A biological survey remote control method and system based on Internet of Things monitoring

Abstract

The present invention relates to the field of remote control technology, and specifically discloses a biological survey remote control method and system based on Internet of Things monitoring, wherein the method comprises the following steps: S1: obtaining an initial path by connecting a front-end acquisition device through a straight line, and screening a target path according to the distance of the initial path; S2: determining a monitoring area according to the location of the front-end acquisition device on the target path, and calculating the number of targets in the monitoring area; generating data points according to the number of targets, and obtaining a target image by connecting the data points through a straight line; calculating the migration coefficient of the target path according to the target image, and determining the migration path according to the migration coefficient; S3: determining whether it is a migration state according to the migration path, and sending a sleep command to the front-end acquisition device according to the predicted time when it is a migration state. The present invention can determine the sleep time of the front-end acquisition device according to the fluctuation of the number of animals, and avoid the situation of missing monitoring data and / or waste of electricity.

Description

Technical Field

[0001] The present invention relates to the technical field of remote control, and in particular to a biological investigation remote control method and system based on Internet of Things monitoring. Background Art

[0002] The Internet of Things monitoring is a comprehensive system with strong prevention capabilities. It is mainly composed of three parts: front-end acquisition equipment, transmission network, and monitoring operation platform. It plays an important role in many fields, such as biological survey. Biological survey is a scientific method for collecting and analyzing information on biological species in a specific area or environment. Through biological surveys, researchers can understand the biodiversity of the region, the health of the ecosystem, the distribution of species, and the impact of human activities on the environment.

[0003] Animal migration refers to the phenomenon that animal groups move from one area to another driven by seasonal changes or changes in environmental conditions. This behavior is usually cyclical and involves large-scale group movement. When animal migration occurs, the front-end collection equipment in the animal's original area can enter a dormant state to reduce power consumption and extend the service life of the front-end collection equipment.

[0004] In actual situations, the time when the front-end acquisition equipment enters the sleep state is mostly a pre-set fixed time, and the specific time of animal migration may fluctuate around the pre-set time, which may cause the front-end acquisition equipment to enter the sleep state too early or too late. When the front-end acquisition equipment enters the sleep state too early, it may miss key monitoring data, resulting in incomplete or inaccurate data; when the front-end acquisition equipment enters the sleep state too late, it will cause unnecessary power consumption. Summary of the invention

[0005] The purpose of the present invention is to provide a biological survey remote control method and system based on Internet of Things monitoring to solve the following technical problems:

[0006] The time when the front-end acquisition equipment enters the sleep state is mostly a pre-set fixed time, and the specific time of animal migration may fluctuate around the pre-set time, which may cause the front-end acquisition equipment to enter the sleep state too early or too late. When the front-end acquisition equipment enters the sleep state too early, it may miss key monitoring data, resulting in incomplete or inaccurate data; when the front-end acquisition equipment enters the sleep state too late, it will cause unnecessary power consumption.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A biological survey remote control method based on Internet of Things monitoring includes the following steps:

[0009] S1: In the preset monitoring area, obtain the location information of the front-end acquisition device, and take the Ath front-end acquisition device as the starting point, connect n front-end acquisition devices by straight lines to obtain an initial path, where n is a preset number, and sort the initial paths according to the length of the distance. The longer the distance, the lower the ranking, and take the first n' initial paths in the sorting as the target path, where n'=0.3n;

[0010] S2: Taking the location of the front-end acquisition device on the target path as the center and the preset radius r as the circle center, a circle is drawn to obtain the monitoring area, and the number of animals in the monitoring area is periodically obtained within a preset acquisition period, and the mean is calculated as the target number;

[0011] Generate the data point corresponding to the i-th acquisition cycle (ij, D ij ), D ij represents the number of targets in the jth monitoring area on the target path in the i-th acquisition cycle, i∈[1,m], m represents the total number of acquisition cycles, and connects the data points corresponding to two adjacent monitoring areas on the target path through a straight line to obtain the target image;

[0012] Calculating a migration coefficient K of a target path according to the target image, and marking the target path as a migration path when the migration coefficient K≥K', where K' represents a preset migration coefficient threshold;

[0013] S3: When the proportion of the migration path to the target path C ≥ 0.7, it is marked as a migration state, and the number of animals in the monitoring area is predicted to reach the preset dormancy number D Dor time T, and sends a sleep command to the front-end acquisition device in the monitoring area at Tcur+T, where Tcur represents the time point marked as the migration state.

[0014] As a further solution of the present invention: in step S2, the process of calculating the migration coefficient K of the target path specifically includes:

[0015] Sorting the target images in time sequence, and determining a functional relationship f(x) of the target images in the sorting;

[0016] The migration coefficient K is calculated by the following formula:

[0017]

[0018] Among them, ε is the preset correction coefficient, f m (x) represents the functional relationship of the target curve corresponding to the mth acquisition cycle, P a Indicates the similarity between the target image a in the sorting and the next adjacent target image a+1.

[0019] As a further solution of the present invention: the process of determining the similarity between the target image a in the sorting and the adjacent next target image a+1 specifically includes:

[0020] The data point corresponding to the front-end acquisition device on the target image a is taken as the target point, and the tangent slope of the target point is determined, and the target value F is calculated. b =F b *f a (b), F b represents the tangent slope of the b-th target point, generating the target vector Y=(F1', F2', ..., F n ');

[0021] Determine the target vector Y' corresponding to the target image a+1, and calculate the cosine value cosθ of the angle between the target vector Y and the target vector Y';

[0022] The similarity P is calculated by the following formula a :

[0023]

[0024] Among them, ga(x)=|fa(x)-fa+1(x)|.

[0025] As a further solution of the present invention: in the step S3, it is predicted that the number of animals in the monitoring area reaches a preset dormant number D Dor The process of time T specifically includes:

[0026] Generate the coordinate points (i, Di) corresponding to the same monitoring area, Di represents the number of targets in the i-th acquisition cycle of the monitoring area, and fit the m coordinate points to obtain the fitting curve h (x);

[0027] The sleep number D Dor Substitute into the fitting curve h(x) and obtain the time T.

[0028] As a further solution of the present invention: in the step S2, the process of calculating the target quantity further includes the following steps:

[0029] The difference between the number of animals in the monitoring area obtained for the Bth time and the target number is calculated, and when the difference is greater than or equal to a preset difference threshold, the number of animals in the monitoring area obtained for the Bth time is removed, and the target number is calculated again.

[0030] As a further solution of the present invention: in the step S1, when the total number of the initial paths is less than or equal to a preset number threshold, all the initial paths are used as target paths.

[0031] As a further solution of the present invention: in the step S2, the number of animals in the monitoring area is obtained based on the camera of the front-end acquisition device.

[0032] A biological survey remote control system based on Internet of Things monitoring, comprising:

[0033] Initial module: In the preset monitoring area, obtain the location information of the front-end acquisition device, and take the Ath front-end acquisition device as the starting point, connect n front-end acquisition devices by straight lines to obtain the initial path, n is a preset number, and sort the initial paths according to the length of the distance. The longer the distance, the lower the ranking, and take the first n' initial paths in the sorting as the target path, n'=0.3n;

[0034] Evaluation module: With the location of the front-end acquisition device on the target path as the center, a circle with a preset radius r is drawn to obtain the monitoring area, and the number of animals in the monitoring area is periodically obtained within a preset acquisition period, and the mean is calculated as the target number;

[0035] Generate the data point corresponding to the i-th acquisition cycle (ij, D ij ), D ij represents the number of targets in the jth monitoring area on the target path in the i-th acquisition cycle, i∈[1,m], m represents the total number of acquisition cycles, and connects the data points corresponding to two adjacent monitoring areas on the target path through a straight line to obtain the target image;

[0036] Calculating a migration coefficient K of a target path according to the target image, and marking the target path as a migration path when the migration coefficient K≥K', where K' represents a preset migration coefficient threshold;

[0037] Adjustment module: When the proportion of migration path to target path C ≥ 0.7, it is marked as migration state, and the number of animals in the monitoring area is predicted to reach the preset dormancy number D Dor time T, and sends a sleep command to the front-end acquisition device in the monitoring area at Tcur+T, where Tcur represents the time point marked as the migration state.

[0038] Beneficial effects of the present invention: In the present invention, firstly, an initial path is obtained by connecting the front-end acquisition devices in a straight line, and a target path is screened according to the distance of the initial path; it is worth noting that a front-end acquisition device appears only once in an initial path, and the longer the distance of the initial path, the greater the possibility that the animal deviates from the initial path when migrating, and it may happen that the animal is actually in a migration state, but the distance is too long, resulting in errors in subsequent judgments, so it is necessary to screen the target path; then the monitoring area is planned and the number of targets in the monitoring area is determined, and a target image is obtained according to the number of targets; it can be understood that the target image reflects the distribution of the number of animals on the target path at a certain point in time. In actual situations, animals generally have fixed territories, and the number of animals in the monitoring area is relatively stable. When animal migration occurs, the number of animals on the target path is relatively stable. The distribution of animal numbers will change, which will lead to changes in the target image. Therefore, the migration coefficient is determined through the target image. Then, the migration path is determined according to the migration coefficient, and whether it is in an adjustment state is determined according to the proportion of the migration path to the target path. It should be noted that the change in the number of animals in a single monitoring area may be affected by a variety of external factors, which may lead to misjudgment. In the present invention, by considering the time series changes and the overall changes, it is possible to more comprehensively and accurately judge whether there is a migration situation. Finally, the time when the number of animals reaches the preset dormant number is predicted, and a dormancy command is sent to the front-end acquisition device in the monitoring area according to the predicted time. When the number of animals in the monitoring area is lower than a certain range, there is no need to continue monitoring. Therefore, by predicting the time when the number of animals reaches the preset dormant number, a dormancy command is sent to the front-end acquisition device. The present invention can determine the dormancy time of the front-end acquisition device according to the fluctuation of the number of animals, avoiding the situation of missing monitoring data and / or waste of electricity. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The present invention will be further described below in conjunction with the accompanying drawings.

[0040] Figure 1 The present invention is a flowchart of a biological investigation remote control method based on Internet of Things monitoring. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of 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.

[0042] The following will be combined with the drawings in the embodiments of 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] See also Figure 1 As shown, the present invention is a biological survey remote control method based on Internet of Things monitoring, comprising the following steps:

[0044] S1: In the preset monitoring area, obtain the location information of the front-end acquisition device, and take the Ath front-end acquisition device as the starting point, connect n front-end acquisition devices by straight lines to obtain an initial path, where n is a preset number, and sort the initial paths according to the length of the distance. The longer the distance, the lower the ranking, and take the first n' initial paths in the sorting as the target path, where n'=0.3n;

[0045] S2: Taking the location of the front-end acquisition device on the target path as the center and the preset radius r as the circle center, a circle is drawn to obtain the monitoring area, and the number of animals in the monitoring area is periodically obtained within a preset acquisition period, and the mean is calculated as the target number;

[0046] Generate the data point corresponding to the i-th acquisition cycle (ij, D ij ), D ij represents the number of targets in the jth monitoring area on the target path in the i-th acquisition cycle, i∈[1,m], m represents the total number of acquisition cycles, and connects the data points corresponding to two adjacent monitoring areas on the target path through a straight line to obtain the target image;

[0047] Calculating a migration coefficient K of a target path according to the target image, and marking the target path as a migration path when the migration coefficient K≥K', where K' represents a preset migration coefficient threshold;

[0048] S3: When the proportion of the migration path to the target path C ≥ 0.7, it is marked as a migration state, and the number of animals in the monitoring area is predicted to reach the preset dormancy number D Dor time T, and sends a sleep command to the front-end acquisition device in the monitoring area at Tcur+T, where Tcur represents the time point marked as the migration state.

[0049] It should be noted that first, the initial path is obtained by connecting the front-end acquisition devices in a straight line, and the target path is screened according to the distance of the initial path; it is worth noting that a front-end acquisition device only appears once in an initial path. The longer the distance of the initial path, the greater the possibility that the animal will deviate from the initial path when migrating. It may happen that the animal is actually in a migration state, but the distance is too long, resulting in errors in subsequent judgments. Therefore, it is necessary to screen the target path; then plan the monitoring area and determine the number of targets in the monitoring area, and obtain the target image based on the number of targets; it is understandable that the target image reflects the distribution of the number of animals on the target path at a certain point in time. In actual situations, animals generally have fixed territories, and the number of animals in the monitoring area is relatively stable. When animal migration occurs, the number of animals on the target path is The distribution will change, which will lead to changes in the target image. Therefore, the migration coefficient is determined by the target image; then, the migration path is determined according to the migration coefficient, and whether it is in an adjustment state is determined according to the proportion of the migration path to the target path. It should be noted that the change in the number of animals in a single monitoring area may be affected by a variety of external factors, and misjudgment may occur. In the present invention, by considering the timing changes and the overall changes, it is possible to more comprehensively and accurately judge whether there is a migration situation; finally, the time when the number of animals reaches the preset dormant number is predicted, and a dormancy command is sent to the front-end acquisition device in the monitoring area according to the predicted time; when the number of animals in the monitoring area is lower than a certain range, there is no need to continue monitoring, so a dormancy command is sent to the front-end acquisition device by predicting the time when the number of animals reaches the preset dormant number.

[0050] In another preferred implementation of the present invention, in step S2, the process of calculating the migration coefficient K of the target path specifically includes:

[0051] Sorting the target images in time sequence, and determining a functional relationship f(x) of the target images in the sorting;

[0052] The migration coefficient K is calculated by the following formula: ;

[0053] Among them, ε is the preset correction coefficient, f m (x) represents the functional relationship of the target curve corresponding to the mth acquisition cycle, P a Indicates the similarity between the target image a in the sorting and the next adjacent target image a+1.

[0054] It is worth noting that max(f1(x)-f m The larger the value of (x), the greater the change in the number of animals in the same monitoring area, so the migration coefficient should be higher. The relationship between the other parameters and the migration coefficient K can be referred to the above ideas and will not be repeated here.

[0055] In another preferred implementation of the present invention, the process of determining the similarity between the target image a in the sorting and the adjacent next target image a+1 specifically includes:

[0056] The data point corresponding to the front-end acquisition device on the target image a is taken as the target point, and the tangent slope of the target point is determined, and the target value F is calculated. b =F b *f a (b), F b represents the tangent slope of the b-th target point, generating the target vector Y=(F1', F2', ..., F n ');

[0057] Determine the target vector Y' corresponding to the target image a+1, and calculate the cosine value cosθ of the angle between the target vector Y and the target vector Y';

[0058] The similarity P is calculated by the following formula a : ;

[0059] Among them, g a (x)=|f a (x)-f a+1 (x) |.

[0060] It can be understood that the closer the cosine value of the angle is to 1, the closer the two vectors are, that is, the more similar they are, which further indicates that the function values ​​and tangent slopes of the corresponding target points on the target image a and the target image a+1 are closer, so the similarity should be higher.

[0061] In another preferred embodiment of the present invention, in step S3, it is predicted that the number of animals in the monitoring area reaches a preset dormant number D Dor The process of time T specifically includes:

[0062] Generate the coordinate points (i, Di) corresponding to the same monitoring area, Di represents the number of targets in the i-th acquisition cycle of the monitoring area, and fit the m coordinate points to obtain the fitting curve h (x);

[0063] The sleep number D Dor Substitute into the fitting curve h(x) and obtain the time T.

[0064] It should be noted that directly relying on current data may lead to misjudgment due to accidental fluctuations or external interference. Curve fitting can smooth out these short-term fluctuations by analyzing data over a period of time, thereby providing more reliable prediction results.

[0065] By generating coordinate points of the number of animals within the collection period and performing curve fitting, the present invention can more accurately predict the time T when the number of animals reaches a preset dormant number. This method can reduce errors, realize intelligent decision-making, and adapt to dynamic environmental changes, ultimately improving prediction accuracy.

[0066] In another preferred implementation of the present invention, in the step S2, the process of calculating the target quantity further includes the following steps:

[0067] The difference between the number of animals in the monitoring area obtained for the Bth time and the target number is calculated, and when the difference is greater than or equal to a preset difference threshold, the number of animals in the monitoring area obtained for the Bth time is removed, and the target number is calculated again.

[0068] It should be noted that in actual monitoring, the number of animals obtained in certain collection cycles may be abnormal due to various reasons (such as equipment failure, accidental animal aggregation or external interference), which may lead to misleading results;

[0069] If these abnormal data are not removed, the calculation of the target number may be biased due to a few extreme values. For example, the target number may be significantly raised or lowered, which may lead to misjudgment of the migration status or incorrect sending of the sleep command.

[0070] In another preferred implementation of the present invention, in the step S1, when the total number of the initial paths is less than or equal to a preset number threshold, all the initial paths are used as target paths.

[0071] It can be understood that when the number of initial paths is less than or equal to the preset number threshold, all initial paths are used as target paths. The main purpose is to ensure the adequacy of the data, avoid data loss caused by screening, maintain the flexibility and adaptability of the system, and ensure full coverage of the entire monitoring area. When the number of paths is limited, all available monitoring data are maximized to ensure the normal operation of the system and the accuracy of the analysis.

[0072] In another preferred implementation of the present invention, in step S2, the number of animals in the monitoring area is obtained based on the camera of the front-end acquisition device.

[0073] It is worth noting that the camera collects images or videos in the area in real time, identifies animal objects in the image through image processing techniques such as background subtraction and target segmentation, and uses classification algorithms to further identify and filter non-animal targets, and finally derives the number of animals.

[0074] A biological survey remote control system based on Internet of Things monitoring, comprising:

[0075] Initial module: In the preset monitoring area, obtain the location information of the front-end acquisition device, and take the Ath front-end acquisition device as the starting point, connect n front-end acquisition devices by straight lines to obtain the initial path, n is a preset number, and sort the initial paths according to the length of the distance. The longer the distance, the lower the ranking, and take the first n' initial paths in the sorting as the target path, n'=0.3n;

[0076] Evaluation module: With the location of the front-end acquisition device on the target path as the center, a circle with a preset radius r is drawn to obtain the monitoring area, and the number of animals in the monitoring area is periodically obtained within a preset acquisition period, and the mean is calculated as the target number;

[0077] Generate the data point corresponding to the i-th acquisition cycle (ij, D ij ), D ij represents the number of targets in the jth monitoring area on the target path in the i-th acquisition cycle, i∈[1,m], m represents the total number of acquisition cycles, and connects the data points corresponding to two adjacent monitoring areas on the target path through a straight line to obtain the target image;

[0078] Calculating a migration coefficient K of a target path according to the target image, and marking the target path as a migration path when the migration coefficient K≥K', where K' represents a preset migration coefficient threshold;

[0079] Adjustment module: When the proportion of migration path to target path C ≥ 0.7, it is marked as migration state, and the number of animals in the monitoring area is predicted to reach the preset dormancy number D Dor time T, and sends a sleep command to the front-end acquisition device in the monitoring area at Tcur+T, where Tcur represents the time point marked as the migration state.

[0080] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A biological survey remote control method based on Internet of Things monitoring, characterized in that: The following steps are involved: S1: In the preset monitoring area, the location information of the front-end acquisition device is obtained, and the Ath front-end acquisition device is used as the starting point, and n front-end acquisition devices are connected by straight lines to obtain an initial path, where n is a preset number, and the initial paths are sorted according to the length of the distance, the longer the distance, the lower the ranking, and the first n' initial paths in the sorting are used as the target path, where n'=0.3n; S2: Taking the location of the front-end acquisition device on the target path as the center and the preset radius r as the circle center, a circle is drawn to obtain the monitoring area, and the number of animals in the monitoring area is periodically obtained within a preset acquisition period, and the mean is calculated as the target number; Generate the data point corresponding to the i-th acquisition cycle (ij, D ij ), D ij represents the number of targets in the jth monitoring area on the target path in the i-th acquisition cycle, i∈[1,m], m represents the total number of acquisition cycles, and connects the data points corresponding to two adjacent monitoring areas on the target path through a straight line to obtain the target image; Calculating a migration coefficient K of a target path according to the target image, and marking the target path as a migration path when the migration coefficient K≥K', where K' represents a preset migration coefficient threshold; S3: When the proportion of the migration path to the target path C ≥ 0.7, it is marked as a migration state, and the number of animals in the monitoring area is predicted to reach the preset dormancy number D Dor time T, and sends a sleep command to the front-end acquisition device in the monitoring area at Tcur+T, where Tcur represents the time point marked as the migration state; The process of calculating the migration coefficient K of the target path specifically includes: Sorting the target images in time sequence, and determining a functional relationship f(x) of the target images in the sorting; The migration coefficient K is calculated by the following formula: Among them, ε is the preset correction coefficient, f m (x) represents the functional relationship of the target curve corresponding to the mth acquisition cycle, P a Indicates the similarity between the target image a in the sorting and the next adjacent target image a+1.

2. According to claim 1, a biological survey remote control method based on Internet of Things monitoring is characterized in that: The process of determining the similarity between the target image a in the sorting and the adjacent next target image a+1 specifically includes: The data point corresponding to the front-end acquisition device on the target image a is taken as the target point, and the tangent slope of the target point is determined, and the target value F is calculated. b '=F b *f a (b), F b represents the tangent slope of the b-th target point, generating the target vector Y=(F1', F2', ..., F n '); Determine the target vector Y' corresponding to the target image a+1, and calculate the cosine value cosθ of the angle between the target vector Y and the target vector Y'; The similarity P is calculated by the following formula a : Among them, g a (x)=|f a (x)-f a+1 (x)|.

3. According to claim 1, a biological survey remote control method based on Internet of Things monitoring is characterized in that: In step S3, it is predicted that the number of animals in the monitoring area reaches the preset dormant number D Dor The process of time T specifically includes: Generate the coordinate points (i, Di) corresponding to the same monitoring area, Di represents the number of targets in the i-th acquisition cycle of the monitoring area, and fit the m coordinate points to obtain the fitting curve h(x); The sleep number D Dor Substitute into the fitting curve h(x) and obtain the time T.

4. The biological survey remote control method based on Internet of Things monitoring according to claim 1 is characterized in that: In the step S2, the process of calculating the target quantity further includes the following steps: The difference between the number of animals in the monitoring area obtained for the Bth time and the target number is calculated, and when the difference is greater than or equal to a preset difference threshold, the number of animals in the monitoring area obtained for the Bth time is removed, and the target number is calculated again.

5. The biological survey remote control method based on Internet of Things monitoring according to claim 1 is characterized in that: In the step S1, when the total number of the initial paths is less than or equal to a preset number threshold, all the initial paths are used as target paths.

6. The biological survey remote control method based on Internet of Things monitoring according to claim 1 is characterized in that: In the step S2, the number of animals in the monitoring area is obtained based on the camera of the front-end acquisition device.

7. A biological survey remote control system based on Internet of Things monitoring, characterized in that: include: Initial module: in the preset monitoring area, obtain the location information of the front-end acquisition device, and take the Ath front-end acquisition device as the starting point, connect n front-end acquisition devices by straight lines to obtain the initial path, n is a preset number, and sort the initial paths according to the length of the distance. The longer the distance, the lower the ranking, and take the first n' initial paths in the sorting as the target path, n'=0.3n; Evaluation module: With the location of the front-end acquisition device on the target path as the center, a circle with a preset radius r is drawn to obtain the monitoring area, and the number of animals in the monitoring area is periodically obtained within a preset acquisition period, and the mean is calculated as the target number; Generate the data point corresponding to the i-th acquisition cycle (ij, D ij ), D ij represents the number of targets in the jth monitoring area on the target path in the i-th acquisition cycle, i∈[1,m], m represents the total number of acquisition cycles, and connects the data points corresponding to two adjacent monitoring areas on the target path through a straight line to obtain the target image; Calculating a migration coefficient K of a target path according to the target image, and marking the target path as a migration path when the migration coefficient K≥K', where K' represents a preset migration coefficient threshold; Adjustment module: When the proportion of migration path to target path C ≥ 0.7, it is marked as migration state, and the number of animals in the monitoring area is predicted to reach the preset dormancy number D Dor time T, and sends a sleep command to the front-end acquisition device in the monitoring area at Tcur+T, where Tcur represents the time point marked as the migration state; The process of calculating the migration coefficient K of the target path specifically includes: Sorting the target images in time sequence, and determining a functional relationship f(x) of the target images in the sorting; The migration coefficient K is calculated by the following formula: Among them, ε is the preset correction coefficient, f m (x) represents the functional relationship of the target curve corresponding to the mth acquisition cycle, P a Indicates the similarity between the target image a in the sorting and the next adjacent target image a+1.

Images