Multi-region-oriented biodiversity intelligent monitoring system

Abstract

The invention relates to the technical field of biodiversity monitoring, and discloses a multi-region-oriented biodiversity intelligent monitoring system, which comprises a data acquisition module, a data transmission module, a data processing module, a data analysis module and a rule analysis module, infrared cameras are widely deployed in multiple areas, comprehensive biological activity data containing timestamps and area numbers are collected according to fixed intervals, and then the low-power-consumption LoRa technology is used for compression transmission, so that data richness is guaranteed, transmission efficiency is improved, and energy consumption is reduced; in the aspects of feature extraction and species identification, key features are scientifically and accurately extracted and species are identified by comprehensively applying a form and behavior feature extraction method and combining a constructed species feature database and an Euclidean distance algorithm, and the database is perfected by means of expert judgment on the condition that the species cannot be identified; and on the aspect of activity law analysis, not only can the active region and time of the same species be determined, but also the immigration and emigration periods of seasonal migration species can be accurately judged.

Description

Technical Field

[0001] The present invention relates to the technical field of biodiversity monitoring, and in particular to a multi-regional intelligent biodiversity monitoring system. Background Art

[0002] In the early days, biodiversity monitoring relied primarily on manual field surveys. This method not only consumed significant manpower, material resources, and time, but also had a very limited monitoring range, making it difficult to conduct comprehensive, real-time monitoring of biodiversity across large, multi-region areas. With the advent of camera technology, infrared cameras began to be used in biomonitoring, partially addressing the shortcomings of manual surveys. However, traditional camera-based monitoring systems lacked systematic data integration and analysis mechanisms, resulting in low data transmission and processing efficiency, making them unable to meet the demand for real-time monitoring of dynamic biodiversity changes.

[0003] In terms of data processing and species identification, previous monitoring systems often relied on simple image recognition algorithms, which lacked comprehensive and accurate extraction of biological features, resulting in low species identification accuracy. Furthermore, the lack of a unified species feature database made it difficult to effectively compare and identify species across different regions.

[0004] In terms of analyzing the patterns of biological activities, existing monitoring systems can often only conduct simple analysis of the activities of a single region or a single species. They are unable to conduct in-depth exploration of the changing patterns of biological activities in multiple regions and multiple species from the perspective of time series and spatial distribution, making it difficult to provide a comprehensive and scientific decision-making basis for biodiversity conservation.

[0005] In summary, existing biodiversity monitoring technologies have numerous shortcomings in monitoring scope, data processing, species identification, and activity pattern analysis. This invention aims to overcome these shortcomings by providing a multi-regional intelligent biodiversity monitoring system that enables comprehensive, real-time, and accurate monitoring of biodiversity, providing strong technical support for biodiversity conservation. Summary of the Invention

[0006] The purpose of the present invention is to provide a multi-regional intelligent biodiversity monitoring system to solve the technical problems raised in the background technology.

[0007] The purpose of the present invention can be achieved through the following technical solutions: A multi-regional biodiversity smart monitoring system, including: Data acquisition module, used to widely deploy infrared cameras in multiple monitoring areas and collect biological activity data in real time; The data transmission module is used to compress the collected biological activity data and transmit it to the data processing module through the low-power wireless communication technology LoRa; a data processing module, configured to perform feature extraction processing on the biological activity data received from the data transmission module and determine key features extracted from the biological activity data; The data analysis module is used to determine the species type of biological species in each monitoring area based on the key features obtained through feature extraction processing; The pattern analysis module analyzes the activity change patterns of biological species of various types based on time series and spatial distribution.

[0008] As a further solution of the present invention: in each monitoring area, various sensors collect data at fixed time intervals e; The biological activity data collected each time includes image data and its corresponding collection timestamp; All collected biological activity data are attached with the pre-set area number for each monitoring area and the collection time stamp of the biological activity data; Infrared cameras are used to capture image data of biological species.

[0009] As a further solution of the present invention: the feature extraction processing method is as follows: Step A1, morphological characteristics: StepA1.1. Use the edge detection algorithm corresponding to the Canny algorithm to detect the edge information of the image data; Among them, the Canny algorithm calculates the magnitude and direction of the image gradient corresponding to the image data, then suppresses the gradient magnitude to non-maximum values, and then connects the edges according to the pre-set high and low thresholds to obtain the species outline in the image data; Step A1.2: For the obtained species outline, traverse all pixels on the species outline, count the number of pixels contained in the outline, and use this as the outline area of ​​the species outline; Step A1.3: Determine the perimeter of the species outline based on the number of all pixels and their corresponding pixel coordinates. The specific method is as follows: Get the pixel coordinates on the species outline and mark them as (x j ,y j ), j = 1, 2, ... m, where m represents the number of all pixels on the species outline; Then the distance between adjacent pixels is calculated using the Euclidean distance formula; The formula is as follows: Among them, D j+1，j Indicates the distance between adjacent pixels; Then sum the distances between all adjacent pixels to get the contour perimeter; The formula is as follows: Where C represents the contour perimeter; Step A1.4. Calculate the circularity of the species outlines in the relevant image data using the circularity formula. The circularity formula is: ; Where Y represents the circularity of the species outline, and S is the outline area of ​​the species outline; Among them, the circularity of the species outline is used to judge the closeness of the species shape to a circle; if the circularity value is close to 1, it is preliminarily judged that the species shape is closer to a circle; Step A2, Behavioral characteristics: Step A2.1. In a monitoring area, extract the designated coordinate points of the species outline in the image data corresponding to the acquisition time stamps and mark them as (xz t , yz t ), t=1, 2, ... v, v represents the number of acquisition timestamps; Among them, the specified coordinate points of the species outline in the image data are ( , ); Then, according to the time trend, the designated coordinate points of the species outline in the image data corresponding to each adjacent acquisition time stamp are connected to generate the movement trajectory of the species; Step A2.2, through: ; Calculate the movement speed V of the species between adjacent collection timestamps t，t+1 ; Then calculate the corresponding movement speed V of the species between all adjacent collection timestamps t，t+1 The mean value and standard deviation are marked as the speed mean Vp and speed dispersion Vb respectively; The calculation formula is: .

[0010] As a further solution of the present invention: the species type is determined as follows: Step B1, database construction: Establish a species characteristics database and enter the species names, morphological characteristics and behavioral characteristics of known species into the species characteristics database; The species characteristics database is entered as follows: Acquire multiple image data of the same known species through the acquisition method of the data acquisition module; Then, the key features are extracted from multiple image data corresponding to known species through the feature extraction processing method of the data processing module; Then the key features of the same known species are imported into the species feature database as a feature sample package; Step B2, similarity calculation: During the identification process, the data processing module extracts the key features of the species to be identified and compares them one by one with each feature sample package in the species feature database; Among them, the Euclidean distance algorithm is used to calculate the similarity between the key features corresponding to the species to be identified and the feature sample package; The Euclidean distance formula is:

[0011] Where U is the similarity between the key features of the species to be identified and the feature sample package, C y is the perimeter of the contour of the species to be identified in the feature sample package, S y is the contour area of ​​the species to be identified in the feature sample package, Y y is the circularity of the outline of the species to be identified in the feature sample package, V py is the mean velocity of the species profile to be identified in the feature sample package, V by is the velocity discrete value of the species outline to be identified in the feature sample package; Step B3, species type determination: The similarity U between the key features of the species to be identified and each feature sample package r Compare with a preset similarity threshold UY respectively: Where r = 1, 2, ... s, s represents the number of feature sample packages; Then select U from r The species name corresponding to a characteristic sample package with Ur ≤ UY and the smallest Ur is used to determine the species type of the species to be identified.

[0012] As a further solution of the present invention: wherein, if all calculated U r If all are greater than UY, it means that the species cannot be accurately identified and is determined with the help of the judgment of professional biologists. At the same time, based on the judgment results of professional biologists, the key characteristics of the species to be identified are generated into a feature sample package and imported into the species feature database.

[0013] As a further solution of the present invention: the activity change pattern analysis method is as follows: The pre-specified first observation period is divided into several first standard time periods. Then, at multiple specified same collection timestamps within each first standard time period, the number of species of the same species type in each monitoring area is counted and marked as N1. g1，f1，z ; Where g1=1, 2, ..., a1, a1 represents the number of the first standard time period, f1=1, 2, ..., h1, h1 is the number of acquisition timestamps in the first standard time period, z=1, 2, ..., c, c represents the number of monitoring areas; Then through:

[0014] Calculate the activity judgment value NA of each monitoring area at the same collection time stamp f，z ; For a collection timestamp, select NA f1，z The monitoring area with the largest value is marked as the active area of ​​the species at the collection timestamp; For a monitoring area, select NA f1，z The collection timestamp with the largest value is marked as the active timestamp of the species in the monitoring area.

[0015] As a further solution of the present invention: For species with seasonal migration habits, the activity change pattern analysis method is also as follows: The pre-specified second observation period is divided into several second standard time periods. Then, at multiple specified same collection timestamps within each second standard time period, the number of species of the same species type in all monitoring areas is counted and marked as N2. g2，f2 ; Where g2 = 1, 2, ..., a2, where a2 represents the number of the second standard time period, and f2 = 1, 2, ..., h2, where h2 represents the number of acquisition timestamps within the second standard time period. Then through:

[0016] Calculate the migration judgment value NB of all monitoring areas in each second standard time period g2 ; Afterwards via:

[0017] Calculate the difference NB between each adjacent second standard time period g2，g2-1 ; In the formula, NB g2 For the adjacent second standard time period, NB g2-1 is the adjacent previous second standard time period; Then, the difference NB between all adjacent second standard time periods is g2，g2-1 Extract all NBs that are positive g2，g2-1 , and extract the NB with the largest value g2，g2-1 and its corresponding second standard time period, and determining the second standard time period as the migration period of species with seasonal migration habits; At the same time, the difference NB between all adjacent second standard time periods g2，g2-1 Extract all NBs that are negative g2，g2-1 , and extract the NB with the smallest value g2，g2-1 and its corresponding second standard time period, and determining the second standard time period as the migration period of species with seasonal migration habits.

[0018] Beneficial effects of the present invention: Wide-ranging monitoring and comprehensive data collection: The system collects biodiversity data from multiple regions by deploying infrared cameras across multiple monitoring areas. Biological activity data is collected at fixed intervals in each monitoring area, including image data and corresponding acquisition timestamps. This data is then assigned a region number, ensuring spatiotemporal identification. This data comprehensively reflects the temporal dynamics of biological activity across different regions, providing a rich and accurate data foundation for subsequent analysis.

[0019] Efficient and stable data transmission: The low-power wireless communication technology LoRa is used, and the collected data is compressed and transmitted, which not only reduces the energy consumption of data transmission, but also reduces the amount of transmitted data, improves transmission efficiency, and ensures that data can be transmitted to the data processing module stably and promptly, meeting the real-time requirements of the system.

[0020] Scientific and Accurate Feature Extraction: The system uses the Canny algorithm to detect image edges and extracts key features based on morphological features such as contour area, perimeter, and circularity, as well as behavioral characteristics such as movement trajectory, velocity mean, and discrete values. This approach characterizes biological species from multiple dimensions, accurately reflecting their appearance and behavioral characteristics, and providing a reliable basis for species identification.

[0021] Accurate and reliable species identification: By building a database of species characteristics, using the Euclidean distance algorithm to calculate similarity, and then determining species type based on a similarity threshold, this method ensures accurate identification. For species that cannot be accurately identified, professional biologists are used to determine and update the database, further improving the reliability of species identification and continuously optimizing the system's recognition capabilities.

[0022] Comprehensive and in-depth analysis of activity patterns: The system can not only analyze the activity of the same species in different monitoring areas and collection timestamps, determine the active areas and active timestamps, but also for species with seasonal migration habits, by counting the number of species in different time periods, calculating the difference between the migration judgment value and the adjacent time periods, accurately determining the species' immigration period and out-migration period, and comprehensively revealing the activity change patterns of biological species, providing detailed decision-making basis for biodiversity conservation.

[0023] Assisting in biodiversity conservation: The system provides multi-regional and multi-dimensional biological data monitoring and analysis, providing comprehensive and accurate data support for biologists and ecological conservation workers to understand the current status of biodiversity and formulate scientific and reasonable conservation strategies, thereby effectively promoting the conservation and sustainable development of biodiversity. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The present invention will be further described below with reference to the accompanying drawings.

[0025] Figure 1 This is a system block diagram of the multi-regional biodiversity intelligent monitoring system of the present invention.

[0026] Figure 2 It is a flow chart of the data processing module of the multi-regional biodiversity intelligent monitoring system of the present invention.

[0027] Figure 3 It is a flow chart of the data analysis module of the multi-regional biodiversity intelligent monitoring system of the present invention. DETAILED DESCRIPTION

[0028] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making any creative efforts shall fall within the scope of protection of the present invention.

[0029] Example 1

[0030] See also Figure 1 、 Figure 2 、 Figure 3 As shown, the present invention is a multi-regional biodiversity intelligent monitoring system, comprising: Data acquisition module, used to widely deploy infrared cameras in multiple monitoring areas and collect biological activity data in real time; In each monitoring area, various sensors collect data at fixed time intervals e; Among them, each collected biological activity data includes image data and its corresponding collection timestamp, and the infrared camera is used to capture the image data of biological species; All collected biological activity data are attached with the pre-set area number for each monitoring area and the collection time stamp of the biological activity data; The data transmission module is used to compress the collected biological activity data and transmit it to the data processing module through the low-power wireless communication technology LoRa; a data processing module, configured to perform feature extraction processing on the biological activity data received from the data transmission module and determine key features extracted from the biological activity data; The feature extraction process is as follows: Step A1, morphological characteristics: StepA1.1. Use the edge detection algorithm corresponding to the Canny algorithm to detect the edge information of the image data; Among them, the Canny algorithm calculates the magnitude and direction of the image gradient corresponding to the image data, then suppresses the gradient magnitude to non-maximum values, and then connects the edges according to the pre-set high and low thresholds to obtain the species outline in the image data; Step A1.2: For the obtained species outline, traverse all pixels on the species outline, count the number of pixels contained in the outline, and use this as the outline area of ​​the species outline; Step A1.3: Determine the perimeter of the species outline based on the number of all pixels and their corresponding pixel coordinates. The specific method is as follows: Get the pixel coordinates on the species outline and mark them as (x j ,y j ), j = 1, 2, ... m, where m represents the number of all pixels on the species outline; Then the distance between adjacent pixels is calculated using the Euclidean distance formula; The formula is as follows: Among them, D j+1，j Indicates the distance between adjacent pixels; Then sum the distances between all adjacent pixels to get the contour perimeter; The formula is as follows: Where C represents the contour perimeter; Step A1.4. Calculate the circularity of the species outlines in the relevant image data using the circularity formula. The circularity formula is: ; Where Y represents the circularity of the species outline, and S is the outline area of ​​the species outline; Among them, the circularity of the species outline is used to judge the closeness of the species shape to a circle; if the circularity value is close to 1, it is preliminarily judged that the species shape is closer to a circle; Step A2, Behavioral characteristics: Step A2.1. In a monitoring area, extract the designated coordinate points of the species outline in the image data corresponding to the acquisition time stamps and mark them as (xz t , yz t), t=1, 2, ... v, v represents the number of acquisition timestamps; Among them, the specified coordinate points of the species outline in the image data are ( , ); Then, according to the time trend, the designated coordinate points of the species outline in the image data corresponding to each adjacent acquisition time stamp are connected to generate the movement trajectory of the species; Step A2.2, through: ; Calculate the movement speed V of the species between adjacent collection timestamps t，t+1 ; Then calculate the corresponding movement speed V of the species between all adjacent collection timestamps t，t+1 The mean value and standard deviation are marked as the speed mean Vp and speed dispersion Vb respectively; The calculation formula is: ; Data analysis module, used to determine species type based on key features obtained through feature extraction; Species type is determined as follows: Step B1, database construction: Establish a species characteristics database and enter the species names, morphological characteristics and behavioral characteristics of known species into the species characteristics database; The species characteristics database is entered as follows: Acquire multiple image data of the same known species through the acquisition method of the data acquisition module; Then, the key features are extracted from multiple image data corresponding to known species through the feature extraction processing method of the data processing module; Then the key features of the same known species are imported into the species feature database as a feature sample package; Step B2, similarity calculation: During the identification process, the data processing module extracts the key features of the species to be identified and compares them one by one with each feature sample package in the species feature database; Among them, the Euclidean distance algorithm is used to calculate the similarity between the key features corresponding to the species to be identified and the feature sample package; The Euclidean distance formula is: Where U is the similarity between the key features of the species to be identified and the feature sample package, C y is the perimeter of the contour of the species to be identified in the feature sample package, S y is the contour area of ​​the species to be identified in the feature sample package, Y yis the circularity of the outline of the species to be identified in the feature sample package, V py is the mean velocity of the species profile to be identified in the feature sample package, V by is the velocity discrete value of the species outline to be identified in the feature sample package; Step B3, species type determination: The similarity U between the key features of the species to be identified and each feature sample package r Compare with a preset similarity threshold UY respectively: Where r = 1, 2, ... s, s represents the number of feature sample packages; Then select U from r The species name corresponding to a characteristic sample package with a value less than or equal to UY and the smallest Ur is used to determine the species type to be identified; In this embodiment, if all calculated U r If all are greater than UY, it means that the species cannot be accurately identified and the determination is made with the help of the judgment of professional biologists; At the same time, based on the judgment results of professional biologists, the key characteristics of the species to be identified are used to generate a feature sample package and imported into the species feature database.

[0031] This embodiment constructs a multi-regional intelligent biodiversity monitoring system. The data acquisition module deploys infrared cameras across multiple monitoring areas, enabling real-time collection of biological activity data. The data transmission module utilizes LoRa low-power wireless communication technology to ensure efficient data transmission. The data processing module uses multiple algorithms to comprehensively extract features from biological activity data, providing key data support for species identification. The data analysis module accurately identifies species by building a species feature database and using the Euclidean distance algorithm to compare key features. When the system cannot accurately identify a species, it can leverage the judgment of professional biologists to improve the database, effectively improving the system's practicality and recognition accuracy, providing a complete and efficient solution for biodiversity monitoring.

[0032] Example 2

[0033] See also Figure 1 、 Figure 2 、 Figure 3 As shown, as the second embodiment of the present invention, when the present application is specifically implemented, compared with the first embodiment, the technical solution of this embodiment is different from that of the first embodiment only in that this embodiment further includes: The regularity analysis module analyzes the activity change patterns of biological species based on time series and spatial distribution; The analysis method of activity change pattern is as follows: The pre-specified first observation period is divided into several first standard time periods. Then, at multiple specified same collection timestamps within each first standard time period, the number of species of the same species type in each monitoring area is counted and marked as N1. g1，f1，z ; Where g1=1, 2, ..., a1, a1 represents the number of the first standard time period, f1=1, 2, ..., h1, h1 is the number of acquisition timestamps in the first standard time period, z=1, 2, ..., c, c represents the number of monitoring areas; Then through: Calculate the activity judgment value NA of each monitoring area at the same collection time stamp f，z ; For a collection timestamp, select NA f1，z The monitoring area with the largest value is marked as the active area of ​​the species at the collection timestamp; For a monitoring area, select NA f1，z The collection timestamp with the largest value is marked as the active timestamp of the species in the monitoring area; Based on Example 1, this example adds a new regularity analysis module to analyze the activity change patterns of biological species based on time series and spatial distribution. By counting the number of the same species type in each monitoring area at the same collection timestamp in multiple standard time periods within a pre-specified observation period, the activity judgment value is calculated to determine the active areas of the species at each collection timestamp and the active timestamps in each monitoring area. This helps to gain a deeper understanding of the activity characteristics of organisms in different regions and times, further enriches the dimensions of biodiversity monitoring, and provides data support with more valuable temporal and spatial dimension analysis for biological behavior research and ecological protection.

[0034] Example 3

[0035] See also Figure 1 、 Figure 2 、 Figure 3 As shown, as the third embodiment of the present invention, when the present application is specifically implemented, compared with the first and second embodiments, the technical solution of this embodiment is to combine the solutions of the first and second embodiments. The technical solution of this embodiment is different from that of the first and second embodiments only in that in this embodiment: for species with seasonal migration habits, the activity change law analysis method is also as follows: The pre-specified second observation period is divided into several second standard time periods. Then, at multiple specified same collection timestamps within each second standard time period, the number of species of the same species type in all monitoring areas is counted and marked as N2. g2，f2 ; Where g2 = 1, 2, ..., a2, where a2 represents the number of the second standard time period, and f2 = 1, 2, ..., h2, where h2 represents the number of acquisition timestamps within the second standard time period. Then through: Calculate the migration judgment value NB of all monitoring areas in each second standard time period g2 ; Afterwards via: Calculate the difference NB between each adjacent second standard time period g2，g2-1 ; In the formula, NB g2 For the adjacent second standard time period, NB g2-1 is the adjacent previous second standard time period; Then, the difference NB between all adjacent second standard time periods is g2，g2-1 Extract all NBs that are positive g2，g2-1 , and extract the NB with the largest value g2，g2-1 and its corresponding second standard time period, and determining the second standard time period as the migration period of species with seasonal migration habits; At the same time, the difference NB between all adjacent second standard time periods g2，g2-1 Extract all NBs that are negative g2，g2-1 , and extract the NB with the smallest value g2，g2-1 and its corresponding second standard time period, and determining the second standard time period as the migration period of species with seasonal migration habits.

[0036] This embodiment combines the solutions of embodiment one and embodiment two, and proposes a special method for analyzing activity change patterns for species with seasonal migration habits. By dividing the second observation period, counting the number of species in all monitoring areas at the same acquisition timestamp in each second standard time period, and calculating the difference between the migration judgment value and the adjacent time period, the species' entry and exit periods can be accurately determined. This method fills the gap in the monitoring and analysis of seasonal migratory species, enabling the system to more comprehensively monitor and analyze the dynamic changes of biodiversity, and providing strong data support and technical guarantees for studying the seasonal migration patterns of organisms, protecting biological habitats, and formulating targeted ecological protection strategies.

[0037] Example 4

[0038] See also Figure 1 、 Figure 2 、 Figure 3 As shown, as the fourth embodiment of the present invention, when this application is specifically implemented, compared with the first, second and third embodiments, the technical solution of this embodiment is to combine the solutions of the above-mentioned first, second and third embodiments.

[0039] This embodiment integrates the technical solutions of the previous three embodiments, enabling the collection, transmission, processing, and species identification of biological activity data. It also analyzes the activity patterns of species based on time series and spatial distribution, and can determine the arrival and departure periods of species with seasonal migration habits. This fully integrated approach empowers the system with greater functionality and broader applicability, meeting the diverse needs of biodiversity monitoring in different scenarios. It significantly improves the efficiency and quality of biodiversity monitoring, providing a comprehensive, integrated intelligent monitoring solution for biodiversity conservation and ecological and environmental research.

[0040] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field according to actual conditions.

[0041] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A multi-regional biodiversity intelligent monitoring system, characterized by: include: The data acquisition module is used to deploy multiple infrared cameras in multiple monitoring areas and collect biological activity data in the corresponding monitoring areas in real time; The data transmission module is used to compress the collected biological activity data and transmit it to the data processing module through the low-power wireless communication technology LoRa; A data processing module is used to perform feature extraction processing on the biological activity data received from the data transmission module and determine key features extracted from the biological activity data, wherein the key features include morphological features and behavioral features; The data analysis module is used to determine the species type of biological species in each monitoring area based on the key features obtained through feature extraction processing; The pattern analysis module analyzes the activity change patterns of biological species of various types based on time series and spatial distribution.

2. The multi-regional biodiversity intelligent monitoring system according to claim 1 is characterized in that: The feature extraction processing method of morphological features is as follows: The edge detection algorithm corresponding to the Canny algorithm is used to detect the edge information of the image data. The Canny algorithm calculates the amplitude and direction of the image gradient corresponding to the image data, then suppresses the gradient amplitude to non-maximum values, and then connects the edges according to pre-set high and low thresholds to obtain the species outline in the image data. For the obtained species outline, all pixels on the species outline are traversed, the number of pixels contained in the outline is counted, and the number is used as the outline area of ​​the species outline; the perimeter of the species outline is determined based on the number of all pixels and their corresponding pixel coordinates; The circularity of the species outline in the relevant image data is calculated using the circularity formula; The circularity formula is: ; Where Y represents the circularity of the species outline, S is the contour area of ​​the species outline, and C is the contour perimeter of the species outline.

3. The multi-regional biodiversity intelligent monitoring system according to claim 2, characterized in that: The contour perimeter is derived as follows: Get the pixel coordinates on the species outline and mark them as (x j ,y j ), j = 1, 2, ... m, where m represents the number of all pixels on the species contour; the distance between adjacent pixels is then calculated using the Euclidean distance formula; and the distances between all adjacent pixels are summed to obtain the contour perimeter.

4. The multi-regional biodiversity intelligent monitoring system according to claim 2, characterized in that: The feature extraction process of behavioral features is as follows: In a monitoring area, extract the designated coordinate points of the species outline in the image data corresponding to the acquisition time stamps and mark them as (xz t , yz t ), t=1, 2, ... v, v represents the number of acquisition timestamps; Among them, the specified coordinate points of the species outline in the image data are ( , ); Then, according to the time trend, the designated coordinate points of the species outline in the image data corresponding to each adjacent acquisition time stamp are connected to generate the movement trajectory of the species; Then through: Calculate the movement speed V of the species between adjacent collection timestamps t，t+1 ; Where, e is the time interval between adjacent acquisition timestamps; Then calculate the corresponding movement speed V of the species between all adjacent collection timestamps t，t+1 The mean value and standard deviation of the speed are marked as the speed mean Vp and the speed dispersion value Vb respectively.

5. The multi-regional biodiversity intelligent monitoring system according to claim 4 is characterized in that: Species type is determined as follows: Establish a species characteristics database and enter the species names, morphological characteristics and behavioral characteristics of known species into the species characteristics database; During the identification process, the data processing module extracts the key features of the species to be identified and compares them one by one with each feature sample package in the species feature database; Among them, the Euclidean distance algorithm is used to calculate the similarity between the key features corresponding to the species to be identified and the feature sample package; Step B3, species type determination: The similarity U between the key features of the species to be identified and each feature sample package r Compare with a preset similarity threshold UY respectively: Where r = 1, 2, ... s, s represents the number of feature sample packages; Then select U from r The species name corresponding to a characteristic sample package with Ur ≤ UY and the smallest Ur is used to determine the species type of the species to be identified.

6. The multi-regional biodiversity intelligent monitoring system according to claim 5, characterized in that: The Euclidean distance formula is: Where U is the similarity between the key features of the species to be identified and the feature sample package, C y is the perimeter of the contour of the species to be identified in the feature sample package, S y is the contour area of ​​the species to be identified in the feature sample package, Y y is the circularity of the outline of the species to be identified in the feature sample package, V py is the mean velocity of the species profile to be identified in the feature sample package, V by is the discrete value of the velocity of the species outline to be identified in the feature sample package.

7. The multi-regional biodiversity intelligent monitoring system according to claim 5, characterized in that: in, The method for entering species characteristics into the database is as follows: Acquire multiple image data of the same known species through the acquisition method of the data acquisition module; Then, the key features are extracted from multiple image data corresponding to known species through the feature extraction processing method of the data processing module; The key features of the same known species are then imported into the species feature database as a feature sample package.

8. The multi-regional biodiversity intelligent monitoring system according to claim 5, characterized in that: in, If all the calculated U r If all are greater than UY, it means that the species cannot be accurately identified and is determined with the help of the judgment of professional biologists. At the same time, based on the judgment results of professional biologists, the key characteristics of the species to be identified are generated into a feature sample package and imported into the species feature database.

9. The multi-regional biodiversity intelligent monitoring system according to claim 1, characterized in that: The analysis method of activity change pattern is as follows: The pre-specified first observation period is divided into several first standard time periods. Then, at multiple specified same collection timestamps within each first standard time period, the number of species of the same species type in each monitoring area is counted and marked as N1. g1，f1，z ; Where g1=1, 2, ..., a1, a1 represents the number of the first standard time period, f1=1, 2, ..., h1, h1 is the number of acquisition timestamps in the first standard time period, z=1, 2, ..., c, c represents the number of monitoring areas; Then through: Calculate the activity judgment value NA of each monitoring area at the same collection time stamp f，z ; For a collection timestamp, select NA f1，z The monitoring area with the largest value is marked as the active area of ​​the species at the collection timestamp; For a monitoring area, select NA f1，z The collection timestamp with the largest value is marked as the active timestamp of the species in the monitoring area.

10. The multi-regional biodiversity intelligent monitoring system according to claim 9, characterized in that: For species with seasonal migration habits, the analysis method for activity changes is as follows: The pre-specified second observation period is divided into several second standard time periods. Then, at multiple specified same collection timestamps within each second standard time period, the number of species of the same species type in all monitoring areas is counted and marked as N2. g2，f2 ; Where g2 = 1, 2, ..., a2, where a2 represents the number of the second standard time period, and f2 = 1, 2, ..., h2, where h2 represents the number of acquisition timestamps within the second standard time period. Then through: Calculate the migration judgment value NB of all monitoring areas in each second standard time period g2 ; Afterwards via: Calculate the difference NB between each adjacent second standard time period g2，g2-1 ; In the formula, NB g2 For the adjacent second standard time period, NB g2-1 is the adjacent previous second standard time period; Then, the difference NB between all adjacent second standard time periods is g2，g2-1 Extract all NBs that are positive g2，g2-1 , and extract the NB with the largest value g2，g2-1 and its corresponding second standard time period, and determining the second standard time period as the migration period of species with seasonal migration habits; At the same time, the difference NB between all adjacent second standard time periods g2，g2-1 Extract all NBs that are negative g2，g2-1 , and extract the NB with the smallest value g2，g2-1 and its corresponding second standard time period, and determining the second standard time period as the migration period of species with seasonal migration habits.

Images