Agricultural insurance unmanned aerial vehicle and dynamic path optimization method

Abstract

The invention discloses an agricultural protection unmanned aerial vehicle and a dynamic path optimization method, and belongs to the technical field of unmanned aerial vehicles. The agricultural protection unmanned aerial vehicle comprises an unmanned aerial vehicle body, a mounting part is arranged on the unmanned aerial vehicle body, a pesticide barrel which is vertically arranged and used for containing pesticide liquid is arranged in the mounting part, and a plurality of nozzles which are annularly arranged and used for pesticide spraying operation are arranged at the bottom of the pesticide barrel; the unmanned aerial vehicle body is further provided with an acquisition module used for acquiring a whole area image of a plant operation area and an image processing module used for identifying the acquired image according to plants and marking plant growth density in a partitioned mode. Compared with the prior art, the self-adaptive pesticide spraying device has the advantages that self-adaptive pesticide distribution and spraying are realized according to different growth densities, the conditions of excessive spraying and insufficient spraying of the plants in areas with lower growth densities are avoided, and the spraying efficiency of the plants is improved. The crop planting quality is improved, and the high use value is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of observation and strike unmanned aerial vehicles, and more specifically, to an agricultural protection unmanned aerial vehicle and a dynamic path optimization method. Background Art

[0002] An agricultural protection unmanned aerial vehicle refers to a vehicle equipped with technical devices such as a high-definition camera and an infrared camera, which can accurately monitor the growth of crops in the operation area, promptly detect the occurrence of pests and diseases. Once pests and diseases are detected, the unmanned aerial vehicle can quickly and accurately spray pesticides, insecticides, etc. to prevent the spread and spread of pests and diseases.

[0003] At present, the existing agricultural protection unmanned aerial vehicles can only spray the operation area according to the preset spraying amount. This spraying method is difficult to adaptively adjust the growth density of plants, resulting in over-spraying of plants in areas with low growth density and insufficient spraying of plants in areas with high growth density. Summary of the Invention

[0004] The present invention provides an agricultural protection unmanned aerial vehicle and a dynamic path optimization method, which can overcome certain or some defects of the prior art.

[0005] According to an agricultural protection unmanned aerial vehicle of the present invention, it includes a unmanned aerial vehicle body. An installation part is provided on the unmanned aerial vehicle body. A medicine cylinder for holding liquid medicine is vertically arranged in the installation part. A plurality of nozzles for spraying operations are annularly arranged at the bottom of the medicine cylinder; the unmanned aerial vehicle body is also provided with a collection module for collecting the full-area images of the plant operation area, an image processing module for identifying the plants in the collected images and marking the growth density of the plants in different zones, and a spraying amount control module for spraying different amounts of liquid medicine according to the plants marked in different growth density zones.

[0006] Preferably, the image processing module includes an image processing module and a judgment module;

[0007] The image processing module is used for identifying and marking the images collected by the collection module;

[0008] The judgment module is used for judging the growth density of the plants at the marked points and the density zones of the growth density of the plants at the marked points, obtaining the growth density zone information of the plants at the marked points and sending the information to the spraying amount control module.

[0009] The plants in the full-area images of the plant operation area collected are identified and marked, and are divided into different zones according to the growth density of the plants, so that the spraying amount control module can accurately control the spraying amount according to the growth density of the plants, avoiding waste of liquid medicine and environmental pollution.

[0010] Preferably, before the image processing module identifies and marks the images collected by the acquisition module, there is also an image preprocessing operation. Specifically, the image preprocessing is to denoise the image and enhance the image contrast.

[0011] Through the above operations, the quality of the images can be greatly improved, the recognizability of the image information can be enhanced, and strong support can be provided for subsequent image identification and marking.

[0012] Preferably, the preprocessed image is marked as the input image, a rapid plant image recognition model is constructed, the input image is used as the input data of the rapid plant image recognition model, and the position map of the plant in the plant operation area image is obtained.

[0013] Through the above operations, plants at different growth stages, of different varieties, and under different lighting conditions can be quickly identified and marked.

[0014] Preferably, the way for the judgment module to judge the growth density of the plants at the marked positions is as follows: a plant image analysis model is constructed, the position map of the plant in the plant operation area image obtained is equally divided and marked as the input photo, the input photo is used as the input data of the plant image analysis and recognition model, the output data is marked as the target label, and the target label is marked as Ms; when the target label Ms ∈ [0, 3], the target position is marked as green, when the target label Ms ∈ (3, 7], the target position is marked as yellow, and when the target label Ms ∈ (7, 10], the target position is marked as red, forming a plant position growth density map.

[0015] Through the above operations, the growth density of the plant positions can be marked and a plant position growth density map can be generated, which can not only accurately identify and partition the plant growth density, facilitating precise spraying on the plant positions during drone spraying, but also be more efficient and accurate compared to traditional on-site sampling and laboratory analysis, greatly saving manpower and material resources;

[0016] In fact, the generated plant position growth density map can visually display the plant growth density, which is not only convenient for managers to quickly understand the plant growth situation, but also serves as a decision-making support tool to help them formulate more scientific and reasonable planting management plans.

[0017] Preferably, the zoning method for the judgment module to judge the growth density of the plants at the marked positions is: set the target label coefficient as Dh; use the formula

[0018]

[0019] Obtain the plant density value Yj, where i = 1, 2... n and n is the number of equal parts of the photographed images. When the plant density value Yj ≥ the density threshold, mark the grape plant position as a key area; otherwise, mark it as a normal area.

[0020] Through the above operations, the growth density of plants can be clearly divided into normal areas and key areas, which can not only clarify the secondary focus of the UAV spraying but also help managers clarify the management priorities and improve the management efficiency.

[0021] Preferably, the plant image quick recognition model is obtained through the following steps: Obtain n pieces of plant image data, mark the plant image data as training images, assign image labels to the training images, divide the training images into a training set and a validation set according to a set ratio, construct a neural network model, and perform iterative training on the neural network model through the training set and the validation set. When the number of iterative training times is greater than the iteration threshold, it is determined that the neural network model is trained. Mark the trained neural network model as the plant image analysis model. The value range of the image label is [0 - 10], where the larger the value of the image label, the greater the growth density of the position.

[0022] Through the above operations, the automatic and intelligent recognition of plant images can be realized, and the growth density of plant positions can be quickly and accurately recognized, greatly improving the efficiency and accuracy of recognition. The growth density of plant positions can be accurately recognized, enabling the UAV to more precisely control the spraying amount for areas with different growth densities.

[0023] Preferably, the plant image data includes images under different varieties, different growth stages, and different lighting conditions.

[0024] Through the above operations, the model can learn more diverse and complex image features, thereby improving the accuracy and generalization ability of recognition.

[0025] Preferably, the spraying amount control module includes a liquid pump installed in the medicine barrel and a control unit. The control unit is used to receive the data sent by the judgment module and control the flow rate of the liquid pump. When the UAV body flies above the plants in the key area, increase the spraying amount; when the UAV body flies above ordinary plants, keep the original amount or reduce it.

[0026] Through the intelligent adjustment method of the spraying amount control module in the above operations, when flying above plants with high growth density, ensure that the liquid medicine can fully cover the plants to achieve a better prevention and control effect; when flying above plants with low growth density, keep the original amount to avoid waste of pesticides, greatly improving the spraying efficiency and accuracy, optimizing the crop growth environment, and reducing labor costs.

[0027] Preferably, a connecting pipe for installing a plurality of nozzles is fixedly provided at the output end of the liquid pump.

[0028] With the above structure, the spraying coverage is improved. At the same time, by adopting the method of centralized liquid supply and transporting the liquid medicine to each nozzle through the liquid pump, the spraying reliability can be improved. Even if a certain nozzle fails, it will not affect the operation of the entire spraying operation.

[0029] A dynamic path optimization method for an agricultural protection unmanned aerial vehicle includes the following steps:

[0030] Step 1: Obtain the information of the plant growth density partition of the marked points.

[0031] Step 2: Combine the plant growth density partition information with the geographical coordinate information of the orthophoto map established by the aerial survey software, and perform coordinate transformation through the coordinate transformation method to obtain the geographical coordinate information of the plant growth density partition.

[0032] Step 3: Generate a map of the plant growth density partition based on the geographical coordinate information of the plant growth density partition.

[0033] Step 4: Obtain the relevant parameter information of the agricultural protection unmanned aerial vehicle and the boundary information of the plant operation area.

[0034] Step 5: Combine the above-mentioned map of the plant growth density partition, the relevant parameter information of the agricultural protection unmanned aerial vehicle, and the boundary information of the plant operation area, and plan the flight path through the aerial survey software.

[0035] The beneficial effects of the present invention are as follows:

[0036] 1. The present invention can realize the adaptive drug dosage distribution and spraying according to the different growth densities of plants during the operation process, can greatly achieve the precise spraying of drugs, avoid the over-spraying of plants in the area with lower growth density and the under-spraying of plants in the area with higher growth density, improve the crop planting quality, and has high use value.

[0037] 2. By identifying and marking key and ordinary areas and using them as path points, the path planning of the unmanned aerial vehicle is realized, and the path planning of the flight path can be realized according to different plant growth states, which greatly improves the precision of the unmanned aerial vehicle spraying, is conducive to the precise management of spraying, and provides a new solution and new idea for the current research on unmanned aerial vehicle spraying. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a schematic diagram of the overall structure of an agricultural protection unmanned aerial vehicle;

[0039] Figure 2 is a schematic side view structure diagram of an agricultural protection unmanned aerial vehicle;

[0040] Figure 3 It is the recognition and marking flowchart of the image processing module;

[0041] Figure 4 It is the judgment density partition flowchart of the judgment module;

[0042] Figure 5 It is the flight route generation flowchart. Specific implementation manner

[0043] To further understand the content of the present invention, the present invention will be described in detail in combination with embodiments. It should be understood that the embodiments are only for explaining the present invention rather than limiting it.

[0044] Embodiment 1

[0045] Please refer to Figures 1 - 4 , this embodiment provides an agricultural insurance UAV, including a UAV main body 100. An installation part is provided on the UAV main body 100. A medicine cylinder 210 for containing liquid medicine is vertically arranged in the installation part. A plurality of nozzles 220 for spraying operations are annularly arranged at the bottom of the medicine cylinder 210; a collection module for collecting the full-area images of the plant operation area is also provided on the UAV main body 100, an image processing module for identifying the plants in the collected images and partitioning and marking the plant growth density, and a spraying amount control module for spraying different spraying amounts according to the plants marked with different growth density partitions.

[0046] When this embodiment is in use, the user controls the UAV main body 100 to fly to collect the full-area images of the plant operation area. After the collection is completed, the collected images are processed, identified and marked by the image processing module to generate a plant point growth density map and divide the key area and the ordinary area according to the plant point growth density map. During the operation process, the spraying amount control module controls the liquid pump flow rate to perform adaptive spraying according to the key area and the ordinary area divided by the plant point growth density map. Compared with the prior art, the present invention can, during the operation process, according to the plant growth density in the operation area, realize the adaptive drug dosage distribution and spraying according to different growth densities, can greatly achieve the precise spraying of drugs, avoid the over-spraying of the plants in the area with lower growth density and the insufficient spraying of the plants in the area with higher growth density, improve the crop planting quality, and has high use value.

[0047] The image processing module includes an image processing module and a judgment module;

[0048] The image processing module is used to identify and mark the images collected by the collection module;

[0049] The judgment module is used to judge the growth density of the plants at the marked points and the density zoning of the growth density of the plants at the marked points, obtain the information on the growth density zoning of the plants at the marked points, and send the information to the pesticide spraying amount control module.

[0050] Identify and mark the plants in the full-area image of the plant operation area collected, and divide them into zones according to the growth density of the plants, so that the pesticide spraying amount control module can accurately control the pesticide spraying amount according to the growth density of the plants, avoiding waste of liquid medicine and environmental pollution.

[0051] Before the image processing module identifies and marks the image collected by the collection module, there is also an image preprocessing operation. Specifically, the image is denoised and the contrast of the image is enhanced.

[0052] Through the above operations, the quality of the image can be greatly improved, the recognizability of the image information can be enhanced, and strong support can be provided for subsequent image recognition and marking.

[0053] Mark the preprocessed image as the input image, construct a rapid plant image recognition model, use the input image as the input data of the rapid plant image recognition model, and obtain the point map of the plants in the plant operation area image.

[0054] Among them, the plant image is linearly marked and supervised learning training is carried out in advance to obtain the weights of the model, and then a rapid plant recognition model is constructed. By constructing the rapid plant recognition model, the plant information in the plant image can be recognized and marked.

[0055] Through the above operations, plants at different growth stages, of different varieties and under different lighting conditions can be quickly recognized and marked.

[0056] The method for the judgment module to judge the growth density of the plants at the marked points is as follows: construct a plant image analysis model, divide the obtained point map of the plants in the plant operation area image equally and mark it as the input photo, use the input photo as the input data of the plant image analysis and recognition model, mark the output data as the target label, and mark the target label as Ms; when the target label Ms ∈ [0, 3], then mark the target position as green, when the target label Ms ∈ (3, 7], then mark the target position as yellow, when the target label Ms ∈ (7, 10], then mark the target position as red, and form a plant point growth density map.

[0057] Among them, green represents the low-density area, yellow represents the medium-density area, and red represents the high-density area. The formed plant point growth density map can be received and viewed through the display terminal.

[0058] Through the above operations, the growth density of plant positions can be marked, and a growth density map of plant positions can be generated, which can not only accurately identify and partition the growth density of plants, facilitate precise spraying at plant positions during drone spraying, but also be more efficient and accurate than traditional on-site sampling and laboratory analysis, greatly saving manpower and material resources;

[0059] In fact, the generated growth density map of plant positions can visually display the growth density of plants, which is not only convenient for managers to quickly understand the growth situation of plants, but also serves as a decision-making support tool to help them formulate more scientific and reasonable planting management plans.

[0060] The judgment module is used to judge the partitioning method of the growth density of plants at the marked positions as follows: set the target label coefficient as Dh; use the formula

[0061]

[0062] Obtain the plant density value Yj, where i = 1, 2...n, and n is the number of equal parts of the photographed image. When the plant density value Yj ≥ the density threshold, mark the position of the grape plant as a key area, otherwise mark the position of the grape plant as an ordinary area.

[0063] Through the above operations, the growth density of plants can be clearly divided into ordinary areas and key areas, which can not only clarify the secondary focus of spraying by drones, but also help managers clarify the management priorities and improve management efficiency.

[0064] Specifically, after equally dividing the position map of the plants in the plant operation area image obtained, data standardization can be achieved, the calculation amount can be reduced, local features can be captured, and the plant image analysis model can maintain the same performance and stability when processing different data.

[0065] In fact, clearly dividing the growth state of plants into ordinary areas and key areas can provide a basis for the main path points and auxiliary path points of the subsequent flight path, so as to realize the path planning with key areas as the main path points and ordinary areas as the auxiliary path points, greatly improving the accuracy of drone spraying.

[0066] The plant image rapid recognition model is obtained through the following steps: Obtain n pieces of plant image data, mark the plant image data as training images, assign image labels to the training images, divide the training images into a training set and a validation set according to a set ratio, construct a neural network model, and perform iterative training on the neural network model through the training set and the validation set. When the number of iterative training times is greater than the iterative training times threshold, it is determined that the neural network model is trained. Mark the trained neural network model as the plant image analysis model. The value range of the image label is [0 - 10], where the larger the value of the image label, the greater the growth density of the point position.

[0067] Through the above operations, the automatic and intelligent recognition of plant images can be realized, the growth density of the point positions of plants can be quickly and accurately recognized, the efficiency and accuracy of recognition are greatly improved, the growth density of the point positions of plants can be accurately recognized, and the UAV can more accurately control the spraying amount for areas with different growth densities.

[0068] The plant image data includes images under different varieties, different growth stages, and different lighting conditions.

[0069] Through the above operations, the model can learn more diverse and complex image features, thereby improving the accuracy and generalization ability of recognition.

[0070] The spraying amount control module includes a liquid pump provided in the medicine barrel 210 and a control unit. The control unit is used to receive the data sent by the judgment module to control the flow rate of the liquid pump. When the UAV main body 100 flies above the plants in the key area, the spraying amount is increased. When the UAV main body 100 flies above ordinary plants, the original amount is maintained or reduced.

[0071] Through the intelligent adjustment method of the spraying amount control module in the above operations, when flying above plants with high growth density, it ensures that the liquid medicine can fully cover the plants to achieve a better prevention and control effect. When flying above plants with low growth density, the original amount is maintained to avoid waste of pesticides, greatly improving the spraying efficiency and accuracy, optimizing the crop growth environment, and reducing labor costs.

[0072] In this embodiment, a connecting pipe for installing a plurality of nozzles 220 is fixedly provided at the output end of the liquid pump.

[0073] Through the above structure, the spraying coverage range is improved. At the same time, by adopting the method of centralized liquid supply and transporting the liquid medicine to each nozzle 220 through the liquid pump, the reliability of spraying can be improved. Even if a certain nozzle 220 fails, it will not affect the operation of the entire spraying operation.

[0074] Embodiment 2

[0075] Please refer toFigure 5 , based on Embodiment 1, this embodiment provides a dynamic path optimization method for an agricultural insurance UAV, including the following steps:

[0076] Step 1: Obtain the information on the division of plant growth density in the marked points.

[0077] Step 2: Combine the information on the division of plant growth density with the geographic coordinate information of the orthophoto map established by the aerial survey software, and perform coordinate transformation through the coordinate transformation method to obtain the geographic coordinate information of the division of plant growth density.

[0078] Step 3: Generate a map of the division of plant growth density based on the geographic coordinate information of the division of plant growth density.

[0079] Step 4: Obtain the relevant parameter information of the agricultural insurance UAV and the boundary information of the plant operation area.

[0080] Step 5: Combine the above map of the division of plant growth density, the relevant parameter information of the agricultural insurance UAV, and the boundary information of the plant operation area, and plan the flight path through the aerial survey software.

[0081] By identifying and marking key and ordinary areas and using them as path points to realize path planning, it is possible to plan the flight path according to different plant growth states, greatly improving the accuracy of UAV spraying, facilitating precise spraying management, and putting forward new solutions and ideas for the current research on UAV spraying.

[0082] Among them, the information on the division of plant growth density is the central pixel coordinates of the key area and the ordinary area of plant growth density.

[0083] The process of coordinate transformation through the coordinate transformation method is as follows: where xi is the longitude of the central coordinate point of the geographic coordinate information of the key area of plant growth density, yi is the latitude of the central coordinate point of the geographic coordinate information of the key area of plant growth density, X0 is the longitude of the first pixel point in the upper left corner of the tiff image, Y0 is the latitude of the first pixel point in the upper left corner of the tiff image, A is the longitude length corresponding to one pixel point, B is the latitude length corresponding to one pixel point; (iu, iv) is the pixel coordinate of the central coordinate point of the key area of plant growth density in the tiff image, which is obtained by conversion through the corresponding pixel coordinates in the full-area image of the plant operation area.

[0084] Through the above operations, accurate geographic coordinate information of the division of plant growth density can be obtained, facilitating the subsequent formulation of the flight path.

[0085] Among them, the relevant parameter information of the agricultural insurance UAV in Step 5 includes the spraying width and the turning radius.

[0086] Among them, in step one, a high-definition camera is used to collect the full-area image of the plant operation area.

[0087] Among them, the tiff image is an image synthesized from the full-area image of the plant operation area through aerial survey software. The tiff image carries geographic coordinate information, while the full-area image of the plant operation area does not carry geographic coordinate information. The position information of the plants can be obtained from the full-area image of the plant operation area through the plant image rapid recognition model. The position information is represented by a series of pixel coordinates. There is a corresponding relationship between the central pixel coordinates of the key and ordinary areas of the plant growth density in the full-area image of the plant operation area and the central coordinate points of the key and ordinary areas of the plant growth density in the tiff image.

[0088] Use aerial survey software to import the map of the plant growth density partition, and set the key areas in the map of the plant growth density partition as the main path points and the ordinary areas as the auxiliary path points. Combine with the boundary of the operation area to generate the flight path.

[0089] Through the above operations, the flight path can be planned according to the actual growth status and position of the plants, covering the areas of the distorted-growing plants, ensuring no omission during spraying. At the same time, this path is more accurate, facilitating the precise management of spraying, and providing a new solution and new idea for the current research on UAV spraying.

[0090] It is easy to understand that those skilled in the art can combine, split, reorganize, etc. the embodiments of the present application based on one or several embodiments provided by the present application to obtain other embodiments, and these embodiments do not exceed the protection scope of the present application.

[0091] The above schematically describes the present invention and its implementation manners. This description is not restrictive. What is shown in the embodiments is only part of the implementation manners of the present invention, and the actual structure is not limited thereto. Therefore, if those of ordinary skill in the art are inspired by it and design, without creative work, a structural manner and embodiments similar to the technical solution without departing from the purpose of the present invention, they shall fall within the protection scope of the present invention.

Claims

1. An agricultural insurance UAV, characterized in that, It includes a drone body (100). An installation part is provided on the drone body (100). Inside the installation part, there is a vertically arranged medicine cylinder (210) for containing liquid medicine. At the bottom of the medicine cylinder (210), there are multiple nozzles (220) arranged in a ring for spraying operations. On the drone body (100), there is also an acquisition module for acquiring full-area images of the plant operation area, an image processing module for identifying plants in the acquired images and marking the growth density of plants in different zones, and a spraying amount control module for spraying different amounts of liquid medicine according to plants marked with different growth density zones.

2. The agricultural insurance drone according to claim 1, wherein: The image processing module includes an image processing module and a judgment module; The image processing module is used to identify and mark the images acquired by the acquisition module; The judgment module is used to judge the growth density of the plants at the marked points and the density zone of the growth density of the plants at the marked points, obtain the growth density zone information of the plants at the marked points and send the information to the spraying amount control module.

3. The agricultural insurance UAV according to claim 2, characterized in that: Before the image processing module identifies and marks the images acquired by the acquisition module, there is also an image preprocessing operation. The image preprocessing specifically includes denoising the image and enhancing the image contrast.

4. The agricultural insurance drone according to claim 3, wherein: Mark the preprocessed image as the input image, construct a rapid plant image recognition model, use the input image as the input data of the rapid plant image recognition model, and obtain the point map of the plants in the plant operation area image.

5. The agricultural insurance drone according to claim 2, characterized in that, The way for the judgment module to judge the growth density of the plants at the marked points is as follows: construct a plant image analysis model, equally divide the obtained point map of the plants in the plant operation area image and mark it as the input photo, use the input photo as the input data of the plant image analysis and recognition model, mark the output data as the target label, and mark the target label as Ms; when the target label Ms ∈ [0, 3], then mark the target position as green, when the target label Ms ∈ (3, 7], then mark the target position as yellow, when the target label Ms ∈ (7, 10], then mark the target position as red, to form a plant point growth density map.

6. The agricultural insurance UAV according to claim 2, wherein The way for the judgment module to judge the growth density zone of the plants at the marked points is as follows: set the target label coefficient as Dh; use the formula Obtain the plant density value Yj, i = 1, 2... n, where n is the number of equal divisions of the photographed photos. When the plant density value Yj ≥ the density threshold, then mark the plant point as the key area, otherwise mark the plant point as the ordinary area.

7. The agricultural insurance drone according to claim 1, characterized in that: The plant image analysis model is obtained through the following steps: obtain n pieces of plant image data, mark the plant image data as the training images, assign image labels to the training images, divide the training images obtained according to a set ratio into a training set and a validation set, construct a neural network model, iteratively train the neural network model through the training set and the validation set. When the number of iterative training times is greater than the iterative times threshold, then determine that the neural network model is trained, and mark the trained neural network model as the plant image analysis model. The value range of the image label is [0 - 10], where the larger the value of the image label, the greater the growth density of the point.

8. The agricultural insurance drone according to claim 7, characterized in that: The plant image data includes images under different varieties, different growth stages and different lighting conditions.

9. The agricultural insurance UAV according to claim 1, characterized in that: The spraying amount control module includes a liquid pump disposed in the medicine barrel (210) and a control unit. The control unit is used to increase the spraying amount when the UAV main body (100) flies above the plants in the key area, and keep the original amount when the UAV main body (100) flies above the plants in the ordinary area.

10. A dynamic path optimization method for an agricultural insurance UAV, characterized in that, It includes the following steps: Step 1: Obtain the information on the growth density zoning of the plants at the marked points. Step 2: Combine the information on the growth density zoning of the plants with the geographical coordinate information of the orthophoto map established by the aerial survey software, and perform coordinate transformation through the coordinate transformation method to obtain the geographical coordinate information of the growth density zoning of the plants. Step 3: Generate a map of the growth density zoning of the plants through the geographical coordinate information of the growth density zoning of the plants. Step 4: Obtain the relevant parameter information of the agricultural protection UAV and the boundary information of the plant operation area. Step 5: Combine the above-mentioned map of the growth density zoning of the plants, the relevant parameter information of the agricultural protection UAV and the boundary information of the plant operation area, and plan the flight path through the aerial survey software.

Images