Bird distance assessment method and system based on bird detection radar

Abstract

The present invention relates to the technical field of bird distance assessment, and specifically to a bird distance assessment method and system based on a bird detection radar. In the present invention, a data acquisition module obtains the center position of a key area, the type of bird, and the bird position, bird speed and bird acceleration at past time points; a past fluctuation analysis module calculates the past distance between the key area and the bird, and assesses the fluctuation of the bird's motion trajectory. If it is abnormal, a distance correction module is introduced for adjustment; a future fluctuation analysis module uses the bird's speed and bird acceleration to predict the bird's future position, calculates the future distance and performs stability analysis; if the future fluctuation analysis is abnormal, the distance correction module is also used for correction, and a correction factor is calculated by a linear regression function combined with maximum speed and type characteristics to ensure consistency between the prediction and actual data; this scheme effectively solves the problem of real-time monitoring of the dynamic behavior of birds and the accuracy of its prediction, and improves the stability and responsiveness of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of bird distance assessment, and in particular to a bird distance assessment method and system based on a bird detection radar. Background Art

[0002] Traditional bird distance assessment methods and systems have several technical bottlenecks and deficiencies, which affect their application and accuracy in dynamic environments.

[0003] First, traditional systems usually rely on fixed or static data collection mechanisms, which can only provide instant distance measurements but cannot perform dynamic behavior analysis. This limits the system's ability to respond to fast-moving or irregular trajectories of birds, resulting in poor accuracy in identifying rapidly changing scenes. In addition, these systems rely on simple geometric measurements to assess the distance between birds and key areas without fully considering the changes in motion over time, such as speed and acceleration, which makes the prediction of future positions lack scientific basis and rationality. In other words, it is impossible to provide effective warnings or adjust strategies when birds deviate from their established tracks.

[0004] More importantly, traditional distance assessment methods often ignore the diversity of bird species and the movement characteristics of different species, which leads to insufficient adaptability when dealing with cross-species data or interactive ecological dynamics. The lack of a model based on the actual behavioral characteristics of birds means that it is difficult for the system to provide refined analysis and customized responses in data processing and decision support. Once there are errors or anomalies in the data, traditional methods usually lack an effective correction mechanism and cannot correct prediction errors by analyzing historical behavior patterns. It is even more difficult to cope with dynamic differences caused by environmental changes or autonomous behavior of birds. Therefore, this requires more advanced systems that can analyze and adjust in multiple dimensions and levels to overcome the above technical difficulties and improve real-time performance, prediction capabilities and data correction accuracy. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for evaluating the distance of flying birds based on a bird detecting radar, so as to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, one of the objects of the present invention is to provide a flying bird distance assessment system based on a bird detection radar, comprising a data acquisition module, a past fluctuation analysis module, a future fluctuation analysis module and a distance correction module, wherein:

[0007] The data acquisition module acquires the radar data of the flying bird and the center position of the key area in real time, and performs feature extraction on the radar data, wherein the feature extraction results include the flying bird type and the flying bird position, flying bird speed and flying bird acceleration at a past time point;

[0008] The past fluctuation analysis module calculates the distance between the center position of the key area and the position of the bird at the past time point according to the position of the bird at the past time point, as the distance at the past time point, and performs past fluctuation analysis on the distance at the past time point, wherein: if the past fluctuation analysis is normal, then enter the future fluctuation analysis module, otherwise, then enter the distance correction module, and re-perform past fluctuation analysis on the corrected result, and if an abnormality occurs at this time, then issue a past fluctuation warning;

[0009] The future fluctuation analysis module predicts the position of the bird at a future time point based on the bird's position, bird speed and bird acceleration at a past time point, and calculates the distance between the center position of the key area and the bird's position at a future time point as the distance at the future time point, and performs future fluctuation analysis on the distance at the future time point, wherein: if the future fluctuation analysis is normal, the future time point distance is output; otherwise, the distance correction module is entered, and the future fluctuation analysis is performed again on the corrected result, and if an abnormality occurs at this time, a future fluctuation warning is issued;

[0010] The distance correction module calculates the linear regression function of the bird using a linear regression algorithm based on the bird's position at a past time point; obtains the maximum bird speed based on the bird's speed at a past time point; calculates a correction factor based on the linear regression function of the bird, the maximum bird speed and the bird type; and corrects the distance of a past time point or a future time point based on the correction factor.

[0011] As a further improvement of this technical solution, the past fluctuation analysis module uses the Euclidean distance calculation method to calculate the distance between the center position of the key area and the bird position at the past time point as the past time point distance. Indicates that Indicates a point in time in the past.

[0012] As a further improvement of this technical solution, the past fluctuation analysis module analyzes the distance between past time points. The process of performing fluctuation analysis is as follows:

[0013] Assess volatility by calculating standard deviation and calculate distance from past time points Standard Deviation The formula is as follows:

[0014] ,in is the total number of past time points, is the average distance of past time points;

[0015] Setting volatility thresholds ,like , it means that the past fluctuation analysis is normal, then enter the future fluctuation analysis module; if , it means that the past fluctuation analysis is abnormal, then enter the distance correction module, and the distance of the corrected past time point The same past fluctuation analysis is performed. If the conditions for abnormal past fluctuation analysis are met again, a past fluctuation warning is issued.

[0016] As a further improvement of the technical solution, the process of the future fluctuation analysis module calculating the distance between the center position of the key area and the position of the bird at a future time point specifically includes:

[0017] Set a future time point , based on the bird's position at a past time point , Bird Speed and bird acceleration , predict the position of the bird at a future time point ,in The calculation formula is as follows:

[0018] ;

[0019] For each future time point , calculate the position of the bird at the future time point Central location to key areas Euclidean distance ,in As the distance to a future time point.

[0020] As a further improvement of the technical solution, the process of the future fluctuation analysis module performing future fluctuation analysis specifically includes:

[0021] Distance to future time point Conduct future volatility analysis, evaluate volatility by calculating standard deviation, and calculate distance to future time points Standard Deviation The formula is as follows:

[0022] ,in is the total number of future time points, is the average distance to future time points;

[0023] Analysis of volatility thresholds in the module based on past volatility , conduct future volatility analysis, where: , it means that the future fluctuation analysis is normal, then the distance of the future time point Output as the output result of the system; if , it means that the future fluctuation analysis is abnormal, then enter the distance correction module, and the distance of the future time point after correction The same future volatility analysis is performed. If the conditions for abnormal future volatility analysis are met again, a future volatility warning is issued.

[0024] As a further improvement of the technical solution, the process of the distance correction module correcting the distance of the past time point or the distance of the future time point using the correction factor specifically includes:

[0025] Based on the position of the bird at a past time point , establish a linear regression function for each coordinate component, including , and ,in , , , , and They represent the parameters of the linear regression function established for the corresponding coordinate components respectively;

[0026] The parameters of the linear regression function are calculated using the least square method, and the parameters of the linear regression function on the corresponding coordinate components are determined by minimizing the sum of squares of the error between the output value of the linear regression function and the actual bird position at the past time point;

[0027] Based on the speed of the bird at a past time point , get the maximum speed ,in ;

[0028] Based on the linear regression function of the bird Parameters, maximum bird speed and bird species, calculate the correction factor , the specific formula is as follows:

[0029] ,in , It is a kind of flying bird;

[0030] Finally, the correction factor is multiplied by the distance of the past time point or the distance of the future time point to perform correction.

[0031] The second object of the present invention is to provide a method for using a flying bird distance assessment system based on a bird detection radar, comprising the following method steps:

[0032] S1. Obtain radar data of flying birds and the center position of key areas in real time, and perform feature extraction on the radar data. The feature extraction results include the flying bird species and the flying bird position, flying bird speed, and flying bird acceleration at past time points.

[0033] S2. According to the bird position at the past time point, the distance between the center position of the key area and the bird position at the past time point is calculated as the past time point distance, and the past fluctuation analysis is performed on the past time point distance. If the past fluctuation analysis is normal, enter S3; otherwise, enter S4, and re-analyze the past fluctuation on the corrected result. If an abnormality occurs at this time, a past fluctuation warning is issued;

[0034] S3, predict the position of the bird at the future time point based on the bird's position, speed and acceleration at the past time point, and calculate the distance between the center of the key area and the bird's position at the future time point as the distance at the future time point, and perform future fluctuation analysis on the distance at the future time point, wherein: if the future fluctuation analysis is normal, the future time point distance is output; otherwise, enter S4, and re-perform future fluctuation analysis on the corrected result, and if an abnormality occurs at this time, a future fluctuation warning is issued;

[0035] S4. According to the position of the bird at the past time point, the linear regression function of the bird is calculated using a linear regression algorithm; and according to the speed of the bird at the past time point, the maximum speed of the bird is obtained; based on the linear regression function of the bird, the maximum speed of the bird and the type of bird, a correction factor is calculated; and the distance of the past time point or the distance of the future time point is corrected according to the correction factor.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The bird distance assessment method and system based on bird detection radar can track the position changes and dynamic behaviors of birds in real time by calculating the distance between the center position of the key area and the position of the bird at the past time point. This distance calculation is used to detect the stability and consistency of the bird's motion trajectory in the past fluctuation analysis to ensure the accuracy of the monitoring data. On this basis, by considering the speed and acceleration of the bird at the past time point to predict the future position, the system can not only verify the current data, but also predict the movement state of the bird in the future in advance. This prediction function is particularly important for the management and risk warning of key areas, and can provide early warning when the bird approaches or deviates from the key area, helping to formulate more appropriate management and intervention strategies. This ability to make future predictions based on historical data improves the dynamic response capability and foresight of the system.

[0038] In addition, by using the bird positions at past time points, as well as the bird speeds and bird species at past time points, a mechanism for data correction is jointly constructed using the bird linear regression function obtained by the linear regression algorithm and the correction factor calculated by combining the speed and bird species. When the system identifies a deviation between the predicted and actual data, the linear regression analysis uses the motion trend model established by past data to adjust the accuracy of the prediction method, while taking into account the maximum motion capacity and species characteristics of specific birds. Through the application of correction factors, it is ensured that data deviations can be effectively corrected even during critical fluctuation periods. In this way, even if there are errors in the data collection process or the birds move abnormally due to the influence of the external environment, the system can still maintain a high degree of reliability. The multi-level nature of this method ensures that whether it is past data analysis, future dynamic prediction or large-scale data correction, it can adapt to the unique behavioral characteristics of different bird species and improve the accuracy and adaptability of the overall data evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0040] Figure 2 It is a schematic diagram of the overall module of the present invention.

[0041] In the figure: 100, data acquisition module; 200, past fluctuation analysis module; 300, future fluctuation analysis module; 400, distance correction module. DETAILED DESCRIPTION

[0042] The following will be combined with the accompanying drawings in the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Next, see Figure 1-Figure 2 One of the purposes of this embodiment is to provide a flying bird distance assessment system based on a bird detection radar, which includes a data acquisition module 100, a past fluctuation analysis module 200, a future fluctuation analysis module 300 and a distance correction module 400.

[0044] The data acquisition module 100 acquires the radar data of the flying bird and the center position of the key area in real time, wherein the center position of the key area is , the key areas are pre-set by the system as reference points for monitoring.

[0045] The data acquisition module 100 extracts features from the radar data. The feature extraction results include the bird species and the bird position, bird speed and bird acceleration at the past time point. The bird species are identified by the bird detection radar target recognition system. Indicates; the position of the bird at a past time point , Bird Speed and bird acceleration In Indicates a point in time in the past.

[0046] The past fluctuation analysis module 200 calculates the distance between the center position of the key area and the bird position at the past time point according to the bird position at the past time point as the past time point distance, which specifically includes:

[0047] For every past time point , calculate the bird's position Central location to key areas Euclidean distance ,in As the distance of a point in time in the past.

[0048] The past fluctuation analysis module 200 performs past fluctuation analysis on the distance of past time points, wherein: if the past fluctuation analysis is normal, the future fluctuation analysis module 300 is entered; otherwise, the distance correction module 400 is entered, and the past fluctuation analysis is performed again on the corrected result, i.e., the secondary past fluctuation analysis. If an abnormality occurs at this time, a past fluctuation warning is performed, which specifically includes:

[0049] Distance to past time points Perform past volatility analysis, evaluate volatility by calculating standard deviation, and calculate distances from past time points Standard Deviation The formula is as follows:

[0050] ,in is the total number of past time points, is the average distance of past time points;

[0051] Setting volatility thresholds ,like , it means that the past fluctuation analysis is normal, then enter the future fluctuation analysis module 300; if , it indicates that the past fluctuation analysis is abnormal, then the distance correction module 400 is entered to correct the past time point distance The same past fluctuation analysis is performed. If the conditions for past fluctuation analysis anomaly are met again, a past fluctuation warning is issued. The past fluctuation warning is used to point out that the behavior of the bird in its past movement is significantly different from the normal pattern, and is used for the application of the learning model.

[0052] The future fluctuation analysis module 300 predicts the position of the bird at a future time point based on the bird's position, bird speed, and bird acceleration at a past time point, and calculates the distance between the center position of the key area and the bird's position at the future time point as the future time point distance, specifically including:

[0053] Set a future time point , based on the bird's position at a past time point , Bird Speed and bird acceleration , predict the position of the bird at a future time point ,in The calculation formula is as follows:

[0054] ;

[0055] For each future time point , calculate the position of the bird at the future time point Central location to key areas Euclidean distance ,in As the distance to a future time point.

[0056] The future fluctuation analysis module 300 performs future fluctuation analysis on the distance at the future time point, wherein: if the future fluctuation analysis is normal, the distance at the future time point is output; otherwise, the module enters the distance correction module 400, and re-analyzes the future fluctuation on the corrected result, i.e., the secondary future fluctuation analysis. If an abnormality occurs at this time, a future fluctuation warning is performed, which specifically includes:

[0057] Distance to future time point Perform future volatility analysis, also evaluate volatility by calculating standard deviation, and calculate the distance to future time points Standard Deviation The formula is as follows:

[0058] ,in is the total number of future time points, is the average distance to future time points;

[0059] Based on the fluctuation threshold in the past fluctuation analysis module 200 , conduct future volatility analysis, where: , it means that the future fluctuation analysis is normal, then the distance of the future time point Output as the output result of the system; if , it indicates that the future fluctuation analysis is abnormal, then the distance correction module 400 is entered, and the distance of the corrected future time point is The same future volatility analysis is performed. If the conditions for abnormal future volatility analysis are met again, a future volatility warning is issued. The future volatility warning indicates that the behavior of the bird will deviate from the normal trajectory in the future, and related risks need to be prevented in advance.

[0060] The past fluctuation analysis module 200 and the future fluctuation analysis module 300 in the system analyze the distance data of the past and future time points through the set fluctuation threshold to determine the stability of the bird's movement; when the past fluctuation analysis results are normal, the system continues to perform future fluctuation analysis to ensure the consistency of the data and the accuracy of the prediction; if the past analysis is abnormal, the system will immediately enter the distance correction module 400 to adjust the data through linear regression and speed correction factors to make it more consistent with the movement law of the bird; if it is still abnormal after adjustment, the system triggers a past fluctuation warning, indicating that further attention and processing are needed;

[0061] Similarly, during the future fluctuation analysis process, if the test results are normal, the data will be directly output to provide a basis for management and decision-making; but if an abnormality occurs, the system will apply the same correction mechanism to re-evaluate the rationality of the data and trigger future fluctuation warnings when necessary. This branching mechanism and feedback loop not only improves the accuracy and response speed of data evaluation, but also provides timely and effective adjustment methods for abnormal situations, ensuring that the system remains efficient and reliable when facing different bird species and uncertain environmental factors.

[0062] The distance correction module 400 calculates the linear regression function of the bird using a linear regression algorithm according to the bird's position at a past time point, specifically including:

[0063] Based on the position of the bird at a past time point , respectively establish a linear regression function for each coordinate component, including , and ,in , , , , and They represent the parameters of the linear regression function established for the corresponding coordinate components;

[0064] The least squares method is used to calculate the parameters of the linear regression function, and the parameters of the linear regression function on the corresponding coordinate components are determined by minimizing the sum of squares of the errors between the output value of the linear regression function and the actual bird position at past time points.

[0065] The distance correction module 400 obtains the maximum bird speed according to the bird speed at the past time point; calculates the correction factor based on the linear regression function of the bird, the maximum bird speed and the bird type; and corrects the distance of the past time point or the distance of the future time point according to the correction factor, specifically including:

[0066] Based on the speed of the bird at a past time point , get the maximum speed ,in ;

[0067] Based on the linear regression function of the bird Parameters, maximum bird speed and bird species, calculate the correction factor , the specific formula is as follows:

[0068] ,in , It is a kind of flying bird;

[0069] Finally, the correction factor is multiplied by the distance of the past time point or the distance of the future time point to perform correction.

[0070] Correction Factor The introduction of is to adjust the abnormal distance data that may exist in the bird's movement process. By adjusting these abnormal data, the overall data set can be made more consistent with the normal movement pattern of the bird and the reliability of the data can be improved. The formula includes the characteristics of specific bird species. , which means that the correction factor is customized to adapt to the specific movement characteristics of different bird species, thereby more accurately reflecting the differences in speed and acceleration between different bird species; maximum speed The introduction of takes into account the maximum speed that the bird can reach during its movement, and combines the parameters in the linear regression function , where the parameters Represents the slope of the function. The correction formula compensates for the abnormal conditions of excessive or insufficient speed through these two factors to maintain an appropriate ratio.

[0071] The second purpose of this embodiment is to provide a method for using a flying bird distance assessment system based on a bird detection radar, comprising the following method steps:

[0072] S1. Obtain radar data of flying birds and the center position of key areas in real time, and perform feature extraction on the radar data. The feature extraction results include the flying bird species and the flying bird position, flying bird speed, and flying bird acceleration at past time points.

[0073] S2. According to the bird position at the past time point, the distance between the center position of the key area and the bird position at the past time point is calculated as the past time point distance, and the past fluctuation analysis is performed on the past time point distance. If the past fluctuation analysis is normal, enter S3; otherwise, enter S4, and re-analyze the past fluctuation on the corrected result. If an abnormality occurs at this time, a past fluctuation warning is issued;

[0074] S3, predict the position of the bird at the future time point based on the bird's position, speed and acceleration at the past time point, and calculate the distance between the center of the key area and the bird's position at the future time point as the distance at the future time point, and perform future fluctuation analysis on the distance at the future time point, wherein: if the future fluctuation analysis is normal, the future time point distance is output; otherwise, enter S4, and re-perform future fluctuation analysis on the corrected result, and if an abnormality occurs at this time, a future fluctuation warning is issued;

[0075] S4. According to the position of the bird at the past time point, the linear regression function of the bird is calculated using a linear regression algorithm; and according to the speed of the bird at the past time point, the maximum speed of the bird is obtained; based on the linear regression function of the bird, the maximum speed of the bird and the type of bird, a correction factor is calculated; and the distance of the past time point or the distance of the future time point is corrected according to the correction factor.

[0076] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A bird distance assessment system based on bird detection radar, characterized in that: The system comprises a data acquisition module (100), a past fluctuation analysis module (200), a future fluctuation analysis module (300) and a distance correction module (400), wherein: The data acquisition module (100) acquires radar data of flying birds and the center position of key areas in real time, and performs feature extraction on the radar data, wherein the feature extraction results include the flying bird species and the flying bird position, flying bird speed and flying bird acceleration at a past time point; The past fluctuation analysis module (200) calculates the distance between the center position of the key area and the position of the bird at the past time point according to the position of the bird at the past time point, as the distance at the past time point, and performs past fluctuation analysis on the distance at the past time point, wherein: if the past fluctuation analysis is normal, then enter the future fluctuation analysis module; otherwise, then enter the distance correction module, and re-perform past fluctuation analysis on the corrected result; if an abnormality occurs at this time, then issue a past fluctuation warning; wherein: The past fluctuation analysis module (200) analyzes the distance between past time points The process of performing fluctuation analysis is as follows: Assess volatility by calculating standard deviation and calculate distance from past time points Standard Deviation The formula is as follows: ,in is the total number of past time points, is the average distance of past time points; Setting volatility thresholds ,like , it means that the past fluctuation analysis is normal, then enter the future fluctuation analysis module (300); if , it indicates that the past fluctuation analysis is abnormal, then the distance correction module (400) is entered to correct the distance of the past time point after correction. Perform the same past fluctuation analysis. If the abnormal conditions of past fluctuation analysis are met again, issue a past fluctuation warning. The future fluctuation analysis module (300) predicts the position of the bird at a future time point based on the bird's position, bird speed and bird acceleration at a past time point, calculates the distance between the center position of the key area and the bird's position at the future time point as the distance at the future time point, and performs future fluctuation analysis on the distance at the future time point, wherein: if the future fluctuation analysis is normal, the future time point distance is output; otherwise, the distance correction module is entered, and the future fluctuation analysis is performed again on the corrected result, and if an abnormality occurs at this time, a future fluctuation warning is issued; The distance correction module (400) calculates the linear regression function of the bird using a linear regression algorithm based on the bird's position at a past time point; obtains the maximum bird speed based on the bird's speed at a past time point; calculates a correction factor based on the linear regression function of the bird, the maximum bird speed and the bird type; and corrects the distance at a past time point or the distance at a future time point based on the correction factor.

2. The bird distance assessment system based on bird detection radar according to claim 1 is characterized in that: The past fluctuation analysis module (200) uses the Euclidean distance calculation method to calculate the distance between the center position of the key area and the position of the bird at a past time point as the past time point distance. Indicates that Indicates a point in time in the past.

3. The bird distance assessment system based on bird detection radar according to claim 2 is characterized in that: The process of the future fluctuation analysis module (300) calculating the distance between the center position of the key area and the position of the bird at a future time point specifically includes: Set a future time point , based on the bird's position at a past time point , Bird Speed and bird acceleration , predict the position of the bird at a future time point ,in The calculation formula is as follows: ； For each future time point , calculate the position of the bird at the future time point Central location to key areas Euclidean distance ,in As the distance to a future time point.

4. The bird distance assessment system based on bird detection radar according to claim 3 is characterized in that: The process of the future fluctuation analysis module (300) performing future fluctuation analysis specifically includes: Distance to future time point Conduct future volatility analysis, evaluate volatility by calculating standard deviation, and calculate distance to future time points Standard Deviation The formula is as follows: ,in is the total number of future time points, is the average distance to future time points; The fluctuation threshold in the module (200) is analyzed based on the past fluctuations , conduct future volatility analysis, where: , it means that the future fluctuation analysis is normal, then the distance of the future time point Output as the output result of the system; if , it indicates that the future fluctuation analysis is abnormal, then the distance correction module (400) is entered, and the distance of the corrected future time point is The same future volatility analysis is performed. If the conditions for abnormal future volatility analysis are met again, a future volatility warning is issued.

5. The bird distance assessment system based on bird detection radar according to claim 4 is characterized in that: The process of the distance correction module (400) correcting the distance of past time points or the distance of future time points using the correction factor specifically includes: Based on the position of the bird at a past time point , establish a linear regression function for each coordinate component, including , and ,in , , , , and They represent the parameters of the linear regression function established for the corresponding coordinate components respectively; The parameters of the linear regression function are calculated using the least square method, and the parameters of the linear regression function on the corresponding coordinate components are determined by minimizing the sum of squares of the error between the output value of the linear regression function and the actual bird position at the past time point; Based on the speed of the bird at a past time point , get the maximum speed ,in ; Based on the linear regression function of the bird Parameters, maximum bird speed and bird species, calculate the correction factor , the specific formula is as follows: ,in , It is a kind of flying bird; Finally, the correction factor is multiplied by the distance of the past time point or the distance of the future time point to perform correction.

6. A method for using the bird distance assessment system based on bird detection radar according to any one of claims 1 to 5, characterized in that: The method comprises the following steps: S1. Obtain radar data of flying birds and the center position of key areas in real time, and perform feature extraction on the radar data. The feature extraction results include the flying bird species and the flying bird position, flying bird speed, and flying bird acceleration at past time points. S2. According to the bird position at the past time point, the distance between the center position of the key area and the bird position at the past time point is calculated as the past time point distance, and the past fluctuation analysis is performed on the past time point distance. If the past fluctuation analysis is normal, enter S3; otherwise, enter S4, and re-analyze the past fluctuation on the corrected result. If an abnormality occurs at this time, a past fluctuation warning is issued; S3, predict the position of the bird at the future time point based on the bird's position, speed and acceleration at the past time point, and calculate the distance between the center of the key area and the bird's position at the future time point as the distance at the future time point, and perform future fluctuation analysis on the distance at the future time point, wherein: if the future fluctuation analysis is normal, the future time point distance is output; otherwise, enter S4, and re-perform future fluctuation analysis on the corrected result, and if an abnormality occurs at this time, a future fluctuation warning is issued; S4. According to the position of the bird at the past time point, the linear regression function of the bird is calculated using a linear regression algorithm; and according to the speed of the bird at the past time point, the maximum speed of the bird is obtained; based on the linear regression function of the bird, the maximum speed of the bird and the type of bird, a correction factor is calculated; and the distance of the past time point or the distance of the future time point is corrected according to the correction factor.

Images