阅读说明：本技术 一种基于5g网络的植保无人机精准对靶喷洒方法及装置 (Plant protection unmanned aerial vehicle accurate target-aiming spraying method and device based on 5G network ) 是由 邓继忠 谢尧庆 张建瓴 严智威 杨畅 黄康华 霍静朗 叶家杭 雷落成 罗明达 于 2021-06-21 设计创作，主要内容包括：本发明公开一种基于5G网络的植保无人机精准对靶喷洒方法及装置,该方法包括：首先无人机按照预先规划的航线飞行,安装在无人机上的相机实时采集地块中无人机飞行前方的目标区域的图像；无人机通过5G网络将图像发送至地面端；接着地面端应用已有深度学习模型对图像进行处理,生成处方图；地面端通过5G网络将处方图发送至无人机；无人机根据处方图,对该目标区域进行实时对靶喷洒；在喷洒过程中实时采集地块中下一目标区域的图像,重复上述步骤,直至完成整块地块的喷洒。该方法中的无人机只需执行一次飞行,即可完成对地块草害的信息采集并对地块草害进行对靶喷洒,操作流程简单,作业效率高,能够实时对地块草害进行对靶喷洒,减少了资源浪费。(The invention discloses a plant protection unmanned aerial vehicle accurate target-aiming spraying method and device based on a 5G network, wherein the method comprises the following steps: firstly, an unmanned aerial vehicle flies according to a pre-planned air route, and a camera arranged on the unmanned aerial vehicle collects images of a target area in a plot in front of the flying of the unmanned aerial vehicle in real time; the unmanned aerial vehicle sends the image to a ground end through a 5G network; processing the image by applying an existing deep learning model to the ground terminal to generate a prescription map; the ground end sends the prescription chart to the unmanned aerial vehicle through a 5G network; the unmanned aerial vehicle sprays the target to the target area in real time according to the prescription chart; and acquiring an image of a next target area in the plot in real time in the spraying process, and repeating the steps until the spraying of the whole plot is finished. The unmanned aerial vehicle in the method can complete information acquisition of the land weeds and spray the targets of the land weeds only by performing one-time flight, the operation flow is simple, the operation efficiency is high, the land weeds can be sprayed to the targets in real time, and the resource waste is reduced.)

1. The plant protection unmanned aerial vehicle accurate target-aiming spraying method based on the 5G network is characterized by comprising the following steps of:

(1) the unmanned aerial vehicle flies according to a planned air route in advance, and images of a target area in front of the unmanned aerial vehicle in the plot are collected in real time;

(2) the unmanned aerial vehicle sends the image to a ground end through a 5G network;

(3) adding a deep learning model at the ground end, processing the image and generating a prescription map;

(4) the ground end sends the prescription chart to the unmanned aerial vehicle through a 5G network;

(5) the unmanned aerial vehicle sprays the target to the target area in real time according to the prescription chart;

(6) the unmanned aerial vehicle gathers the image of the next target area in the unmanned aerial vehicle flight the place ahead in real time at the in-process that sprays, repeats above-mentioned step (2) - (5), carries out image acquisition and sprays the target to every target area in real time.

2. The plant protection unmanned aerial vehicle accurate target spraying method based on the 5G network as claimed in claim 1, wherein in step (1), the current position of the unmanned aerial vehicle is read in real time, and when the unmanned aerial vehicle does not reach the target area, the target area is subjected to image acquisition in advance.

3. The plant protection unmanned aerial vehicle accurate target spraying method based on the 5G network according to claim 2, wherein when the horizontal distance from the flight of the unmanned aerial vehicle to the starting boundary of the target area is L, the unmanned aerial vehicle performs image acquisition on the target area, and when the unmanned aerial vehicle reaches the starting boundary of the target area, the target spraying is started on the target area, wherein the horizontal distance L from the unmanned aerial vehicle to the starting boundary of the target area is calculated according to the following formula:

L＝V_{fly away}(T₁+T₂+T₃)

Wherein, V in the formula_{Fly away}Indicating the current flight speed, T, of the drone₁Time, T, required for the unmanned aerial vehicle to send the image to the ground₂Time, T, required to generate a treatment map for each image processed at the surface₃The time required for the ground to send the prescription map to the drone.

4. The plant protection unmanned aerial vehicle accurate target spraying method based on the 5G network as claimed in claim 1, wherein in the step (1), the unmanned aerial vehicle photographs the target area through a camera at the airborne end, so as to obtain the image of the target area.

5. The plant protection unmanned aerial vehicle accurate target spraying method based on 5G network as claimed in claim 1, in steps (2) and (4), the unmanned aerial vehicle and the ground terminal both adopt an embedded development board Jetson TX2 as a core, and perform kernel compiling on Jetson TX2, so that the 5G module is connected with Jetson TX2, and the 5G module is implanted with a 5G phone card, so that the unmanned aerial vehicle and the ground terminal can receive the 5G network.

6. The plant protection unmanned aerial vehicle accurate target spraying method based on 5G network as claimed in claim 5, in step (2), the image is sent by downloading a program for capturing and sending Qt Creator and implanted image in an embedded development board Jetson TX2, and the Qt Creator is used for automatic capturing and image sending and receiving;

in the step (4), the receiving of the image and the sending of the prescription map are also operated by adopting the Qt Creator, the ground side configures an automatic sending program in the Qt Creator, and the Qt Creator automatically sends the image to the unmanned aerial vehicle after the prescription map is processed and generated by the ground side.

7. The plant protection unmanned aerial vehicle precise target spraying method based on the 5G network as claimed in claim 1, 3, 5 or 6, wherein the unmanned aerial vehicle is in communication connection with the ground end through a transit server.

8. The plant protection unmanned aerial vehicle accurate target spraying method based on the 5G network as claimed in claim 1, wherein in step (5), the width of the target area shot by the unmanned aerial vehicle is equal to the width of the target area when the unmanned aerial vehicle sprays.

9. The plant protection unmanned aerial vehicle accurate target spraying method based on the 5G network as claimed in claim 3, wherein in the step (1), the specific process that the unmanned aerial vehicle flies according to the planned route is as follows: unmanned aerial vehicle takes off when the horizontal distance apart from the originated border of plot is L at least, flies to when the distance that exceeds the terminal point border of plot is L, and unmanned aerial vehicle turns around and continues to fly towards the originated border of plot, and when unmanned aerial vehicle exceeded the distance on the originated border of plot and was L, unmanned aerial vehicle turned around, repeats above-mentioned step, covers whole plot until unmanned aerial vehicle's course.

10. The plant protection unmanned aerial vehicle accurate target-aiming spraying device based on the 5G network is applied to the plant protection unmanned aerial vehicle accurate target-aiming spraying method based on the 5G network, which is characterized by comprising an unmanned aerial vehicle, a ground end, a transfer server and a 5G module, wherein the unmanned aerial vehicle and the ground end server are both in communication connection with the transfer server, the 5G module provides the 5G network for the unmanned aerial vehicle and the ground end server, wherein,

the airborne end of the unmanned aerial vehicle is provided with a camera for shooting a target area image, and the ground end is provided with a ground end server; the unmanned aerial vehicle sends the shot target area image to a ground server through a 5G network, the ground server is implanted with a deep learning model, processes the image to generate a prescription, the ground server sends the prescription map to the unmanned aerial vehicle through the 5G network, and the unmanned aerial vehicle sprays the target to the target area in real time according to the prescription map.

Technical Field

The invention relates to the technical field of agricultural aviation plant protection mechanical equipment, in particular to a plant protection unmanned aerial vehicle accurate target spraying method and device based on a 5G network.

Background

In the rice planting process, the weed is an important factor influencing the growth of rice, and the crop yield is reduced. When the plant protection unmanned aerial vehicle is used for controlling weeds, the plant protection unmanned aerial vehicle is generally uniformly sprayed on the whole field, and weeds are generally distributed in a point or sheet shape due to uneven distribution; the operation mode does not consider the distribution difference condition of the weeds in the operation area, so that the dosage of the effective components of the pesticide is insufficient in the area with serious weeds and excessive in the area with slight or no weeds, and the targeted spraying of the weeds is particularly important.

At present, adopt plant protection unmanned aerial vehicle to spray to the target, utilize advantages such as unmanned aerial vehicle easy operation, efficient, atomization effect are good, realize the accurate spraying to the crop. For example, the invention patent application with application publication number CN105173085A discloses an automatic control system and method for variable pesticide application of an unmanned aerial vehicle, in the method, an unmanned aerial vehicle acquires field images, then an information management decision system works out a variable pesticide application prescription according to the analysis result of the disease condition of crops, finally, an unmanned aerial vehicle variable pesticide application control system obtains the variable pesticide application prescription, and a plant protection multi-rotor unmanned aerial vehicle carries out automatic variable pesticide application operation according to a preset route. However, the above method has the following disadvantages:

when the unmanned aerial vehicle collects field images, the unmanned aerial vehicle is required to carry out one-time flight, multiple or even dozens of low-altitude remote sensing images capable of covering the whole field are collected according to a planned route, then the collected images are subjected to off-line analysis, in the analysis process, the images of all the areas in the field are spliced into a complete field orthographic image by using an image splicing technology, and then a field disease distribution map and a variable pesticide application working diagram are obtained; then the unmanned aerial vehicle carries out the second flight, and carries out automatic variable pesticide application operation according to a preset route; in the operation process, unmanned aerial vehicle need carry out twice flight, and the operating efficiency is low, resource-wasting moreover.

Disclosure of Invention

The invention aims to overcome the existing problems and provides a 5G network-based plant protection unmanned aerial vehicle accurate target spraying method, the unmanned aerial vehicle in the method can complete information acquisition of the land weeds and target spraying of the land weeds only by performing one-time flight, the operation flow is simple, the operation efficiency is high, the land weeds can be sprayed on the targets in real time, and the resource waste is reduced.

The invention further aims to provide a device applied to the plant protection unmanned aerial vehicle accurate target spraying method based on the 5G network.

The purpose of the invention is realized by the following technical scheme:

a plant protection unmanned aerial vehicle accurate target spraying method based on a 5G network comprises the following steps:

(1) the unmanned aerial vehicle flies according to a planned air route in advance, and images of a target area in front of the unmanned aerial vehicle in the plot are collected in real time;

(2) the unmanned aerial vehicle sends the image to a ground end through a 5G network;

(3) adding a deep learning model at the ground end, processing the image and generating a prescription map;

(4) the ground end sends the prescription chart to the unmanned aerial vehicle through a 5G network;

(5) the unmanned aerial vehicle sprays the target to the target area in real time according to the prescription chart;

(6) the unmanned aerial vehicle gathers the image of the next target area in the unmanned aerial vehicle flight the place ahead in real time at the in-process that sprays, repeats above-mentioned step (2) - (5), carries out image acquisition and sprays the target to every target area in real time.

In the step (1), the current position of the unmanned aerial vehicle is read in real time, and when the unmanned aerial vehicle does not reach the target area, the image acquisition is performed on the target area in advance. Because unmanned aerial vehicle need send the image to ground end, ground end to image processing generation prescription picture and ground end send unmanned aerial vehicle's in-process with prescription picture after gathering the image, there is time consumption, consequently, when unmanned aerial vehicle has not arrived the target area yet, carry out image acquisition to this target area, can make the accurate realization of unmanned aerial vehicle spray the target for unmanned aerial vehicle can accurately spray in the target area.

Preferably, when the horizontal distance between the unmanned aerial vehicle and the starting boundary of the target area is L, the unmanned aerial vehicle acquires images of the target area, and when the unmanned aerial vehicle reaches the starting boundary of the target area, the target area starts to be sprayed, wherein the horizontal distance L between the unmanned aerial vehicle and the starting boundary of the target area is calculated according to the following formula:

L＝V_{fly away}(T₁+T₂+T₃)

Wherein, V in the formula_{Fly away}Indicating the current flight speed, T, of the drone₁Time, T, required for the unmanned aerial vehicle to send the image to the ground₂Time, T, required to generate a treatment map for each image processed at the surface₃The time required for the ground to send the prescription map to the drone. Because the communication between ground end and unmanned aerial vehicle and ground end exist time delay when handling the image, prejudge the aircraft in advanceThe position, when the horizontal distance of unmanned aerial vehicle flight distance target area's the initial boundary is L, carry out the image to this target area and adopt, when unmanned aerial vehicle flies to the initial boundary of target area, just spray the target according to the square to this target area for it is higher to spray the precision.

Preferably, in step (1), the unmanned aerial vehicle photographs the target area through a camera at the airborne end, so as to acquire an image of the target area.

Preferably, in the steps (2) and (4), the unmanned aerial vehicle and the ground terminal both adopt an embedded development board Jetson TX2 as a core, kernel compilation is performed on Jetson TX2, so that the 5G module is connected with Jetson TX2, and a 5G phone card is implanted in the 5G module, so that the unmanned aerial vehicle and the ground terminal can receive a 5G network. The 5G network is adopted for transmission, so that the speed is higher, and the spraying efficiency is improved.

Preferably, in the step (2), the image is sent by downloading a program for capturing and sending the Qt Creator and the implant image in the Jetson TX2 of the embedded development board, and the Qt Creator is used for automatic capturing and sending and receiving of the image. Specifically, a trigger instruction for automatic shooting and sending needs to be added in the Qt Creator, and when the unmanned aerial vehicle needs to shoot a target area, the trigger instruction is triggered to shoot the target area, and then the image is sent to the ground.

Preferably, in step (4), the receiving of the image and the sending of the prescription map are also operated by using the Qt Creator, the ground configures an automatic sending program in the Qt Creator, and the Qt Creator automatically sends the image to the drone after the prescription map is generated by the ground processing.

Preferably, the unmanned aerial vehicle and the ground terminal are in communication connection through a transfer server. Because unmanned aerial vehicle can not direct communication with the network that ground end was located, can realize communicating between unmanned aerial vehicle and the ground end through the transfer server to realize information transfer between unmanned aerial vehicle and the ground end.

Preferably, Socket programming is adopted between the unmanned aerial vehicle and the ground terminal to realize information transmission and sending, and a transmission protocol is a TCP (transmission control protocol); the method aims to realize the bidirectional transmission between the unmanned aerial vehicle and the ground end by setting a TCP (transmission control protocol); socket is written on the Qt Creator of the unmanned aerial vehicle and the ground end, and transmission and sending of images are achieved.

Preferably, in the step (5), the width of the target area shot by the unmanned aerial vehicle is equal to the width of the target area when the unmanned aerial vehicle sprays. The air route planning is carried out through the broad width of the unmanned aerial vehicle when the unmanned aerial vehicle sprays, and the re-spraying can not occur when the unmanned aerial vehicle turns back.

Further, in the step (4), the prescription chart includes pesticide application information of the target area, the pesticide application information is a weed condition of the target area, the weed condition is a proportion of weeds in the shot target area, and the unmanned aerial vehicle receives the proportion information and then controls a spray head in the pesticide spraying device to respond.

Preferably, in the step (1), the specific process of the unmanned aerial vehicle flying according to the planned route is as follows: unmanned aerial vehicle takes off when the horizontal distance apart from the originated border of plot is L at least, flies to when the distance that exceeds the terminal point border of plot is L, and unmanned aerial vehicle turns around and continues to fly towards the originated border of plot, and when unmanned aerial vehicle exceeded the distance on the originated border of plot and was L, unmanned aerial vehicle turned around, repeats above-mentioned step, covers whole plot until unmanned aerial vehicle's course. When unmanned aerial vehicle flies according to this air route, the target area who shoots does not overlap, sets for unmanned aerial vehicle and takes off in advance for unmanned aerial vehicle can acquire the image of target area in advance, guarantees to realize spraying to the monoblock landmass, when unmanned aerial vehicle reachd the initial boundary of landmass, begins to spray the landmass, can effectively reduce the waste of medicine.

A plant protection unmanned aerial vehicle accurate targeting spraying device based on a 5G network comprises an unmanned aerial vehicle, a ground end, a transfer server and a 5G module, wherein the unmanned aerial vehicle and the ground end server are both in communication connection with the transfer server, the 5G module provides the 5G network for the unmanned aerial vehicle and the ground end server, wherein,

the airborne end of the unmanned aerial vehicle is provided with a camera for shooting a target area image, and the ground end is provided with a ground end server; the unmanned aerial vehicle sends the shot target area image to a ground server through a 5G network, the ground server is implanted with a deep learning model, processes the image to generate a prescription, the ground server sends the prescription map to the unmanned aerial vehicle through the 5G network, and the unmanned aerial vehicle sprays the target to the target area in real time according to the prescription map.

Compared with the prior art, the invention has the following beneficial effects:

according to the plant protection unmanned aerial vehicle accurate target-aiming spraying method and device based on the 5G network, the unmanned aerial vehicle sends the image to the ground end through the 5G network, the ground end sends the prescription map to the unmanned aerial vehicle through the 5G network, rapid transmission of the image and the prescription map is achieved, image collection and spraying of a target area can be combined to one unmanned aerial vehicle, the image of the current target area is collected and immediately transmitted to the ground end for analysis, the prescription map is generated and then transmitted to the unmanned aerial vehicle, accurate target-aiming spraying is immediately executed, compared with the prior art that two times of unmanned aerial vehicles are adopted for flying, the image splicing process is omitted, one-time flying is adopted, the operation flow is simple, the unmanned aerial vehicle operation efficiency is greatly improved, and meanwhile resource waste is reduced.

Drawings

Fig. 1 is a schematic process diagram of a plant protection unmanned aerial vehicle accurate target-aiming spraying method based on a 5G network in the invention.

Fig. 2 is a schematic view of image acquisition of the unmanned aerial vehicle in the present invention.

FIG. 3 is a flow chart of sending and receiving images and processing diagrams in the present invention.

Fig. 4 is a flight route map of the unmanned aerial vehicle in the present invention.

Detailed Description

In order to make those skilled in the art understand the technical solutions of the present invention well, the following description of the present invention is provided with reference to the embodiments and the accompanying drawings, but the embodiments of the present invention are not limited thereto.

Referring to fig. 1, the embodiment discloses a plant protection unmanned aerial vehicle accurate target spraying method based on a 5G network, including the following steps:

(1) the unmanned aerial vehicle flies above a plot according to a pre-planned route, the plot is divided into a plurality of target areas, and the unmanned aerial vehicle acquires images of the target areas in front of the unmanned aerial vehicle in the plot in real time;

(2) the unmanned aerial vehicle sends the image to a ground end through a 5G network;

(3) adding a deep learning model at the ground end, processing the image and generating a prescription map;

(4) the ground end sends the prescription chart to the unmanned aerial vehicle through a 5G network;

(5) the unmanned aerial vehicle sprays the target to the target area in real time according to the prescription chart;

(6) the unmanned aerial vehicle gathers the image of the next target area in the unmanned aerial vehicle flight the place ahead in real time at the in-process that sprays, repeats above-mentioned step (2) - (5), carries out image acquisition and sprays the target to every target area in real time.

Referring to fig. 2 and 4, in step (1), the current position of the drone is read in real time, and when the drone does not reach the target area, image acquisition is performed on the target area in advance. Because unmanned aerial vehicle need send the image to ground end, ground end to image processing generation prescription picture and ground end send unmanned aerial vehicle's in-process with prescription picture after gathering the image, there is time consumption, consequently, when unmanned aerial vehicle has not arrived the target area yet, carry out image acquisition to this target area, can make the accurate realization of unmanned aerial vehicle spray the target for unmanned aerial vehicle can accurately spray in the target area.

Referring to fig. 2 and 4, when the horizontal distance between the flight distance of the unmanned aerial vehicle and the starting boundary of the target area is L, the unmanned aerial vehicle acquires an image of the target area, and when the unmanned aerial vehicle reaches the starting boundary of the target area, the target area starts to be sprayed, wherein the horizontal distance L between the unmanned aerial vehicle and the starting boundary of the target area is calculated according to the following formula:

L＝V_{fly away}(T₁+T₂+T₃)

Wherein, V in the formula_{Fly away}Indicating the current flight speed, T, of the drone₁Time, T, required for the unmanned aerial vehicle to send the image to the ground₂Time, T, required to generate a treatment map for each image processed at the surface₃Time required for ground-side to send prescription chart to unmanned aerial vehicleAnd (3) removing the solvent. Because communication and ground end exist time delay between ground end and the unmanned aerial vehicle when handling the image, prejudge the position of aircraft in advance, when the horizontal distance of unmanned aerial vehicle flight distance target area's the initial boundary is L, carry out the image to this target area and adopt, when unmanned aerial vehicle flies to target area initial boundary, spray the target to this target area according to the prescription picture just for it is higher to spray the precision.

Referring to fig. 2 and 4, in the present embodiment, the current flying speed V of the drone_{Fly away}The time T required for the unmanned aerial vehicle to send the image to the ground end when flying at the speed of 5m/s₁50-100 ms, the time T required for processing the image at the ground end to generate a treatment chart₂The time T required for the ground end to send the prescription chart to the unmanned aerial vehicle is 150-200 ms₃And (2) calculating the maximum delay time by referring to the formula to obtain the horizontal distance L of the unmanned aerial vehicle from the initial boundary of the target area, wherein the horizontal distance L is 200-300 ms:

5m/s×(100ms+200ms+300ms)＝3m

when the horizontal distance between the unmanned aerial vehicle and the starting boundary of the target area is 3m, image acquisition is carried out on the target area, and when the unmanned aerial vehicle flies to the target area, spraying is just carried out according to the prescription map of the target area.

Referring to fig. 1, in step (1), the unmanned aerial vehicle shoots the target area through the camera at the aircraft nose machine-mounted end, thereby obtains the image of the target area, and the size of the image of gathering is 512 × 512 pixels, and during the collection, the image head of camera needs to aim at the target area in front of the unmanned aerial vehicle, and when the horizontal distance between the unmanned aerial vehicle and the starting boundary of the target area is 3m, the camera shoots the target area.

Referring to fig. 3, in steps (2) and (4), the unmanned aerial vehicle and the ground terminal both adopt an embedded development board Jetson TX2 as a core, and perform kernel compilation on Jetson TX2, so that the 5G module is connected with Jetson TX2, and a 5G phone card is implanted in the 5G module, so that the unmanned aerial vehicle and the ground terminal can receive a 5G network. The 5G network is adopted for transmission, so that the speed is higher, and the spraying efficiency is improved. In addition, the ground terminal can also access the 5G network through the 5G WiFi module.

Referring to fig. 3, in step (2), the image is sent by downloading a Qt Creator and a program for capturing and sending the implant image in the Jetson TX2 of the embedded development board, and the Qt Creator is used for automatic capturing and sending and receiving of the image. Specifically, a trigger instruction for automatic shooting and sending needs to be added in the Qt Creator, and when the unmanned aerial vehicle needs to shoot a target area, the trigger instruction is triggered to shoot the target area, and then the image is sent to the ground.

Referring to fig. 3, in step (4), the receiving of the image and the sending of the prescription map are also operated by the Qt Creator, the ground configures an automatic sending program in the Qt Creator, and after the prescription map is processed and generated by the ground, the ground is automatically sent to the drone by the Qt Creator.

Specifically, the unmanned aerial vehicle and the ground terminal are in communication connection through a transfer server. Because unmanned aerial vehicle can not direct communication with the network that ground end was located, can realize communicating between unmanned aerial vehicle and the ground end through the transfer server to realize information transfer between unmanned aerial vehicle and the ground end.

Referring to fig. 3, Socket programming is adopted between the unmanned aerial vehicle and the ground end to realize transmission and sending of information, and a transmission protocol is a TCP protocol; the method aims to realize the bidirectional transmission between the unmanned aerial vehicle and the ground end by setting a TCP (transmission control protocol); socket is written on the Qt Creator of the unmanned aerial vehicle and the ground end, and transmission and sending of images are achieved.

Referring to fig. 4, in step (5), the width of the target area shot by the unmanned aerial vehicle is equal to the width of the target area when the unmanned aerial vehicle sprays. The air route planning is carried out through the broad width of the unmanned aerial vehicle when the unmanned aerial vehicle sprays, and the re-spraying can not occur when the unmanned aerial vehicle turns back.

Specifically, in the step (4), the prescription chart includes pesticide application information of the target area, the pesticide application information is a weed condition of the target area, the weed condition is a proportion of weeds in the shot target area, and the unmanned aerial vehicle receives the proportion information and then controls a spray head in the pesticide spraying device to respond.

Referring to fig. 4, in step (1), the specific process of the unmanned aerial vehicle flying according to the planned route is as follows: unmanned aerial vehicle takes off when the horizontal distance apart from the originated border of plot is L at least, flies to when the distance that exceeds the terminal point border of plot is L, and unmanned aerial vehicle turns around and continues to fly towards the originated border of plot, and when unmanned aerial vehicle exceeded the distance on the originated border of plot and was L, unmanned aerial vehicle turned around, repeats above-mentioned step, covers whole plot until unmanned aerial vehicle's course. A plurality of target areas in the parcel are arranged along the planned air route, and when unmanned aerial vehicle flies according to this air route, the target area who shoots does not overlap, sets for unmanned aerial vehicle and takes off in advance for unmanned aerial vehicle can acquire the image of target area in advance, guarantees to realize spraying to the monoblock parcel, and when unmanned aerial vehicle reachd the originated border in parcel, begin to spray the parcel, can effectively reduce the waste of medicine. Specifically, the unmanned aerial vehicle takes off when the horizontal distance from the starting boundary of the plot is at least 3m, and takes a picture of the target area.

The land parcel in the implementation is a paddy field, the unmanned aerial vehicle acquires the images of the weeds in the target area through a camera at the airborne end, and the unmanned aerial vehicle sends the images of the weeds to the ground end through a 5G network; and adding a deep learning weed identification model at the ground end, and processing the weed image to generate a prescription map.

Referring to fig. 1, the embodiment further discloses a plant protection unmanned aerial vehicle accurate targeting spraying device based on a 5G network, comprising an unmanned aerial vehicle, a ground terminal, a transit server and a 5G module, wherein the unmanned aerial vehicle and the ground terminal server are both in communication connection with the transit server, the 5G module provides the 5G network for the unmanned aerial vehicle and the ground terminal server, wherein,

the airborne end of the unmanned aerial vehicle is provided with a camera for shooting a target area image, and the ground end is provided with a ground end server; the unmanned aerial vehicle sends the shot target area image to a ground server through a 5G network, the ground server is implanted with a deep learning model, processes the image to generate a prescription, the ground server sends the prescription map to the unmanned aerial vehicle through the 5G network, and the unmanned aerial vehicle sprays the target to the target area in real time according to the prescription map.

The present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents and are included in the scope of the present invention.