CN114005050A - Agricultural pest information acquisition system and method - Google Patents

Abstract

The invention discloses an agricultural pest information acquisition system and method, wherein a disturbance unit is adopted to apply disturbance to crops, when the crops are exposed out of positions with pests, images of the crops can be acquired, and pest information can be acquired according to the acquired images. Not only utilize unmanned aerial vehicle to improve information acquisition efficiency, the image of gathering moreover contains harmful biological information, has improved the help of image collection harmful biological information.

Description

Agricultural pest information acquisition system and method

Technical Field

The invention relates to the technical field of agricultural equipment, in particular to an agricultural pest information acquisition system and method.

Background

China is a big agricultural country, and the breeding production of various agricultural plants, animals and the like provides a large number of high-quality products for China and even abroad, so the agricultural health development is always an important strategy for China.

During the growth of crops, the crops are inevitably affected by various pests and diseases, so that the production condition of the crops needs to be monitored at certain specific time, the pests and the diseases are discovered as early as possible, and corresponding control measures are taken. The traditional monitoring means is mainly that people go to the field on the spot, the pest species are identified by means of visual observation, and the information of the species, the quantity, the position and the like of the pests is recorded manually. Although the mode can obtain more detailed information, the efficiency is very low, and the mode is seriously dependent on the experience of personnel and is not suitable for modern large-scale agricultural planting.

Under the application of modern technology, many planting areas begin to use unmanned aerial vehicles to collect the information of crops, mainly collect the image of crops, then adopt the information of automatic mode discernment harmful organism, greatly reduced the human input, improved efficiency greatly. However, the way of collecting information by the unmanned aerial vehicle is also limited, for collecting image information, the unmanned aerial vehicle can only look down or look at the crops at a relatively far position, and for crops with relatively high planting density, such as herbaceous and shrub crops, the unmanned aerial vehicle can only collect images in a looking down posture, and the images collected in the posture cannot provide meaningful information. For example, when some harmful organisms gather at the back of the leaves, branches and the like of crops, images containing harmful organism information cannot be acquired by adopting a top-view posture, so that the images cannot provide effective assistance for acquiring the harmful organism information.

Disclosure of Invention

The embodiment of the invention provides an agricultural pest information acquisition system and method, which are used for solving the problem that crop images acquired by adopting a overlooking posture in the prior art cannot provide effective help.

In one aspect, embodiments of the present invention provide an agricultural pest information collecting system, including:

unmanned aerial vehicle, unmanned aerial vehicle includes:

the rotor wing unit is used for providing flight power;

the disturbance unit is used for applying disturbance to the crops so as to expose the positions of the pests on the crops;

the image acquisition unit is used for acquiring images of harmful organisms on crops;

and the data acquisition server is used for identifying the information of the pests in the image.

In another aspect, an embodiment of the present invention provides an agricultural pest information collecting method, including:

applying a perturbation to the crop to expose the crop to a location where pests are present;

collecting images of pests on crops;

information identifying the pest in the image.

The system and the method for acquiring the agricultural pest information have the following advantages:

the disturbance unit is adopted to apply disturbance to the crops, when the crops are exposed out of the position with the pests, the images of the crops can be collected, and then pest information can be obtained according to the collected images. Not only utilize unmanned aerial vehicle to improve information acquisition efficiency, the image of gathering moreover contains harmful biological information, has improved the help of image collection harmful biological information.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic diagram of the components of an agricultural pest information collection system provided by an embodiment of the invention;

fig. 2 is a schematic structural diagram of an unmanned aerial vehicle according to an embodiment of the present invention;

fig. 3 is a flowchart of an agricultural pest information collection method according to an embodiment of the present invention.

Description of reference numerals: 100-an unmanned aerial vehicle, 110-a protection unit, 120-a rotor unit, 121-a support rod, 130-an image acquisition arm, 131-an image acquisition unit, 140-a disturbance arm, 141-a disturbance unit, 200-a mobile terminal and 300-a data acquisition server.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Fig. 1 is a schematic composition diagram of an agricultural pest information acquisition system provided by an embodiment of the present invention, and fig. 2 is a schematic structural diagram of an unmanned aerial vehicle provided by an embodiment of the present invention. The invention provides an agricultural pest information acquisition system, which comprises:

drone 100, drone 100 includes:

a rotor unit 120, the rotor unit 120 being configured to provide flight power;

a disturbance unit 141, wherein the disturbance unit 141 is used for applying disturbance to the crops to expose the crops to the position where the harmful organisms exist;

the image acquisition unit 131, the image acquisition unit 131 is used for acquiring the image of the harmful organism on the crops;

and the data acquisition server 300 is used for identifying the information of the harmful organisms in the image by the data acquisition server 300.

Exemplarily, the rotor unit 120 includes a driving device and a blade, the driving device is preferably a motor, the blade is disposed on a rotating shaft of the driving device, and the blade can be driven to rotate after the driving device is powered on to operate, so as to generate power for driving the unmanned aerial vehicle 100 to fly. The number of the paddles arranged on the rotating shaft of the driving device can be any number, such as one, two, and the like, and the embodiment of the invention is not particularly limited.

The disturbance unit 141 applies a disturbance to the crop preferably by spraying a fluid, which may be air, water, insecticide, etc. Accordingly, the disturbing unit 141 may adopt a fan, a spray head, or the like for the difference of the sprayed media. When the disturbance unit 141 ejects the fluid and the fluid acts on the crops, the leaves of the crops move under the influence of the fluid, and the originally shielded positions, such as the back of the leaves, the branches and the like, are exposed, and the images of the positions can be acquired by the image acquisition unit 131. Since the crops are exposed to the position with the pests under the action of the fluid sprayed by the disturbance unit 141, the images collected by the image collection unit 131 contain the pest information, so that the collected images have practical significance.

In addition to the above-described rotor unit 120, disturbance unit 141, and image acquisition unit 131, the drone 100 may further include a control unit, a storage unit, and a power supply unit. The control unit is used for controlling the working state of the rotor unit 120, so that the unmanned aerial vehicle 100 can perform various flight actions such as forward movement, backward movement, lifting, lowering and the like, and can also control the working states of other units, such as the disturbance unit 141, the image acquisition unit 131 and the like. The storage unit is used for storing the image acquired by the image acquisition unit 131, and the power supply unit is used for supplying power to each electronic device in the unmanned aerial vehicle 100.

Additionally, the drone 100 may further include a communication unit configured to perform data communication with the data acquisition server 300, so as to transmit the acquired and stored image to the data acquisition server 300, or receive various control instructions sent by the data acquisition server 300, such as returning, going to the next acquisition point, and the like.

In the embodiment of the present invention, the data collecting server 300 stores the pest identification model, and when the image is input into the pest identification model, the corresponding pest information, including the type and the amount of the pest, can be obtained. The pest identification model preferably employs a neural network model that requires training and testing to adapt the neural network model to the identification of one or more pests prior to identifying the pest information. The number of the neural network models can be one or more, if one neural network model has high identification accuracy for one pest or one type of pest and low identification accuracy for other pests, the neural network model can be used for identifying only one pest or one type of pest, and other pests can be identified by using other neural network models. If a neural network model has a high recognition accuracy for multiple types of pests, the neural network model may be used to simultaneously recognize multiple types of pests.

In a possible embodiment, further comprising: and the mobile terminal 200 is used for acquiring the image and sending the image to the data acquisition server 300 by the mobile terminal 200.

For example, the mobile terminal 200 may adopt a general electronic device, such as a mobile phone or a tablet computer, and then install a special APP (Application program) in the general electronic device, through which the mobile terminal can be communicatively connected to the drone 100 and the data acquisition server 300. In some practical scenarios, when collecting information of crops in a relatively large range, the distance between the drone 100 and the data collection server 300 is relatively long, resulting in a relatively large communication cost, so that the mobile terminal 200 can be held by a worker to move near the drone 100, so as to receive an image sent by the drone 100 in real time or after the drone 100 completes a phase of image collection. The worker may upload the image stored in the mobile terminal 200 to the data collection server 300 after one image collection job is completed. In the process of uploading the image, the worker may also input image related information including a collection location, a collection person, a crop type, and the like in the mobile terminal 200, so that the data collection server 300 may perform subsequent analysis.

In the embodiment of the present invention, the mobile terminal 200 is connected to both the drone 100 and the data acquisition server 300 in a wireless communication manner. The wireless communication can be WiFi, GPRS and other technologies. Of course, the mobile terminal 200 may also be connected with the drone 100 and/or the data collection server 300 by way of wired communication.

In one possible embodiment, the drone 100 further comprises: and a protective unit 110, wherein the protective unit 110 is arranged outside the rotor unit 120, and the protective unit 110 protects the rotor unit 120.

Illustratively, the protection unit 110 is preferably a cylindrical hollow ring, which encloses the rotor unit 120 inside, and can prevent external objects, such as branches and leaves, animals, etc., from colliding with the rotor unit 120 to damage the rotor unit 120.

In the embodiment of the present invention, the inside of the protection unit 110 is provided with a strut 121, and the strut 121 is preferably a central symmetrical structure, and the symmetrical center is the center of the protection unit 110. The arrangement of the struts 121 inside the protective unit 110 not only provides reinforcement for the protective unit 110, but also provides a mounting location for the rotor unit 120, i.e. the rotor unit 120 can be mounted on the struts 121, so that the rotor unit 120 is protected by the protective unit 110.

In the embodiment of the present invention, the unmanned aerial vehicle 100 employs a plurality of rotor units 120, so that the unmanned aerial vehicle 100 can perform a flight action more quickly and accurately. When the drone has a plurality of rotor units 120, the protection unit 110 may be disposed outside the plurality of rotor units 120, in which case the number of protection units 110 may be one. Alternatively, the protection unit 110 may be disposed outside each rotor unit 120, and the number of the protection units 110 is equal to that of the rotor units 120.

In one possible embodiment, the drone 100 further comprises: an image acquisition arm 130, one end of the image acquisition arm 130 is arranged on the protection unit 120, and an image acquisition unit 131 is arranged at the other end of the image acquisition arm 130.

For example, the image capturing arm 130 may be a mechanical arm formed by sequentially connecting a plurality of image capturing arms, and a driving motor is disposed between two adjacent image capturing arms, and each driving motor on the image capturing arm 130 is controlled by the control unit to adjust the posture and position of the image capturing unit 131 under the control of the control unit, so that the image capturing unit 131 is located at an optimal position at any time, and the quality of the acquired image is improved. Moreover, the image capturing arm 130 may be folded and stowed to the bottom of the drone 100 when images do not need to be captured to reduce flight impacts on the drone 100.

In one possible embodiment, the drone 100 further comprises: the disturbance arm 140, one end of the disturbance arm 140 is disposed on the shielding unit 120, and the disturbance unit 141 is disposed at the other end of the disturbance arm 140.

For example, the disturbance arm 140 may also be a mechanical arm, which is formed by sequentially connecting a plurality of disturbance arms, and a driving motor is provided between two adjacent disturbance arms, and each driving motor on the disturbance arm 140 is controlled by the control unit to adjust the posture and position of the disturbance unit 141 under the control of the control unit, so that the disturbance unit 141 applies disturbance to the optimal position of the crop, and the quality of the image acquired by the image acquisition unit 131 is improved. Also, the disturbance arm 140 may fold and stow to the bottom of the drone 100 when it is not necessary to apply a disturbance to reduce the flight impact on the drone 100.

In the embodiment of the present invention, the above-mentioned control of the image capturing arm 130 and the disturbance arm 140 may be performed autonomously by a control unit in the drone 100, or may be manually controlled by a worker holding the mobile terminal 200. Furthermore, when the disturbance unit 141 applies disturbance to the crop, the position of the crop exposed by the pest is also often toward the disturbance unit 141, so the image capturing arm 130 and the disturbance arm 140 need to be arranged at a short distance so that the image capturing unit 131 can capture a high-quality image. However, when the image capturing arm 130 and the disturbance arm 140 are disposed at a short distance, the center of gravity of the drone 100 may deviate from the geometric center thereof, and thus the center of gravity of the drone 100 may be located at the geometric center thereof as much as possible by adjusting the positions of units in the drone 100, such as a power supply unit.

Moreover, when the disturbance unit 141 ejects the fluid to apply disturbance to the crop, the disturbance unit 141 also applies a reaction force to the unmanned aerial vehicle 100, so that the flight state of the unmanned aerial vehicle 100 is disturbed. In order to reduce the interference to the flight state 100 of the unmanned aerial vehicle, an anti-disturbance unit may be disposed on the protection unit 110, and the anti-disturbance unit may provide an acting force opposite to a reaction force generated by the disturbance unit 141 when the disturbance unit 141 operates, so as to counteract the influence of the reaction force of the disturbance unit 141 on the state of the unmanned aerial vehicle 100 as much as possible.

Based on the above, an embodiment of the present invention further provides an agricultural pest information collecting method, as shown in fig. 3, the method includes:

s300, applying disturbance to the crops to expose the positions of the pests on the crops;

s310, collecting images of pests on crops;

and S320, identifying the information of the pests in the image.

Illustratively, the pest identification model stored in the data collection server 300 may be utilized in identifying pest information in an image. In particular, the pest identification model may employ a neural network model, such as CNN, RNN, or the like. The neural network models need to be trained before being used, after images with specific types of pest information are input into the neural network models, various parameters in the neural network models are adjusted according to the difference between the output result and the pest information, such as node threshold values, connection weight values and the like, so that the neural network models are adaptive to the specific types of pest information, and the accuracy of identifying the specific types of pests is improved. And after the neural network model is trained, performance testing is required, and the neural network model can be put into use only when the testing accuracy reaches the requirement.

While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (10)

1. An agricultural pest information collection system, comprising:

a drone (100), the drone (100) comprising:

a rotor unit (120), the rotor unit (120) for providing flight power;

a disturbance unit (141), the disturbance unit (141) being used for applying disturbance to the crops to expose the crops to the position where the harmful organisms exist;

the image acquisition unit (131), the said image acquisition unit (131) is used for gathering the image of the pest on the crops;

a data acquisition server (300), the data acquisition server (300) being configured to identify information of pests in the image.

2. An agricultural pest information collection system according to claim 1, further comprising:

the mobile terminal (200) is used for acquiring the image and sending the image to the data acquisition server (300).

3. The agricultural pest information collection system of claim 2, wherein the mobile terminal (200) is in wireless communication with both the drone (100) and the data collection server (300).

4. The agricultural pest information collection system of claim 1, wherein the drone (100) further includes:

a protective unit (110), the protective unit (110) being disposed outside the rotor unit (120), the protective unit (110) being for user protection of the rotor unit (120).

5. An agricultural pest information collection system according to claim 4, wherein the number of the rotor units (120) is plural, and the guard unit (110) is provided outside the plural rotor units (120).

6. An agricultural pest information collection system according to claim 4, wherein the number of the rotor units (120) is plural, the number of the guard units (110) is the same as the number of the rotor units (120), and the guard units (110) are provided outside each of the rotor units (120).

7. The agricultural pest information collection system of claim 4, wherein the drone (100) further includes:

the protection device comprises an image acquisition arm (130), wherein one end of the image acquisition arm (130) is arranged on the protection unit (120), and the image acquisition unit (131) is arranged at the other end of the image acquisition arm (130).

8. The agricultural pest information collection system of claim 4, wherein the drone (100) further includes:

the disturbance arm (140), one end of the disturbance arm (140) is arranged on the protection unit (120), and the disturbance unit (141) is arranged at the other end of the disturbance arm (140).

9. A method of applying the agricultural pest information collection system of any one of claims 1-8, comprising:

applying a perturbation to the crop to expose the crop to a location where pests are present;

collecting images of pests on crops;

information identifying a pest in the image.

10. An agricultural pest information collection method according to claim 9, wherein said identifying information of pests in said image includes:

inputting the image into a pest identification model;

information about the pest is obtained.

Images