CN116806791A - Crop pest identification method and system based on deep learning - Google Patents
Crop pest identification method and system based on deep learning Download PDFInfo
- Publication number
- CN116806791A CN116806791A CN202310760659.6A CN202310760659A CN116806791A CN 116806791 A CN116806791 A CN 116806791A CN 202310760659 A CN202310760659 A CN 202310760659A CN 116806791 A CN116806791 A CN 116806791A
- Authority
- CN
- China
- Prior art keywords
- crop
- crops
- insect pests
- moving mechanism
- image data
- 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.)
- Granted
Links
- 241000607479 Yersinia pestis Species 0.000 title claims abstract description 85
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000013135 deep learning Methods 0.000 title claims abstract description 15
- 201000010099 disease Diseases 0.000 claims abstract description 71
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 71
- 241000238631 Hexapoda Species 0.000 claims abstract description 54
- 239000000575 pesticide Substances 0.000 claims description 62
- 238000005507 spraying Methods 0.000 claims description 53
- 238000012545 processing Methods 0.000 claims description 40
- 230000011664 signaling Effects 0.000 claims description 26
- 239000007921 spray Substances 0.000 claims description 22
- 239000007788 liquid Substances 0.000 claims description 20
- 239000003814 drug Substances 0.000 claims description 18
- 238000007689 inspection Methods 0.000 claims description 16
- 238000003860 storage Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 6
- 230000000149 penetrating effect Effects 0.000 claims description 5
- 229940079593 drug Drugs 0.000 claims description 4
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 claims description 3
- 238000013499 data model Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012271 agricultural production Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000003750 conditioning effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- -1 re Species 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Abstract
The invention discloses a method and a system for identifying crop diseases and insect pests based on deep learning, which relate to the technical field of agricultural disease and insect pest identification and are used for solving the problems that the existing crop disease and insect pest identification system is small in identification range of crops, can not accurately identify different parts of the crops and can not carry out emergency treatment for the first time after identifying the disease and insect pests; the invention can collect the image data of the crops by using the identification system arranged in the crop area, judges whether the crop disease and pest conditions exist by using the image data, can collect different parts of the crops in a targeted way when the image data of the crops are collected, automatically patrols the growth conditions of the crops, and greatly reduces the manpower requirement.
Description
Technical Field
The invention relates to the technical field of agricultural pest identification, in particular to a crop pest identification method and system based on deep learning.
Background
In agricultural production, pest and disease damage is a common problem, and if timely identification and treatment are performed, loss can be effectively reduced, and yield can be improved.
The traditional crop pest and disease damage identification method mainly relies on manual observation and experience judgment, and has the problems of low accuracy, low efficiency and the like.
And some are used in the recognition system of crop diseases and insect pests and are little to the recognition range of crops, can't carry out accurate discernment to the different positions of crops to can't carry out emergency treatment in the very first time after discerning the disease and insect pest, the effect is relatively poor.
In order to solve the above-mentioned defect, a technical scheme is provided.
Disclosure of Invention
The invention aims to solve the problems that the existing crop pest and disease damage identification system is small in identification range of crops, different parts of the crops cannot be accurately identified, and emergency treatment cannot be carried out for the first time after the pests and the diseases are identified, and provides a method and a system for identifying the crop pest and disease damage based on deep learning.
The aim of the invention can be achieved by the following technical scheme:
the crop pest identification method based on deep learning comprises the following steps:
step S1: firstly, driving a collection assembly to move in multiple directions through a lifting structure, a transverse moving mechanism and a longitudinal moving mechanism, collecting crop growth data in an area, and establishing a crop model according to the crop growth data;
step S2: controlling the lifting structure, the transverse moving mechanism and the longitudinal moving mechanism according to the established crop model to drive the collecting assembly to collect image data of different crop leaves, stems and roots;
step S3: the collected crop image data is imported into a preset training model, whether the crop has diseases or insect pests is judged, and when the crop has no diseases or insect pests, the inspection time is reset;
step S4: when crop diseases and insect pests exist, acquiring crops with specific diseases and insect pests and positions of the crops, analyzing the area proportion of the diseases and insect pests and comparing the area proportion with a preset area proportion, and when the analyzed real-time area proportion is smaller than or larger than the preset area proportion, generating single-point processing signaling and area processing signaling respectively;
step S5: after receiving the single-point processing signaling or the regional processing signaling, controlling the pesticide spraying assembly to switch to a single-point pesticide spraying mode or a single-row pesticide spraying mode respectively to spray pesticide to the region where crop diseases and insect pests occur.
Further, a crop pest identification system based on deep learning, comprising:
the collection assembly is used for collecting image data of crop growth;
the collecting assembly comprises a second threaded sleeve and a guide sleeve, the second threaded sleeve and the guide sleeve are connected to the longitudinal moving mechanism, the second threaded sleeve is fixed with the guide sleeve, an L-shaped hanging plate is fixed at the bottom of the second threaded sleeve, a reversing motor is mounted on the first screw rod, the output end of the reversing motor penetrates through the L-shaped hanging plate and is fixedly provided with a connecting column, the lower end of the connecting column is provided with a collecting camera, and the collecting camera is provided with a downward inclined angle;
the pesticide spraying component is used for spraying the pesticides with different modes on crops with diseases and insect pests;
the pesticide spraying assembly comprises a pesticide storage tank, the pesticide storage tank is arranged on the outer side of a transverse moving mechanism, pesticide liquid is arranged in the pesticide storage tank, a booster pump is arranged on the surface of the pesticide storage tank, an outlet of the booster pump is connected with a connecting hose, the other end of the connecting hose is connected with a fixed pipe, the top of the vertical plate is fixed with, the fixed pipe is arranged on the top of the vertical plate, the other end of the connecting hose is hoisted with an electric switching valve, the electric switching valve is positioned above the middle of a second screw rod, the other end of the fixed pipe is connected with an inlet of the electric switching valve, an extension pipe is connected at an outlet of the electric switching valve, a convex plate is fixed on the surface of the connecting column, a joint pipe is arranged on the convex plate, the joint pipe is connected with the other end of the extension pipe, the other end of the joint pipe is connected with a one-way sprayer through a diagonal pipe, two electric hanging rods are arranged on the electric switching valve, the other ends of the two electric hanging rods are respectively communicated with two liquid draining pipes through penetrating pipes, and the other two liquid draining pipes penetrate through the two liquid penetrating pipes and are respectively distributed on the surfaces of the two hanging rods at equal intervals;
the longitudinal moving mechanism is used for driving the whole collecting assembly to longitudinally move so as to collect the influence data of crops in different longitudinal directions; the transverse moving mechanism is used for driving the collecting assembly to transversely move to collect image data of crops in different transverse directions; the lifting structure is used for supporting the whole system, changing the collection height of the collection assembly and adapting to the collection of different parts of crops;
the adjusting module is used for acquiring crop planting data, establishing a crop growth model, controlling the collecting assembly according to the model to collect crop and crop part image data in different areas, analyzing the collected image data, judging whether pest and disease exist or not, and carrying out emergency treatment on the crop with the pest and disease.
Further, the adjustment module includes:
the modeling unit is used for establishing a model for the growth data of crops;
the inspection unit is used for controlling the collection assembly to collect image data of crops and crop parts in different areas according to the established crop growth model;
the analysis unit is used for receiving the image data of the crop growth condition obtained by the inspection unit and judging whether the crop has diseases and insect pests according to the trained data model;
the specific analysis process comprises the following steps:
retrieving and importing image data of normal states of crops and states of diseases and insect pests at different positions;
the image enhancement, denoising and cutting operation are carried out on the imported image data, so that the accuracy and the efficiency of subsequent processing are convenient;
training the preprocessed image data by using a convolutional neural network, and performing model selection and parameter adjustment in a data training process by a cross validation method;
the trained model is applied to crop growth condition image data acquired by the inspection unit, and whether crop diseases and insect pests exist or not is judged;
when the crop diseases and insect pests do not exist, resetting the inspection time, and inspecting the crops again when the next period appears;
when the crop diseases and insect pests exist, analyzing the ratio of the disease and insect pests areas, and when the ratio of the disease and insect pests is smaller than a preset value, generating a single-point processing signaling, and simultaneously acquiring the positions of the disease and insect pests areas and sending the positions to a processing unit;
when the ratio of the pest and disease damage area is larger than a preset value, generating an area processing signaling, and simultaneously acquiring the position of the pest and disease damage area and sending the position to a processing unit;
and the processing unit is used for receiving the single-point processing signaling and the regional processing signaling and respectively spraying the medicines to the pest crops through different modes.
Further, the specific process of emergency treatment of crops by the treatment unit comprises the following steps:
when single-point processing signaling is received, the pesticide spraying component is controlled to switch to a single-point pesticide spraying mode to spray pesticides at a certain position in a targeted manner, the lifting structure, the transverse moving mechanism and the longitudinal moving mechanism are used for controlling the whole collecting component to drive the unidirectional spray head to move to spray pesticides at different positions, and the collecting component is matched to collect images of the pesticide spraying process of crops in real time;
when the region processing signaling is received, the pesticide spraying assembly is controlled to be switched to a single-row spraying mode to spray pesticides on crops in a certain row of crops, and the lifting structure, the transverse moving mechanism and the longitudinal moving mechanism are utilized to drive the whole pesticide spraying assembly to move, so that the positions of the crops in different rows are changed to spray the pesticides.
Further, the vertical moving mechanism comprises two vertical plates, the two vertical plates are respectively connected to the horizontal moving mechanism, the two vertical plates are respectively fixed with an extension plate on the side wall, a second screw rod is rotationally connected between the two vertical plates and is in threaded connection with a second threaded sleeve, a guide rod is fixed between the two extension plates, the guide sleeve is in sliding connection with the surface of the guide rod, a second motor is mounted on the side wall of the vertical plate, and the output end of the second motor penetrates through the vertical plate and is fixed with the second screw rod.
Further, the lateral shifting mechanism includes two first screw sleeve, two first screw sleeve is fixed in the bottom of two risers respectively, be connected with two spill backup pads on the elevation structure, two all rotate in the spill backup pad and be connected with first lead screw, and first lead screw and first screw sleeve threaded connection, two first motor is all installed to the lateral wall of spill backup pad, and the output of first motor runs through the spill backup pad and is fixed with first lead screw.
Further, the elevation structure includes two hydraulic cylinder, two hydraulic cylinder connects in the bottom of two spill backup pads respectively, and the spill backup pad is connected on hydraulic cylinder's extension, two hydraulic cylinder's bottom all is fixed with inserts, two hydraulic cylinder's surface all is fixed with the balance board that a plurality of equidistance distributes, a plurality of the bottom of balance board all is fixed with the grounding piece.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, the identification system arranged in the crop area can be used for collecting crop image data, judging whether crop diseases and insect pests exist or not by using the image data, and collecting different parts of crops in a targeted manner when the crop image data are collected, so that the crop growth condition is automatically inspected, and the manpower requirement is greatly reduced;
when the crop disease and insect pest situation occurs, the invention can automatically spray and treat the crop disease and insect pest area, stop damage in time, avoid the disease and insect pest situation of large-area crops, and simultaneously treat the pesticide in different modes according to the occupation ratio situation of the crop disease and insect pest area, thereby avoiding the situations of large manual inspection manpower requirement and inaccurate inspection.
Drawings
For the convenience of those skilled in the art, the present invention will be further described with reference to the accompanying drawings;
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a general block diagram of a conditioning module according to the present invention;
FIG. 3 is a schematic diagram of the structure of the present invention;
FIG. 4 is a partial view of the portion A of FIG. 3 in accordance with the present invention;
FIG. 5 is a second view of the present invention shown in FIG. 3;
fig. 6 is a partial view of fig. 5 at B in accordance with the present invention.
In the drawings, the list of components represented by the various numbers is as follows:
1. a lifting structure; 101. a hydraulic cylinder; 102. performing ground insertion; 103. a balance plate; 104. a grounding member; 2. a lateral movement mechanism; 201. a concave support plate; 202. a first motor; 203. a first screw rod; 204. a first threaded sleeve; 3. a longitudinal movement mechanism; 301. a vertical plate; 302. an extension plate; 303. a second motor; 304. a second screw rod; 305. a guide rod; 4. a spraying component; 401. a drug storage tank; 402. a booster pump; 403. a connecting hose; 404. a fixed tube; 405. an electric switching valve; 406. a boom; 407. a liquid discharge pipe; 408. a liquid ejecting head; 409. an extension tube; 410. a convex plate; 411. a joint pipe; 412. a diagonal bracing tube; 413. a unidirectional spray head; 5. a collection assembly; 501. a second threaded sleeve; 502. guide sleeve; 503. l-shaped hanging plates; 504. reversing the motor; 505. a connecting column; 506. and (5) collecting a camera.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the present disclosure is for the purpose of describing particular embodiments only, and is not intended to be limiting of the disclosure. As used in the specification and claims of this disclosure, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should be further understood that the term "and/or" as used in the present disclosure and claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
As shown in fig. 1-6, a crop pest identification system based on deep learning comprises a lifting structure 1, a transverse moving mechanism 2, a longitudinal moving mechanism 3, a pesticide spraying component 4 and a collecting component 5; the collecting assembly 5 is connected to the longitudinal moving mechanism 3, the longitudinal moving mechanism 3 is connected to the transverse moving mechanism 2, the transverse moving mechanism 2 is connected to the lifting structure 1, the mutual matching of the lifting structure 1, the transverse moving mechanism 2 and the longitudinal moving mechanism 3 can drive the collecting assembly 5 to collect image data of crops in different areas, the pesticide spraying assembly 4 and the collecting assembly 5 are in a linkage state, and the position of the pesticide spraying assembly can be changed when the collecting assembly 5 moves, so that the pesticide spraying treatment range of the crops is improved;
wherein: the lifting structure 1 is used for supporting the whole system, changing the collection height of the collection assembly 5 and adapting to the collection of different parts of crops; the lifting structure 1 comprises two hydraulic cylinders 101, wherein the two hydraulic cylinders 101 are respectively connected to the bottoms of two concave supporting plates 201, the concave supporting plates 201 are connected to the extending ends of the hydraulic cylinders 101, ground inserts 102 are fixedly arranged at the bottoms of the two hydraulic cylinders 101, a plurality of balancing plates 103 which are distributed at equal intervals are fixedly arranged on the surfaces of the two hydraulic cylinders 101, and grounding pieces 104 are fixedly arranged at the bottoms of the plurality of balancing plates 103; the lifting structure 1 adopts two hydraulic cylinders 101 to control the height of the whole collection assembly 5 to change, the lifting structure 1 can be supported by inserting a ground insert 102 at the bottom of the hydraulic cylinders 101 into the ground, and the lifting structure 1 is matched with a plurality of grounding pieces 104 at the bottom of a balance plate 103 so as to improve the stability of the whole lifting structure 1;
the transverse moving mechanism 2 is used for driving the collecting assembly 5 to transversely move to collect image data of crops in different transverse directions, the transverse moving mechanism 2 comprises two first threaded sleeves 204, the two first threaded sleeves 204 are respectively fixed at the bottoms of the two vertical plates 301, the lifting structure 1 is connected with two concave supporting plates 201, the two concave supporting plates 201 are internally and rotatably connected with first screw rods 203, the first screw rods 203 are in threaded connection with the first threaded sleeves 204, the side walls of the two concave supporting plates 201 are respectively provided with a first motor 202, and the output ends of the first motors 202 penetrate through the concave supporting plates 201 and are fixed with the first screw rods 203; the first motor 202 in the transverse moving mechanism 2 can drive the first screw rod 203 to rotate, so that the first threaded sleeve 204 in threaded connection with the first screw rod 203 drives the whole longitudinal moving mechanism 3 and the collecting assembly 5 to move left and right, and the effect of controlling the transverse movement of the collecting assembly 5 is realized;
the longitudinal moving mechanism 3 is used for driving the whole collecting assembly 5 to longitudinally move to collect influence data of crops in different longitudinal directions, the longitudinal moving mechanism 3 comprises two vertical plates 301, the two vertical plates 301 are respectively connected to the transverse moving mechanism 2, extension plates 302 are fixed on the side walls of the two vertical plates 301, a second screw rod 304 is rotatably connected between the two vertical plates 301, the second screw rod 304 is in threaded connection with a second threaded sleeve 501, a guide rod 305 is fixed between the two extension plates 302, the guide sleeve 502 is slidably connected to the surface of the guide rod 305, a second motor 303 is mounted on the side wall of the vertical plate 301, and the output end of the second motor 303 penetrates through the vertical plate 301 and is fixed with the second screw rod 304; the second motor 303 is utilized in the longitudinal moving mechanism 3 to drive the second screw rod 304 connected between the two vertical plates 301 to rotate, so that the second threaded sleeve 501 in threaded connection with the second screw rod 304 drives the whole collecting assembly 5 to move back and forth, wherein the guide sleeve 502 fixed on the side edge of the second threaded sleeve 501 can slide on the surface of the guide rod 305, and the stability of the back and forth movement of the collecting assembly 5 is improved;
the pesticide spraying component 4 is used for spraying the pesticides with different modes on crops with diseases and insect pests; the spraying component 4 comprises a medicine storage tank 401, the medicine storage tank 401 is arranged on the outer side of the transverse moving mechanism 2, pesticide liquid is arranged in the medicine storage tank 401, a booster pump 402 is arranged on the surface of the medicine storage tank 401, an outlet of the booster pump 402 is connected with a connecting hose 403, the other end of the connecting hose 403 is connected with a fixed pipe 404, the top of a vertical plate 301 is fixed with a 414, the fixed pipe 404 is arranged on the 414, the other end of the 414 is suspended with an electric switching valve 405, the electric switching valve 405 is positioned above the middle of a second screw rod 304, the other end of the fixed pipe 404 is connected with an inlet of the electric switching valve 405, an outlet of the electric switching valve 405 is connected with an extension pipe 409, the surface of a connecting column 505 is fixed with a convex plate 410, a joint pipe 411 is arranged on the convex plate 410, the joint pipe 411 is connected with the other end of the extension pipe 409, the other end of the joint pipe 411 is connected with a one-way sprayer through an inclined supporting pipe 412, two electric suspenders 406 are arranged on the electric switching valve 405, the other ends of the two electric suspenders 406 are both fixed with drain pipes 407, the other two outlets of the electric switching valve 405 are both communicated with the two drain pipes 408 through penetrating pipes respectively, and the two drain pipes 408 are distributed on the surfaces of the two drain pipes which are distributed at equal intervals; the medicine spraying component 4 changes the medicine spraying position through the lifting structure 1, the transverse moving mechanism 2 and the longitudinal moving mechanism 3, and can also switch different medicine spraying modes by utilizing the electric switching valve 405;
when the single-point medicine spraying mode is switched, the pesticide liquid in the medicine storage tank 401 is led into the sequentially connected hose 403, the fixed pipe 404 and the electric switching valve 405 by the pressurization of the booster pump 402, and the liquid outlet of the electric switching valve 405 is controlled to the port of the extension pipe 409, so that the pesticide liquid is discharged through the extension pipe 409, the joint pipe 411 and the inclined support pipe 412 by utilizing the unidirectional spray nozzle 413, the unidirectional spray nozzle 413 is provided with an inclination angle matched with the acquisition camera 506, the acquisition camera 506 can monitor the medicine spraying condition, and meanwhile, the joint pipe 411 is connected to the connecting column 505 by the convex plate 410, so that the change of the angle of the acquisition camera 506 can be carried out, and a certain allowance is provided by utilizing the extension pipe 409 for the wider spraying range of the whole unidirectional spray nozzle 413;
when the single-row pesticide spraying mode is switched, pesticide liquid is discharged to the liquid discharge pipe 407 and a plurality of liquid spraying heads 408 which are arranged at equal intervals through the through pipe by controlling the liquid discharge port of the electric switching valve 405 to the through pipe in the electric suspender 406, so that pesticide spraying is uniformly performed on one row of pesticide, the pesticide spraying efficiency is improved, and the two liquid discharge pipes 407 are symmetrically arranged, so that the pesticide spraying on the left side and the right side of the single-row crops can be flexibly switched, and different use conditions are adapted;
the collection assembly 5 is used for collecting image data of crop growth; the collecting assembly 5 comprises a second threaded sleeve 501 and a guide sleeve 502, the second threaded sleeve 501 and the guide sleeve 502 are connected to the longitudinal moving mechanism 3, the second threaded sleeve 501 is fixed with the guide sleeve 502, an L-shaped hanging plate 503 is fixed at the bottom of the second threaded sleeve 501, a reversing motor 504 is mounted on the first screw 203, the output end of the reversing motor 504 penetrates through the L-shaped hanging plate 503 and is fixedly provided with a connecting post 505, the lower end of the connecting post 505 is provided with a collecting camera 506, and the collecting camera 506 is provided with a downward inclined angle; the acquisition camera 506 can drive the acquisition camera to change the shooting visual angle through the reversing motor 504, so that the range of the whole collection assembly 5 for crop image data acquisition is increased, and different acquisition requirements are adapted;
further comprises: the adjusting module is used for acquiring crop planting data, establishing a crop growth model, controlling the collecting assembly 5 according to the model to collect crop and crop part image data in different areas, analyzing the collected image data, judging whether diseases and insect pests exist or not, and carrying out emergency treatment on the crops with the diseases and insect pests; the adjusting module comprises a modeling unit, a patrol unit, an analysis unit and a processing unit.
The modeling unit is used for establishing a model for the growth data of crops;
the whole collection assembly 5 is driven to move in multiple angles and directions by matching of the lifting structure 1, the transverse moving mechanism 2 and the longitudinal moving mechanism 3 in the identification system, and growth data of each row of crops are collected; specific crop growth data include: the planting arrangement and the number of rows of crops, the spacing of the crops in each row and each row, the height of the crops and the precise heights of the leaves, stems and roots of the crops;
after the obtained crop growth data are analyzed, the data are subjected to preliminary treatment, including cleaning, screening and conversion of the data, so that the data are convenient for the use of subsequent modeling;
the data of preliminary treatment is imported, and a model is built by the planting arrangement and the number of rows of crops, the spacing between each row and each row of crops, the height of the crops and the accurate height data of the leaves, the stems and the roots of the crops;
automatically detecting the established model, and automatically repairing the detected problems such as irregular surfaces, overlapping and holes to form a crop growth model;
the inspection unit is used for controlling the collection assembly 5 to collect image data of crops in different areas and crop parts according to the crop growth model established by the modeling unit;
the method comprises the steps of obtaining the interval and arrangement interval between each crop in modeling data, driving a collecting assembly 5 to transversely and longitudinally move through a transverse moving mechanism 2 and a longitudinal moving mechanism 3, and driving the whole collecting assembly 5 to adjust the height through the height of crops, the precise height of blades, stems and roots in the modeling data and a lifting structure 1, so that image data of the growth condition of the crops are obtained;
the analysis unit is used for receiving the image data of the crop growth condition obtained by the inspection unit and judging whether the crop has diseases and insect pests according to the data analysis;
retrieving and importing image data of normal states of crops and states of diseases and insect pests at different positions;
the image enhancement, denoising and cutting operation are carried out on the imported image data, so that the accuracy and the efficiency of subsequent processing are convenient;
training the preprocessed image data by using a convolutional neural network to improve the accuracy of recognition; the data training process carries out model selection and parameter adjustment through a cross validation method;
the trained model is applied to crop growth condition image data acquired by the inspection unit, and whether crop diseases and insect pests exist or not is judged;
when the crop diseases and insect pests do not exist, resetting the inspection time, and inspecting the crops again when the next period appears;
when the crop diseases and insect pests exist, analyzing the ratio of the disease and insect pests areas, and when the ratio of the disease and insect pests is smaller than a preset value, generating a single-point processing signaling, and simultaneously acquiring the positions of the disease and insect pests areas and sending the positions to a processing unit;
when the ratio of the pest and disease damage area is larger than a preset value, generating an area processing signaling, and simultaneously acquiring the position of the pest and disease damage area and sending the position to a processing unit;
the processing unit is used for receiving the single-point processing signaling and the area processing signaling and respectively spraying the medicines to the pest crops through different modes;
when single-point processing signaling is received, the pesticide spraying component 4 is controlled to switch to a single-point pesticide spraying mode to spray pesticides at a certain position in a targeted manner, the lifting structure 1, the transverse moving mechanism 2 and the longitudinal moving mechanism 3 are used for controlling the whole collecting component 5 to drive the unidirectional spray nozzle 413 to move to spray pesticides at different positions, and the image acquisition can be carried out on the pesticide spraying process of crops in real time by matching with the collecting component 5;
when receiving the region processing signaling, the pesticide spraying assembly 4 is controlled to switch to a single-row spraying mode to spray pesticides on crops in a certain row of crops, and the lifting structure 1, the transverse moving mechanism 2 and the longitudinal moving mechanism 3 are utilized to drive the whole pesticide spraying assembly 4 to move, so that the positions of the crops in different rows are changed to spray the pesticides.
The preferred embodiments of the invention disclosed above are intended only to assist in the explanation of the invention. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best understand and utilize the invention. The invention is limited only by the claims and the full scope and equivalents thereof.
Claims (7)
1. The crop pest identification method based on deep learning is characterized by comprising the following steps of:
step S1: firstly, a collection assembly (5) is driven to move in multiple directions through a lifting structure (1), a transverse moving mechanism (2) and a longitudinal moving mechanism (3), crop growth data in an area are collected, and a crop model is built according to the crop growth data;
step S2: controlling the lifting structure (1), the transverse moving mechanism (2) and the longitudinal moving mechanism (3) to drive the collecting assembly (5) to collect image data of different crop leaves, stems and roots according to the established crop model;
step S3: the collected crop image data is imported into a preset training model, whether the crop has diseases or insect pests is judged, and when the crop has no diseases or insect pests, the inspection time is reset;
step S4: when crop diseases and insect pests exist, acquiring crops with specific diseases and insect pests and positions of the crops, analyzing the area proportion of the diseases and insect pests and comparing the area proportion with a preset area proportion, and when the analyzed real-time area proportion is smaller than or larger than the preset area proportion, generating single-point processing signaling and area processing signaling respectively;
step S5: after receiving the single-point processing signaling or the regional processing signaling, the pesticide spraying component (4) is controlled to be switched to a single-point pesticide spraying mode or a single-row pesticide spraying mode respectively to spray pesticide to the region where crop diseases and insect pests occur.
2. Crop pest identification system based on deep learning, characterized by comprising:
the collection assembly (5) is used for collecting image data of crop growth;
the collecting assembly (5) comprises a second threaded sleeve (501) and a guide sleeve (502), the second threaded sleeve (501) and the guide sleeve (502) are connected to the longitudinal moving mechanism (3), the second threaded sleeve (501) is fixed with the guide sleeve (502), an L-shaped hanging plate (503) is fixed at the bottom of the second threaded sleeve (501), a reversing motor (504) is mounted on the first screw rod (203), the output end of the reversing motor (504) penetrates through the L-shaped hanging plate (503) and is fixedly provided with a connecting column (505), the lower end of the connecting column (505) is provided with a collecting camera (506), and the collecting camera (506) is provided with a downward inclined angle;
the pesticide spraying component (4) is used for spraying the pesticides with different modes on crops with diseases and insect pests;
the spraying component (4) comprises a medicine storage tank (401), the medicine storage tank (401) is arranged on the outer side of a transverse moving mechanism (2), pesticide liquid is arranged in the medicine storage tank (401), a booster pump (402) is arranged on the surface of the medicine storage tank (401), a connecting hose (403) is connected to an outlet of the booster pump (402), a fixing pipe (404) is connected to the other end of the connecting hose (403), a top of the vertical plate (301) is fixed with a fixing pipe (414), the fixing pipe (404) is arranged on the fixing pipe (414), an electric switching valve (405) is hung at the other end of the fixing pipe (414), the electric switching valve (405) is arranged above the second screw rod (304), the other end of the fixing pipe (404) is connected to an inlet of the electric switching valve (405), an extension pipe (409) is connected to an outlet of the electric switching valve (405), a convex plate (410) is fixed on the surface of the connecting column (403), a joint pipe 411 is arranged on the convex plate (410), the joint pipe (411) and the other end of the joint pipe (409) and the electric switching valve (405) are connected with the other end of the electric switching valve (405) through the electric switching pipe (405), the other ends of the two electric suspenders (406) are respectively fixed with a liquid discharge pipe (407), the other two outlets of the electric switching valve (405) are respectively communicated with the two liquid discharge pipes (407) through penetrating pipes, the two penetrating pipes respectively penetrate through the two electric suspenders (406), and the surfaces of the two liquid discharge pipes (407) are respectively provided with a plurality of liquid spraying heads (408) which are distributed at equal intervals;
the longitudinal moving mechanism (3) is used for driving the whole collecting assembly (5) to longitudinally move so as to collect influence data of crops in different longitudinal directions; the transverse moving mechanism (2) is used for driving the collecting assembly (5) to transversely move so as to collect image data of crops in different transverse directions; the lifting structure (1) is used for supporting the whole system and changing the collection height of the collection assembly (5) so as to adapt to the collection of different parts of crops;
the adjusting module is used for acquiring crop planting data, establishing a crop growth model, collecting crop and crop part image data in different areas according to the model control collecting assembly (5), analyzing the collected image data, judging whether diseases and insect pests exist or not, and carrying out emergency treatment on crops with the diseases and insect pests.
3. The deep learning based crop pest identification system of claim 2, wherein the adjustment module comprises:
the modeling unit is used for establishing a model for the growth data of crops;
the inspection unit is used for controlling the collection assembly (5) to collect image data of crops and crop parts in different areas according to the established crop growth model;
the analysis unit is used for receiving the image data of the crop growth condition obtained by the inspection unit and judging whether the crop has diseases and insect pests according to the trained data model;
the specific analysis process comprises the following steps:
retrieving and importing image data of normal states of crops and states of diseases and insect pests at different positions;
the image enhancement, denoising and cutting operation are carried out on the imported image data, so that the accuracy and the efficiency of subsequent processing are convenient;
training the preprocessed image data by using a convolutional neural network, and performing model selection and parameter adjustment in a data training process by a cross validation method;
the trained model is applied to crop growth condition image data acquired by the inspection unit, and whether crop diseases and insect pests exist or not is judged;
when the crop diseases and insect pests do not exist, resetting the inspection time, and inspecting the crops again when the next period appears;
when the crop diseases and insect pests exist, analyzing the ratio of the disease and insect pests areas, and when the ratio of the disease and insect pests is smaller than a preset value, generating a single-point processing signaling, and simultaneously acquiring the positions of the disease and insect pests areas and sending the positions to a processing unit;
when the ratio of the pest and disease damage area is larger than a preset value, generating an area processing signaling, and simultaneously acquiring the position of the pest and disease damage area and sending the position to a processing unit;
and the processing unit is used for receiving the single-point processing signaling and the regional processing signaling and respectively spraying the medicines to the pest crops through different modes.
4. The crop pest identification system based on deep learning of claim 2, wherein the specific process of emergency treatment of crops by the processing unit comprises the following steps:
when single-point processing signaling is received, the pesticide spraying component (4) is controlled to switch to a single-point pesticide spraying mode to spray pesticides at a certain position in a targeted manner, the lifting structure (1), the transverse moving mechanism (2) and the longitudinal moving mechanism (3) are used for controlling the whole collecting component (5) to drive the unidirectional spray head (413) to move to spray pesticides at different positions, and the collecting component (5) is matched to collect images of the pesticide spraying process of crops in real time;
when receiving the regional processing signaling, the pesticide spraying assembly (4) is controlled to be switched to a single-row spraying mode to spray pesticides on crops in a certain row of crops, and the lifting structure (1), the transverse moving mechanism (2) and the longitudinal moving mechanism (3) are utilized to drive the whole pesticide spraying assembly (4) to move, so that the positions of the crops in different rows are changed to spray the pesticides.
5. The crop pest and disease damage identification system based on deep learning according to claim 2, wherein the longitudinal moving mechanism (3) comprises two vertical plates (301), the two vertical plates (301) are respectively connected to the transverse moving mechanism (2), the side walls of the two vertical plates (301) are respectively fixed with an extension plate (302), a second screw rod (304) is rotatably connected between the two vertical plates (301), the second screw rod (304) is in threaded connection with the second threaded sleeve (501), a guide rod (305) is fixed between the two extension plates (302), the guide sleeve (502) is slidably connected to the surface of the guide rod (305), a second motor (303) is installed on the side wall of the vertical plate (301), and the output end of the second motor (303) penetrates through the vertical plate (301) and is fixed with the second screw rod (304).
6. The crop pest and disease damage identification system based on deep learning according to claim 2, wherein the lateral movement mechanism (2) comprises two first threaded sleeves (204), the two first threaded sleeves (204) are respectively fixed at the bottoms of the two vertical plates (301), the two concave supporting plates (201) are connected to the lifting structure (1), the two concave supporting plates (201) are internally and respectively connected with a first screw rod (203) in a rotating manner, the first screw rods (203) are in threaded connection with the first threaded sleeves (204), the side walls of the two concave supporting plates (201) are respectively provided with a first motor (202), and the output ends of the first motors (202) penetrate through the concave supporting plates (201) and are fixed with the first screw rods (203).
7. The crop pest and disease damage identification system based on deep learning according to claim 2, wherein the lifting structure (1) comprises two hydraulic cylinders (101), the two hydraulic cylinders (101) are respectively connected to the bottoms of the two concave supporting plates (201), the concave supporting plates (201) are connected to the extending ends of the hydraulic cylinders (101), the bottoms of the two hydraulic cylinders (101) are fixedly provided with ground inserts (102), the surfaces of the two hydraulic cylinders (101) are fixedly provided with a plurality of equally distributed balance plates (103), and the bottoms of the plurality of balance plates (103) are fixedly provided with grounding pieces (104).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310760659.6A CN116806791B (en) | 2023-06-27 | 2023-06-27 | Crop pest identification method and system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310760659.6A CN116806791B (en) | 2023-06-27 | 2023-06-27 | Crop pest identification method and system based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116806791A true CN116806791A (en) | 2023-09-29 |
| CN116806791B CN116806791B (en) | 2024-02-02 |
Family
ID=88142354
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310760659.6A Active CN116806791B (en) | 2023-06-27 | 2023-06-27 | Crop pest identification method and system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116806791B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| KR20180130386A (en) * | 2017-05-29 | 2018-12-07 | 주식회사 대길 | Pesticide spraying apparatus for green house facility |
| KR20190112540A (en) * | 2018-03-26 | 2019-10-07 | 주식회사 더블유피 | Agricultural drone system using individual nozzle control for efficient pesticide application |
| CN113100210A (en) * | 2021-04-25 | 2021-07-13 | 山西农业大学 | Pesticide atomization spraying unmanned aerial vehicle and control method thereof |
| CN113155192A (en) * | 2021-04-19 | 2021-07-23 | 河南省农业科学院农业经济与信息研究所 | Intelligent crop table type real-time acquisition device |
| CN115024300A (en) * | 2022-06-30 | 2022-09-09 | 昆明理工大学 | Intelligent pesticide spraying device, method and system of panax notoginseng plant protection unmanned aerial vehicle for preventing and controlling diseases and pests |
| CN115956549A (en) * | 2022-08-23 | 2023-04-14 | 天津农学院 | Automatic medicine spraying robot based on machine vision |
-
2023
- 2023-06-27 CN CN202310760659.6A patent/CN116806791B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| KR20180130386A (en) * | 2017-05-29 | 2018-12-07 | 주식회사 대길 | Pesticide spraying apparatus for green house facility |
| KR20190112540A (en) * | 2018-03-26 | 2019-10-07 | 주식회사 더블유피 | Agricultural drone system using individual nozzle control for efficient pesticide application |
| CN113155192A (en) * | 2021-04-19 | 2021-07-23 | 河南省农业科学院农业经济与信息研究所 | Intelligent crop table type real-time acquisition device |
| CN113100210A (en) * | 2021-04-25 | 2021-07-13 | 山西农业大学 | Pesticide atomization spraying unmanned aerial vehicle and control method thereof |
| CN115024300A (en) * | 2022-06-30 | 2022-09-09 | 昆明理工大学 | Intelligent pesticide spraying device, method and system of panax notoginseng plant protection unmanned aerial vehicle for preventing and controlling diseases and pests |
| CN115956549A (en) * | 2022-08-23 | 2023-04-14 | 天津农学院 | Automatic medicine spraying robot based on machine vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116806791B (en) | 2024-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110235882B (en) | Accurate variable fruit tree pesticide application robot based on multiple sensors | |
| CN103999841B (en) | A kind of automatic target detection spraying system | |
| CN104115809B (en) | A kind of field crops sprays machine and spraying method to target | |
| CN105230224A (en) | Intelligent weeding machine and weed removing method thereof | |
| CN104186451A (en) | Insect-killing weeding pesticide spraying robot based on machine vision | |
| CN204104581U (en) | A kind of automatic target detection spraying system based on binocular vision technology | |
| CN207284200U (en) | Intelligent translational formula water-fertilizer-pesticide integration sprinkling irrigation equipment | |
| CN111587872A (en) | Robot for spraying pesticide | |
| CN116806791B (en) | Crop pest identification method and system based on deep learning | |
| CN106171080A (en) | A kind of booth vegetable intelligent cultivation robot | |
| CN109005693B (en) | Open-close type intelligent weeding device based on binocular vision for obtaining seedling and grass information | |
| DE60028232T3 (en) | METHOD AND DEVICE FOR MILKING ANIMALS | |
| CN110999680A (en) | Green house device with automatic irrigation system | |
| CN204964213U (en) | Vertical type spraying volume spray deposition uniformity detector | |
| CN204090848U (en) | A kind of deinsectization weeding spraying machine device people based on machine vision | |
| CN107027550A (en) | A kind of multi-functional plant protection device and its application method based on growth cabinet | |
| CN205082548U (en) | Intelligence weeder | |
| CN105145527A (en) | Multifunctional mulberry tree field managing system | |
| CN210900625U (en) | Drip irrigation device for greenhouse planting | |
| CN115956549A (en) | Automatic medicine spraying robot based on machine vision | |
| CN214339604U (en) | Wa mandarin orange is planted and is used pesticide spraying apparatus | |
| CN110262364B (en) | Agricultural planting environment monitoring device based on internet | |
| Tóth et al. | Automatic cluster remover with and without follow-up milking | |
| CN205082547U (en) | Inrow weeding device of intelligence weeder | |
| CN111357520B (en) | Cutting mechanism for topping crops, topping mechanism and topping vehicle |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |