A voice bird repelling method, device, terminal and storage medium

Abstract

This application relates to a voice-based bird deterrence method, device, terminal, and storage medium, belonging to the technical field of bird control. The method includes: acquiring target birds; acquiring historical monitoring data based on the target birds, including occurrence time, occurrence location, occurrence frequency, deterrence audio type, and the escape speed corresponding to each type of deterrence audio; sorting the deterrence audios in descending order of escape speed to obtain a candidate set; calculating a first score F1 for each deterrence audio in the candidate set based on occurrence time; calculating a second score F2 for each deterrence audio in the candidate set based on occurrence location; calculating a third score F3 for each deterrence audio in the candidate set based on occurrence frequency; acquiring preset first weight W1, second weight W2, and third weight W3; calculating the total score D for each deterrence audio in the candidate set D = F1*W1 + F2*W2 + F3*W3; and adjusting the order of the deterrence audios in the candidate set according to the total score to obtain a target set. This application can improve bird deterrence effectiveness.

Description

Technical Field

[0001] This application relates to the technical field of bird control, and in particular to a voice-based bird deterrence method, device, terminal, and storage medium. Background Technology

[0002] Bird activity around power transmission lines poses safety hazards to the power transmission system. On the one hand, the branches or wires used by birds to build nests can easily cause short circuits and trip the power lines; on the other hand, bird droppings left on the power lines can corrode metal components and cause power outages during rain.

[0003] To promptly eliminate the threat posed by bird activity to railway lines, commonly used bird control measures include:

[0004] (1) Install bird deterrent spikes. After long-term operation, bird deterrent spikes will rust and age, reducing their bird deterrent effect and making them inconvenient for personnel to maintain. In addition, their bird deterrent effect is not obvious against large birds with long legs.

[0005] (2) Set up windmill-style bird deterrents: Because birds are highly adaptable, after a period of time, they become accustomed to such devices and are no longer afraid of them.

[0006] (3) Pulse bird deterrent: It is easily affected by the strong electromagnetic field environment of the power transmission line, and it is difficult to solve the energy problem. The problem of rust prevention and anti-aging of the device is also difficult to solve.

[0007] (4) Set up a single sound bird repeller: Since different bird species have different fears of sounds, and birds can easily adapt to a single sound for a long time, the bird repeller effect will decrease over time.

[0008] (5) Whistle bird deterrent: It uses wind to blow the reflector and whistle installed on the blade to make a sound. It cannot deter birds in windless weather, and the bird deterrent effect will decrease greatly over time.

[0009] (6) Ultrasonic bird repeller: It is slow to take effect, expensive, and must be equipped with a rain protection device when used in the field, making maintenance difficult.

[0010] (7) Chemical bird repellents: have a short effective time and can also cause environmental pollution.

[0011] (8) Insulating bird-proof partition: It must be removed before the replacement of insulators can be carried out. It is suitable for small-scale installation and is easy to break, which brings inconvenience to maintenance.

[0012] (8) Crossarm sealing cap: Applied to old utility poles, which are difficult to modify and have a small protection range.

[0013] (9) Bird protection cover on pole top: The protection range is small and the effect is not obvious.

[0014] (10) New pole and tower structure: only applicable to newly built lines and has a high cost.

[0015] Therefore, it can be seen that the above-mentioned bird-repelling measures are all inadequate and cannot meet the needs of effective bird repelling. Summary of the Invention

[0016] This application provides a voice-based bird deterrence method, device, terminal, and storage medium, which features improved bird deterrence effectiveness.

[0017] The purpose of this application is to provide a method for repelling birds using voice commands.

[0018] The aforementioned objective of this application is achieved through the following technical solution:

[0019] A method for scare away birds using voice commands includes:

[0020] Acquire the target birds;

[0021] Extract historical monitoring data corresponding to the target bird from the database. The historical monitoring data includes the time of occurrence, location of occurrence, frequency of occurrence, type of expulsion audio, and escape speed corresponding to each type of expulsion audio.

[0022] Extract the escape speed corresponding to each type of expulsion audio, and sort the expulsion audio in descending order of escape speed to obtain a candidate set;

[0023] Calculate the first score F1 for each type of expulsion audio in the candidate set based on its occurrence time;

[0024] Calculate the second score F2 for each type of expulsion audio in the candidate set based on its location;

[0025] Calculate the third score F3 for each type of expulsion audio in the candidate set based on its frequency of occurrence;

[0026] Obtain the preset first weight W1, second weight W2, and third weight W3, and calculate the total score D for each type of expulsion audio in the candidate set: D=F1*W1+F2*W2+F3*W3;

[0027] The target set is obtained by rearranging the expulsion audio from the candidate set in descending order of total score.

[0028] By adopting the above technical solution, this application uses historical monitoring data of target birds to provide data support for the current bird control strategy. Specifically, the appearance time, location, frequency, and escape speed of the target birds in history are used as influencing factors to select the bird control audio with the best bird control effect. The selected bird control audio is then used to drive away the target birds, so that this application can achieve the purpose of improving the bird control effect.

[0029] In a preferred embodiment, this application may be further configured such that, before calculating the first score F1 for each expulsion audio in the candidate set based on its occurrence time, the method further includes:

[0030] Obtain the original set, which stores various expulsion audios used to drive away target birds;

[0031] Calculate the time period t for one loop of the expulsion audio in the original set.

[0032] In a preferred embodiment, this application can be further configured such that: calculating the first score F1 for each expulsion audio in the candidate set based on its occurrence time includes:

[0033] When the time period t reaches the preset period T, the expulsion audio is divided into multiple groups according to the original centralized expulsion audio, and each group contains multiple expulsion audio.

[0034] Then sort the resulting groups to obtain a sequence set;

[0035] The expulsion audio that is closest to the current monitoring time will be used as the center audio.

[0036] All expulsion audios within the group containing the central audio have a first score of 0.

[0037] In the sequence set, the groups that are located after the group containing the central audio are sorted in ascending order of their order, with the first score increasing by 2 points each time.

[0038] In the sequence set, the groups that precede the group containing the central audio are assigned a first score that increases by 1 point in order from closest to furthest from the group.

[0039] In a preferred embodiment, this application can be further configured such that: calculating the first score F1 for each expulsion audio in the candidate set based on its occurrence time includes:

[0040] When the time period t is lower than the preset period T, the first score F1 = d;

[0041] Where d is the number of days since the current monitoring time.

[0042] By adopting the above technical solution, the calculation formula for the first score is related to the time period t. Specifically, when the time period t is larger (bird activity is less frequent), the original set of bird-repelling audio is divided into multiple groups before calculating the first score. At this time, since the bird-repelling audio in the original set has not been played for a long time, the bird-repelling effect is not significantly different, so the first score of bird-repelling audio within the same group is set to be equal. When the time period t is smaller (bird activity is more frequent), the number of days since the current monitoring time is used as the first score. Since the loop playback time is faster, the first score of the bird-repelling audio that is furthest from the current monitoring time is set to the highest, which reduces the probability of the target bird hearing the same bird-repelling audio twice in a row, thus ensuring the accuracy of the calculated first score.

[0043] In a preferred embodiment, this application can be further configured such that calculating the second score F2 for each type of expulsion audio in the candidate set based on its location includes:

[0044] Obtain the current monitoring location, the distance L between the location where the incident occurred and the current monitoring location, and the preset value 5, where the unit of 5 is kilometers;

[0045] When L < 5, the second score F2 = 0;

[0046] When L≥5, the second score F2 takes the integer part of L / 5.

[0047] By adopting the above technical solution, since most birds cannot hear the played expulsion audio at a distance of more than 5 kilometers from the playback source, the preset value of 5 kilometers is set, which makes the accuracy of the second score calculated by this application higher.

[0048] In a preferred embodiment, this application can be further configured such that: calculating the third score F3 for each expulsion audio in the candidate set based on its frequency of occurrence includes:

[0049] Count the number of times n is played in the candidate set of expulsion audio;

[0050] When the number of plays n reaches or exceeds the first preset value N, the third score F3 = 0;

[0051] When the number of plays n is lower than the first preset value N, the third score F3 = Nn.

[0052] By adopting the above technical solution, when the number of times the expulsion audio is played is too high, the target birds may become accustomed to the excessive number of times the expulsion audio is played. Therefore, the third score of the expulsion audio that has been played more than the first preset value N is set to 0, so as to ensure the accuracy of the calculated third score.

[0053] In a preferred embodiment, this application can be further configured such that, prior to acquiring the target bird, the method further includes:

[0054] Acquire images of birds in the target area;

[0055] Deep learning models were used to extract the color, texture, and shape of bird images.

[0056] Based on the color, texture, and shape, the corresponding birds are matched in the database, and the birds matched from the database are used as the target birds.

[0057] By adopting the above technical solution, the accuracy of the target birds obtained by matching conditions such as color, texture, and shape is relatively high. After selecting appropriate bird-repelling audio for the target birds, this application can achieve the purpose of improving the bird-repelling effect.

[0058] The second objective of this application is to provide a voice-activated bird deterrent device.

[0059] The second objective of this application is achieved through the following technical solution:

[0060] A voice-activated bird deterrent device for performing any of the above-mentioned voice-activated bird deterrent methods includes a microwave radar sensor, a processor, a wide-angle camera, and a loudspeaker;

[0061] The microwave radar sensor is used to monitor the target area in real time. When birds move to a distance of less than a preset distance between the birds and the microwave radar sensor, the microwave radar sensor outputs a wake-up signal to the processor.

[0062] The processor includes a central processing unit, an AI image recognition sensor, and a memory, wherein the memory stores a database;

[0063] The central processing unit is used to control the wide-angle camera to image birds in the target area when a wake-up signal is received. The wide-angle camera is also used to return the image data to the AI ​​image recognition sensor after imaging.

[0064] The AI ​​image recognition sensor is used to match the birds in the image data with corresponding birds from the memory based on the image data, and to use the birds matched from the memory as target birds;

[0065] The central processing unit includes:

[0066] The data acquisition module is used to acquire data on the target birds.

[0067] The data extraction module is used to extract historical monitoring data corresponding to the target bird from the database. The historical monitoring data includes the time of occurrence, location of occurrence, frequency of occurrence, type of expulsion audio, and escape speed corresponding to each type of expulsion audio.

[0068] The first processing module is used to extract the escape speed corresponding to each type of expulsion audio, and sort the expulsion audio in order of escape speed from high to low to obtain a candidate set;

[0069] The data calculation module is used to calculate the first score F1 for each type of expulsion audio in the candidate set based on the time of occurrence; calculate the second score F2 for each type of expulsion audio in the candidate set based on the location of occurrence; and calculate the third score F3 for each type of expulsion audio in the candidate set based on the frequency of occurrence.

[0070] The second processing module is used to obtain the preset first weight W1, second weight W2, and third weight W3, and calculate the total score D=F1*W1+F2*W2+F3*W3 for each type of expulsion audio in the candidate set.

[0071] The data generation module is used to rearrange the expulsion audios in the candidate set according to the total score from high to low to obtain the target set;

[0072] The loudspeaker is used to play the expulsion audio that is ranked first in the target set.

[0073] By adopting the above technical solution, the voice bird deterrence device can monitor birds in the target area in real time, and identify the type of bird when the distance between the bird and the microwave radar sensor is lower than the preset distance. At the same time, the identified birds are used as target birds, and the historical monitoring data of the target birds are used to provide data support for the current bird deterrence strategy, so that the current bird deterrence effect is better.

[0074] The third objective of this application is to provide a terminal.

[0075] The aforementioned objective three of this application is achieved through the following technical solution:

[0076] A terminal includes a storage component and a processing component, wherein the storage component stores a computer program, and the processing component executes the program to implement any of the above-described voice-based bird deterrence methods.

[0077] The fourth objective of this application is to provide a computer-readable storage medium capable of storing a corresponding program.

[0078] The fourth objective of this application is achieved through the following technical solution:

[0079] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processing component, implements any of the above-described voice-based bird deterrence methods.

[0080] In summary, this application includes at least one of the following beneficial technical effects:

[0081] 1. First, this application identifies the target bird by color, texture, shape, etc., which can ensure the accuracy of the identified target bird;

[0082] 2. Secondly, after identifying the target birds, this application uses historical monitoring data of the target birds to provide data support for the current bird control strategy. Specifically, the historical appearance time, location, frequency, and escape speed of the target birds are used as influencing factors to select the bird control audio with the best bird control effect. The selected bird control audio is then used to drive away the target birds, thereby enabling this application to achieve the goal of improving bird control effectiveness. Attached Figure Description

[0083] Figure 1 This is a schematic diagram of an exemplary operating environment according to an embodiment of this application.

[0084] Figure 2 This is a schematic diagram of the structure of a voice-activated bird deterrent device according to an embodiment of this application.

[0085] Figure 3 This is an exploded view of a voice-activated bird deterrent device according to an embodiment of this application.

[0086] Figure 4 This is a system diagram of a bird deterrent component of a voice-activated bird deterrent device according to an embodiment of this application.

[0087] Figure 5 This is a flowchart of a voice-based bird-repelling method according to an embodiment of this application.

[0088] Figure 6 This is a storage example diagram of the data stored in the database according to an embodiment of this application.

[0089] Figure 7 This is a block diagram of a voice-based bird deterrent system according to an embodiment of this application.

[0090] Explanation of reference numerals in the attached drawings: 100, utility pole; 200, power line; 300, bird deterrent device; 310, housing; 311, support plate; 312, PCB board; 313, connecting plate; 3131, mainboard; 3132, handle; 320, microwave radar sensor; 330, processor; 331, central processing unit; 3311, data acquisition module; 3312, data extraction module; 3313, first processing module; 3314, data calculation module; 3315, second processing module; 3316, data generation module; 332, AI image recognition sensor; 333, memory; 340, wide-angle camera; 350, speaker. Detailed Implementation

[0091] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0092] Figure 1 This is a schematic diagram of an exemplary operating environment according to an embodiment of this application. (Refer to...) Figure 1 The operating environment includes a utility pole 100 and an electric wire 200 and a voice-activated bird deterrent device 300 installed on the utility pole 100. One end of the utility pole 100 is buried underground, and the other end of the utility pole 100 is fixedly connected to the electric wire 200 and the voice-activated bird deterrent device 300, respectively.

[0093] Reference Figure 2 , Figure 3 and Figure 4The voice-activated bird deterrent device 300 includes a housing 310, a microwave radar sensor 320, a processor 330, a speaker 350 disposed within the housing 310, and a wide-angle camera 340 disposed on the top of the housing 310. In a specific example, the housing 310 is a square column shape and is injection molded from ABS material. A support plate 311 is disposed inside the housing 310 and is fixedly connected to the inner wall of the housing 310. The support plate 311 is also made of ABS material. A PCB board 312 is disposed on the support plate 311. In addition to mounting the microwave radar sensor 320, the processor 330, and the speaker 350, the PCB board 312 also has a power supply module for powering the PCB board 312 on its surface facing away from the support plate 311. To improve the power module's battery life, this application configures the power module with a photovoltaic panel. The photovoltaic panel is connected to the power module via a cable and is fixed to the utility pole 100 near the end where the wire 200 is disposed, which is not shown in the accompanying drawings. Photovoltaic panels convert solar energy into electrical energy, which is then transmitted to the battery via cables. The power module can use a lithium battery or a rechargeable battery; however, for compatibility with the housing 310, the power module in this application uses a ternary lithium battery. The power module connects to a power interface on the PCB, thereby supplying power to the PCB board 312. The circuitry on the PCB board 312 then delivers the necessary operating voltages to the microwave radar sensor 320, processor 330, wide-angle camera 340, and speaker 350, respectively.

[0094] Additionally, through holes corresponding to the microwave radar sensor 320 and the speaker 350 are respectively provided on the side wall of the housing 310, so that the microwave radar sensor 320 can collect the required data through the through holes, and the speaker 350 can play the repulsion audio through the through holes. It should be noted that other functional devices may also be provided on the PCB board 312, such as an Internet of Things module for communicating with other voice bird deterrent devices 300, and a photosensitive sensor for monitoring day or night, etc., which will not be described in detail here.

[0095] A connecting plate 313 is provided at the bottom of the housing 310. The connecting plate 313 consists of two parts: a main plate 3131 and a handle 3132. The main plate 3131 is rectangular and fits snugly against the bottom of the housing 310. To enhance the stability of the connection between the housing 310 and the connecting plate 313, threaded holes are evenly spaced around the perimeter of the main plate 3131. Screws pass through the threaded holes and connect to the housing 310, thereby fixing the housing 310 to the main plate 3131. A magnet is provided at the center of the main plate 3131. The magnet can be magnetically connected to a steel bar made of magnetic material. Since the end of the utility pole 100 with the wire 200 is usually equipped with a steel bar, the housing 310 can be fixed to the utility pole 100. In addition, the handle 3132 also has threaded holes. Therefore, screws can also be used to fix the handle 3132 to the utility pole 100 through these threaded holes, further enhancing the stability of the housing 310 located on the utility pole 100.

[0096] The side wall of housing 310 away from connecting plate 313 forms the top of housing 310. A wide-angle camera 340 is installed on the top of housing 310. The wide-angle camera 340 uses a servo system, enabling it to continuously track the bird's position and acquire images of the bird from multiple angles. Simultaneously, a water channel is provided on the top of housing 310 to drain water accumulated on the housing 310 and prevent water from seeping into the interior of housing 310.

[0097] Reference Figure 4 The processor 330 is connected to the microwave radar sensor 320, the wide-angle camera 340, and the speaker 350. The processor 330 includes a central processing unit 331, an AI image recognition sensor 332, and a memory 333. The memory 333 stores various bird types crawled from the internet and audio recordings used to scare away birds. A database is built based on these bird types and bird-scare audio recordings, which can be supplemented by bird experts. During the use of the voice-activated bird deterrent device 300, the microwave radar sensor 320 monitors the target area in real time. When birds move to a distance below a preset distance from the microwave radar sensor 320, the microwave radar sensor 320 outputs a wake-up signal to the processor 330. The processor 330 then controls the wide-angle camera 340 to image the birds in the target area and outputs the image data to the processor 330. The processor 330 identifies the birds in the image data and selects a bird-scare audio recording from the database based on the identified birds to scare away the birds moving in the target area.

[0098] The target area mentioned above refers to the spatial area centered on the microwave radar sensor 320 and with a preset distance as the radius. In this example, when the distance between the bird and the microwave radar sensor 320 is lower than the preset distance, the bird may perch on the utility pole 100 or the power line 200. This preset distance can be determined through a limited number of trials. Therefore, when birds move to a distance below the preset distance between themselves and the microwave radar sensor 320, they need to be driven away.

[0099] To better deter birds that are less than a preset distance from the microwave radar sensor 320, this application proposes a voice-based bird deterrence method, which is executed by the processor 330. (See reference...) Figure 5 The main process of the voice-based bird deterrence method is described below.

[0100] Step S1: Obtain the target bird.

[0101] The target bird refers to a bird whose distance from the microwave radar sensor 320 is less than a preset distance. The process of acquiring the target bird includes: First, the microwave radar sensor 320 monitors the target area in real time. When a bird moves to a distance less than the preset distance from the microwave radar sensor 320, the microwave radar sensor 320 outputs a wake-up signal to the processor 330. The processor 330 then controls the wide-angle camera 340 to image the birds in the target area; that is, the wide-angle camera 340 captures images of the birds in the target area and generates bird images, which are then sent to the processor 330 for identification. The processor 330 uses a deep learning model to extract the color, markings, and shape of the birds in the bird images, and selects the birds with the highest similarity in color, markings, and shape to those in the bird images from the database as the target birds. Figure 6 As shown, after inputting image data, the color, texture, and shape of the birds in the image data are identified respectively.

[0102] Step S2: Extract historical monitoring data corresponding to the target bird from the database. The historical monitoring data includes the time of occurrence, location of occurrence, frequency of occurrence, type of expulsion audio, and escape speed corresponding to each type of expulsion audio.

[0103] like Figure 6 As shown, the database also stores historical monitoring data corresponding to the target bird, including the time of appearance, location of appearance, frequency of appearance, type of deterrent audio, and escape speed corresponding to each type of deterrent audio. Therefore, when the processor 330 identifies the target bird based on the bird image, it can directly retrieve the historical monitoring data corresponding to the target bird from the database.

[0104] The aforementioned occurrence time refers to the moment when the processor 330 receives the wake-up signal output by the microwave radar sensor 320, which is recorded by the clock chip built into the processor 330. Regarding the location of occurrence, since multiple voice-activated bird deterrent devices 300 can be installed on the power transmission line 200, and these devices can communicate via an IoT module, their historical monitoring data is interconnected. Therefore, with multiple voice-activated bird deterrent devices 300 sharing their historical monitoring data, the target bird can appear in multiple locations.

[0105] The aforementioned frequency of occurrence refers to the number of times that a specific type of bird, representing the target bird, moves to a distance below a preset distance from the microwave radar sensor 320. This frequency of occurrence is not limited to a single target bird, but rather includes the total number of occurrences for the entire type of target bird. For example, if there is a magpie E in the current target area, and before magpie E was driven away, magpies A, B, C, and D were also driven away (A, B, C, D, and E are five different magpies), then when retrieving the historical monitoring data for magpie E, the frequency of magpie E's occurrence is five times, which includes the occurrences of magpies A, B, C, and D.

[0106] The aforementioned bird-repelling audio type refers to the bird-repelling audio played by the speaker 350 controlled by the processor 330 when birds move to a distance below a preset distance between the birds and the microwave radar sensor 320. This bird-repelling audio is also called bird-repelling audio.

[0107] It should be noted that the database also combines bird-repelling audio used to drive away target birds into a raw set, such as... Figure 6 In this example, the expulsion audios 1 to i are all used to expel target birds, and i ≥ 1. To facilitate the explanation of the source of the expulsion audios, this example takes the source of the expulsion audios of magpies as an example: Magpies have natural enemies such as goshawks and peregrine eagles, and magpies are also afraid of rhythmic audio. Therefore, for magpies, the expulsion audios include the sound of goshawks, the sound of peregrine eagles, and various rhythmic sound waves.

[0108] Escape speed refers to the speed at which a bird moves away from the microwave radar sensor 320 when the loudspeaker 350 plays a dispersal audio signal. Specifically, when the loudspeaker 350 plays the audio, the microwave radar sensor 320 collects several nodes in the bird's flight path. The processor 330 determines the path based on these nodes and calculates the bird's escape speed based on the time the bird reaches each node. Of course, other methods can also be used to calculate the bird's escape speed when being dispersed; this calculation is not limited here. Furthermore, since a dispersal audio signal may be played multiple times by the loudspeaker 350, one type of dispersal audio signal may correspond to multiple escape speeds, such as... Figure 6 In the text, the expulsion audio corresponds to escape speeds from 1 to j, where j ≥ 1.

[0109] Step S3: Extract the escape speed corresponding to each type of expulsion audio, and sort the expulsion audio in descending order of escape speed to obtain a candidate set.

[0110] like Figure 6 As shown, once the target bird species is identified, each type of expulsion audio corresponding to the target bird species can be determined. Then, the escape speed corresponding to each type of expulsion audio is extracted. If a type of expulsion audio corresponds to multiple escape speeds, the average value of the expulsion audio is calculated and used as the escape speed of that expulsion audio. Finally, the expulsion audio is sorted in descending order of escape speed to obtain a candidate set.

[0111] It should be noted that, to avoid birds hearing only one type of deterrent audio for an extended period, and thus, through prolonged adaptation, becoming less afraid of the deterrent audio played by speaker 350, this application aims to play the deterrent audio from the original set in rotation. After one or more cycles, the bird deterrent method provided in this application is used as the basis for deterring birds at that moment. Therefore, this application can obtain multiple types of deterrent audio and the escape speed corresponding to each type of deterrent audio, and then sort the deterrent audio in descending order of escape speed to obtain a candidate set.

[0112] Step S4: Calculate the first score F1 for each type of expulsion audio in the candidate set based on the occurrence time; calculate the second score F2 for each type of expulsion audio in the candidate set based on the occurrence location; calculate the third score F3 for each type of expulsion audio in the candidate set based on the occurrence frequency.

[0113] First, the further away the occurrence time is from the current monitoring time, the higher the first score. The current monitoring time refers to the moment when the processor 330 receives the wake-up signal output by the microwave radar sensor 320. In one possible implementation, the process of calculating the first score includes: first, obtaining the original set, and then calculating the time period t for the expulsion audio in the original set to play once. If the time period t is longer, it indicates that the frequency of bird activity is lower. In this case, the cyclical order can be used as the unit of score. Specifically: set a preset period T. When the time period t reaches the preset period T, divide the expulsion audio into multiple groups according to the order of the expulsion audio in the original set. Each group contains multiple expulsion audio. Then sort the groups to obtain a sequence set. The expulsion audio closest to the current monitoring time is taken as the center audio. The first score of all expulsion audio in the group containing the center audio is 0. In the sequence set, the first score of the groups after the group containing the center audio increases by 2 points in order from closest to furthest. In the sequence set, the first score of the groups before the group containing the center audio increases by 1 point in order from closest to furthest. For example, if the original set contains expulsion audios a, b, c, d, e, f, g, and h, and the time period t required for speaker 350 to play these expulsion audios is 5 days, while the set period is T=4 days, then the expulsion audios can be divided into three groups: (a, b, c), (d, e, f), and (g, h). If the current monitoring time should cycle to expulsion audio d, i.e., d is the center audio, then the first score for expulsion audios d, e, and f is 0, while the first score for expulsion audios g and h is 2, and the first score for expulsion audios a, b, and c is 1. It should be noted that in this application, when dividing the groups, the more expulsion audios there are, the fewer the number of groups; conversely, the fewer the expulsion audios, the more groups. However, the number of groups should not exceed 10 to avoid excessive differences in the first score between groups, which would reduce the accuracy of the calculated first score.

[0114] In this application, if the time period t is lower than the preset period T, it indicates that the target bird's activity is frequent. In this case, the first score F1 = d, where d is the number of days since the current monitoring time. For example, if there are expulsion audio A and expulsion audio B, and expulsion audio A has been monitored for 15 days since the current monitoring time, while expulsion audio B has been monitored for 22 days, then the first score for expulsion audio A is 15 points, and the first score for expulsion audio B is 22 points. In other examples, other calculation methods can be used to calculate the first score, and this application does not impose any restrictions on this.

[0115] Regarding the second score, the farther the location in the historical monitoring data is from the current monitoring location, the higher the second score. The current monitoring location is where the microwave radar sensor 320 detected the target bird. In one possible implementation, the process of calculating the second score is as follows: First, obtain the location of each type of deterrent audio. If there are multiple locations for the deterrent audio, select the common center of the multiple locations as the location of the deterrent audio. Then calculate the second score F2 = L / 5, where L is the distance from the location of the deterrent audio to the current monitoring location, and 5 is a preset value in kilometers. When L < 5, the second score F2 = 0; when L ≥ 5, the second score F2 is the integer part of L / 5. This application sets the preset value to 5 kilometers because most birds cannot hear the played deterrent audio beyond 5 kilometers from the playback source, hence the preset value of 5 kilometers. Therefore, in practical applications, an appropriate preset value can be selected based on the hearing range of birds in different areas.

[0116] Regarding the third score, the lower the frequency of occurrence, the higher the score. In one possible implementation, the number of times the expulsion audio in the candidate set is played is counted, and a first preset value N is set. When the number of plays n reaches or exceeds the first preset value N, the third score F3 = 0; when the number of plays n is less than the first preset value N, the third score F3 = Nn, where the first preset value N is 10. It should be noted that since the expulsion audio in the original set is played in a loop, the difference in the number of plays between any two expulsion audio will not exceed one.

[0117] Step S5: Obtain the preset first weight W1, second weight W2, and third weight W3, and calculate the total score D for each type of expulsion audio in the candidate set: D = F1*W1 + F2*W2 + F3*W3.

[0118] The total score D = F1*W1 + F2*W2 + F3*W3, indicating that the first weight W1 corresponds to the first score F1, the second weight W2 corresponds to the second score F2, and the third weight W3 corresponds to the third score F3. In this example, the first weight W1 = the second weight W2 = 40%, and the third weight is 20%. The third weight W3 is set lower than the first weight W1 and the second weight W2 because the difference in the number of playbacks of any two expulsion audios does not exceed one. Therefore, the difference in the value of the third score F3 calculated based on the number of playbacks is not large. Thus, in order to reduce the impact of the third score F3 on the total score D, the third weight W3 is set lower than the first weight W1 and the second weight W2.

[0119] Step S6: Rearrange the expulsion audios in the candidate set according to the total score from high to low to obtain the target set.

[0120] After calculating the total score, the expulsion audios in the candidate set are readjusted in descending order of total score to obtain the target set. The processor 330 will prioritize the speaker 350 to play the expulsion audio with the highest score according to the order of the expulsion audios in the target set.

[0121] It should be noted that, in order to reduce the computational load on the processor 330, this application sets a second preset value M. When the number of times the processor 330 controls the speaker 350 to play the repulsion audio according to the order of the target set reaches the second preset value M, the processor 330 recalculates the target set. The setting of the second preset value M is related to the frequency of bird activity. The more frequent the bird activity, the smaller the second preset value M; conversely, the less frequent the bird activity, the larger the second preset value M. Based on this, the processor 330 can iteratively play the repulsion audio for frequently active birds, while for birds with low activity frequency, it can maintain a good bird-repelling effect while reducing the computational load, thereby improving the service life of the processor 330.

[0122] This application also provides a voice-activated bird deterrent system, which is located within processor 330, specifically within central processing unit 331. Figure 7 As shown, a voice-based bird deterrence system includes a data acquisition module 3311, a data extraction module 3312, a first processing module 3313, a data calculation module 3314, a second processing module 3315, and a data generation module 3316. The data acquisition module 3311, data extraction module 3312, first processing module 3313, data calculation module 3314, second processing module 3315, and data generation module 3316 are connected sequentially.

[0123] Specifically, the data acquisition module 3311 is used to acquire target birds. The data extraction module 3312 is used to extract historical monitoring data corresponding to the target birds from the database. The historical monitoring data includes the occurrence time, occurrence location, occurrence frequency, type of deterrent audio, and escape speed corresponding to each type of deterrent audio. The first processing module 3313 is used to extract the escape speed corresponding to each type of deterrent audio and sort the deterrent audio in descending order of escape speed to obtain a candidate set. The data calculation module 3314 is used to calculate the first score F1 of each deterrent audio in the candidate set based on the occurrence time; calculate the second score F2 of each deterrent audio in the candidate set based on the occurrence location; and calculate the third score F3 of each deterrent audio in the candidate set based on the occurrence frequency. The second processing module 3315 is used to acquire the preset first weight W1, second weight W2, and third weight W3, and calculate the total score D of each deterrent audio in the candidate set D=F1*W1+F2*W2+F3*W3. The data generation module 3316 is used to readjust the expulsion audio in the candidate set according to the total score from high to low to obtain the target set.

[0124] To better execute the above-described method, this application also provides a terminal, which includes a storage component and a processing component.

[0125] The storage component can be used to store instructions, programs, code, code sets, or instruction sets. The storage component may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the aforementioned voice-based bird-repelling method; the data storage area may store data involved in the aforementioned voice-based bird-repelling method.

[0126] The processing component may include one or more processing cores. The processing component executes or runs instructions, programs, code sets, or instruction sets stored in the storage component, calls data stored in the storage component, and performs various functions and processes data as described in this application. The processing component may be at least one of a specific application-specific integrated circuit, a digital signal processor, a digital signal processing device, a programmable logic device, a field-programmable gate array, a controller, a microcontroller, and a microprocessor. It is understood that, for different devices, the electronic devices used to implement the functions of the above-described processing component may also be other types, and the embodiments of this application do not specifically limit this.

[0127] This application also provides a computer-readable storage medium, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and other media capable of storing program code. This computer-readable storage medium stores a computer program that can be loaded by a processing component and executed using the aforementioned voice-based bird-repelling method.

[0128] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for repelling birds using voice commands, characterized in that, include: Acquire the target birds; Extract historical monitoring data corresponding to the target bird from the database. The historical monitoring data includes the time of occurrence, location of occurrence, frequency of occurrence, type of expulsion audio, and escape speed corresponding to each type of expulsion audio. Extract the escape speed corresponding to each type of expulsion audio, and sort the expulsion audio in descending order of escape speed to obtain a candidate set; Calculate the first score F1 for each type of expulsion audio in the candidate set based on its occurrence time; The calculation of the second score F2 for each type of expulsion audio in the candidate set based on the location of occurrence includes: obtaining the current monitoring location, the distance L between the location of occurrence and the current monitoring location, and a preset value of 5, where 5 is in kilometers; when L < 5, the second score F2 = 0; when L ≥ 5, the second score F2 is the integer part of L / 5; The third score F3 for each type of expulsion audio in the candidate set is calculated based on its frequency of occurrence, including: counting the number of times n of expulsion audio in the candidate set is played; when the number of times n is played reaches or exceeds the first preset value N, the third score F3 = 0; when the number of times n is played is less than the first preset value N, the third score F3 = Nn. Obtain the preset first weight W1, second weight W2, and third weight W3, and calculate the total score D for each type of expulsion audio in the candidate set: D=F1*W1+F2*W2+F3*W3; The target set is obtained by rearranging the expulsion audio from the candidate set in descending order of total score; Play the first expulsion audio in the target set according to the order of the expulsion audios; when the number of times the expulsion audios in the target set are played reaches the second preset value M, the target set is recalculated; the setting of the second preset value M is related to the frequency of bird activity. The more frequent the bird activity, the smaller the second preset value M; the lower the frequency of bird activity, the larger the second preset value M.

2. The voice-based bird-repelling method according to claim 1, characterized in that, Before calculating the first score F1 for each expulsion audio in the candidate set based on its occurrence time, the method further includes: Obtain the original set, which stores various expulsion audios used to drive away target birds; Calculate the time period t for one loop of the expulsion audio in the original set.

3. The voice-based bird-repelling method according to claim 2, characterized in that, The calculation of the first score F1 for each type of expulsion audio in the candidate set based on its occurrence time includes: When the time period t reaches the preset period T, the expulsion audio is divided into multiple groups according to the original centralized expulsion audio, and each group contains multiple expulsion audio. Then sort the resulting groups to obtain a sequence set; The expulsion audio that is closest to the current monitoring time will be used as the center audio. All expulsion audios within the group containing the central audio have a first score of 0. In the sequence set, the groups that are located after the group containing the central audio are sorted in ascending order of their order, with the first score increasing by 2 points each time. In the sequence set, the groups that precede the group containing the central audio are assigned a first score that increases by 1 point in order from closest to furthest from the group.

4. The voice-based bird-repelling method according to claim 2, characterized in that, The calculation of the first score F1 for each type of expulsion audio in the candidate set based on its occurrence time includes: When the time period t is lower than the preset period T, the first score F1 = d; Where d is the number of days since the current monitoring time.

5. The voice-based bird-repelling method according to claim 1, characterized in that, Before acquiring the target bird, the method further includes: Acquire images of birds in the target area; Deep learning models were used to extract the color, texture, and shape of bird images; Based on the color, texture, and shape, the corresponding birds are matched in the database, and the birds matched from the database are used as the target birds.

6. A voice-activated bird deterrent device (300) for performing the method as described in any one of claims 1-5, characterized in that, Includes a microwave radar sensor (320), a processor (330), a wide-angle camera (340), and a speaker (350); The microwave radar sensor (320) is used to monitor the target area in real time. When the distance between the bird and the microwave radar sensor (320) is lower than the preset distance, the microwave radar sensor (320) outputs a wake-up signal to the processor (330). The processor (330) includes a central processing unit (331), an AI image recognition sensor (332), and a memory (333) storing a database; The central processing unit (331) is used to control the wide-angle camera (340) to image birds in the target area when a wake-up signal is received. The wide-angle camera (340) is also used to return the image data to the AI ​​image recognition sensor (332) after imaging. The AI ​​image recognition sensor (332) is used to match the birds in the image data with the corresponding birds from the memory (333) based on the image data, and to use the birds matched from the memory (333) as the target birds; The central processing unit (331) includes: The data acquisition module (3311) is used to acquire data on the target birds; The data extraction module (3312) is used to extract historical monitoring data corresponding to the target bird from the database. The historical monitoring data includes the time of occurrence, the location of occurrence, the frequency of occurrence, the type of expulsion audio, and the escape speed corresponding to each type of expulsion audio. The first processing module (3313) is used to extract the escape speed corresponding to each type of expulsion audio and sort the expulsion audio in order of escape speed from high to low to obtain a candidate set. The data calculation module (3314) is used to calculate the first score F1 of each type of expulsion audio in the candidate set based on the occurrence time; calculate the second score F2 of each type of expulsion audio in the candidate set based on the occurrence location; and calculate the third score F3 of each type of expulsion audio in the candidate set based on the occurrence frequency. The second processing module (3315) is used to obtain the preset first weight W1, second weight W2, and third weight W3, and calculate the total score D=F1*W1+F2*W2+F3*W3 for each type of expulsion audio in the candidate set. The data generation module (3316) is used to readjust the expulsion audio in the candidate set according to the total score from high to low to obtain the target set; The speaker (350) is used to play the expulsion audio that is ranked first in the target set.

7. A terminal, characterized in that, It includes a storage component and a processing component, wherein the storage component stores a computer program, and the processing component executes the program to implement the method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processing component, implements the method as described in any one of claims 1-5.

Images