CN115500334A - Sprayer, plant disease and insect pest identification method and identification equipment - Google Patents
Sprayer, plant disease and insect pest identification method and identification equipment Download PDFInfo
- Publication number
- CN115500334A CN115500334A CN202110631125.4A CN202110631125A CN115500334A CN 115500334 A CN115500334 A CN 115500334A CN 202110631125 A CN202110631125 A CN 202110631125A CN 115500334 A CN115500334 A CN 115500334A
- Authority
- CN
- China
- Prior art keywords
- plant
- image
- spray
- sprayer
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 241000607479 Yersinia pestis Species 0.000 title claims abstract description 115
- 238000000034 method Methods 0.000 title claims abstract description 63
- 201000010099 disease Diseases 0.000 title claims description 91
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims description 91
- 241000238631 Hexapoda Species 0.000 title claims description 49
- 239000007921 spray Substances 0.000 claims abstract description 129
- 238000005507 spraying Methods 0.000 claims abstract description 79
- 238000004891 communication Methods 0.000 claims abstract description 59
- 239000007788 liquid Substances 0.000 claims abstract description 54
- 238000012545 processing Methods 0.000 claims description 67
- 239000000575 pesticide Substances 0.000 claims description 22
- 238000003062 neural network model Methods 0.000 claims description 21
- 238000012549 training Methods 0.000 claims description 19
- 230000007613 environmental effect Effects 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 11
- 239000006199 nebulizer Substances 0.000 claims description 10
- 239000000126 substance Substances 0.000 claims description 9
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 2
- 241000196324 Embryophyta Species 0.000 description 204
- 230000001276 controlling effect Effects 0.000 description 27
- 239000003814 drug Substances 0.000 description 18
- 230000000694 effects Effects 0.000 description 16
- 230000015654 memory Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 9
- 239000004009 herbicide Substances 0.000 description 7
- 238000001556 precipitation Methods 0.000 description 7
- 239000005648 plant growth regulator Substances 0.000 description 6
- 230000002363 herbicidal effect Effects 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 230000002265 prevention Effects 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000000844 anti-bacterial effect Effects 0.000 description 3
- 239000003899 bactericide agent Substances 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000008635 plant growth Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000012530 fluid Substances 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000012010 growth Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 244000000010 microbial pathogen Species 0.000 description 2
- 239000003595 mist Substances 0.000 description 2
- 238000010295 mobile communication Methods 0.000 description 2
- 239000004800 polyvinyl chloride Substances 0.000 description 2
- 230000001105 regulatory effect Effects 0.000 description 2
- 241000238421 Arthropoda Species 0.000 description 1
- 244000241235 Citrullus lanatus Species 0.000 description 1
- 235000012828 Citrullus lanatus var citroides Nutrition 0.000 description 1
- 241000282414 Homo sapiens Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 244000061456 Solanum tuberosum Species 0.000 description 1
- 235000002595 Solanum tuberosum Nutrition 0.000 description 1
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 239000003905 agrochemical Substances 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 235000013339 cereals Nutrition 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 235000009508 confectionery Nutrition 0.000 description 1
- 230000034994 death Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 239000000417 fungicide Substances 0.000 description 1
- 230000035784 germination Effects 0.000 description 1
- 239000003112 inhibitor Substances 0.000 description 1
- 239000002917 insecticide Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000010985 leather Substances 0.000 description 1
- 235000015097 nutrients Nutrition 0.000 description 1
- 229920000915 polyvinyl chloride Polymers 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01M—CATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- A01M7/00—Special adaptations or arrangements of liquid-spraying apparatus for purposes covered by this subclass
- A01M7/0025—Mechanical sprayers
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01M—CATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- A01M7/00—Special adaptations or arrangements of liquid-spraying apparatus for purposes covered by this subclass
- A01M7/0089—Regulating or controlling systems
Abstract
The application discloses atomizer, plant pest identification method and identification equipment, wherein, the atomizer includes: the spray rod is arranged on the spray rod; the box body comprises a liquid storage box, a communication module, a control module and a power supply module; a liquid storage tank for storing a spray liquid; the first end of the spray rod is connected with the liquid storage tank, and the second end of the spray rod is provided with a spray head; the first end of the supporting rod is connected with the spray rod, and the second end of the supporting rod is provided with an image collector; the power supply module is respectively connected with the communication module, the control module and the image collector and is used for supplying power to the communication module, the control module and the image collector; the control module is used for controlling the image collector to collect the plant images and controlling the communication module to send the plant images to the identification equipment; the control communication module receives a spraying instruction fed back by the identification equipment aiming at the plant image and responds to the spraying instruction to control the spray head to spray the spraying liquid.
Description
Technical Field
The application relates to the technical field of information, in particular to a sprayer, a plant disease and insect pest identification method and identification equipment.
Background
Diseases and pests are the combined name of diseases and pests, and often have adverse effects on agriculture, forestry, animal husbandry and the like. At present, in the process of preventing and controlling plant diseases and insect pests, a fixed image collector is arranged at a fixed point in a crop planting area to collect plant images, or an unmanned aerial vehicle is used for collecting the plant images; further, the collected plant images are identified to spray the medicine.
However, this method of collecting plant images can only collect leaves on the surface of the plant, it is difficult to collect information on the back of the plant leaves or the stem, crop diseases can be exposed on the back of the stem or the leaves, and some crop diseases can be exposed at the bottom of the plant and blocked by the leaves. Therefore, the current mode of collecting plant images is difficult to find plant diseases and insect pests in time, and further the diseases and insect pests cannot be prevented and controlled in time, so that the plants die due to withering.
Disclosure of Invention
The application provides a atomizer, plant diseases and insect pests identification method and identification equipment, with image collector integration on the atomizer, if set up on the bracing piece be connected with the spray lance, so, along with the removal of spray lance between the plant, image collector can gather the image that contains each position of plant, further, carries out the plant diseases and insect pests discernment and applys the medicine based on the all-round image to plant collection, must realize the prevention and cure effect of better plant diseases and insect pests.
The technical scheme of the application is realized as follows:
in a first aspect, the present application provides a sprayer comprising a tank, a spray bar, and a support bar; wherein the content of the first and second substances,
the box body comprises a liquid storage box, a communication module, a control module and a power supply module; wherein the content of the first and second substances,
the liquid storage tank is used for storing spraying liquid;
the first end of the spray rod is connected with the liquid storage tank, and the second end of the spray rod is provided with a spray head;
the first end of the supporting rod is connected with the spray rod, and the second end of the supporting rod is provided with an image collector;
the power supply module is respectively connected with the communication module, the control module and the image collector and is used for supplying power to the communication module, the control module and the image collector;
the control module is used for controlling the image collector to collect plant images and controlling the communication module to send the plant images to the identification device; and controlling the communication module to receive a spraying instruction fed back by the identification equipment aiming at the plant image, and responding to the spraying instruction to control the spray head to spray the spraying liquid.
In a second aspect, the present application provides a plant pest identification method, applied to an identification device, the method comprising:
receiving a plant image which is sent by a sprayer and collected by an image collector on the sprayer;
inputting the plant image into a multi-task processing model obtained through training to obtain a pest and disease identification result which is output by the multi-task processing model and is obtained by identifying the plant image;
and based on the pest and disease identification result, generating a spraying instruction for controlling a spray head of the sprayer to spray spraying liquid, and sending the spraying instruction to the sprayer.
In a third aspect, the present application provides a method for identifying a plant pest applied to the sprayer of the first aspect, the method comprising:
controlling the image collector to collect plant images through a control module;
sending the plant image to a recognition device through a communication module;
and receiving a spraying instruction fed back after the recognition equipment processes the plant image based on a multitasking model through the communication module, and responding to the spraying instruction to control a spray head to spray spraying liquid.
In a fourth aspect, the present application provides an identification device comprising:
the receiving module is used for receiving the plant images which are sent by the sprayer and collected by the image collector on the sprayer;
the processing module is used for inputting the plant image into a multi-task processing model obtained through training to obtain a pest and disease identification result which is output by the multi-task processing model and is obtained by identifying the plant image;
the processing module is further used for generating a spraying instruction for controlling a spray head of the sprayer to spray the spraying liquid based on the pest and disease identification result;
and the sending module is used for sending the spraying instruction to the sprayer.
In a fifth aspect, the present application provides a plant pest identification system, comprising:
a nebulizer and identification device; wherein the content of the first and second substances,
the sprayer is used for controlling the image collector to collect plant images through the control module; sending the plant image to the identification device through a communication module;
the identification equipment is used for receiving the plant images sent by the sprayer and collected by the image collector on the sprayer; inputting the plant image into a multi-task processing model obtained through training to obtain a pest and disease identification result which is output by the multi-task processing model and is obtained by identifying the plant image; generating a spraying instruction for controlling a spray head of the sprayer to spray spraying liquid based on the pest and disease identification result, and sending the spraying instruction to the sprayer;
the sprayer is further used for receiving a spraying instruction fed back by the identification equipment aiming at the plant image through the communication module and responding to the spraying instruction to control a sprayer to spray spraying liquid.
A computer storage medium storing one or more programs executable by one or more processors to implement the steps of the plant pest identification method of the second or third aspect as described above.
The application provides a atomizer, plant pest identification method and identification equipment, wherein, the atomizer includes: the spray rod is arranged on the spray rod; the box body comprises a liquid storage box, a communication module, a control module and a power supply module; a liquid storage tank for storing spray liquid; the first end of the spray rod is connected with the liquid storage tank, and the second end of the spray rod is provided with a spray head; the first end of the supporting rod is connected with the spray rod, and the second end of the supporting rod is provided with an image collector; the power supply module is respectively connected with the communication module, the control module and the image collector and is used for supplying power to the communication module, the control module and the image collector; the control module is used for controlling the image collector to collect the plant images and controlling the communication module to send the plant images to the identification equipment; the control communication module receives a spraying instruction fed back by the identification equipment aiming at the plant image and responds to the spraying instruction to control the spray head to spray the spraying liquid; that is to say, this application is integrated on the atomizer with image collector, if set up on the bracing piece of being connected with the spray lance, so, along with the removal of spray lance between the plant, image collector can gather the image that contains each position of plant, and further, carry out the plant diseases and insect pests discernment and apply the medicine based on the image of all-round to plant collection, must realize the prevention and cure effect of better plant diseases and insect pests.
Drawings
Fig. 1 is a schematic network architecture diagram of an alternative plant pest identification system provided in an embodiment of the present application;
fig. 2 is a schematic structural diagram of a sprayer provided by an embodiment of the application;
FIG. 3 is a schematic diagram of a portion of an alternative sprayer provided by an embodiment of the present application;
FIG. 4 is a schematic diagram of a portion of an alternative sprayer provided by an embodiment of the present application;
fig. 5 is a schematic diagram of an alternative plant pest identification method provided by an embodiment of the present application;
FIG. 6 is a schematic diagram of an alternative multi-tasking model training method provided by an embodiment of the present application;
fig. 7 is a schematic flow chart of an alternative plant pest identification method provided by an embodiment of the present application;
fig. 8 is a schematic structural diagram of an identification device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.
In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be described in further detail with reference to the attached drawings, the described embodiments should not be considered as limiting the present application, and all other embodiments obtained by a person of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.
In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict.
In the following description, references to the terms "first \ second \ third" are only to distinguish similar objects and do not denote a particular order, but rather the terms "first \ second \ third" are used to interchange specific orders or sequences, where appropriate, so as to enable the embodiments of the application described herein to be practiced in other than the order shown or described herein.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of the present application only and is not intended to be limiting of the application.
Referring to fig. 1, fig. 1 is a schematic diagram of a network architecture for implementing a plant pest identification system provided by the present application, the network architecture at least including a sprayer 10, an identification device 20, and a network 30; wherein the nebulizer 10 and the identification device 20 are connected by a network 30. Here, the network 30 may be a wide area network or a local area network, or a combination of both, and uses wireless links to achieve data transmission. The nebulizer 10 is integrated with an image collector. The sprayer 10 can automatically collect plant images and send the plant images to the identification device 20, and then automatically execute spraying operation according to the spraying instruction fed back by the identification device 20; the identification device 20 may be referred to as a remote device connected to the nebulizer 10, the identification device 20 including, but not limited to, a smart phone, a tablet computer, a laptop computer, a desktop computer, a server, etc.; the server may be a single server, or may be a server cluster or a cloud computing center including a plurality of servers.
Referring to fig. 2, fig. 2 is a schematic structural view of the sprayer 10 provided herein, the sprayer 10 including: a box body 101, a spray rod 102 and a support rod 103; wherein the content of the first and second substances,
the case 101 includes a liquid storage tank 1011, a communication module 1012, a control module 1013, and a power supply module 1014; wherein the content of the first and second substances,
a liquid storage tank 1011 for storing a spray liquid;
the first end of the spray rod 102 is connected with the liquid storage tank 1011, and the second end of the spray rod 102 is provided with the spray head 104;
the first end of the supporting rod 103 is connected with the spray rod 102, and the second end of the supporting rod 103 is provided with an image collector 105; the image collector 105 comprises a single camera or an array of cameras.
The power supply module 1014 is respectively connected with the communication module 1012, the control module 1013 and the image collector 105, and is used for supplying power to the communication module 1012, the control module 1013 and the image collector 105;
a control module 1013 for controlling the image collector 105 to collect the plant image and controlling the communication module 1012 to send the plant image to the identification device 20 shown in fig. 1; the control communication module 1012 receives a spray instruction fed back by the recognition device 20 for the plant image, and controls the spray head 104 to spray the spray liquid in response to the spray instruction.
Illustratively, a first end of the spray bar 102 is connected with the liquid storage tank 1011, the communication module 1012 is a wireless communication module, illustratively, as shown in fig. 2, the communication module 1012 comprises a transmitting part 11 and a receiving part 12, the transmitting part 11 is integrally arranged outside the box body 101, and the receiving part 12 is integrally arranged inside the box body 101. Of course, the transmitting part 11 and the receiving part 12 of the communication module 1012 may also be arranged together (not shown in the figure), for example, both arranged integrally inside the casing 101 or arranged integrally outside the casing 101.
Illustratively, the sprayer 10 captures images of plants at specific intervals, such as 0.3 second intervals, by the image capture device 105, so as to achieve the technical effects of capturing images at each position in the crop area in the ground without causing excessive transmission flow or queuing of images for uploading.
Illustratively, the communication module 1012 includes a fifth generation mobile communication technology (5G) communication module, and the 5G communication module is adopted to transmit images and instructions, so as to achieve the technical effects of low delay, large bandwidth and high uploading speed.
In some embodiments, sprayer 10 may be a knapsack sprayer, including a manual knapsack sprayer and an automatic knapsack sprayer. In the scene of spraying pesticide on crops by using the knapsack manual sprayer, the knapsack manual sprayer consists of a pesticide liquid box welding part, a pump, an air chamber, a water outlet pipe, a handle switch, a spray rod, a spray head, a rocker part and a strap system. The product is rotated by the rocker part to open and close the leather cup in the pump and the air chamber in a rotation way, so that the internal pressure in the air chamber is gradually increased, and the liquid medicine at the bottom of the liquid medicine box passes through the water outlet pipe, the spray rod and the spray head to spray mist. In a scene of spraying pesticide on crops by using the knapsack automatic sprayer, the knapsack automatic sprayer is a high-pressure automatic sprayer, and the knapsack automatic sprayer adopts the Bernoulli principle, namely, in fluid, the flow rate is high, and the pressure is low; small flow speed and high pressure. The fluid will automatically flow from high pressure to low pressure. When passing through the Y-piece, the low-speed water flows to the high-speed air. The water is torn into small droplets by the high-speed air, and the small droplets are sprayed to form mist.
The present application provides a nebulizer 10 comprising: a box body 101, a spray rod 102 and a support rod 103; the case 101 includes a liquid storage tank 1011, a communication module 1012, a control module 1013, and a power supply module 1014; a liquid storage tank 1011 for storing a spray liquid; the first end of the spray rod 102 is connected with the liquid storage tank 1011, and the second end of the spray rod 102 is provided with the spray head 104; the first end of the supporting rod 103 is connected with the spray rod 102, and the second end of the supporting rod 103 is provided with an image collector 105; the power supply module 1014 is respectively connected with the communication module 1012, the control module 1013 and the image collector 105, and is used for supplying power to the communication module 1012, the control module 1013 and the image collector 105; a control module 1013 for controlling the image collector 105 to collect the plant image and controlling the communication module 1012 to send the plant image to the identification device 20; the control communication module 1012 receives a spraying instruction fed back by the recognition device 20 for the plant image, and controls the spray head 104 to spray the spray liquid in response to the spraying instruction; that is to say, this application is integrated on atomizer 10 with image collector 105, if set up on the bracing piece 103 of being connected with spray lance 102, so, along with the removal of spray lance 102 between the plant, image collector 105 can gather the image that contains each position of plant, and further, carry out pest and disease damage discernment and apply the medicine to the image of plant collection based on the all-round, must can realize better prevention and cure effect of pest and disease damage.
In an embodiment of the present application, as shown in fig. 2 and 3, the spray bar 102 has a first section 1021, a second section 1022, and a third section 1023, the second section 1022 of the spray bar 102 connects the first section 1021 of the spray bar 102 and the third section 1023 of the spray bar 102, the first section 1021 of the spray bar 102 is aligned with the support bar 103, the third section 1023 of the spray bar 102 is parallel to the support bar 103, and the distance between the spray head 104 and the image collector 105 is greater than a target distance. In this way, a certain distance is kept between the spray head 104 and the image collector 105, so that the sprayed pesticide is prevented or reduced from being scattered on the image collector 105. Further, in the upper and lower spaces of the three-dimensional space, the spray head 104 is located above the image collector 105, so that the image collector 105 is as close as possible to the ground, and thus high-definition images of the front and back surfaces of the leaves and images on the stems can be collected more easily.
In the embodiment of the present application, referring to fig. 2 and 3, the spray bar 102 is a telescopic structure, and/or the image collector 105 is disposed at the second end of the support bar 103 through a rotatable structure; a rotation button (not shown) is disposed on the spray bar 102 for controlling the rotatable structure to rotate to change the collection angle of the image collector 105. Illustratively, a rotary button may be provided on the handle 13 of the spray bar 102 for ease of operation.
In the embodiment of the present application, referring to fig. 2 and 3, the sprayer 10 further includes a positioning module 106, and the positioning module 106 is connected to the control module 1013; the control module 1013 is further configured to control the positioning module 106 to obtain position information of the plant image acquired by the image acquisition device 105; the control communication module 1012 transmits the location information to the recognition device 20; the control communication module 1012 receives a spray instruction fed back by the recognition device 20 for the plant image and the position information, and controls the spray head to spray the spray liquid in response to the spray instruction. As shown in fig. 3, the positioning module 106 is disposed on the supporting rod 103 and close to the image collector 105, so as to accurately obtain the position information of the image collector 105 when collecting the plant image.
In the embodiment of the present application, the positioning module 106 is adopted to record the position and send the position to the identification device, so that the identification device determines the environmental information of the area based on the position such as longitude and latitude, and the environmental information includes, but is not limited to, altitude, the amount of sky clouds in the weather information, the amount of daily precipitation, and the temperature. Therefore, the identification equipment can accurately position the geographical position of the plant and the surrounding environment of the plant growth, and further, the technical effects of improving the plant disease and insect pest identification accuracy and facilitating the disease and insect pest identification are achieved.
In the embodiment of the present application, referring to fig. 2 and 4, the liquid storage tank 1011 includes at least two independent medicine boxes, the first end of the spray rod is connected to each independent medicine box, and the spraying liquids stored in the at least two independent medicine boxes have different effects on plants. Illustratively, and as shown in connection with fig. 4, the liquid storage tank 1011 includes four separate medicine boxes, such as a first medicine box 14, a second medicine box 15, a third medicine box 16, and a fourth medicine box 17, which respectively contain an insecticide, a bactericide, a plant growth regulator, and a herbicide. Wherein, the pesticide: the pesticide is used for preventing and controlling pests such as agriculture, forestry, pasturing, sanitation, grain storage and the like or other arthropods. And (3) bactericide: are a class of pesticides, commonly referred to as fungicides, used to control plant diseases caused by a variety of pathogenic microorganisms. But internationally, they are generally referred to as agents for controlling various pathogenic microorganisms. Plant growth regulator: the synthetic chemical substances which have regulating effect on the growth and development of plants are called plant growth regulators. A plant growth regulator is a kind of agricultural chemicals for regulating the growth of plant. People can promote, inhibit, delay and other regulation activities on plants through specific plant growth regulators, so that the plants can grow and develop according to the direction required by human beings. For example, wheat stalks are made shorter and stronger through a retardant to enhance the lodging resistance; people promote more nutrient substances to be transported to fruits through the accelerant, and cultivate the sweet watermelon; people inhibit the potato germination through an inhibitor, prolong the storage period and the like. Herbicide: pesticides used to kill or control the growth of weeds are called herbicides. That is to say, this application adopts automatic knapsack sprayer design scheme, has reached automatic 4 kinds of pesticides of spraying, the automatic analysis plant's plant diseases and insect pests, the even technological effect of promotion pesticide spraying.
In an achievable scene, an operator uses a knapsack sprayer with a camera, a communication module, a positioning module and four pesticides to patrol in a crop planting area and automatically collects plant images. Furthermore, the knapsack sprayer sends the collected plant images and the collected position information to the identification device, the identification device feeds back a spraying instruction to the knapsack sprayer according to the plant images and the position information, and the knapsack sprayer responds to the instruction to realize reasonable pesticide application.
In the present embodiment, the sprayer 10 is a high-pressure automatic sprayer. Illustratively, as shown in fig. 4, the sprayer 10 further includes a solenoid valve 18, a pressure pump 19, a Polyvinyl chloride (PVC) line 41, a PVC line and electric wire 42, and a medicine box cover 43.
From the above, the present application provides a plant disease and pest identification system, and the plant disease and pest identification system includes atomizer and identification equipment. The sprayer is provided with a camera as a knapsack sprayer, and the knapsack box of the knapsack sprayer contains four kinds of pesticides: the method comprises the steps that an operator uses a sprayer to perform patrol inspection in the agricultural land, in the patrol inspection process, a camera on the sprayer acquires images of the front surface, the back surface and the stem of high-definition plant leaves every certain time such as 0.3 second, then the images are transmitted to identification equipment through a fifth generation mobile communication technology (5G) network to be stored and analyzed to determine whether diseases and insect pests exist and what types of diseases and insect pests exist, the identification equipment generates a spraying instruction according to an identification result and sends the spraying instruction to a receiving end of a knapsack sprayer, and the knapsack sprayer responds to the spraying instruction to automatically select pesticides and spray the pesticides.
Referring to fig. 5, fig. 5 is a schematic diagram of an implementation flow of a plant pest identification method provided in an embodiment of the present application, where the plant pest identification method may be applied to the identification device 20 shown in fig. 1; the plant disease and insect pest identification method comprises the following steps:
And 202, inputting the plant images into the trained multi-task processing model to obtain a pest and disease identification result which is output by the multi-task processing model and is obtained by identifying the plant images.
And 203, generating a spraying instruction for controlling a spray head of the sprayer to spray the spraying liquid based on the pest and disease identification result, and sending the spraying instruction to the sprayer.
The plant disease and insect pest identification method provided by the application receives plant images sent by a sprayer and collected by an image collector on the sprayer; inputting the plant image into a multi-task processing model obtained through training to obtain a pest and disease identification result which is output by the multi-task processing model and is obtained by identifying the plant image; generating a spraying instruction for controlling a spray head of the sprayer to spray the spraying liquid based on the pest and disease identification result, and sending the spraying instruction to the sprayer; namely, the identification equipment identifies the plant image collected by the camera of the sprayer by using the multitask processing model to obtain a pest identification result; and then the spraying instruction of controlling the spray head to spray the medicine is generated based on the recognition result, the medicine is discharged according to the symptom, and meanwhile, the effect of controlling the sprayer to automatically spray is achieved.
In the embodiment of the application, at the initial stage of the multitasking model training, after the identification device receives the plant image and the position information, the altitude, the weather cloud amount, the daily precipitation amount and the temperature in the altitude and weather information of the corresponding region are collected from the internet based on the position information. Further, the plant image may be labeled based on manual or automatic labeling, for example, the identification device presents the received plant image to an expert for labeling, the expert labels whether the plant image has a pest, if so, what pest is indicated, and how a solution is given to treat the pest, the solution includes a pesticide, a bactericide, a plant growth regulator, a herbicide, and if the plant image does not have a pest but the plant is withered or dead, the expert needs to label whether it is due to environmental factors. The plant image marked by the expert, the position information corresponding to the plant image and the environment information corresponding to the plant image are used as sample data of the multi-task processing model; the environmental information includes, but is not limited to, altitude of the corresponding region, sky cloud amount in weather information, daily precipitation amount, and temperature.
In other embodiments of the present application, in step 202, before inputting the plant image into the trained multitasking model to obtain the pest and disease identification result output by the multitasking model and obtained by identifying the plant image, the identification device 20 may further perform the following steps: and receiving the position information of the plant image collected by the image collector sent by the sprayer. In this embodiment, in step 202, the plant image is input into the trained multitask processing model to obtain a pest and disease identification result, which is output by the multitask processing model and is obtained by identifying the plant image, and the pest and disease identification result can be obtained through the following steps: and inputting the plant image and the position information into the multitask processing model to obtain a pest and disease identification result output by the multitask processing model. That is to say, in this embodiment, the multitask processing model combines plant images and position information, adopts the multitask analysis mode, improves the plant pest identification accuracy rate and the effect of accurate spraying control.
In other embodiments of the present application, in step 202, before inputting the plant image into the trained multitasking model to obtain the plant disease and insect pest identification result output by the multitasking model and obtained by identifying the plant image, the identification device 20 may further perform the following steps: receiving position information sent by a sprayer when an image collector collects plant images; environmental information associated with the location information is obtained. In this embodiment, in step 202, the plant image is input into the trained multitask processing model to obtain a pest and disease identification result, which is output by the multitask processing model and is obtained by identifying the plant image, and the pest and disease identification result can be obtained through the following steps: and inputting the plant image, the position information and the environment information into the multi-task processing model to obtain a pest and disease identification result output by the multi-task processing model. That is to say, in this embodiment, the multitask processing model combines the plant image, the location information, and the environmental information determined based on the location information, and adopts a multitask analysis mode, so as to further improve the plant pest identification accuracy and the effect of accurate spraying control.
In other embodiments of the present application, in step 202, before inputting the plant image into the trained multitasking model to obtain the pest and disease identification result output by the multitasking model and obtained by identifying the plant image, the identification device 20 may further perform the following steps:
firstly, obtaining a plurality of sample images, wherein the sample images comprise label information, and the label information comprises a real result expected to be obtained from the sample images by at least one task in the plurality of tasks;
secondly, inputting the plurality of sample images into the deep neural network model, and updating the shared parameters and the task parameters of the deep neural network model according to the output result of the deep neural network model and the label information to obtain a multi-task processing model.
Further description is made herein of the training process for the multitasking model,
after acquiring the plurality of sample images, the identification device 20 may further perform the steps of: the method comprises the steps of obtaining a plurality of sample data, wherein the sample data comprises sample position information corresponding to a plurality of sample images and sample environment information related to the sample position information.
At this time, the step of inputting the plurality of sample images into the deep neural network model, and updating the shared parameter and the task parameter of the deep neural network model according to the output result of the deep neural network model and the tag information to obtain the multitask processing model may be implemented by the following three steps:
firstly, inputting a plurality of sample images into a convolutional neural network model to obtain characteristic information of the plurality of sample images output by the convolutional neural network model.
And secondly, determining a first initial weight corresponding to the characteristic information, a second initial weight corresponding to the sample position information and a third initial weight corresponding to the sample environment information in the deep neural network model.
That is, before the multitask processing model is trained, the initialization weights corresponding to the feature information, the sample position information, and the sample environment information are set. In practical application, the influence degree of the information on pest and disease identification in an application scene can be combined, and the weight corresponding to each information is reasonably set.
And thirdly, inputting the plurality of sample images, the plurality of sample position information and the plurality of sample environment information into the deep neural network model, and updating the shared parameters and the task parameters of the network model according to the output result of the network model and the label information to obtain a multi-task processing model.
The task parameters comprise a first initial weight, a second initial weight and a third initial weight. Here, during the training of the multitask model, the weights corresponding to the above information can be continuously adjusted through inverse gradient propagation.
In the embodiment of the application, a multitask mutual promotion scheme in a multitask processing model is adopted, and the effect of improving the accuracy of identifying the plant diseases and insect pests by the model is achieved.
Referring to fig. 6, the training process of the recognition device on the multitasking model is as shown in fig. 6:
first, sample data is obtained.
The sample data comprises plant disease and insect pest images marked by experts, the geographical position of the image, such as longitude and latitude, the altitude corresponding to the geographical position, the amount of sky clouds in weather information, the daily precipitation and the temperature.
And secondly, the characteristics of the sample data have different weights in the model, and weight parameters are initialized randomly before the model begins to train.
Here, the weight parameter corresponding to each feature is continuously adjusted through inverse gradient propagation in the model training process. Namely, the weight parameters corresponding to the characteristics of the plant disease and insect pest image, the longitude and latitude, the altitude, the sky cloud cover, the daily precipitation and the temperature are adjusted.
In this embodiment, the plant disease and insect pest image may be encoded by a neural network model. The Neural network model includes a Convolutional Neural Network (CNN).
The geographical position of the image is represented by longitude and latitude, two continuous values are input into the weight parameters, and the weight parameters are initialized randomly before the model is trained.
The corresponding altitude of the geographic position of the image is also called absolute height, namely the height difference between a certain place and the sea level, the average sea level is usually used as a standard for calculation, a continuous value is input into weight parameters, and the weight parameters are initialized randomly before the model is trained.
The geographical position of the image corresponds to the sky cloud amount in the weather information, and the sky cloud amount generally consists of sunny, cloudy and cloudy. These four discrete variables are input to the weight parameters, which are all initialized at random prior to model training.
The daily precipitation corresponding to the geographic position of the image is input into the weight parameters, and the weight parameters are initialized randomly before the model training.
The temperature corresponding to the geographical location where the image is located, a continuous value is input to the weight parameters, and the weight parameters are initialized randomly before the model is trained.
And thirdly, fusing the characteristics of all the data on a multilayer Deep Neural Network (DNN).
And fourthly, the multitasking model at least comprises three tasks, such as judging whether pests and diseases exist, judging what types of pests and diseases exist, and judging whether weeds exist around the model. Compared with a common training mode, the multi-task processing model trained in the multi-task mode can be promoted to judge whether plant withering and death caused by environmental factors or plant diseases and insect pests are caused by the model, and the plant disease and insect pest identification accuracy rate is improved.
The accuracy rate of judging whether plant images have plant diseases and insect pests or not can be improved through the multi-task mode training model, and the accuracy rate of judging whether the plant images are withered or dead due to which plant diseases and insect pests is improved. The model can judge whether the plant suffers from plant diseases and insect pests, if the plant does not suffer from plant diseases and insect pests, the plant will not wither or die due to environmental factors, and if the plant suffers from plant diseases and insect pests, the model can judge which kind of plant diseases and insect pests result in. The information provides more accurate guidance for the operator, and the operator can conveniently and timely save the loss.
The multitask processing model is mainly used for judging whether a plant image is withered or dead due to the plant diseases and insect pests, and is mainly used for judging which type of plant diseases and insect pests and is a multi-classification problem.
The auxiliary task of the multitasking model is to judge whether the plant image has the plant diseases and insect pests, only whether the plant image has the plant diseases and insect pests or not is judged, the task is easy to judge, the main task is promoted to judge, and if the auxiliary task judges that the plant diseases and insect pests occur, the main task is promoted to learn and judge what types of plant diseases and insect pests occur.
The auxiliary task is to model whether the plant image withers or dies due to environmental factors. Environmental factors judge whether weeds exist around, and if weeds exist, the weeds are fed back to the automatic sprayer through the system to be directly sprayed with the herbicide. By adopting the weed identification scheme, the effect of identifying weeds around plants and accurately spraying the herbicide is achieved.
In other embodiments of the present application, taking as an example that the plant disease and insect pest identification method can be applied to a plant disease and insect pest identification system, the plant disease and insect pest identification system includes:
a nebulizer and identification device; wherein the content of the first and second substances,
the sprayer is used for controlling the image collector to collect plant images through the control module; sending the plant image to the identification device through the communication module;
the identification equipment is used for receiving the plant images which are sent by the sprayer and collected by the image collector on the sprayer; inputting the plant image into a multi-task processing model obtained through training to obtain a pest and disease identification result which is output by the multi-task processing model and is obtained by identifying the plant image; generating a spraying instruction for controlling a spray head of the sprayer to spray spraying liquid based on the pest and disease identification result, and sending the spraying instruction to the sprayer;
the sprayer is also used for receiving a spraying instruction fed back by the identification equipment aiming at the plant image through the communication module and controlling the spray head to spray the spraying liquid in response to the spraying instruction.
In an implementation scenario, taking the plant pest identification method as an example, which can be applied to a plant pest identification system, the method includes the following steps:
step1: an operator carries the sprayer on the back to the crop area to pass through the crops, and a camera on the sprayer is utilized to acquire plant images of the plants.
step1.1: the camera on the atomizer is to plant acquisition image, utilizes 5G communication module to send plant pest identification equipment, uses 5G communication to have low time delay, big bandwidth, uploads fast characteristics.
step1.2: be provided with orientation module such as Global Positioning System (GPS) on the atomizer, send plant pest identification equipment after acquireing positional information.
Step2: and the identification equipment processes the data after receiving the plant image and the GPS position information.
Step2.1: the identification device collects the current altitude, sky cloud cover, daily precipitation and temperature of the corresponding area from the network through the received GPS position information.
Step2.2: the recognition device inputs the data features referred to in step2.1 into the multitasking model.
Step 2.3: and the multitask processing model in the identification equipment judges whether the plant has plant diseases and insect pests, what types of plant diseases and insect pests are suffered by the plant and whether weeds around the plant have influences corresponding to plant growth through the input data characteristics. After the multi-task processing model finds the plant diseases and insect pests and determines the types of the plant diseases and insect pests, the multi-task processing model sends a spraying instruction to the knapsack sprayer in real time.
Step 2.4: 4 kinds of pesticides are contained in a pesticide box of the sprayer, and the pesticides are automatically sprayed according to a received instruction, for example, weeds around plants are collected by a camera, and then the herbicides are automatically sprayed.
In other embodiments of the present application, a method for identifying plant diseases and insect pests is provided, which is applied to the sprayer 10, and the method includes:
And 403, receiving a spraying instruction fed back by the recognition equipment after processing the plant image based on the multitasking model through the communication module, and responding to the spraying instruction to control the spray head to spray the spraying liquid.
This application is integrated on the atomizer with image collector, if set up on the bracing piece of being connected with the spray lance, so, along with the removal of spray lance between the plant, image collector can gather the image that contains each position of plant, further, carries out the plant diseases and insect pests discernment and applys the medicine to the image of plant collection based on the all-round, must realize the prevention and cure effect of better plant diseases and insect pests.
Referring to fig. 8, the present application provides an identification device 20, the identification device 20 comprising:
the receiving module 501 is used for receiving the plant image sent by the sprayer and collected by the image collector on the sprayer;
the processing module 502 is configured to input the plant image into the trained multitask processing model to obtain a pest identification result output by the multitask processing model and obtained by identifying the plant image;
the processing module 502 is further configured to generate a spraying instruction for controlling a spray head of the sprayer to spray the spraying liquid based on the pest and disease identification result;
a sending module 503, configured to send a spraying instruction to the sprayer.
In other embodiments of the present application, the receiving module 501 is further configured to receive position information, sent by the sprayer, of a plant image acquired by the image acquirer.
The processing module 502 is further configured to input the plant image and the position information into the multitask processing model to obtain a pest identification result output by the multitask processing model.
In other embodiments of the present application, the receiving module 501 is further configured to receive position information sent by the sprayer when the image collector collects the plant image; environmental information associated with the location information is obtained.
The processing module 502 is further configured to input the plant image, the position information, and the environment information into the multitask processing model to obtain a pest identification result output by the multitask processing model.
In other embodiments of the present application, the receiving module 501 is further configured to obtain a plurality of sample images, where the sample images include label information, and the label information includes a real result that at least one task of the plurality of tasks is expected to obtain from the sample images.
The processing module 502 is further configured to input the multiple sample images into the deep neural network model, and update the shared parameter and the task parameter of the deep neural network model according to the output result of the deep neural network model and the tag information, so as to obtain a multi-task processing model.
The above description of the apparatus embodiment is similar to the above description of the method embodiment, with similar beneficial effects as the method embodiment. For technical details not disclosed in the embodiments of the apparatus of the present application, reference is made to the description of the embodiments of the method of the present application for understanding.
It should be noted that, in the embodiment of the present application, if the plant disease and insect pest identification method is implemented in the form of a software functional module and is sold or used as an independent product, the method may also be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a terminal device to execute all or part of the methods of the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read Only Memory (ROM), a magnetic disk, or an optical disk. Thus, embodiments of the present application are not limited to any specific combination of hardware and software.
Embodiments of the application provide a computer readable storage medium storing one or more programs, which are executable by one or more processors to implement the steps shown in fig. 5 or fig. 7.
Here, it should be noted that: the above description of the storage medium and device embodiments, similar to the description of the method embodiments above, has similar beneficial effects as the method embodiments. For technical details not disclosed in the embodiments of the storage medium and apparatus of the present application, reference is made to the description of the embodiments of the method of the present application for understanding.
The computer storage medium/Memory may be a Memory such as a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmable Read Only Memory (EPROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a magnetic Random Access Memory (FRAM), a Flash Memory (Flash Memory), a magnetic surface Memory, an optical Disc, or a Read Only Disc (CD-ROM); but may also be various terminals such as mobile phones, computers, tablet devices, personal digital assistants, etc., that include one or any combination of the above-mentioned memories.
It should be appreciated that reference throughout this specification to "one embodiment" or "an embodiment of the present application" or "a previous embodiment" or "some embodiments" or "some implementations" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present application. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" or "an embodiment of the present application" or "the preceding embodiments" or "some implementations" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. It should be understood that, in the various embodiments of the present application, the sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application. The above-mentioned serial numbers of the embodiments of the present application are merely for description, and do not represent the advantages and disadvantages of the embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. The above-described device embodiments are merely illustrative, for example, the division of a unit is only one logical function division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units; can be located in one place or distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit may be implemented in the form of hardware, or in the form of hardware plus a software functional unit.
The methods disclosed in the several method embodiments provided in the present application may be combined arbitrarily without conflict to obtain new method embodiments.
The features disclosed in the several product embodiments presented in this application can be combined arbitrarily, without conflict, to arrive at new product embodiments.
The features disclosed in the several method or apparatus embodiments provided in the present application may be combined arbitrarily, without conflict, to arrive at new method embodiments or apparatus embodiments.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: various media that can store program codes, such as a removable Memory device, a Read Only Memory (ROM), a magnetic disk, or an optical disk.
Alternatively, the integrated units described above in the present application may be stored in a computer-readable storage medium if they are implemented in the form of software functional modules and sold or used as independent products. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially implemented or portions thereof contributing to the related art may be embodied in the form of a software product stored in a storage medium, and including several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods of the embodiments of the present application. And the aforementioned storage medium includes: various media that can store program code, such as removable storage devices, ROMs, magnetic or optical disks, etc.
It should be noted that the drawings in the embodiments of the present application are only for illustrating schematic positions of the respective devices on the terminal equipment, and do not represent actual positions in the terminal equipment, actual positions of the respective devices or the respective areas may be changed or shifted according to actual situations (for example, structures of the terminal equipment), and the proportions of different parts in the terminal equipment in the drawings do not represent actual proportions.
The above embodiments are only examples of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (14)
1. The sprayer is characterized by comprising a box body, a spray rod and a support rod; wherein, the first and the second end of the pipe are connected with each other,
the box body comprises a liquid storage box, a communication module, a control module and a power supply module; wherein the content of the first and second substances,
the liquid storage tank is used for storing spraying liquid;
the first end of the spray rod is connected with the liquid storage tank, and the second end of the spray rod is provided with a spray head;
the first end of the supporting rod is connected with the spray rod, and the second end of the supporting rod is provided with an image collector;
the power supply module is respectively connected with the communication module, the control module and the image collector and is used for supplying power to the communication module, the control module and the image collector;
the control module is used for controlling the image collector to collect plant images and controlling the communication module to send the plant images to the identification device; and controlling the communication module to receive a spraying instruction fed back by the identification equipment aiming at the plant image, and responding to the spraying instruction to control the spray head to spray the spraying liquid.
2. The sprayer of claim 1, wherein the spray bar has a first section, a second section, and a third section, the second section of the spray bar connecting the first section of the spray bar and the third section of the spray bar, the first section of the spray bar being in line with the support bar, the third section of the spray bar being parallel to the support bar, the distance between the spray head and the image collector being greater than a target distance.
3. The sprayer according to claim 1, wherein the spray bar is of a telescopic structure, and/or the image collector is arranged at the second end of the support bar through a rotatable structure; and a rotary key is arranged on the spray rod and used for controlling the rotatable structure to rotate so as to change the acquisition angle of the image acquisition device.
4. The nebulizer of claim 1, further comprising a positioning module, the positioning module being connected to the control module;
the control module is also used for controlling the positioning module to acquire the position information of the plant image acquired by the image acquisition device; controlling the communication module to send the location information to the identification device; and controlling the communication module to receive a spraying instruction fed back by the identification equipment aiming at the plant image and the position information, and responding to the spraying instruction to control the spray head to spray the spraying liquid.
5. The sprayer according to claim 1, wherein said reservoir comprises at least two separate pesticide boxes, said first end of said spray bar being connected to each of said separate pesticide boxes, respectively, said spray liquids stored in said at least two separate pesticide boxes acting differently on the plants.
6. The nebulizer of any one of claims 1 to 5, wherein the nebulizer is a high pressure automatic nebulizer.
7. A plant pest and disease identification method is applied to identification equipment, and comprises the following steps:
receiving a plant image which is sent by a sprayer and collected by an image collector on the sprayer;
inputting the plant image into a multi-task processing model obtained through training to obtain a pest and disease identification result which is output by the multi-task processing model and is obtained by identifying the plant image;
and generating a spraying instruction for controlling a spray head of the sprayer to spray spraying liquid based on the pest and disease identification result, and sending the spraying instruction to the sprayer.
8. The method of claim 7, further comprising:
receiving position information sent by a sprayer when the image collector collects the plant image;
correspondingly, the step of inputting the plant image into a multi-task processing model obtained through training to obtain a pest and disease identification result output by the multi-task processing model and obtained by identifying the plant image includes:
and inputting the plant image and the position information into the multitask processing model to obtain the pest and disease identification result output by the multitask processing model.
9. The method of claim 7, further comprising:
receiving position information sent by the sprayer when the image collector collects the plant image;
acquiring environmental information associated with the position information;
correspondingly, the step of inputting the plant image into a multi-task processing model obtained through training to obtain a pest and disease identification result output by the multi-task processing model and obtained by identifying the plant image includes:
and inputting the plant image, the position information and the environment information into the multitask processing model to obtain the pest and disease identification result output by the multitask processing model.
10. The method according to any one of claims 7 to 9, wherein the receiving the image of the plant collected by the image collector on the sprayer sent by the sprayer is preceded by:
obtaining a plurality of sample images, wherein the sample images comprise label information, and the label information comprises real results expected to be obtained from the sample images by at least one task in a plurality of tasks;
and inputting the plurality of sample images into a deep neural network model, and updating the shared parameters and the task parameters of the deep neural network model according to the output result of the deep neural network model and the label information to obtain the multi-task processing model.
11. The method of claim 10, wherein after the obtaining the plurality of sample images, the method further comprises:
acquiring a plurality of sample data, wherein the sample data comprises sample position information corresponding to the plurality of sample images and sample environment information associated with the sample position information;
correspondingly, the inputting the plurality of sample images into a deep neural network model, and updating the shared parameters and the task parameters of the deep neural network model according to the output result of the deep neural network model and the tag information to obtain the multitask processing model includes:
inputting the sample images into a convolutional neural network model to obtain characteristic information of the sample images output by the convolutional neural network model;
determining a first initial weight corresponding to the characteristic information, a second initial weight corresponding to the sample position information and a third initial weight corresponding to the sample environment information in the deep neural network model;
inputting the plurality of sample images, the plurality of sample position information and the plurality of sample environment information into the deep neural network model, and updating the shared parameters and the task parameters of the network model according to the output result of the network model and the label information to obtain the multitask processing model; wherein the task parameters include the first initial weight, the second initial weight, and the third initial weight.
12. A method for identifying plant diseases and insect pests, which is applied to the sprayer according to any one of claims 1 to 6, comprising:
controlling the image collector to collect plant images through a control module;
sending the plant image to a recognition device through a communication module;
and receiving a spraying instruction fed back after the recognition equipment processes the plant image based on a multitasking model through the communication module, and responding to the spraying instruction to control a spray head to spray spraying liquid.
13. The method of claim 12, further comprising:
the control module controls a positioning module to obtain the position information of the image collector when the image collector collects the plant image;
sending the location information to the identification device through the communication module;
correspondingly, the receiving, by the communication module, a spraying instruction fed back by the identification device for the plant image includes:
receiving, by the communication module, the spraying instruction fed back by the identification device for the plant image and the position information.
14. An identification device, characterized in that the identification device comprises:
the receiving module is used for receiving the plant images which are sent by the sprayer and collected by the image collector on the sprayer;
the processing module is used for inputting the plant images into a multi-task processing model obtained through training to obtain a pest and disease identification result which is output by the multi-task processing model and is obtained by identifying the plant images;
the processing module is further used for generating a spraying instruction for controlling a spray head of the sprayer to spray spraying liquid based on the pest and disease identification result;
and the sending module is used for sending the spraying instruction to the sprayer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110631125.4A CN115500334A (en) | 2021-06-07 | 2021-06-07 | Sprayer, plant disease and insect pest identification method and identification equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110631125.4A CN115500334A (en) | 2021-06-07 | 2021-06-07 | Sprayer, plant disease and insect pest identification method and identification equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115500334A true CN115500334A (en) | 2022-12-23 |
Family
ID=84499420
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110631125.4A Pending CN115500334A (en) | 2021-06-07 | 2021-06-07 | Sprayer, plant disease and insect pest identification method and identification equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115500334A (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103999841A (en) * | 2014-04-18 | 2014-08-27 | 浙江大学 | Automatic targeting and spraying system |
| CN205284756U (en) * | 2016-01-04 | 2016-06-08 | 安徽农业大学 | Crops atomizer based on intelligence image processing |
| CN106585992A (en) * | 2016-12-15 | 2017-04-26 | 上海土是宝农业科技有限公司 | Method and system for intelligent identification and accurate pesticide spraying using unmanned aerial vehicles |
| CN106991619A (en) * | 2017-06-05 | 2017-07-28 | 河北哲瀚网络科技有限公司 | A kind of diseases and pests of agronomic crop intelligent diagnosis system and diagnostic method |
| CN108549869A (en) * | 2018-04-13 | 2018-09-18 | 哈尔滨理工大学 | A kind of adaptive operational method of plant protection drone based on expert system |
| US20180330166A1 (en) * | 2017-05-09 | 2018-11-15 | Blue River Technology Inc. | Automated plant detection using image data |
| CN109122633A (en) * | 2018-06-25 | 2019-01-04 | 华南农业大学 | The accurate variable-rate spraying device of the plant protection drone of Decision of Neural Network and control method |
| CN109863916A (en) * | 2019-01-10 | 2019-06-11 | 康程 | Method based on intelligent full physical schemes deinsectization weeding |
| CN111914914A (en) * | 2020-07-21 | 2020-11-10 | 上海理想信息产业（集团）有限公司 | Method, device, equipment and storage medium for identifying plant diseases and insect pests |
| US20200375172A1 (en) * | 2019-05-31 | 2020-12-03 | Milar Agro Tech S.r.l. | Device to detect and exercise control over weeds applied on agricultural machinery |
| EP3804488A1 (en) * | 2019-10-08 | 2021-04-14 | Bayer AG | Apparatus for plant disease and pest detection |
-
2021
- 2021-06-07 CN CN202110631125.4A patent/CN115500334A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103999841A (en) * | 2014-04-18 | 2014-08-27 | 浙江大学 | Automatic targeting and spraying system |
| CN205284756U (en) * | 2016-01-04 | 2016-06-08 | 安徽农业大学 | Crops atomizer based on intelligence image processing |
| CN106585992A (en) * | 2016-12-15 | 2017-04-26 | 上海土是宝农业科技有限公司 | Method and system for intelligent identification and accurate pesticide spraying using unmanned aerial vehicles |
| US20180330166A1 (en) * | 2017-05-09 | 2018-11-15 | Blue River Technology Inc. | Automated plant detection using image data |
| CN106991619A (en) * | 2017-06-05 | 2017-07-28 | 河北哲瀚网络科技有限公司 | A kind of diseases and pests of agronomic crop intelligent diagnosis system and diagnostic method |
| CN108549869A (en) * | 2018-04-13 | 2018-09-18 | 哈尔滨理工大学 | A kind of adaptive operational method of plant protection drone based on expert system |
| CN109122633A (en) * | 2018-06-25 | 2019-01-04 | 华南农业大学 | The accurate variable-rate spraying device of the plant protection drone of Decision of Neural Network and control method |
| CN109863916A (en) * | 2019-01-10 | 2019-06-11 | 康程 | Method based on intelligent full physical schemes deinsectization weeding |
| US20200375172A1 (en) * | 2019-05-31 | 2020-12-03 | Milar Agro Tech S.r.l. | Device to detect and exercise control over weeds applied on agricultural machinery |
| EP3804488A1 (en) * | 2019-10-08 | 2021-04-14 | Bayer AG | Apparatus for plant disease and pest detection |
| CN111914914A (en) * | 2020-07-21 | 2020-11-10 | 上海理想信息产业（集团）有限公司 | Method, device, equipment and storage medium for identifying plant diseases and insect pests |
Non-Patent Citations (1)
| Title |
|---|
| 谢能付等: "《智能农业—智能时代的农业生产方式变革》", vol. 1, 中国铁道出版社, pages: 56 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Subeesh et al. | Automation and digitization of agriculture using artificial intelligence and internet of things | |
| US20220167605A1 (en) | Method for plantation treatment of a plantation field | |
| US20220167546A1 (en) | Method for plantation treatment of a plantation field with a variable application rate | |
| Oliveira et al. | Agricultural robotics: A state of the art survey | |
| Edan et al. | Agriculture automation | |
| Sivarethinamohan et al. | Captivating profitable applications of artificial intelligence in agriculture management | |
| Jain et al. | Agriculture assistant for crop prediction and farming selection using machine learning model with real-time data using imaging through uav drone | |
| CN115500334A (en) | Sprayer, plant disease and insect pest identification method and identification equipment | |
| Patil et al. | Review on automatic variable-rate spraying systems based on orchard canopy characterization | |
| Phillips | Precision agriculture: supporting global food security. | |
| Kuppusamy et al. | Machine Learning-Enabled Internet of Things Solution for Smart Agriculture Operations | |
| Esau et al. | Artificial intelligence and deep learning applications for agriculture | |
| Talaviya et al. | Artificial Intelligence in Agriculture | |
| Phade et al. | IoT‐Enabled Unmanned Aerial Vehicle: An Emerging Trend in Precision Farming | |
| TUĞRUL | ARTIFICIAL INTELLIGENCE AND SMART FARMING APPLICATIONS IN AGRICULTURAL MACHINERY | |
| Joy et al. | Agriculture 4.0 in Bangladesh: issues and challenges | |
| Zaman | Precision Agriculture: Evolution, Insights and Emerging Trends | |
| Subeesh et al. | Artificial Intelligence in Agriculture | |
| US20240000002A1 (en) | Reduced residual for smart spray | |
| Agrawal et al. | Mechanizing Indian Agriculture with precision farming technologies: Challenges and perspective | |
| US20230360149A1 (en) | Computer implemented method for providing test design and test instruction data for comparative tests for yield, gross margin, efficacy and/or effects on vegetation indices on a field for different rates or application modes of one product | |
| GOHIL | Feasibility study of mechanization of agricultural robots in the harvesting of artichokes | |
| Wójcik-Czerniawska | The role of Artificial Intelligence (AI) in agriculture and its impact on economy | |
| Muralidharan et al. | Internet of Agro Drones for Precision Agriculture | |
| Vyas et al. | A Review on Application of Drone System in Precision Agriculture |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |