A bird target recognition method and system based on bird detection radar

Abstract

The present invention relates to the technical field of bird target identification, and specifically, to a bird target identification method and system based on bird detection radar. It includes obtaining monitoring data fed back by bird detection radars at each monitoring point, capturing bird flight target items; configuring expulsion points in an orchard area, and matching the expulsion response time and expulsion frequency of expulsion points at different locations. The present invention improves the accuracy of bird type identification by capturing multiple bird flight target items for multi-faceted synchronous identification, and performs data feedback comparison on the changed target items that have not been processed in advance, and uses the successfully compared changed target items as regular target items for identifying the current bird, further enriching the reference basis for identifying the bird type, and further improving the accuracy of bird type identification. At the same time, the expulsion response time and expulsion frequency of the expulsion point are set according to the predicted probability of the flight destination, which not only avoids the continuous activation of the expulsion equipment, but also improves the expulsion effect.

Description

Technical Field

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

[0002] The working principle of bird-detecting radar is mainly based on the reflection and reception of electromagnetic waves. Bird-detecting radar transmits electromagnetic waves of a certain frequency. When these electromagnetic waves encounter birds, they will be reflected back and received by the radar receiver. Bird-detecting radar is widely used in large orchards and other places where bird activities need to be monitored. It can provide real-time information on the activity level and accurate location of harmful birds around the orchard, helping orchard personnel take corresponding measures to prevent birds from damaging orchard fruit trees.

[0003] Traditional bird-detecting radars mainly use the acquired flight status and flight trajectory to predict bird types during the bird-detecting process. However, due to the wide variety of bird species in bird habitats, most birds use orchards as feeding places, so their sizes are not much different, and there are also cases where birds have similar flight trajectories. Therefore, it is difficult to accurately locate bird targets simply by obtaining the corresponding flight trajectory. At the same time, during the process of expelling birds, if they are continuously started, it will not only cause a waste of resources, but also be detrimental to the pollination of fruit trees. Therefore, it is necessary to formulate the expulsion response time and expulsion frequency of the expulsion equipment in advance according to the prediction results to adapt to the bird expulsion work under different conditions.

[0004] In order to address the above problems, there is an urgent need for a bird target recognition method and system based on bird detection radar. Summary of the invention

[0005] The object of the present invention is to provide a bird target recognition method and system based on bird detection radar to solve the problems raised in the above background technology.

[0006] To achieve the above object, one of the objects of the present invention is to provide a bird target recognition method based on a bird detection radar, comprising the following steps:

[0007] S1. Get the geographical location of the orchard area and the geographical location of bird habitats , configure bird-detecting radar monitoring points;

[0008] S2. Obtain the monitoring data fed back by the bird-detecting radar at each monitoring point and capture the bird flight target items ;

[0009] S3. Collect the flight status and eating habits of different birds in the current bird habitat and establish a database of bird routine habits ;

[0010] S4. Compare the currently captured bird flight target items Bird habit database , mark the target item of capturing bird flight Unconventional items in ;

[0011] S5. Predict the type of bird caught and the probability of its flight destination , generate a flight trajectory simulation diagram;

[0012] S6. Configure the eviction points in the orchard area , according to the predicted probability of flight destination And the flight trajectory simulation diagram, matching the eviction response time of eviction points at different locations and the frequency of evictions ;

[0013] S7. Obtain expulsion result information and establish non-routine items Weight distribution strategy to determine unconventional items Whether to include bird flight targets .

[0014] As a further improvement of the technical solution, the method for configuring the bird detection radar monitoring point in S1 comprises the following steps:

[0015] S1.1. Obtain the monitoring airspace range of the bird detection radar ;

[0016] S1.2. Use airspace monitoring equipment to obtain the span of airspace between the orchard area and the bird habitat;

[0017] S1.3, combined with the bird routine habit database in S4 , get the flight range of regular birds ;

[0018] S1.4. According to the monitoring airspace range of bird detection radar And the flight range of conventional birds , bird-detecting radar monitoring points are deployed along the span of the airspace to fully capture the flight status of birds.

[0019] As a further improvement of this technical solution, the bird flight target item in S2 Including bird flight speed , Birds flying direction , Bird flight height difference , Stopover points in the airspace and time spent in the airspace ;

[0020] The flight speed of birds is the average speed of birds flying within the span of airspace;

[0021] Birds flying direction is the flight angle of the bird in the span airspace, that is, a spatial rectangular coordinate system is established in the span airspace, and the angle formed by the path formed by the bird's flight and the axes in different directions is the flight direction of the bird. ;

[0022] Bird flight altitude difference The difference between the flight heights of birds captured at two monitoring time points;

[0023] Stopover point in the airspace The target location point within the span airspace where the bird stays during flight;

[0024] Time spent in the airspace It is the time that the bird stays at the target location within the span of the airspace.

[0025] As a further improvement of this technical solution, the bird routine habit database is established in S3 The method comprises the following steps:

[0026] S3.1. Collect information on bird life habits and food distribution in bird habitats;

[0027] S3.2. Obtain the fruit types in the orchard area and predict the dietary preferences of birds in the bird habitat for different fruit types;

[0028] S3.3. Use video surveillance equipment to capture images of birds in flight, identify bird types, and use bird detection radar to capture bird flight status;

[0029] S3.4. Bind bird types, bird flight status, and dietary preferences for different fruit types to generate a database of bird routine habits .

[0030] As a further improvement of the technical solution, the target item of bird flight capture in S4 is marked Unconventional items in The method comprises the following steps:

[0031] S4.1. Change the flight speed of birds and the difference in the flight altitude of birds Define as a regular target item and get the range value of each regular target item ;

[0032] S4.2. Stop points in the airspace , Birds flying direction and time spent in the airspace Define as a change target item and obtain the monitoring value of each change target item , as an unconventional item The tag value of

[0033] S4.3. Set the range value of the general target item As a numerical basis for predicting bird types.

[0034] As a further improvement of the technical solution, the eviction response time of the eviction points at different locations is matched in S6. and the frequency of evictions The method comprises the following steps:

[0035] S6.1. Combine the flight trajectory simulation diagram to obtain the destination corresponding to each flight trajectory, and identify the target expulsion point of the current destination;

[0036] S6.2. Combine the predicted bird type, corresponding flight time and flight trajectory distance to predict the time range for the bird to arrive at the expulsion point, and take the shortest time within the time range as the expulsion response time of the expulsion point ;

[0037] S6.3, combining the predicted eating preferences of birds in the bird habitats for different types of fruits in S3.2, and ranking the fruit preferences corresponding to the destinations;

[0038] S6.4. Match the corresponding expulsion frequency according to the fruit tendency .

[0039] As a further improvement of this technical solution, the non-routine item is established in S7 The method of weight allocation strategy includes the following steps:

[0040] S7.1. Locate the destination of successful expulsion based on the expulsion result information;

[0041] S7.2. Obtain the flight destination probability corresponding to the destination where the expulsion was successful ;

[0042] S7.3. Calculation of corresponding non-routine items Weight score ,in, The same irregular items appear in the flight trajectory The number of birds, For non-routine items The general constant of .

[0043] The second object of the present invention is to provide a system for realizing a bird target identification method based on a bird detection radar, including a bird detection radar monitoring point control module, a flight target item capture module, a bird flight state prediction module, a flight target item classification module and an expulsion point expulsion state response module:

[0044] The bird-detecting radar monitoring point control module is used to obtain the geographical location of the orchard area and the geographical location of the bird habitat, and configure the bird-detecting radar monitoring points;

[0045] The flight target item capture module is used to obtain the monitoring data fed back by the bird detection radar at each monitoring point and capture the bird flight target item;

[0046] The bird flight state prediction module compares the currently captured bird flight target item with the bird routine habit database, predicts the captured bird type and the probability of the flight destination, and generates a flight trajectory simulation diagram;

[0047] The flight target item classification module is used to mark the unconventional items in the bird capture flight target items;

[0048] The expulsion point expulsion status response module is used to configure the expulsion points in the orchard area, match the expulsion response time and expulsion frequency of the expulsion points in different positions according to the predicted flight destination probability and the flight trajectory simulation diagram, and feed back the expulsion result information to the flight target item classification module. The flight target item classification module establishes an unconventional item weight allocation strategy to determine whether the unconventional item is included in the bird flight target item.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] In the bird detection radar-based bird target identification method and system, multiple bird flight target items are captured for multi-faceted simultaneous identification to improve the accuracy of bird type identification, and data feedback comparison is performed on the changed target items that have not been processed in advance. The successfully compared changed target items are used as regular target items for identifying the current bird, further enriching the reference basis for identifying the bird type and further improving the accuracy of bird type identification. At the same time, the expulsion point expulsion response time and expulsion frequency are set according to the predicted flight destination probability, which not only avoids the continuous activation of the expulsion equipment, but also improves the expulsion effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a diagram of the overall method steps of the present invention;

[0052] Figure 2 A method step diagram for configuring bird detection radar monitoring points of the present invention;

[0053] Figure 3 A method step diagram for establishing a bird routine habit database of the present invention;

[0054] Figure 4 A method step diagram for capturing unconventional items in bird flight target items for marking of the present invention;

[0055] Figure 5 A method step diagram for matching eviction response time and eviction frequency of eviction points at different locations according to the present invention;

[0056] Figure 6 A method step diagram for establishing a non-routine item weight allocation strategy of the present invention;

[0057] Figure 7 It is a flowchart of the overall system of the present invention. DETAILED DESCRIPTION

[0058] 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.

[0059] See also Figure 1 As shown, one of the purposes of the present invention is to provide a bird target recognition method based on a bird detection radar, comprising the following steps:

[0060] S1. Get the geographical location of the orchard area and the geographical location of bird habitats , configure bird-detecting radar monitoring points;

[0061] S2. Obtain the monitoring data fed back by the bird-detecting radar at each monitoring point and capture the bird flight target items ;

[0062] S3. Collect the flight status and eating habits of different birds in the current bird habitat and establish a database of bird routine habits ;

[0063] S4. Compare the currently captured bird flight target items Bird habit database , mark the target item of capturing bird flight Unconventional items in ;

[0064] S5. Predict the type of bird caught and the probability of its flight destination , generate a flight trajectory simulation diagram;

[0065] S6. Configure the eviction points in the orchard area , according to the predicted probability of flight destination And the flight trajectory simulation diagram, matching the eviction response time of eviction points at different locations and the frequency of evictions ;

[0066] S7. Obtain expulsion result information and establish non-routine items Weight distribution strategy to determine unconventional items Whether to include bird flight targets .

[0067] In specific use, during the process of bird target identification, since there is a certain distance between the orchard area and the bird habitat, in order to ensure that the bird detection radar can adapt to the airspace corresponding to the distance between the two, it is necessary to obtain the geographical location of the orchard area. and the geographical location of bird habitats , configure bird-detecting radar monitoring points to ensure that the bird-detecting radar can capture birds in time when they pass by. After completing the deployment of bird-detecting radar monitoring points, capture bird flight target items by obtaining the monitoring data fed back by the bird-detecting radars at each monitoring point. , such as bird flight paths and flight altitudes, as a reference for later prediction of bird types;

[0068] Since different birds have different eating habits, that is, they have different preferences for fruits from different fruit trees, birds will develop corresponding flight routes based on their own eating habits. In order to further distinguish bird types and improve the accuracy of bird identification, it is necessary to collect the flight status and eating habits of different birds in the current bird habitat and establish a bird routine habit database. For example, by using video surveillance equipment in conjunction with bird-detecting radar for simultaneous identification and comparing the currently captured bird flight target items Bird habit database , mark the target item of capturing bird flight Unconventional items in , which is related to the bird habit database Different target items in the monitoring area, such as birds resting on poles, branches and other targets in the monitoring area, these unconventional items Some are habitual movements of birds in their foraging process. Most of them are caused by man-made obstacles on their foraging routes, such as observation towers, power lines and other buildings built on their routes. Birds make habitual movements to avoid or temporarily stop. Some are non-routine. For accidental events, such as avoiding natural enemies, so non-routine items It needs to be verified later to determine whether it can be used as a target item for identifying bird types;

[0069] In the later monitoring process, in order to effectively cooperate with the response of the expulsion equipment and carry out the expulsion work in a timely and effective manner, it is necessary to predict the type of birds captured and the probability of their flight destination. , generate a flight trajectory simulation map, that is, after predicting the current bird type, combine its eating habits and the corresponding flight trajectory to predict the probability of the flight destination For example, the currently predicted bird type prefers fruits of type A fruit trees, type B fruit trees, and type C fruit trees. The corresponding flight trajectory direction may pass through the area where the fruits of type B fruit trees and type C fruit trees are located. Therefore, combined with the area where the fruits of type B fruit trees and type C fruit trees are located and the currently monitored flight trajectory, two corresponding flight trajectory simulation diagrams are predicted. Both situations are possible, but the probability of occurrence is determined by the degree to which the predicted bird type prefers the fruits of type B fruit trees and type C fruit trees. Therefore, the expulsion frequency of the corresponding orchard area needs to be matched and adjusted. At this time, by configuring the expulsion points in the orchard area , according to the predicted probability of flight destination And the flight trajectory simulation diagram, matching the eviction response time of eviction points at different locations and the frequency of evictions , such as eviction response time It is determined by the bird's flight speed and the distance to the destination, corresponding to the expulsion frequency Probability of flight destination After the prediction work is completed, the unconventional items are established by obtaining the eviction result information. Weight distribution strategy to determine unconventional items Whether to include bird flight targets , that is, whether the birds are detected passing by the expulsion point and whether the expulsion is successful. If successful, it means that the prediction work is successful. The corresponding unconventional items That is, it may become the target item for current bird type prediction.

[0070] In addition, if Figure 2 As shown, the method for configuring the bird detection radar monitoring point in S1 includes the following steps:

[0071] S1.1. Obtain the monitoring airspace range of the bird detection radar ;

[0072] S1.2. Use airspace monitoring equipment to obtain the span of airspace between the orchard area and the bird habitat;

[0073] S1.3. Combined with the bird routine database in S4 , get the flight range of regular birds ;

[0074] S1.4. According to the monitoring airspace range of bird detection radar And the flight range of conventional birds , bird-detecting radar monitoring points are deployed along the span of the airspace to fully capture the flight status of birds.

[0075] Furthermore, the bird flight target item in S2 Including bird flight speed , Birds flying direction , Bird flight height difference , Stopover points in the airspace and time spent in the airspace ;

[0076] The flight speed of birds is the average speed of birds flying within the span of airspace;

[0077] Birds flying direction is the flight angle of the bird in the span airspace, that is, a spatial rectangular coordinate system is established in the span airspace, and the angle formed by the path formed by the bird's flight and the axes in different directions is the flight direction of the bird. ;

[0078] Bird flight altitude difference The difference between the flight heights of birds captured at two monitoring time points;

[0079] Stopover point in the airspace The target location point within the span airspace where the bird stays during flight;

[0080] Time spent in the airspace It is the time that the bird stays at the target location within the span of the airspace.

[0081] When using bird radar to identify bird targets, the area between the orchard area and the bird habitat often exceeds the monitoring range of a single bird radar. Therefore, in order to ensure that the monitoring point can achieve full coverage of bird monitoring, it is first necessary to obtain the monitoring airspace range of the bird radar. , that is, the bird-detecting radar can monitor the effective range, and then use the airspace monitoring equipment to obtain the span airspace range formed between the orchard area and the bird habitat. In the division process, it is necessary to first determine the edge area of ​​the span airspace. The present invention mainly obtains the length of the edge line of the orchard area, the length of the edge line of the bird habitat, and the area formed by the spacing between the two areas as the plane area, and takes the highest point detected by the bird-detecting radar as the vertical area, and the area formed by the plane area is the span airspace range;

[0082] It is worth noting that the flight status of some birds tends to remain unchanged. Therefore, in the actual classification process, it is necessary to combine the bird routine habit database in S4. , get the flight range of regular birds , using bird-detecting radar to capture the flight paths of birds in real time, counting the flight areas they form, and using the corresponding flight areas as the flight airspace range of regular birds Of course, the flight airspace range It belongs to the span airspace, which includes the flight airspace. The remaining areas are generally difficult for birds to reach, so the flight range of conventional birds is As the target area for later monitoring, finally according to the monitoring airspace range of the bird detection radar And the flight range of conventional birds , bird-detecting radar monitoring points are deployed along the span of the airspace to fully capture the flight status of birds, that is, the flight airspace range of conventional birds is monitored through various bird-detecting radar monitoring points different areas within.

[0083] Furthermore, if Figure 3 As shown, a bird routine database is established in S3 The method comprises the following steps:

[0084] S3.1. Collect information on bird life habits and food distribution in bird habitats;

[0085] S3.2. Obtain the fruit types in the orchard area and predict the dietary preferences of birds in the bird habitat for different fruit types;

[0086] S3.3. Use video surveillance equipment to capture images of birds in flight, identify bird types, and use bird detection radar to capture bird flight status;

[0087] S3.4. Bind bird types, bird flight status, and dietary preferences for different fruit types to generate a database of bird routine habits .

[0088] In specific use, since different types of birds have different food requirements, and are related to the geographical environment of the area where they are located, in order to obtain the needs of birds in the current bird habitat for different fruits, it is necessary to collect the bird's living habits and the food distribution in the bird habitat in advance, that is, the bird's dietary preference for different foods. Since the present invention is to protect the fruits in the orchard, it is only necessary to consider the preference of the fruit types in the orchard area for different birds, and it is also necessary to consider the food distribution in the bird habitat. This is because for birds, although the food in the habitat can meet their living needs, for specific foods, such as fruits that do not appear in the habitat in the orchard, the birds' preference will increase. Therefore, the next step is to obtain the fruit types in the orchard area and predict the dietary preference of birds in the bird habitat for different fruit types, which can be obtained through biological research results and regional bird monitoring;

[0089] The flight status of birds will change during different activities. For example, there will be obvious differences in the flight status when foraging and courting. Therefore, in order to distinguish the activity status of birds, it is necessary to use video surveillance equipment to capture images of birds in flight, identify bird types, and use bird detection radar to capture the flight status of birds. That is, the current activity status of birds can be judged by the flight status. Finally, the bird type, bird flight status and eating preference for different types of fruits are combined to generate a database of birds' regular habits. , as the bird flight target item for later comparison The basis.

[0090] Specifically, Figure 4 As shown, the target item of bird flight capture is marked in S4 Unconventional items in The method comprises the following steps:

[0091] S4.1. Change the flight speed of birds and the difference in the flight altitude of birds Define as a regular target item and get the range value of each regular target item ;

[0092] S4.2. Stop points in the airspace , Birds flying direction and time spent in the airspace Define as a change target item and obtain the monitoring value of each change target item , as an unconventional item The tag value of

[0093] S4.3. Set the range value of the general target item As a numerical basis for predicting bird types.

[0094] When used specifically, mark unusual items During the process, due to different bird flight target items The value ranges are different, and some bird flight target items No pattern to follow, such as stop points within the airspace These bird flight targets can only be used as a reference for identifying bird types in specific areas and after long-term verification. Therefore, it is necessary to identify the bird flight targets. Classification processing is performed, in which the flight speed of birds and the difference in the flight altitude of birds Define as a regular target item and get the range value of each regular target item , that is, there are certain rules that can be used as a reference for predicting bird types. For example, the flight speed range of a type A flock is , while the flight speed range of the B-type bird flock is , so when the flight speed detected at a certain time is , then the bird may be a Class A flock, and the remaining conventional target items need to be compared numerically to obtain the prediction result. The corresponding stop points in the airspace with no regular pattern will be , Birds flying direction and time spent in the airspace Define as a change target item and obtain the range value of each change target item , as an unconventional item The mark value of the eviction point is used to determine the corresponding prediction result. After being evaluated as a regular target item, the corresponding tag value is used as the regular target item range value corresponding to the regular target item evaluated as a regular target item. .

[0095] In addition, if Figure 5 As shown, the eviction response time of eviction points at different locations in S6 and the frequency of evictions The method comprises the following steps:

[0096] S6.1. Combine the flight trajectory simulation diagram to obtain the destination corresponding to each flight trajectory, and identify the target expulsion point of the current destination;

[0097] S6.2. Combine the predicted bird type, corresponding flight time and flight trajectory distance to predict the time range for the bird to arrive at the expulsion point, and take the shortest time within the time range as the expulsion response time of the expulsion point ;

[0098] S6.3, combining the predicted eating preferences of birds in the bird habitats for different fruit types in S3.2, and ranking the fruit preferences corresponding to the destinations;

[0099] S6.4. Match the corresponding expulsion frequency according to the fruit tendency .

[0100] In specific use, in most cases, there are multiple predicted flight trajectories of birds. This is because there are multiple preferred foods for birds in the orchard area. Although the corresponding preferences are different, they may still go to the destination where the preferred foods exist. At this time, by combining the flight trajectory simulation diagram, the destination corresponding to each flight trajectory is obtained, and the target expulsion point of the current destination is identified. The present invention mainly uses a shock wave sound pressure bird repellent to expel birds. Its working principle is to release strong sound pressure energy at intervals to stimulate the eardrum and brain nerves of birds, so that the birds feel uncomfortable and leave the orchard. Therefore, the location where the shock wave sound pressure bird repellent is deployed is the target expulsion point;

[0101] In order to reduce the startup duration, it is necessary to combine the predicted bird type, the corresponding flight time and the flight trajectory distance to predict the time range of the bird's arrival at the expulsion point, and take the shortest time within the time range as the expulsion response time of the expulsion point. For example, according to the predicted bird flight speed and flight trajectory distance, the time range required under different flight trajectories is predicted. In order to be able to expel the bird in time, the present invention takes the shortest time within the time range as the expulsion response time of the expulsion point. ;

[0102] For the eviction frequency In general, the present invention mainly determines the preference of birds for the destination food, that is, combining the predicted dietary preference of birds in the bird habitat in S3.2 for different fruit types, sorting the fruit preference corresponding to the destination, and matching the corresponding expulsion frequency according to the fruit preference. ,For example , as well as They are the destinations corresponding to the three predicted flight trajectories. The fruit types of the three destinations are different, and the highest corresponding tendency is The fruit of the destination, followed by The fruit of the destination, finally The fruit of the destination, so the fruit preference ranking of the three destinations is , and the eviction frequency of the eviction device at the eviction point It has been determined in advance that , and the corresponding frequency is , at this time, the expulsion frequency is determined according to the inclination of each destination Match, that is The eviction frequency corresponding to the destination for , The eviction frequency corresponding to the destination for , The eviction frequency corresponding to the destination for .

[0103] Further, such as Figure 6 As shown, S7 creates non-routine items The method of weight allocation strategy includes the following steps:

[0104] S7.1. Locate the destination of successful expulsion based on the expulsion result information;

[0105] S7.2. Obtain the flight destination probability corresponding to the destination where the expulsion was successful ;

[0106] S7.3. Calculation of corresponding non-routine items Weight score ,in, The same irregular items appear in the flight trajectory The number of birds, For non-routine items The general constant of .

[0107] When used in a non-routine manner In the weight allocation process, we first need to combine the expulsion result information to locate the destination where the expulsion was successful. The greater the probability, the corresponding unconventional item The higher the probability of being judged as a regular target item, the higher the probability of being judged as a regular target item. To define each non-routine item The weight score corresponding to the unconventional item Weight score , where the conventional constant and non-routine items This is related to the actual monitoring values ​​of birds. Although there is no regularity, there is a certain range of values, such as the time spent in the airspace. , the birds will not stay at a stop longer than the time they need to spend foraging, so when the corresponding unconventional items When the corresponding monitoring value exceeds its value range, the corresponding conventional constant will decrease, that is, inversely proportional. In the actual calculation process, when the monitored value is within the value range, the conventional constant The value is 1. The larger the excess, the larger the normal constant. The smaller the value, the more weight score threshold is configured. When the weight score threshold is exceeded, the currently monitored unusual items It will be marked as a regular target item and used as a reference item for later identification of bird types. When the weight score threshold is not exceeded, the currently monitored unconventional items Still classified as a change target item.

[0108] The second object of the present invention is to provide a system for bird target identification method based on bird detection radar, such as Figure 7 As shown, it includes a bird detection radar monitoring point control module, a flight target item capture module, a bird flight status prediction module, a flight target item classification module, and an expulsion point expulsion status response module:

[0109] Among them, the bird-detecting radar monitoring point control module is used to obtain the geographical location of the orchard area and the geographical location of the bird habitat, and configure the bird-detecting radar monitoring points;

[0110] The flight target item capture module is used to obtain the monitoring data fed back by the bird detection radar at each monitoring point and capture the bird flight target items;

[0111] The bird flight status prediction module compares the currently captured bird flight target items with the bird routine habit database, predicts the captured bird type and the probability of flight destination, and generates a flight trajectory simulation diagram;

[0112] The flight target item classification module is used to mark the unconventional items in the bird capture flight target items;

[0113] The expulsion point and expulsion status response module is used to configure the expulsion points in the orchard area. According to the predicted flight destination probability and the flight trajectory simulation diagram, the expulsion response time and expulsion frequency of the expulsion points in different locations are matched. At the same time, the expulsion result information is fed back to the flight target item classification module. The flight target item classification module establishes an unconventional item weight allocation strategy to determine whether the unconventional item should be included in the bird flight target item.

[0114] When used specifically, during the process of bird target identification, the geographical location of the orchard area and the geographical location of the bird habitat are first obtained through the bird detection radar monitoring point control module, and the regional state formed between the two is obtained. According to the monitoring range of each bird detection radar, the bird detection radar monitoring points are configured, that is, the location of each bird detection radar. Then, the bird detection radars at each monitoring point will feed the monitoring data back to the flight target item capture module, and the bird flight target items are captured through the flight target item capture module, that is, the bird flight target items include the bird flight speed, the bird flight direction, the bird flight altitude difference, the stop point in the airspace, and the stop time in the airspace, and the bird routine habit database stores the flight status and eating habits of different birds in the current bird habitat;

[0115] In order to identify the current flight status of birds, the bird flight status prediction module compares the currently captured bird flight target items with the bird routine habit database, predicts the captured bird type and the probability of the flight destination, generates a flight trajectory simulation diagram, and marks the unconventional items in the captured bird flight target items through the flight target item classification module, namely, the bird flight direction, the stop point in the domain, and the stay time in the airspace. Finally, the expulsion point in the orchard area is configured through the expulsion point expulsion status response module. According to the predicted flight destination probability and the flight trajectory simulation diagram, the expulsion response time and expulsion frequency of the expulsion points in different positions are matched. At the same time, the expulsion result information is fed back to the flight target item classification module. The flight target item classification module establishes an unconventional item weight allocation strategy to determine whether the unconventional item is included in the bird flight target item, that is, by calculating the weight score of each unconventional item and comparing the weight score threshold to make a judgment.

[0116] 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 target recognition method based on bird detection radar, characterized in that: The steps include: S1. Get the geographical location of the orchard area and the geographical location of bird habitats , configure bird-detecting radar monitoring points; S2. Obtain the monitoring data fed back by the bird-detecting radar at each monitoring point and capture the bird flight target items ; S3. Collect the flight status and eating habits of different birds in the current bird habitat and establish a database of bird routine habits ; S4. Compare the currently captured bird flight target items Bird habit database , mark the target item of capturing bird flight Unconventional items in ; S5. Predict the type of bird caught and the probability of its flight destination , generate a flight trajectory simulation diagram; S6. Configure the eviction points in the orchard area , according to the predicted probability of flight destination And the flight trajectory simulation diagram, matching the eviction response time of eviction points at different locations and the frequency of evictions ; The eviction response time of the eviction points at different locations is matched in S6 and the frequency of evictions The method comprises the following steps: S6.

1. Combine the flight trajectory simulation diagram to obtain the destination corresponding to each flight trajectory, and identify the target expulsion point of the current destination; S6.

2. Combine the predicted bird type, corresponding flight time and flight trajectory distance to predict the time range for the bird to arrive at the expulsion point, and take the shortest time within the time range as the expulsion response time of the expulsion point ; S6.3, combining the predicted eating preferences of birds in the bird habitats for different types of fruits in S3.2, and ranking the fruit preferences corresponding to the destinations; S6.

4. Match the corresponding expulsion frequency according to the fruit tendency ; S7. Obtain expulsion result information and establish non-routine items Weight distribution strategy to determine unconventional items Whether to include bird flight targets ; The S7 establishes non-routine items The method of weight allocation strategy includes the following steps: S7.

1. Locate the destination of successful expulsion based on the expulsion result information; S7.

2. Obtain the flight destination probability corresponding to the destination where the expulsion was successful ; S7.

3. Calculation of corresponding non-routine items Weight score ,in, The same irregular items appear in the flight trajectory The number of birds, For non-routine items The general constant of .

2. The bird target identification method based on bird detection radar according to claim 1 is characterized in that: The method for configuring the bird detection radar monitoring point in S1 comprises the following steps: S1.

1. Obtain the monitoring airspace range of the bird detection radar ; S1.

2. Use airspace monitoring equipment to obtain the span of airspace between the orchard area and the bird habitat; S1.3, combined with the bird routine habit database in S4 , get the flight range of regular birds ; S1.

4. According to the monitoring airspace range of bird detection radar And the flight range of conventional birds , bird-detecting radar monitoring points are deployed along the span of the airspace to fully capture the flight status of birds.

3. The bird target identification method based on bird detection radar according to claim 1 is characterized in that: The bird flight target item in S2 Including bird flight speed , Birds flying direction , Bird flight height difference , Stopover points in the airspace and time spent in the airspace ; The flight speed of birds is the average speed of birds flying within the span of airspace; Birds flying direction is the flight angle of the bird in the span airspace, that is, a spatial rectangular coordinate system is established in the span airspace, and the angle formed by the path formed by the bird's flight and the axes in different directions is the flight direction of the bird. ; Bird flight altitude difference The difference between the flight heights of birds captured at two monitoring time points; Stopover point in the airspace The target location point within the span airspace where the bird stays during flight; Time spent in the airspace It is the time that the bird stays at the target location within the span of the airspace.

4. The bird target identification method based on bird detection radar according to claim 1 is characterized in that: The bird routine habit database is established in S3 The method comprises the following steps: S3.

1. Collect information on bird life habits and food distribution in bird habitats; S3.

2. Obtain the fruit types in the orchard area and predict the dietary preferences of birds in the bird habitat for different fruit types; S3.

3. Use video surveillance equipment to capture images of birds in flight, identify bird types, and use bird detection radar to capture bird flight status; S3.

4. Bind bird types, bird flight status, and dietary preferences for different fruit types to generate a database of bird routine habits .

5. The bird target identification method based on bird detection radar according to claim 3 is characterized in that: The S4 marks the target item of capturing bird flight Unconventional items in The method comprises the following steps: S4.

1. Change the flight speed of birds and the difference in the flight altitude of birds Define as a regular target item and get the range value of each regular target item ; S4.

2. Stop points in the airspace , Birds flying direction and time spent in the airspace Define as a change target item and obtain the monitoring value of each change target item , as an unconventional item The tag value of S4.

3. Set the range value of the general target item As a numerical basis for predicting bird types.

6. A system for implementing the bird target identification method based on bird detection radar according to claim 1, characterized in that: It includes bird-detecting radar monitoring point control module, flight target item capture module, bird flight status prediction module, flight target item classification module and expulsion point expulsion status response module: The bird-detecting radar monitoring point control module is used to obtain the geographical location of the orchard area and the geographical location of the bird habitat, and configure the bird-detecting radar monitoring points; The flight target item capture module is used to obtain the monitoring data fed back by the bird detection radar at each monitoring point and capture the bird flight target item; The bird flight state prediction module compares the currently captured bird flight target item with the bird routine habit database, predicts the captured bird type and the probability of the flight destination, and generates a flight trajectory simulation diagram; The flight target item classification module is used to mark the unconventional items in the bird capture flight target items; The expulsion point expulsion status response module is used to configure the expulsion points in the orchard area, match the expulsion response time and expulsion frequency of the expulsion points in different positions according to the predicted flight destination probability and the flight trajectory simulation diagram, and feed back the expulsion result information to the flight target item classification module. The flight target item classification module establishes an unconventional item weight allocation strategy to determine whether the unconventional item is included in the bird flight target item.

Images