the candidate trajectory dataset into a spatiotemporal Transformer model configured with rotational position embedding. Based on the rotational geographic angular distance encoding, long-span sequence alignment, bird individual identity identification, and missing segment interpolation are performed to obtain a continuous and complete set of flight paths, thus generating a continuous flight path dataset.

[0012] S6. Perform error evaluation on the continuous flight path dataset and update the weight parameters based on the evaluation results, then write them into the trajectory database to form a stored continuous flight path dataset. Associate the stored continuous flight path dataset with the bird ecological knowledge map and real-time environmental data to output a set of cross-regional bird migration monitoring results.

[0013] Optionally, S2 includes the following steps:

[0014] S21. Perform bird target detection and recognition processing on the image and video data in the original bird observation dataset, and output a set of spatial coordinates of all detected bird targets in each image or video frame. Each spatial coordinate in the set consists of a horizontal coordinate, a vertical coordinate, and a corresponding time tag. The number of spatial coordinate sets is equal to the number of bird targets detected in the image or video frame.

[0015] S22. Use the audio recognition model to identify bird species from the sound data in the original bird observation dataset. Extract the voiceprint features of each sound data segment and match the voiceprint features with known bird species labels to generate an audio recognition result set. Each element in the audio recognition result set includes the bird species label and time label corresponding to the sound data. The number of audio recognition result sets is equal to the number of sound data segments.

[0016] S23. Based on the GPS tags in the original bird observation dataset, perform time normalization and spatial coordinate normalization on all observation data. After each observation data is processed through the spatial coordinate normalization function and the time normalization function, it is unified into standardized observation data. The standardized observation data includes the normalized horizontal coordinate, vertical coordinate, normalized time tag, and observation data category. The number of standardized observation data sets is equal to the total number of observation data.

[0017] S24. Confidence screening and duplicate observation elimination are performed on the standardized observation data set. A confidence threshold is set, and only observation data with a confidence score not less than the confidence threshold is retained. Duplicate observation records with the same time label and spatial coordinates are eliminated. Each element in the confidence screening set is a unique observation data with a confidence score not less than the confidence threshold.

[0018] S25. Clean the confidence filter set of outliers, calculate the Euclidean distance between each observation data and the mean vector in the confidence filter set, retain only the observation data whose Euclidean distance is not higher than the outlier discrimination threshold, and finally output the standardized bird observation dataset.

[0019] Optionally, S3 includes the following steps:

[0020] S31. For a standardized bird observation dataset, use a multi-target tracking algorithm to assign an identity to each bird target in consecutive video frames. In each frame, process each spatial coordinate in the spatial coordinate set based on its abscissa, ordinate, and time tag. By comparing the correlation between spatial coordinates and time tags in adjacent frames, associate the spatial coordinates of the same bird in different frames to form an initial set of temporal trajectories. Each temporal trajectory in the initial set consists of several temporally associated spatial coordinates and time tags. The number of temporal trajectories in the initial set equals the total number of trajectories.

[0021] S32. Extract the bird species label and time label corresponding to each segment of sound data from the audio recognition result set in the standardized bird observation dataset, match and associate the bird species labels identified under the same time label with the matching spatial coordinates in the video frame, fuse the image, video, and sound recognition results, and construct a fused observation segment set containing spatial coordinates, time labels, and bird species labels. Each fused observation segment in the fused observation segment set consists of spatial coordinates, time labels, and bird species labels. The number of fused observation segment sets is equal to the number of fused segments.

[0022] S33. Perform trajectory fusion on the initial time-series trajectory set and the fused observation segment set based on a trajectory similarity index function. The trajectory similarity index function is used to measure the comprehensive similarity between the initial time-series trajectory and the fused observation segment in terms of spatial coordinates and species labels. Each trajectory segment in the trajectory segment set consists of a temporally associated spatial coordinate, a time label, and a bird species label. The number of trajectory segment sets is equal to the number of trajectory segments.

[0023] S34. All the trajectory segments in the trajectory segment set are uniformly formatted so that each trajectory segment contains temporally associated spatial coordinates, time labels, and bird species labels, and outputs a bird trajectory segment dataset.

[0024] Optionally, the trajectory similarity index function value is obtained by calculating the Euclidean distance between the initial time series trajectory and the fused observation segment in spatial coordinates, and then calculating the consistency measure between the initial time series trajectory and the fused observation segment in species labels. The Euclidean distance and the consistency measure are multiplied by the corresponding normalized weight coefficients and then accumulated. The result is obtained by taking the inverse of the accumulated value plus one. By maximizing the trajectory similarity index function value, the trajectory and observation segments of the same bird body are fused to obtain a trajectory segment set.

[0025] Optionally, the S4 includes the following steps:

[0026] S41. Build a lion population optimization search model based on cross-regional migration constraints, taking a bird track segment dataset as input. The bird track segment dataset is partitioned into geographic zones, each of which is assigned a unique number.

[0027] S42. For any two bird track segments, the spatial distance between the two bird track segments, the time tag difference between the two bird track segments, and the consistency parameter between the species tags of the two bird track segments are calculated respectively by assigning spatial distance weights, time difference weights, and species consistency weights. The above three values are weighted and summed according to the spatial distance weights, time difference weights, and species consistency weights to obtain the basic fitness parameter. The basic fitness parameter is used to measure the ecological splicing cost between any two bird track segments. The spatial distance is measured by calculating the geographical distance between the end point of the first bird track segment and the starting point of the second bird track segment. The time tag difference is the absolute value of the end time tag of the first bird track segment and the starting time tag of the second bird track segment. The species consistency parameter is zero when the species tags of the two bird track segments are the same, and one when they are different. The sum of the spatial distance weight, time difference weight, and species consistency weight is one.

[0028] S43. Based on the basic fitness parameters, a wind direction matching parameter, a terrain climbing constraint parameter, and a historical migration record consistency parameter are introduced to reflect the matching of the angle between the wind direction and the flight direction between two bird track segments, the change in terrain elevation between the two bird track segments, and the splicing consistency of the two bird track segments in the historical data, respectively. By assigning correction weights for wind direction matching, terrain climbing, and historical migration record, the wind direction matching parameter, terrain climbing constraint parameter, and historical migration record consistency parameter are added as correction items to the basic fitness parameters to obtain an enhanced fitness parameter. The enhanced fitness parameter is used to measure the comprehensive splicing cost of two bird track segments under ecological and physical constraints. The closer the value of the wind direction matching parameter is to 1, the more consistent the wind direction and flight direction are. The terrain climbing constraint parameter is used to determine whether the terrain elevation difference exceeds the species' flyable range. The historical migration record consistency parameter is used to determine whether the splicing instance exists in the historical data.

[0029] S44. The male lions conduct a global search within the bird track segment dataset. For each geographic zone, all bird track segment pairs within the zone are combined and matched. The enhanced fitness parameter for each bird track segment pair is calculated. The bird track segment pairs with the lowest enhanced fitness parameters are selected as preliminary candidate pairings. The preliminary candidate pairings constitute the global candidate pairing set.

[0030] S45. Lionesses introduce a migration ecological filtering mechanism based on the global candidate pairing set. For each candidate pairing, the mechanism sequentially determines whether its time tag interval is less than or equal to the maximum allowable migration time jump threshold, whether its wind direction compatibility parameter is greater than or equal to the wind assistance effect lower limit, and whether its terrain climbing constraint parameter is equal to zero. Only when all conditions are met is the pairing retained as a locally valid pairing. All locally valid pairs constitute the local candidate pairing set.

[0031] S46. The lion cubs introduce temporal and spatial perturbations to the unmatched bird track segments in the bird track segment dataset, randomly perturbing the time tags and spatial coordinates, respectively, to generate new combinations. The enhanced fitness parameter of each combination is calculated. All new combinations are screened using the same criteria as above, and valid pairs are retained. All valid pairs constitute the perturbation expansion candidate set.

[0032] S47. Splice the local candidate pair set with all valid pairs in the perturbed and expanded candidate set in the order of their time tags to obtain complete candidate trajectories. Each complete candidate trajectory is composed of several sequentially spliced bird trajectory segments. During the trajectory splicing process, the enhanced fitness parameters between each group of bird trajectory segments are recorded. The arithmetic mean of all enhanced fitness parameters is used as the average path splicing cost of the complete candidate trajectory.

[0033] S48. For each complete candidate trajectory, the average path splicing cost is converted into a confidence parameter using a negative normalized exponential function based on the average path splicing cost. The normalization factor is used to adjust the distribution range of the confidence parameter. The confidence parameters of all complete candidate trajectories form a scoring matrix. All complete candidate trajectories and the scoring matrix are output to form a candidate trajectory dataset.

[0034] Optionally, the S5 includes the following steps:

[0035] S51. The candidate trajectory dataset consists of several candidate trajectories. Each candidate trajectory is composed of several sequentially spliced bird trajectory segments. Each bird trajectory segment contains spatial coordinates, a time label, and a bird species label. The candidate trajectory dataset is converted into a sequence structure recognizable by the model to obtain a set of candidate trajectory sequences. Each candidate trajectory sequence is a time-ordered sequence.

[0036] S52. For the candidate trajectory sequence set, calculate the spatial variation based on the longitude and latitude of each trajectory point, and convert the spatial variation into polar coordinates to obtain the geographic argument parameter and geographic modulus parameter, respectively. The geographic argument parameter is used to measure the spatial angular distance between two trajectory points, and the geographic modulus parameter is used to measure the geographic distance between two trajectory points. The geographic argument parameter and geographic modulus parameter are combined with the time tag difference to form a rotated geographic angular distance code. The rotated geographic angular distance code is embedded as the spatial and temporal features of the candidate trajectory sequence, and the rotated geographic angular distance code is one-to-one corresponding to each trajectory point.

[0037] S53. The candidate trajectory sequence and its corresponding rotated geographic angular distance encoding are jointly input into a spatiotemporal Transformer model configured with a rotation position embedding. The spatiotemporal Transformer model directly uses the geographic argument parameters and geographic modulus parameters of the rotated geographic angular distance encoding in each attention head to measure the spatial relationship between trajectory points, and combines the time label difference as a measure of the temporal relationship. During the calculation of the attention weight, the geographic argument parameters and geographic modulus parameters of the rotated geographic angular distance encoding are used to dynamically adjust the attention allocation ratio between different trajectory points, realizing multi-dimensional collaborative perception of space and time, and outputting a multi-head attention weight matrix. All multi-head attention weight matrices correspond one-to-one to the candidate trajectory sequence.

[0038] S54. Based on the multi-head attention weight matrix, long-span sequence alignment is performed on each candidate trajectory sequence. During the sequence alignment process, the temporal order of trajectory points is dynamically adjusted for each trajectory segment and its context segment using the geographic argument parameters and geographic modulus parameters encoded by the rotated geographic angular distance. The spatial connection structure between trajectory points is also optimized to ensure that the aligned trajectory sequence maintains continuity and rationality in both spatial and temporal dimensions, thus generating a set of aligned trajectory sequences.

[0039] S55. Based on the aligned track sequence set, perform bird individual identity identification based on the geographic argument parameters, geographic modulus parameters, and time tag differences encoded by the rotational geographic angular distance of each track point. By comparing the geographic argument parameters, geographic modulus parameters, and time tag differences between different track points within the same track sequence, the spatial proximity, temporal continuity, and motion consistency of the track points are comprehensively determined. Track points belonging to the same bird are aggregated into a unified identity, resulting in a labeled track identity tag set. The track identity tag set corresponds one-to-one with the aligned track sequence set.

[0040] S56. For the missing segments in the aligned trajectory sequence set, using the geographic argument parameters and geographic modulus parameters encoded by the rotated geographic angular distance and their corresponding time tag differences, calculate the spatial coordinate prediction values and time tag prediction values of the missing segments based on the interpolation prediction method, and add the predicted values to the missing segments to achieve continuity repair of the trajectory sequence and generate a continuous and complete flight path set.

[0041] S57. All continuous and complete flight path sets are uniformly formatted. Each continuous and complete flight path contains time-related spatial coordinates, time labels, bird species labels, track identity labels, and rotational geographic angular distance codes corresponding to each track point, and a continuous flight path dataset is output.

[0042] Optionally, the S6 includes the following steps:

[0043] S61. The continuous flight path dataset is written into the trajectory database, which records the spatial coordinates, time tag, bird species tag, trajectory identity tag, rotational geographic angle encoding and weight parameters of each flight path to form a stored continuous flight path dataset;

[0044] S62. Associate the stored continuous flight path dataset with the bird ecological knowledge graph. Based on the spatial coordinates, time label, and bird species label of each flight path, extract the corresponding habitat type, migratory species, migration time window, and ecological risk level parameters from the bird ecological knowledge graph to generate a path-knowledge graph matching parameter set.

[0045] S63. Associate the stored continuous flight path dataset with the real-time environmental data, extract the corresponding meteorological conditions, surface characteristics, and environmental interference parameters from the real-time environmental data based on the spatial coordinates, time tags, and weight parameters of each flight path, and generate a path-environmental parameter set;

[0046] S64. Based on the path-knowledge graph matching parameter set and the path-environment parameter set, output the cross-regional bird migration monitoring result set, and output all cross-regional bird migration monitoring result sets.

[0047] Optionally, the cross-regional bird migration monitoring result set includes the following result types:

[0048] The habitat type identification result is based on the spatial coordinates of the flight path and the habitat type parameters in the knowledge graph, generating a habitat type label corresponding to each flight path;

[0049] The identification results of high-risk migration areas are based on the flight paths and ecological risk level parameters in the knowledge map, the spatiotemporal distribution monitoring results of meteorological species in the environmental data, and the bird species labels, time labels and spatial coordinates of the flight paths to output the spatiotemporal distribution change trends of each species;

[0050] The migration channel identification results output the migration channel segment and channel activity index based on the spatial overlap and temporal consistency parameters of multiple flight paths.

[0051] A bird flight path intelligent tracking and monitoring system based on big data analysis is used to implement a bird flight path intelligent tracking and monitoring method based on big data analysis, including the following modules:

[0052] Multimodal bird observation data acquisition module, used to collect multimodal bird observation data including images, videos and sounds in real time to generate original bird observation datasets;

[0053] The data preprocessing module is used to perform data preprocessing on the original bird observation dataset, locate and identify bird targets in images and videos, identify the bird species corresponding to the sound signals, and output a standardized bird observation dataset;

[0054] The trajectory segment modeling module is used to build the temporal trajectory of the same bird based on the standardized bird observation dataset by calling the multi-target tracking algorithm, fusing the multimodal recognition results, and constructing the trajectory segment dataset;

[0055] Lion population optimization module, which is used to perform lion population optimization search on the trajectory segment dataset based on cross-regional migration constraints, and obtain candidate trajectory datasets and confidence score matrices using spatial distance weights, time difference weights, species consistency weights, and ecological correction parameters;

[0056] The spatiotemporal Transformer fine-grained alignment module is used to input the candidate trajectory dataset into the spatiotemporal Transformer model configured with rotated geographic angular distance encoding. Based on the rotated geographic angular distance encoding, it performs long-span sequence alignment, bird individual identity identification, and missing segment interpolation to generate a continuous flight path dataset.

[0057] The trajectory evaluation and knowledge fusion module is used to perform error evaluation on the continuous flight path dataset and update the weight parameters based on the evaluation results before writing them into the trajectory database. It associates the stored continuous flight path dataset with the bird ecological knowledge graph and real-time environmental data, and outputs a set of cross-regional bird migration monitoring results.

[0058] The beneficial effects of the present invention are:

[0059] (1) In the lion group optimization search model, the present invention combines the ecological characteristics of geographical zone division, wind direction matching, terrain climbing constraints and consistency of historical migration records to perform multi-level constraints and corrections on the global pairing and local screening of trajectory fragments, realizing multi-target and cross-regional intelligent coarse cascading of trajectory fragments. Compared with the traditional splicing method based on heuristic threshold or Euclidean distance, it can dynamically adapt to the migration behavior of different species, geographical environments and climatic conditions, and effectively reduce the probability of mis-splicing and missed splicing.

[0060] (2) The present invention proposes a rotational geographic angular distance encoding method that combines spatial amplitude parameters and geographic modulus parameters with time label differences, embeds it into a multi-head attention mechanism, and realizes dynamic association in the spatial and temporal dimensions of the trajectory sequence. It can accurately model the flight paths of birds over long distances and under complex geographical conditions. The Transformer can capture long-distance dependencies and periodic migration patterns in the sequence alignment, individual identity identification, and missing segment interpolation links, and achieve reconstruction results with controllable errors and coherent trajectories.

[0061] (3) The present invention combines multi-source observation data (images, sounds, trajectories, and weather) with a dynamic weight update mechanism, and cooperates with knowledge graph association and environmental data fusion to output real-time multi-dimensional monitoring results of bird habitat type discrimination, identification of high-risk migration areas, species spatiotemporal distribution monitoring, and migration channel activity. It shows extremely high adaptability and scalability in actual deployment, and can cover the continuous tracking needs of bird migration under different time periods, across multiple geographical zones, and heterogeneous equipment collection conditions, effectively supporting application scenarios of ecological protection, migration research, and airspace control, and has significant engineering promotion and ecological protection value. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0063] Figure 1 This is a flow chart of a method and system for intelligent tracking and monitoring of bird flight paths based on big data analysis proposed by the present invention. DETAILED DESCRIPTION

[0064] The present invention will now be described in further detail with reference to the accompanying drawings, which are simplified schematic diagrams that illustrate the basic structure of the present invention in a schematic manner.

[0065] refer to Figure 1 , an intelligent tracking and monitoring method for bird flight paths based on big data analysis, comprising the following steps:

[0066] S1. Deploy bird monitoring equipment in multiple geographical areas, including plateaus, mountains, wetlands, and coasts, to collect multimodal bird observation data including images, videos, and sounds in real time and generate original bird observation datasets.

[0067] S2. Perform data preprocessing on the original bird observation dataset. Target detection and recognition algorithms are used to locate and identify bird targets in images and videos. Audio recognition models are used to identify the bird species corresponding to the sound signals. Time and space coordinates are unified based on GPS tags. Low-confidence or duplicate observation records are removed, and outliers are cleaned to output a standardized bird observation dataset.

[0068] S3. Based on a standardized bird observation dataset, a multi-target tracking algorithm is used to establish the temporal trajectory of the same bird in consecutive video frames. The image, video, and sound recognition results are integrated to construct a set of trajectory segments, resulting in a bird trajectory segment dataset.

[0069] S4. Build a lion population optimization search model. Input the bird trajectory segment dataset into the lion population optimization search model. Perform multi-target coarse cascade through global search of male lions, local screening of female lions, and diffusion exploration of cubs. Output a set of candidate trajectories and a confidence matrix to form a candidate trajectory dataset.

[0070] S5. Input the candidate trajectory dataset into a spatiotemporal Transformer model configured with rotational position embedding. Based on the rotational geographic angular distance encoding, long-span sequence alignment, bird individual identity identification, and missing segment interpolation are performed to obtain a continuous and complete set of flight paths, thus generating a continuous flight path dataset.

[0071] S6. Perform error evaluation on the continuous flight path dataset and update the weight parameters based on the evaluation results, then write them into the trajectory database to form a stored continuous flight path dataset. Associate the stored continuous flight path dataset with the bird ecological knowledge map and real-time environmental data to output a set of cross-regional bird migration monitoring results.

[0072] In this embodiment, S2 includes the following steps:

[0073] S21. Perform bird target detection and recognition processing on the image and video data in the original bird observation dataset, and output a set of spatial coordinates of all detected bird targets in each image or video frame. Each spatial coordinate in the set consists of a horizontal coordinate, a vertical coordinate, and a corresponding time tag. The number of spatial coordinate sets is equal to the number of bird targets detected in the image or video frame.

[0074] S22. Use the audio recognition model to identify bird species from the sound data in the original bird observation dataset. Extract the voiceprint features of each sound data segment and match the voiceprint features with known bird species labels to generate an audio recognition result set. Each element in the audio recognition result set includes the bird species label and time label corresponding to the sound data. The number of audio recognition result sets is equal to the number of sound data segments.

[0075] S23. Based on the GPS tags in the original bird observation dataset, perform time normalization and spatial coordinate normalization on all observation data. After each observation data is processed through the spatial coordinate normalization function and the time normalization function, it is unified into standardized observation data. The standardized observation data includes the normalized horizontal coordinate, vertical coordinate, normalized time tag, and observation data category. The number of standardized observation data sets is equal to the total number of observation data.

[0076] S24. Confidence screening and duplicate observation elimination are performed on the standardized observation data set. A confidence threshold is set, and only observation data with a confidence score not less than the confidence threshold is retained. Duplicate observation records with the same time label and spatial coordinates are eliminated. Each element in the confidence screening set is a unique observation data with a confidence score not less than the confidence threshold.

[0077] S25. Clean the confidence filter set of outliers, calculate the Euclidean distance between each observation data and the mean vector in the confidence filter set, retain only the observation data whose Euclidean distance is not higher than the outlier discrimination threshold, and finally output the standardized bird observation dataset.

[0078] In this embodiment, S3 includes the following steps:

[0079] S31. For a standardized bird observation dataset, use a multi-target tracking algorithm to assign an identity to each bird target in consecutive video frames. In each frame, process each spatial coordinate in the spatial coordinate set based on its abscissa, ordinate, and time tag. By comparing the correlation between spatial coordinates and time tags in adjacent frames, associate the spatial coordinates of the same bird in different frames to form an initial set of temporal trajectories. Each temporal trajectory in the initial set consists of several temporally associated spatial coordinates and time tags. The number of temporal trajectories in the initial set equals the total number of trajectories.

[0080] S32. Extract the bird species label and time label corresponding to each segment of sound data from the audio recognition result set in the standardized bird observation dataset, match and associate the bird species labels identified under the same time label with the matching spatial coordinates in the video frame, fuse the image, video, and sound recognition results, and construct a fused observation segment set containing spatial coordinates, time labels, and bird species labels. Each fused observation segment in the fused observation segment set consists of spatial coordinates, time labels, and bird species labels. The number of fused observation segment sets is equal to the number of fused segments.

[0081] S33. Perform trajectory fusion on the initial time-series trajectory set and the fused observation segment set based on a trajectory similarity index function. The trajectory similarity index function is used to measure the comprehensive similarity between the initial time-series trajectory and the fused observation segment in terms of spatial coordinates and species labels. Each trajectory segment in the trajectory segment set consists of a temporally associated spatial coordinate, a time label, and a bird species label. The number of trajectory segment sets is equal to the number of trajectory segments.

[0082] S34. All the trajectory segments in the trajectory segment set are uniformly formatted so that each trajectory segment contains temporally associated spatial coordinates, time labels, and bird species labels, and outputs a bird trajectory segment dataset.

[0083] In this embodiment, the trajectory similarity index function value is obtained by calculating the Euclidean distance between the initial time series trajectory and the fused observation segment in the spatial coordinates, and then calculating the consistency measure of the initial time series trajectory and the fused observation segment in the species label. The Euclidean distance and the consistency measure are multiplied by the corresponding normalized weight coefficients and then accumulated. The result is obtained by taking the inverse of the accumulated value plus one. By maximizing the trajectory similarity index function value, the trajectory and observation segments of the same bird body are fused to obtain a trajectory segment set.

[0084] In this embodiment, S4 includes the following steps:

[0085] S41. Build a lion population optimization search model based on cross-regional migration constraints, taking a bird track segment dataset as input. The bird track segment dataset is partitioned into geographic zones, each of which is assigned a unique number.

[0086] S42. For any two bird track segments, the spatial distance between the two bird track segments, the time tag difference between the two bird track segments, and the consistency parameter between the species tags of the two bird track segments are calculated respectively by assigning spatial distance weights, time difference weights, and species consistency weights. The above three values are weighted and summed according to the spatial distance weights, time difference weights, and species consistency weights to obtain the basic fitness parameter. The basic fitness parameter is used to measure the ecological splicing cost between any two bird track segments. The spatial distance is measured by calculating the geographical distance between the end point of the first bird track segment and the starting point of the second bird track segment. The time tag difference is the absolute value of the end time tag of the first bird track segment and the starting time tag of the second bird track segment. The species consistency parameter is zero when the species tags of the two bird track segments are the same, and one when they are different. The sum of the spatial distance weight, time difference weight, and species consistency weight is one.

[0087] S43. Based on the basic fitness parameters, a wind direction matching parameter, a terrain climbing constraint parameter, and a historical migration record consistency parameter are introduced to reflect the matching of the angle between the wind direction and the flight direction between two bird track segments, the change in terrain elevation between the two bird track segments, and the splicing consistency of the two bird track segments in the historical data, respectively. By assigning correction weights for wind direction matching, terrain climbing, and historical migration record, the wind direction matching parameter, terrain climbing constraint parameter, and historical migration record consistency parameter are added as correction items to the basic fitness parameters to obtain an enhanced fitness parameter. The enhanced fitness parameter is used to measure the comprehensive splicing cost of two bird track segments under ecological and physical constraints. The closer the value of the wind direction matching parameter is to 1, the more consistent the wind direction and flight direction are. The terrain climbing constraint parameter is used to determine whether the terrain elevation difference exceeds the species' flyable range. The historical migration record consistency parameter is used to determine whether the splicing instance exists in the historical data.

[0088] S44. The male lions conduct a global search within the bird track segment dataset. For each geographic zone, all bird track segment pairs within the zone are combined and matched. The enhanced fitness parameter for each bird track segment pair is calculated. The bird track segment pairs with the lowest enhanced fitness parameters are selected as preliminary candidate pairings. The preliminary candidate pairings constitute the global candidate pairing set.

[0089] S45. Lionesses introduce a migration ecological filtering mechanism based on the global candidate pairing set. For each candidate pairing, the mechanism sequentially determines whether its time tag interval is less than or equal to the maximum allowable migration time jump threshold, whether its wind direction compatibility parameter is greater than or equal to the wind assistance effect lower limit, and whether its terrain climbing constraint parameter is equal to zero. Only when all conditions are met is the pairing retained as a locally valid pairing. All locally valid pairs constitute the local candidate pairing set.

[0090] S46. The lion cubs introduce temporal and spatial perturbations to the unmatched bird track segments in the bird track segment dataset, randomly perturbing the time tags and spatial coordinates, respectively, to generate new combinations. The enhanced fitness parameter of each combination is calculated. All new combinations are screened using the same criteria as above, and valid pairs are retained. All valid pairs constitute the perturbation expansion candidate set.

[0091] S47. Splice the local candidate pair set with all valid pairs in the perturbed and expanded candidate set in the order of their time tags to obtain complete candidate trajectories. Each complete candidate trajectory is composed of several sequentially spliced bird trajectory segments. During the trajectory splicing process, the enhanced fitness parameters between each group of bird trajectory segments are recorded. The arithmetic mean of all enhanced fitness parameters is used as the average path splicing cost of the complete candidate trajectory.

[0092] S48. For each complete candidate trajectory, the average path splicing cost is converted into a confidence parameter using a negative normalized exponential function based on the average path splicing cost. The normalization factor is used to adjust the distribution range of the confidence parameter. The confidence parameters of all complete candidate trajectories form a scoring matrix. All complete candidate trajectories and the scoring matrix are output to form a candidate trajectory dataset.

[0093] In this embodiment, S5 includes the following steps:

[0094] S51. The candidate trajectory dataset consists of several candidate trajectories. Each candidate trajectory is composed of several sequentially spliced bird trajectory segments. Each bird trajectory segment contains spatial coordinates, a time label, and a bird species label. The candidate trajectory dataset is converted into a sequence structure recognizable by the model to obtain a set of candidate trajectory sequences. Each candidate trajectory sequence is a time-ordered sequence.

[0095] S52. For the candidate trajectory sequence set, calculate the spatial variation based on the longitude and latitude of each trajectory point, and convert the spatial variation into polar coordinates to obtain the geographic argument parameter and geographic modulus parameter, respectively. The geographic argument parameter is used to measure the spatial angular distance between two trajectory points, and the geographic modulus parameter is used to measure the geographic distance between two trajectory points. The geographic argument parameter and geographic modulus parameter are combined with the time tag difference to form a rotated geographic angular distance code. The rotated geographic angular distance code is embedded as the spatial and temporal features of the candidate trajectory sequence, and the rotated geographic angular distance code is one-to-one corresponding to each trajectory point.

[0096] S53. The candidate trajectory sequence and its corresponding rotated geographic angular distance encoding are jointly input into a spatiotemporal Transformer model configured with a rotation position embedding. The spatiotemporal Transformer model directly uses the geographic argument parameters and geographic modulus parameters of the rotated geographic angular distance encoding in each attention head to measure the spatial relationship between trajectory points, and combines the time label difference as a measure of the temporal relationship. During the calculation of the attention weight, the geographic argument parameters and geographic modulus parameters of the rotated geographic angular distance encoding are used to dynamically adjust the attention allocation ratio between different trajectory points, realizing multi-dimensional collaborative perception of space and time, and outputting a multi-head attention weight matrix. All multi-head attention weight matrices correspond one-to-one to the candidate trajectory sequence.

[0097] S54. Based on the multi-head attention weight matrix, long-span sequence alignment is performed on each candidate trajectory sequence. During the sequence alignment process, the temporal order of trajectory points is dynamically adjusted for each trajectory segment and its context segment using the geographic argument parameters and geographic modulus parameters encoded by the rotated geographic angular distance. The spatial connection structure between trajectory points is also optimized to ensure that the aligned trajectory sequence maintains continuity and rationality in both spatial and temporal dimensions, thus generating a set of aligned trajectory sequences.

[0098] S55. Based on the aligned track sequence set, perform bird individual identity identification based on the geographic argument parameters, geographic modulus parameters, and time tag differences encoded by the rotational geographic angular distance of each track point. By comparing the geographic argument parameters, geographic modulus parameters, and time tag differences between different track points within the same track sequence, the spatial proximity, temporal continuity, and motion consistency of the track points are comprehensively determined. Track points belonging to the same bird are aggregated into a unified identity, resulting in a labeled track identity tag set. The track identity tag set corresponds one-to-one with the aligned track sequence set.

[0099] S56. For the missing segments in the aligned trajectory sequence set, using the geographic argument parameters and geographic modulus parameters encoded by the rotated geographic angular distance and their corresponding time tag differences, calculate the spatial coordinate prediction values and time tag prediction values of the missing segments based on the interpolation prediction method, and add the predicted values to the missing segments to achieve continuity repair of the trajectory sequence and generate a continuous and complete flight path set.

[0100] S57. All continuous and complete flight path sets are uniformly formatted. Each continuous and complete flight path contains time-related spatial coordinates, time labels, bird species labels, track identity labels, and rotational geographic angular distance codes corresponding to each track point, and a continuous flight path dataset is output.

[0101] In this embodiment, S6 includes the following steps:

[0102] S61. The continuous flight path dataset is written into the trajectory database, which records the spatial coordinates, time tag, bird species tag, trajectory identity tag, rotational geographic angle encoding and weight parameters of each flight path to form a stored continuous flight path dataset;

[0103] S62. Associate the stored continuous flight path dataset with the bird ecological knowledge graph. Based on the spatial coordinates, time label, and bird species label of each flight path, extract the corresponding habitat type, migratory species, migration time window, and ecological risk level parameters from the bird ecological knowledge graph to generate a path-knowledge graph matching parameter set.

[0104] S63. Associate the stored continuous flight path dataset with the real-time environmental data, extract the corresponding meteorological conditions, surface characteristics, and environmental interference parameters from the real-time environmental data based on the spatial coordinates, time tags, and weight parameters of each flight path, and generate a path-environmental parameter set;

[0105] S64. Based on the path-knowledge graph matching parameter set and the path-environment parameter set, output the cross-regional bird migration monitoring result set, and output all cross-regional bird migration monitoring result sets.

[0106] In this implementation, the cross-regional bird migration monitoring result set includes the following result types:

[0107] The habitat type identification result is based on the spatial coordinates of the flight path and the habitat type parameters in the knowledge graph, generating a habitat type label corresponding to each flight path;

[0108] The identification results of high-risk migration areas are based on the flight paths and ecological risk level parameters in the knowledge map, the spatiotemporal distribution monitoring results of meteorological species in the environmental data, and the bird species labels, time labels and spatial coordinates of the flight paths to output the spatiotemporal distribution change trends of each species;

[0109] The migration channel identification results output the migration channel segment and channel activity index based on the spatial overlap and temporal consistency parameters of multiple flight paths.

[0110] A bird flight path intelligent tracking and monitoring system based on big data analysis is used to implement a bird flight path intelligent tracking and monitoring method based on big data analysis, including the following modules:

[0111] Multimodal bird observation data acquisition module, used to collect multimodal bird observation data including images, videos and sounds in real time to generate original bird observation datasets;

[0112] The data preprocessing module is used to perform data preprocessing on the original bird observation dataset, locate and identify bird targets in images and videos, identify the bird species corresponding to the sound signals, and output a standardized bird observation dataset;

[0113] The trajectory segment modeling module is used to build the temporal trajectory of the same bird based on the standardized bird observation dataset by calling the multi-target tracking algorithm, fusing the multimodal recognition results, and constructing the trajectory segment dataset;

[0114] Lion population optimization module, which is used to perform lion population optimization search on the trajectory segment dataset based on cross-regional migration constraints, and obtain candidate trajectory datasets and confidence score matrices using spatial distance weights, time difference weights, species consistency weights, and ecological correction parameters;

[0115] The spatiotemporal Transformer fine-grained alignment module is used to input the candidate trajectory dataset into the spatiotemporal Transformer model configured with rotated geographic angular distance encoding. Based on the rotated geographic angular distance encoding, it performs long-span sequence alignment, bird individual identity identification, and missing segment interpolation to generate a continuous flight path dataset.

[0116] The trajectory evaluation and knowledge fusion module is used to perform error evaluation on the continuous flight path dataset and update the weight parameters based on the evaluation results before writing them into the trajectory database. It associates the stored continuous flight path dataset with the bird ecological knowledge graph and real-time environmental data, and outputs a set of cross-regional bird migration monitoring results.

[0117] Example 1:

[0118] During continuous monitoring, the system received data samples uploaded by multiple observation terminals. One of the data fragments was an individual bird trajectory data fragment numbered "P000321." The data fragment contained 52 time-series trajectory points. Each trajectory point recorded spatial coordinates ([119.028, 32.108]), a time tag ("2023-03-15 10:23:01"), and the bird species tag "Species-A." The single-point observation confidence level was greater than 0.82. However, the data fragment was missing two time intervals in the uploaded batch, 2.7 hours and 4.3 hours, respectively, indicating spatial drift. Traditional methods, using the nearest neighbor principle, spliced P000321 with two other trajectory fragments numbered "P000347" and "P000412." The resulting overall spatial error reached 2.16 kilometers, and some identity tags were duplicated, resulting in decreased trajectory integrity and individual consistency.

[0119] After using the method of the present invention, the P000321 segment and all uploaded trajectory segments were first evaluated for basic ecological constraint fitness. The system calculated the weighted scores of P000321 and its candidate splicing segments in terms of spatial distance, time tag difference, species tag consistency, wind direction parameter, and terrain climbing constraint. For the splicing attempt of P000321 and P000347, the spatial distance parameter was 0.97 kilometers, the time tag difference was 1.8 hours, the species tag was consistent, the wind direction matching parameter was 0.87, the terrain climbing constraint parameter was 0, and the final enhanced fitness parameter was 0.52. For the splicing attempt of P000321 and P000412, the spatial distance parameter was 2.13 kilometers, the time tag difference was 3.9 hours, the species was consistent, the wind direction matching parameter was 0.51, the terrain climbing constraint parameter was 0.23, and the enhanced fitness parameter was 1.48. The system prioritizes retaining P000347 with the lowest fitness parameter as the next segment for splicing, and automatically eliminates the splicing path of P000412.

[0120] Subsequently, the system interpolated and repaired the 2.7 hours of missing observations in the spliced segments of P000321 and P000347 using a spatiotemporal Transformer module with rotated geographic angular distance encoding. The interpolation module converts the spatial polar coordinate changes of the preceding and succeeding trajectory points into geographic argument parameters (e.g., 62.7 degrees and 64.3 degrees), geographic modulus parameters (e.g., 1.22 kilometers and 1.01 kilometers), and a time tag difference (2.7 hours), dynamically adjusting the spatial position of the interpolated points in the sequence. After interpolation, three trajectory interpolation points were generated, with spatial errors of 0.92 kilometers, 1.08 kilometers, and 1.17 kilometers, respectively, which are lower than the average of 2.01 kilometers for the traditional Transformer method.

[0121] During the identity identification phase, the system compared the complete trajectory with the identity feature vectors of historical migratory individuals in the database. The spatial trajectory similarity threshold was 0.91, and the species label consistency threshold was 1. Ultimately, P000321, P000347, and all trajectory points in the interpolation section were uniformly labeled as the same bird individual "IDX_1876."

[0122] After the full process of splicing and repair is completed, the system automatically evaluates the average spatial error of the continuous flight path, and the result is 1.07 kilometers, which is lower than the upper limit of 2.0 kilometers set by the system. The trajectory confidence weight is increased to 0.93 and stored in the trajectory database.

[0123] The knowledge graph, triggered synchronously with real-time environmental data, detected that the trajectory coincided with a meteorological parameter with wind speeds exceeding 10.5 m / s in a specific section (trajectory points 27-41). The ecological knowledge graph marked this section as a potential high-risk migration area, automatically outputting the high-risk area label in the monitoring report and associating the risk factor with a sudden change in wind speed. Furthermore, based on historical data on similar individuals, the system determined that the habitat type continuously passed through by the trajectory was river wetland-coastal mudflat, and recorded the species' typical migration corridor for the current year.

[0124] Comparative experiments showed that when the traditional nearest neighbor + absolute position encoding Transformer was used to process this batch of data, the average spatial error of the P000321 splicing was 2.23 kilometers, the breakpoint interpolation error was 2.14 kilometers, the number of identity label misassignments was 2, the habitat risk zone recall rate was 0.81, and the splicing time was 37.1 seconds. After processing with the method of the present invention, the average spatial error of the splicing of the same batch was 1.07 kilometers, the breakpoint interpolation error was 1.06 kilometers, the identity label misassignment was 0, the risk zone recall rate was 0.93, and the splicing time was 19.2 seconds.

[0125] Another track fragment, sample number P001765, was spliced using traditional methods. Due to mis-splicing, the overall time series of the track was misaligned and individual identities switched, preventing the system from forming a complete cross-regional path. However, the proposed method automatically eliminated unreasonable paths and ultimately generated a high-confidence path spanning four geographic zones and covering 78 consecutive track points. The mean spatial error was 1.22 kilometers, and identity consistency reached 100%.

[0126] In summary, in the entire continuous monitoring process, the method of the present invention can automatically complete the normalization, fragment splicing, missing repair and risk area identification of bird trajectory data under massive, multi-source and complex conditions. The core indicators are significantly better than traditional solutions. The data of the entire process are recorded in the database and are traceable and reviewable, proving that the present invention not only improves the splicing accuracy, interpolation accuracy and identity consistency, but also greatly enhances the ecological risk identification and real-time monitoring capabilities.

[0127] In the lion group optimization search model, the present invention combines the ecological characteristics of geographical zone zoning, wind direction matching, terrain climbing constraints and consistency of historical migration records to perform multi-level constraints and corrections on the global pairing and local screening of trajectory fragments, realizing multi-target and cross-regional intelligent coarse cascading of trajectory fragments. Compared with traditional splicing methods based on heuristic thresholds or Euclidean distances, it can dynamically adapt to the migration behavior of different species, geographical environments and climatic conditions, and effectively reduce the probability of mis-splicing and missed splicing.

[0128] This paper proposes a rotational geographic angular distance encoding method that combines spatial argument parameters and geographic modulus parameters with time label differences, embeds it into a multi-head attention mechanism, and realizes dynamic association in the spatial and temporal dimensions of trajectory sequences. It can accurately model the flight paths of birds over long distances and under complex geographical conditions. The Transformer can capture long-distance dependencies and periodic migration patterns in sequence alignment, individual identity identification, and missing segment interpolation, achieving reconstruction results with controllable errors and coherent trajectories.

[0129] The present invention combines multi-source observation data (images, sounds, trajectories, and weather) with a dynamic weight update mechanism, and cooperates with knowledge graph association and environmental data fusion to output real-time multi-dimensional monitoring results of bird habitat type discrimination, identification of high-risk migration areas, species spatiotemporal distribution monitoring, and migration channel activity. It shows extremely high adaptability and scalability in actual deployment, and can cover the continuous tracking needs of bird migration under different time periods, across multiple geographical zones, and heterogeneous equipment collection conditions, effectively supporting application scenarios of ecological protection, migration research, and airspace control, and has significant engineering promotion and ecological protection value.

[0130] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A method for intelligent tracking and monitoring of bird flight paths based on big data analysis, characterized in that: The steps include: S1. Collect multimodal bird observation data including images, videos, and sounds in real time to generate an original bird observation dataset; S2. Perform data preprocessing on the original bird observation dataset and output a standardized bird observation dataset; S3. Based on the standardized bird observation dataset, a multi-target tracking algorithm is used to establish the temporal trajectory of the same bird in consecutive video frames, thus obtaining a bird trajectory segment dataset. S4. Build a lion group optimization search model, input the bird trajectory segment dataset into the lion group optimization search model to form a candidate trajectory dataset; S5. Input the candidate trajectory dataset into a spatiotemporal Transformer model configured with rotational position embedding. Based on the rotational geographic angular distance encoding, long-span sequence alignment, bird individual identity identification, and missing segment interpolation are performed to obtain a continuous and complete set of flight paths, generating a continuous flight path dataset. S6. Form a stored continuous flight path dataset based on the continuous flight path dataset, associate the stored continuous flight path dataset with the bird ecological knowledge map and real-time environmental data, and output a set of cross-regional bird migration monitoring results.

2. The method for intelligent tracking and monitoring of bird flight paths based on big data analysis according to claim 1, characterized in that: The S2 comprises the following steps: S21. Perform bird target detection and recognition processing on the image and video data in the original bird observation dataset, and output the spatial coordinate set of all detected bird targets in each image frame or video frame; S22. Use the audio recognition model to identify bird species from the sound data in the original bird observation dataset. Extract the voiceprint features of each sound segment and match the voiceprint features with known bird species labels to generate a set of audio recognition results. S23. Based on the GPS tags in the original bird observation dataset, perform time normalization and spatial coordinate normalization on all observation data to obtain a standardized observation data set; S24. Confidence screening and duplicate observation elimination are performed on the standardized observation data set to obtain a confidence screening set; S25. Clean the confidence filter set of outliers, calculate the Euclidean distance between each observation data and the mean vector in the confidence filter set, retain only the observation data whose Euclidean distance is not higher than the outlier discrimination threshold, and finally output the standardized bird observation dataset.

3. The method for intelligent tracking and monitoring of bird flight paths based on big data analysis according to claim 2, characterized in that: The S3 includes the following steps: S31. For a standardized bird observation dataset, use a multi-target tracking algorithm to assign an identity to each bird target in consecutive video frames. In each frame, process each spatial coordinate in the spatial coordinate set based on its abscissa, ordinate, and time tag. By comparing the correlation between spatial coordinates and time tags in adjacent frames, associate the spatial coordinates of the same bird in different frames with their identities to form an initial time-series trajectory set. S32. Extract the bird species label and time label corresponding to each segment of sound data from the audio recognition result set in the standardized bird observation dataset. Match and associate the bird species labels identified under the same time label with the matching spatial coordinates in the video frame. Fuse the image, video, and sound recognition results to construct a fused observation segment set containing spatial coordinates, time labels, and bird species labels. S33. For the initial time series trajectory set and the fused observation segment set, trajectory fusion is performed according to the trajectory similarity index function to obtain a trajectory segment set; S34. All the trajectory segments in the trajectory segment set are uniformly formatted so that each trajectory segment contains temporally associated spatial coordinates, time labels, and bird species labels, and outputs a bird trajectory segment dataset.

4. The method for intelligent tracking and monitoring of bird flight paths based on big data analysis according to claim 3, characterized in that: The trajectory similarity index function value is obtained by calculating the Euclidean distance between the initial time series trajectory and the fused observation segment in spatial coordinates, calculating the consistency measure of the initial time series trajectory and the fused observation segment in species label, multiplying the Euclidean distance and the consistency measure by the corresponding normalized weight coefficient and then accumulating them. The result is obtained by taking the inverse of the accumulated value plus one. By maximizing the trajectory similarity index function value, the trajectory and observation segments of the same bird body are fused to obtain a trajectory segment set.

5. The method for intelligent tracking and monitoring of bird flight paths based on big data analysis according to claim 1, characterized in that: The S4 comprises the following steps: S41. Build a lion population optimization search model based on cross-regional migration constraints, taking a bird track segment dataset as input. The bird track segment dataset is partitioned into geographic zones, each of which is assigned a unique number. S42. For any two bird track segments, calculate the spatial distance between the two bird track segments, the time label difference between the two bird track segments, and the consistency parameter between the species labels of the two bird track segments by assigning spatial distance weights, time difference weights, and species consistency weights, and perform a weighted sum of these three values according to the spatial distance weights, time difference weights, and species consistency weights to obtain the basic fitness parameter; S43. Based on the basic fitness parameters, by assigning wind direction matching correction weights, terrain climbing correction weights, and historical migration record correction weights respectively, the wind direction matching parameter, terrain climbing constraint parameter, and historical migration record consistency parameter are added as correction items to the basic fitness parameters to obtain enhanced fitness parameters; S44. The male lions conduct a global search within the bird track segment dataset. For each geographic zone, all bird track segment pairs within the zone are combined and matched. The enhanced fitness parameter for each bird track segment pair is calculated. The bird track segment pairs with the lowest enhanced fitness parameters are selected as preliminary candidate pairings. The preliminary candidate pairings constitute the global candidate pairing set. S45. Lionesses introduce a migration ecological filtering mechanism based on the global candidate pairing set. For each candidate pairing, the mechanism sequentially determines whether its time tag interval is less than or equal to the maximum allowable migration time jump threshold, whether its wind direction compatibility parameter is greater than or equal to the wind assistance effect lower limit, and whether its terrain climbing constraint parameter is equal to zero. Only when all conditions are met is the pairing retained as a locally valid pairing. All locally valid pairs constitute the local candidate pairing set. S46. The lion cubs introduce temporal and spatial perturbations to the unmatched bird track segments in the bird track segment dataset, randomly perturbing the time tags and spatial coordinates, respectively, to generate new combinations. The enhanced fitness parameter of each combination is calculated. All new combinations are screened using the same criteria as above, and valid pairs are retained. All valid pairs constitute the perturbation expansion candidate set. S47. Splice the local candidate pair set with all valid pairs in the perturbed and expanded candidate set in the order of their time tags to obtain a complete candidate trajectory. During the trajectory splicing process, record the enhanced fitness parameters between each set of bird trajectory segments. The arithmetic mean of all enhanced fitness parameters is used as the average path splicing cost of the complete candidate trajectory. S48. For each complete candidate trajectory, the average path splicing cost is converted into a confidence parameter using a negative normalized exponential function based on the average path splicing cost. The confidence parameters of all complete candidate trajectories form a scoring matrix to form a candidate trajectory dataset.

6. The method for intelligent tracking and monitoring of bird flight paths based on big data analysis according to claim 5, characterized in that: The S5 comprises the following steps: S51. The candidate trajectory dataset consists of several candidate trajectories. The candidate trajectory dataset is converted into a sequence structure recognizable by the model to obtain a candidate trajectory sequence set. S52. For the candidate trajectory sequence set, calculate the spatial variation based on the longitude and latitude of each trajectory point, and convert the spatial variation into polar coordinates to obtain the geographic argument parameter and geographic modulus parameter, and together with the time tag difference, form the rotational geographic angular distance code; S53. The candidate trajectory sequence and its corresponding rotated geographic angular distance encoding are jointly input into a spatiotemporal Transformer model configured with a rotational position embedding. The spatiotemporal Transformer model directly uses the geographic argument parameters and geographic modulus parameters of the rotated geographic angular distance encoding in each attention head to measure the spatial relationship between trajectory points, and combines the time label difference as a measure of the temporal relationship. During the calculation of the attention weight, the geographic argument parameters and geographic modulus parameters of the rotated geographic angular distance encoding are used to dynamically adjust the attention allocation ratio between different trajectory points, and output a multi-head attention weight matrix. S54. Based on the multi-head attention weight matrix, perform long-span sequence alignment on each candidate trajectory sequence. During the sequence alignment process, for each trajectory segment and its context segment, use the geographic argument parameter and geographic modulus parameter of the rotated geographic angle encoding to dynamically adjust the temporal order of the trajectory points, and optimize the spatial connection structure between the trajectory points to generate an aligned trajectory sequence set. S55. Based on the aligned track sequence set, perform bird individual identity identification. By comparing the geographic argument parameters, geographic modulus parameters, and time tag differences between different track points within the same track sequence, comprehensively determine the spatial proximity, temporal continuity, and motion consistency of the track points. Track points belonging to the same bird are aggregated into a unified identity, resulting in a labeled track identity tag set. S56. For the missing segments in the aligned trajectory sequence set, using the geographic argument parameters and geographic modulus parameters encoded by the rotated geographic angular distance and their corresponding time tag differences, calculate the spatial coordinate prediction values and time tag prediction values of the missing segments based on the interpolation prediction method, and add the predicted values to the missing segments to generate a continuous and complete flight path set. S57. All continuous and complete flight path sets are uniformly formatted and output as a continuous flight path dataset.

7. The method for intelligent tracking and monitoring of bird flight paths based on big data analysis according to claim 6, characterized in that: The S6 comprises the following steps: S61. The continuous flight path dataset is written into the trajectory database, which records the spatial coordinates, time tag, bird species tag, trajectory identity tag, rotational geographic angle encoding and weight parameters of each flight path to form a stored continuous flight path dataset; S62. Associate the stored continuous flight path dataset with the bird ecological knowledge graph. Based on the spatial coordinates, time label, and bird species label of each flight path, extract the corresponding habitat type, migratory species, migration time window, and ecological risk level parameters from the bird ecological knowledge graph to generate a path-knowledge graph matching parameter set. S63. Associate the stored continuous flight path dataset with the real-time environmental data, extract the corresponding meteorological conditions, surface characteristics, and environmental interference parameters from the real-time environmental data based on the spatial coordinates, time tags, and weight parameters of each flight path, and generate a path-environmental parameter set; S64. Based on the path-knowledge graph matching parameter set and the path-environment parameter set, output the cross-regional bird migration monitoring result set, and output all cross-regional bird migration monitoring result sets.

8. The method for intelligent tracking and monitoring of bird flight paths based on big data analysis according to claim 7, characterized in that: The cross-regional bird migration monitoring result set includes the following result types: The habitat type identification result is based on the spatial coordinates of the flight path and the habitat type parameters in the knowledge graph, generating a habitat type label corresponding to each flight path; The identification results of high-risk migration areas are based on the flight paths and ecological risk level parameters in the knowledge map, the spatiotemporal distribution monitoring results of meteorological species in the environmental data, and the bird species labels, time labels and spatial coordinates of the flight paths to output the spatiotemporal distribution change trends of each species; The migration channel identification results output the migration channel segment and channel activity index based on the spatial overlap and temporal consistency parameters of multiple flight paths.

9. A bird flight path intelligent tracking and monitoring system based on big data analysis, used to implement the bird flight path intelligent tracking and monitoring method based on big data analysis according to any one of claims 1 to 8, characterized in that: Includes the following modules: Multimodal bird observation data acquisition module, used to collect multimodal bird observation data including images, videos and sounds in real time to generate original bird observation datasets; The data preprocessing module is used to perform data preprocessing on the original bird observation dataset, locate and identify bird targets in images and videos, identify the bird species corresponding to the sound signals, and output a standardized bird observation dataset; The trajectory segment modeling module is used to build the temporal trajectory of the same bird based on the standardized bird observation dataset by calling the multi-target tracking algorithm, fusing the multimodal recognition results, and constructing the trajectory segment dataset; Lion population optimization module, which is used to perform lion population optimization search on the trajectory segment dataset based on cross-regional migration constraints, and obtain candidate trajectory datasets and confidence score matrices using spatial distance weights, time difference weights, species consistency weights, and ecological correction parameters; The spatiotemporal Transformer fine-grained alignment module is used to input the candidate trajectory dataset into the spatiotemporal Transformer model configured with rotated geographic angular distance encoding. Based on the rotated geographic angular distance encoding, it performs long-span sequence alignment, bird individual identity identification, and missing segment interpolation to generate a continuous flight path dataset. The trajectory evaluation and knowledge fusion module is used to perform error evaluation on the continuous flight path dataset and update the weight parameters based on the evaluation results before writing them into the trajectory database. It associates the stored continuous flight path dataset with the bird ecological knowledge graph and real-time environmental data, and outputs a set of cross-regional bird migration monitoring results.

Images