CN113100207B - Accurate formula pesticide applying robot system based on wheat disease information and pesticide applying method - Google Patents
Accurate formula pesticide applying robot system based on wheat disease information and pesticide applying method Download PDFInfo
- Publication number
- CN113100207B CN113100207B CN202110400904.3A CN202110400904A CN113100207B CN 113100207 B CN113100207 B CN 113100207B CN 202110400904 A CN202110400904 A CN 202110400904A CN 113100207 B CN113100207 B CN 113100207B
- Authority
- CN
- China
- Prior art keywords
- image
- disease
- wheat
- pesticide
- crop
- 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.)
- Active
Links
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Environmental Sciences (AREA)
- Zoology (AREA)
- Wood Science & Technology (AREA)
- Pest Control & Pesticides (AREA)
- Insects & Arthropods (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Mechanical Engineering (AREA)
- Image Processing (AREA)
Abstract
The invention provides a precise formula pesticide applying robot system and a pesticide applying method based on wheat disease information, which are used for solving the technical problems of pesticide waste and environmental pollution caused by poor accuracy of the existing automatic crop pesticide spraying technology. The system includes industry camera, microspur camera, PC, prescription pond, accurate controller and the shower nozzle of giving medicine to the poor free of charge, the PC is connected with industry camera, microspur camera, prescription pond and accurate controller of giving medicine to the poor free of charge respectively, and the prescription pond is connected with accurate controller of giving medicine to the poor free of charge, and accurate controller of giving medicine to the poor free of charge is connected with the shower nozzle of giving medicine to the poor free of charge. The invention has certain advantages in the aspects of disease type identification and formula preparation, has good effect in the aspect of identifying wheat disease type, has stronger advantages in the aspect of giving a formula based on the disease severity, and has stronger adaptability and higher identification accuracy.
Description
Technical Field
The invention relates to the technical field of agricultural engineering, in particular to a precise formula pesticide application robot system and a pesticide application method based on wheat disease information.
Background
In recent years, with the development of computer technology, precise drug delivery robot systems have gained more and more attention. The accurate pesticide application in agricultural production by using computer image processing and computer vision technology is also gradually developed. The digital image processing algorithm has great advantages in the aspects of automatic alignment and automatic identification, and is the basis and key technology of modern precision agriculture. According to the requirement of accurate pesticide application, the identification accuracy of the crop row center line can influence the alignment accuracy of a pesticide spraying nozzle, so that pesticide waste or environmental pollution is easily caused; the accuracy of disease type identification and the disease severity can have great influence on the effect of accurate pesticide application, and the pesticide can be applied according to symptoms only by accurately judging the disease type, so that pesticide waste can be reduced by determining the disease severity, and quantitative pesticide application is carried out. At present, no universal accurate pesticide application algorithm which has adaptability completely meeting the requirements exists at home and abroad, and no large-scale accurate pesticide application machine or accurate pesticide application robot which can realize the functions is available.
Disclosure of Invention
Aiming at the defects in the background technology, the invention provides a precise formula pesticide application robot system and a pesticide application method based on wheat disease information, and solves the technical problems of pesticide waste and environmental pollution caused by poor accuracy of the existing automatic pesticide spraying technology for crops.
The technical scheme of the invention is realized as follows:
the utility model provides an accurate prescription robot system that gives medicine to poor free of charge based on wheat disease information, includes industry camera, microspur camera, PC, prescription pond, accurate controller and the shower nozzle of giving medicine to poor free of charge, the PC is connected with industry camera, microspur camera, prescription pond and accurate controller of giving medicine to poor free of charge respectively, and the prescription pond is connected with accurate controller of giving medicine to poor free of charge, and accurate controller of giving medicine to poor free of charge is connected with the shower nozzle of giving medicine to poor free of charge.
Preferably, the PC comprises a disease recognition system, a crop row recognition system and a formula expert system, wherein the disease recognition system is respectively connected with the macro camera and the formula expert system, the formula expert system is connected with the formula pool, and the crop row recognition system is respectively connected with the industrial camera and the accurate pesticide application controller.
A pesticide application method of a precise formula pesticide application robot system based on wheat disease information comprises the following steps:
s1, acquiring an image of a wheat crop by using an industrial camera, and transmitting the image to a crop row recognition system of a PC (personal computer) to obtain deviation information of a pesticide application spray head and a crop row;
s2, acquiring a leaf disease image of the wheat crop by using a macro camera, and transmitting the leaf disease image to a disease identification system of a PC (personal computer) to obtain the disease type and the disease degree;
s3, transmitting the disease types and the disease degrees in the step S2 to a formula expert system, and preparing pesticide types and pesticide spraying amounts in a formula pool through the formula expert system to obtain a pesticide solution;
and S4, transmitting deviation information of the pesticide applying spray head and the crop row in the step S1 to the accurate pesticide applying controller, and controlling the position of the pesticide applying spray head by the accurate pesticide applying controller to spray the pesticide solution in the step S3 onto the sick wheat crop to realize accurate pesticide applying of the wheat crop.
Preferably, the method for transmitting the image to the crop row recognition system of the PC in step S1 to obtain the deviation information between the pesticide application nozzle and the crop row is as follows:
s11, preprocessing the image of the wheat crop by using an image processing method to obtain a binary image;
s12, removing pinhole noise between lines and on crop lines in the binary image by using a morphological algorithm to obtain a crop line skeleton;
s13, fitting the crop row skeleton by using a rapid Hough transform algorithm based on directional processing to obtain the center line of the crop row;
and S14, calculating the pixel deviation distance between the center line of the crop row and the center of the image according to the transformation relation between the world coordinate system and the image coordinate system to obtain the actual geographic deviation corresponding to the image pixel deviation, and combining the position information of the pesticide application spray head and the industrial camera to obtain the deviation information of the pesticide application spray head and the crop row.
Preferably, the method for preprocessing the image of the wheat crop by using the image processing method to obtain the binary image comprises the following steps:
s11.1, graying the image of the wheat crop by utilizing an improved over-green graying algorithm to obtain a grayscale image;
the grayscale image is:
wherein Gray represents a Gray value of the image, G represents a green component of the image, R represents a red component of the image, and B represents a blue component of the image;
s11.2, carrying out binary filtering processing on the gray level image, and removing noise in the gray level image to obtain a filtered image;
s11.3, carrying out binarization on the filtered image by adopting a maximum inter-class variance method to obtain a binary image.
Preferably, in the step S2, the method of transmitting the leaf defect image to a defect recognition system of a PC to obtain the type and degree of the defect includes:
s21, separating leaves from a background of the leaf disease image by using an image processing method to obtain a leaf disease image;
s22, segmenting the leaf lesion image by adopting a green segmentation algorithm to obtain a lesion segmentation image, and performing image restoration on the lesion segmentation image to obtain a lesion image, wherein the image restoration refers to restoring the transformed gray level image into a color image according to image coordinate position information;
s23, respectively extracting features of the scab image by using a color histogram, a color moment, a contrast ratio, a moment of invariance and an eccentricity ratio calculation method to obtain initial standby features, and screening the initial standby features by using a principal component analysis method to obtain core features for identifying the wheat leaf diseases, wherein the core features comprise a red moment, R-G, a circularity, a contrast ratio, an eccentricity ratio and a scab area number ratio;
and S24, identifying the core characteristics by using an SVM algorithm based on the Fisher criterion, and outputting the type and severity of the disease.
Preferably, the method for separating the leaf from the background of the leaf disease image by using the image processing method to obtain the leaf disease image comprises the following steps:
s21.1, converting the leaf disease image into a gray image, and processing the gray image by using an improved watershed algorithm to obtain a secondary gray image;
s21.2, segmenting the secondary gray image by using a self-adaptive threshold algorithm to obtain a binary image of the blade;
s21.3, cutting the two-value image of the blade subjected to the expansion corrosion filtering to obtain a target image;
and S21.4, converting the target image into a color image according to a spatial relationship according to the invariance principle of image spatial pixels to obtain a leaf disease image, wherein the spatial relationship refers to the correspondence of the coordinate positions of the binary image and the color image.
Preferably, the method for identifying the core features by using the SVM algorithm based on the fisher criterion and outputting the types and severity of the diseases comprises the following steps:
s24.1, constructing disease category space M = V of wheat leaves 1 ∪V 2 ∪…∪V i ∪…∪V N Wherein V is i Indicating disease of type i;
s24.2, dividing the disease category space M into two categories of subspaces, and constructing a linear discriminant function by adopting a Fisher criterion:
wherein, X = (X) 1 ,x 2 ,…,x p ) T I =1,2, \ 8230for core features, p, p represents the number of core features, C is a constant vector i Representing the contribution rate of different core features, and U (X) represents a discriminant function;
s24.3, judging the types of the diseases according to the obtained core characteristics of the wheat leaf diseases by a multi-type recognition algorithm of a support vector machine decision tree based on Fisher criterion; and determining the severity grade of the disease according to the number of the core characteristics of the wheat leaf diseases.
The beneficial effect that this technical scheme can produce: the invention provides a realization method on the configuration of software and hardware, provides a method for automatic alignment of a spraying nozzle, automatic selection of pesticide types and control of spraying amount, has certain advantages in the aspects of disease type identification and formula preparation, has good effect on the aspect of identifying wheat disease types, has strong advantages in the aspect of giving formulas based on disease severity, and has strong adaptability and high identification accuracy.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a block diagram of the system of the present invention;
FIG. 2 is a flow chart of the present invention;
FIG. 3 is an original view of a wheat field according to the present invention;
FIG. 4 is a line center line extraction plot of a wheat crop of the present invention;
FIG. 5 is an original drawing of wheat stripe rust of the present invention;
FIG. 6 is a plot of wheat stripe rust treatment according to the present invention;
FIG. 7 is an original drawing of wheat leaf rust of the present invention;
FIG. 8 is an original drawing of wheat leaf rust with a complex background;
FIG. 9 is a graph showing the result of wheat powdery mildew identification according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art based on the embodiments of the present invention without inventive step, are within the scope of the present invention.
Embodiment 2, as shown in fig. 2, a method for applying a pesticide to a precise formulation pesticide application robot system based on wheat disease information includes the following specific steps:
s1, collecting an RGB color image of a wheat crop by using an industrial camera MV-VD030SC, wherein the image comprises a plurality of crop line information as shown in figure 3, and transmitting the image to a crop line identification system of a PC (personal computer) to obtain deviation information of a pesticide application spray head and a crop line; the method comprises the following steps:
s11, preprocessing the image of the wheat crop by using an image processing method to obtain a binary image; the specific method comprises the following steps:
s11.1, graying the image of the wheat crop by utilizing an improved over-green graying algorithm to obtain a grayscale image;
the grayscale image is:
wherein Gray represents a Gray value of the image, G represents a green component of the image, R represents a red component of the image, and B represents a blue component of the image;
s11.2, carrying out binary filtering processing on the gray level image, and removing noise in the gray level image to obtain a filtered image;
s11.3, carrying out binarization on the filtered image by adopting a maximum inter-class variance method to obtain a binary image. At the moment, only crop row information exists in the binary image, the background is removed, and background segmentation is achieved.
S12, removing pinhole noise between lines and on crop lines in the binary image by using a morphological algorithm, then refining the pinhole noise, removing redundant lines in the image on the premise of considering information quantity and accuracy in order to reduce calculated quantity, and only reserving two lines closest to the center to represent a crop line skeleton, wherein the obtained crop line skeleton is shown in figure 4.
S13, in order to better reflect the trend of the crop row, fitting the crop row skeleton into a straight line by utilizing a rapid Hough transformation algorithm based on directional processing, and taking the straight line as the center line of the crop row;
the principle of the fast Hough transformation algorithm based on directional processing is as follows:
the Hough transform fits crop rows into straight lines, and the basic idea is to convert the straight line detection problem in an image into the detection problem of the midpoint in the parameter space by utilizing the duality of point lines, and complete the straight line detection through the accumulation statistics of the midpoint in the parameter space. There are two methods for improving the speed of Hough transformation, namely reducing the number of participating points and reducing the calculation space. The Hough transform based on directional processing of the present invention is based on a combination of these two methods. In order to reduce the complexity of the calculation, the linear polar coordinate equation is expressed as follows:
ρ＝xcosθ+ysinθ；
where ρ is the distance from the origin to the line and θ is the angle between the x-axis and the line. When calculating, firstly, the whole image is quantized into several subregions, and two points (x) are randomly selected on the near-line segment in the thinned image m ,y m )、(x n ,y n ) A straight line is determined, and the parameters are as follows:
ρ＝x m cosθ+y m sinθ；
and (3) forming all pixel points in each section of the near straight line into a set A, randomly taking two points in the set A to calculate the slope, and if the requirement of the near straight line is not met, reselecting. Because only the middle 2 lines of the crop line image are extracted, the value range of theta can be compressed to 30-150 degrees according to the method
The value of p is calculated, wherein,
and finally, accumulating the accumulated array according to the values of the rho and the theta to obtain the number of collinear points, and determining the size of the accumulator according to the value ranges of the rho and the theta and the resolution ratios of the rho and the theta so as to detect the straight line.
S14, according to the transformation relation between the world coordinate system and the image coordinate system, calculating the pixel deviation distance between the center line of the crop row and the center of the image to obtain the actual geographic deviation corresponding to the image pixel deviation, combining the position information of the pesticide application spray head and the industrial camera to obtain the deviation information of the pesticide application spray head and the crop row, and sending the deviation information to the accurate pesticide application controller through the PC.
S2, acquiring a leaf disease image of the wheat crop by using a macro camera (the model is a Nikon D80 type color digital camera), and transmitting the leaf disease image to a disease identification system of a PC (personal computer) to obtain the disease type and the disease degree; the method comprises the following steps:
s21, separating the leaves from the background of the leaf disease image by using an image processing method to obtain a leaf disease image, as shown in figures 5 and 7; the method comprises the following steps:
s21.1, when the diseased leaves are separated from the complex background, converting the leaf disease image into a gray image, and processing the gray image by using an improved watershed algorithm to obtain a secondary gray image;
s21.2, segmenting the secondary gray image by using a self-adaptive threshold algorithm to obtain a binary image of the blade;
s21.3, cutting the two-value image of the blade subjected to the expansion corrosion filtering to obtain a target image;
and S21.4, converting the target image into a color image according to a spatial relationship according to the invariance principle of image spatial pixels to obtain a leaf disease image, wherein the spatial relationship refers to the correspondence of the coordinate positions of the binary image and the color image.
S22, segmenting the leaf lesion image by adopting a green segmentation algorithm to obtain a lesion segmentation image, and performing image restoration on the lesion segmentation image to obtain a lesion segmentation image with white lesions and black background, as shown in FIGS. 6 and 8, wherein the image restoration refers to restoring the transformed gray-scale image into a color image according to the image coordinate position information;
s23, comprehensively considering different properties of different wheat diseases in various aspects of the image, defining proper characteristics, and obtaining a characteristic set capable of correctly classifying the wheat diseases in the research. And extracting initial standby characteristics for identifying various wheat leaf diseases. Obtaining core characteristics for identifying various wheat leaf diseases by a characteristic selection method. Respectively extracting features of the scab image by using calculation methods of a color histogram, a color moment, a contrast, a moment of invariance and an eccentricity ratio to obtain initial standby features, and screening the initial standby features by using a principal component analysis method to obtain core features for identifying wheat leaf diseases, wherein the core features comprise a red moment, R-G, a circularity, a contrast, an eccentricity ratio and a scab area number ratio;
s24, designing a multiclass recognition algorithm of a support vector machine decision tree based on Fisher criterion, and judging the types of diseases according to the obtained core characteristics of the wheat leaf diseases; and determining the severity grade of the disease according to the number of the core characteristics of the wheat leaf diseases. And sending the information of the type and the severity degree to the precise pesticide application controller by the PC. The core features are identified by using an SVM algorithm based on the Fisher criterion, and the types and severity of diseases are output, as shown in FIG. 9. The SVM algorithm based on the Fisher discriminant method is as follows:
the system sets of core feature combination data can be expressed as:
L N ＝{X i |i＝1,2,…,N}；
in the formula, L N Representing the core feature composition space, vector X i Combining vectors for the core features;
X i ＝(x 1 ,x 2 ,…,x p ) i T ；
wherein the component x k (k =1,2, \8230;, p) represent different core features.
S24.1, dividing the actual disease category space M into N categories, and constructing the disease category space M = V of the wheat leaves 1 ∪V 2 ∪…∪V i ∪…∪V N Wherein V is i Indicating disease of type i; the disease identification is to find that V is satisfied 1 Vector X of j Characteristic value of (c):
s24.2, firstly, dividing the disease category space into subspaces which are only divided into two categories by adopting a decision tree method based on a support vector machine, and obtaining a plurality of subspaces combined by the two categories when identifying the categories. And identifying the subspace of the two types of spatial combinations by adopting the idea of fee pause criterion. Dividing the disease category space M into two categories of subspaces, and constructing a linear discriminant function by adopting a Fisher criterion:
wherein, X = (X) 1 ,x 2 ,…,x p ) T I =1,2, \ 8230for core features, p, p represents the number of core features, C is a constant vector i Representing the contribution rate of different core features, and U (X) represents a discriminant function; under the specified conditions, a characteristic value U is determined 0 When the value is a threshold value, the space is judged as a category 1 space when the value is smaller than the threshold value, and the space is judged as a category 2 space when the value is larger than the threshold value. Fischer discrimination methodUnder the optimal standard condition, the linear discriminant function obtained by only using the knowledge of the subset I and the subset II is more accurate than a method based on the Euclidean distance and the Mahalanobis distance.
S24.3, judging the types of the diseases according to the obtained wheat leaf disease core characteristics by a multi-type recognition algorithm of a decision tree of a support vector machine based on Fisher' S criterion; and determining the severity grade of the disease according to the number of the core characteristics of the wheat leaf diseases.
And by analogy, a category 1 space, a category 2 space, a category 3 space and the like are obtained respectively.
S3, transmitting the disease types and the disease degrees in the step S2 to a formula expert system, and preparing pesticide types and pesticide spraying amounts in a formula pool through the formula expert system to obtain a pesticide solution;
the formula expert system firstly obtains relevant formulas which comprise information such as required pesticide types and the dosage of each pesticide. And extracting a certain amount of pesticides from different pesticide containers in a formula pool respectively, and then mixing the pesticides by stirring to obtain a pesticide solution.
S4, transmitting deviation information of the pesticide applying spray head and the crop row in the step S1 to the accurate pesticide applying controller, and controlling the position of the pesticide applying spray head by the accurate pesticide applying controller to spray the pesticide solution in the step S3 onto the ill-conditioned wheat crop, so that accurate pesticide applying of the wheat crop is realized, pesticide waste is reduced, and pesticide applying efficiency and accuracy are improved.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and should not be taken as limiting the scope of the present invention, which is intended to cover any modifications, equivalents, improvements, etc. within the spirit and scope of the present invention.
Claims (4)
1. An accurate formula pesticide application robot system based on wheat disease information is characterized by comprising an industrial camera, a microspur camera, a PC (personal computer), a formula pool, an accurate pesticide application controller and a pesticide application spray head, wherein the PC is respectively connected with the industrial camera, the microspur camera, the formula pool and the accurate pesticide application controller; the PC comprises a disease identification system, a crop row identification system and a formula expert system, wherein the disease identification system is respectively connected with the microspur camera and the formula expert system, the formula expert system is connected with the formula pool, and the crop row identification system is respectively connected with the industrial camera and the accurate pesticide application controller;
the application method of the accurate formula application robot system based on the wheat disease information comprises the following steps:
s1, collecting an image of a wheat crop by using an industrial camera, and transmitting the image to a crop row recognition system of a PC (personal computer) to obtain deviation information of a pesticide application spray head and a crop row;
s2, acquiring a leaf disease image of the wheat crop by using a macro camera, and transmitting the leaf disease image to a disease identification system of a PC (personal computer) to obtain the disease type and the disease degree;
in step S2, the specific method is:
s21, separating leaves from a background of the leaf disease image by using an image processing method to obtain a leaf disease image;
s22, segmenting the leaf lesion image by adopting a green segmentation algorithm to obtain a lesion segmentation image, and carrying out image restoration on the lesion segmentation image to obtain a lesion image, wherein the image restoration refers to restoring the transformed gray level image into a color image according to image coordinate position information;
s23, respectively extracting features of the scab image by using a color histogram, a color moment, a contrast ratio, a moment of invariance and an eccentricity ratio calculation method to obtain initial standby features, and screening the initial standby features by using a principal component analysis method to obtain core features for identifying the wheat leaf diseases, wherein the core features comprise a red moment, R-G, a circularity, a contrast ratio, an eccentricity ratio and a scab area number ratio;
s24, identifying the core characteristics by using an SVM algorithm based on the Fisher criterion, and outputting the type and severity of the disease;
the implementation method of the step S24 is as follows:
s24.1, constructing disease category space M = V of wheat leaves 1 ∪V 2 ∪…∪V i ∪…∪V N Wherein, V i Indicating the i-th disease;
s24.2, dividing the disease category space M into two categories of subspaces, and constructing a linear discriminant function by adopting a Fisher criterion:
wherein, X = (X) 1 ,x 2 ,…,x p ) T I =1,2, \8230forcore features, p, p represents the number of core features, C is a constant vector, C i Representing the contribution rate of different core features, and U (X) represents a discriminant function;
s24.3, judging the types of the diseases according to the obtained wheat leaf disease core characteristics by a multi-type recognition algorithm of a decision tree of a support vector machine based on Fisher' S criterion; determining the severity level of the disease according to the number of the core characteristics of the disease of the wheat leaves;
s3, transmitting the disease types and the disease degrees in the step S2 to a formula expert system, and preparing a pesticide solution in a formula pool after the pesticide types and the pesticide spraying amount are given by the formula expert system;
and S4, transmitting deviation information of the pesticide applying spray head and the crop row in the step S1 to the accurate pesticide applying controller, and controlling the position of the pesticide applying spray head by the accurate pesticide applying controller to spray the pesticide solution in the step S3 onto the sick wheat crop to realize accurate pesticide applying of the wheat crop.
2. The wheat disease information-based precise formula pesticide application robot system according to claim 1, wherein the method for transmitting the image to the crop row recognition system of the PC in the step S1 to obtain the deviation information of the pesticide application spray head and the crop row comprises the following steps:
s11, preprocessing an image of the wheat crop by using an image processing method to obtain a binary image;
s12, removing pinhole noise between lines and on crop lines in the binary image by using a morphological algorithm to obtain a crop line skeleton;
s13, fitting the crop row skeleton by using a rapid Hough transform algorithm based on directional processing to obtain the center line of the crop row;
s14, calculating the pixel deviation distance between the center line of the crop row and the center of the image according to the transformation relation between the world coordinate system and the image coordinate system to obtain the actual geographic deviation corresponding to the image pixel deviation, and combining the position information of the pesticide application spray head and the industrial camera to obtain the deviation information of the pesticide application spray head and the crop row.
3. The wheat disease information-based precise formula pesticide application robot system according to claim 2, wherein the method for preprocessing the image of the wheat crop by using an image processing method to obtain a binary image comprises the following steps:
s11.1, graying the image of the wheat crop by utilizing an improved over-green graying algorithm to obtain a grayscale image;
the grayscale image is:
wherein Gray represents a Gray value of the image, G represents a green component of the image, R represents a red component of the image, and B represents a blue component of the image;
s11.2, carrying out binary filtering processing on the gray level image, and removing noise in the gray level image to obtain a filtered image;
s11.3, carrying out binarization on the filtered image by adopting a maximum inter-class variance method to obtain a binary image.
4. The wheat disease information-based precise formula pesticide application robot system according to claim 1, wherein the leaf and background separation is performed on the leaf disease image by using an image processing method, and the method for obtaining the leaf disease image comprises the following steps:
s21.1, converting the leaf disease image into a gray image, and processing the gray image by using an improved watershed algorithm to obtain a secondary gray image;
s21.2, segmenting the secondary gray image by using a self-adaptive threshold algorithm to obtain a binary image of the blade;
s21.3, cutting the two-value image of the blade subjected to the expansion corrosion filtering to obtain a target image;
and S21.4, converting the target image into a color image according to a spatial relationship according to the invariance principle of image spatial pixels to obtain a leaf disease image, wherein the spatial relationship refers to the correspondence of the coordinate positions of the binary image and the color image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110400904.3A CN113100207B (en) | 2021-04-14 | 2021-04-14 | Accurate formula pesticide applying robot system based on wheat disease information and pesticide applying method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110400904.3A CN113100207B (en) | 2021-04-14 | 2021-04-14 | Accurate formula pesticide applying robot system based on wheat disease information and pesticide applying method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113100207A CN113100207A (en) | 2021-07-13 |
| CN113100207B true CN113100207B (en) | 2022-11-22 |
Family
ID=76717597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110400904.3A Active CN113100207B (en) | 2021-04-14 | 2021-04-14 | Accurate formula pesticide applying robot system based on wheat disease information and pesticide applying method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113100207B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114514914A (en) * | 2021-12-28 | 2022-05-20 | 中国农业大学 | Intelligent sensing fertilization and pesticide spraying method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR349054A (en) * | 1904-03-04 | 1905-05-10 | Victor Vermorel | Multi-jet, high flow man-backpack sprayer |
| CN101927220A (en) * | 2010-07-05 | 2010-12-29 | 中国农业大学 | Accurate intelligent targeted spraying machine |
| CN102613161A (en) * | 2012-04-06 | 2012-08-01 | 山东农业大学 | Control system for boom sprayer and spraying compensation control method |
| CN105173085A (en) * | 2015-09-18 | 2015-12-23 | 山东农业大学 | Automatic control system and method for variable pesticide spraying of unmanned aerial vehicle |
| CN109964905A (en) * | 2019-03-19 | 2019-07-05 | 安徽农业大学 | Robot and its control method are administered to target based on walking certainly for fruit tree identification positioning |
| CN111627031A (en) * | 2020-05-29 | 2020-09-04 | 郑州轻工业大学 | Tile-house-shaped polygon-based crop root system phenotype analysis device and method |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004161724A (en) * | 2002-11-14 | 2004-06-10 | Masaru Kawai | Pest controlling liquid for plant culture and method for producing the same |
| CN102907406B (en) * | 2012-09-20 | 2014-11-19 | 北京林业大学 | Accurately targeted drug applying device and method for fruit tree rootstock |
| CN103530643A (en) * | 2013-10-11 | 2014-01-22 | 中国科学院合肥物质科学研究院 | Pesticide positioned spraying method and system on basis of crop interline automatic identification technology |
| CN103891697B (en) * | 2014-03-28 | 2015-08-12 | 南通职业大学 | The variable spray method of a kind of indoor autonomous spraying machine device people |
| US10269107B2 (en) * | 2017-02-23 | 2019-04-23 | Global Neighbor Inc | Selective plant detection and treatment using green luminance photometric machine vision scan with real time chromaticity operations and image parameter floors for low processing load |
| US11627736B2 (en) * | 2018-02-15 | 2023-04-18 | Monsanto Technology, Llc | Management of corn through semi-dwarf systems |
| CN108684282B (en) * | 2018-04-11 | 2023-07-07 | 北京麦飞科技有限公司 | Agricultural inspection and threshing integrated machine system and spraying control method |
| CN110089297B (en) * | 2019-05-18 | 2021-11-26 | 安徽大学 | Method and device for diagnosing severity of disease condition of wheat scab in field environment |
| CN110150260A (en) * | 2019-06-11 | 2019-08-23 | 东北农业大学 | A kind of Intelligent target spray weed-eradicating robot based on deep learning |
| CN110235882B (en) * | 2019-06-28 | 2021-09-14 | 南京农业大学 | Accurate variable fruit tree pesticide application robot based on multiple sensors |
| CN110333737A (en) * | 2019-07-05 | 2019-10-15 | 湖北理工学院 | A kind of high unmanned aerial vehicle control system of spraying efficiency for pesticide spraying and method |
| CN110692352B (en) * | 2019-09-19 | 2021-12-07 | 北京农业智能装备技术研究中心 | Intelligent agricultural robot and control method thereof |
| CN112136381A (en) * | 2020-09-29 | 2020-12-29 | 郑州轻工业大学 | Weeding robot for corn field and control system of weeding robot |
-
2021
- 2021-04-14 CN CN202110400904.3A patent/CN113100207B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR349054A (en) * | 1904-03-04 | 1905-05-10 | Victor Vermorel | Multi-jet, high flow man-backpack sprayer |
| CN101927220A (en) * | 2010-07-05 | 2010-12-29 | 中国农业大学 | Accurate intelligent targeted spraying machine |
| CN102613161A (en) * | 2012-04-06 | 2012-08-01 | 山东农业大学 | Control system for boom sprayer and spraying compensation control method |
| CN105173085A (en) * | 2015-09-18 | 2015-12-23 | 山东农业大学 | Automatic control system and method for variable pesticide spraying of unmanned aerial vehicle |
| CN109964905A (en) * | 2019-03-19 | 2019-07-05 | 安徽农业大学 | Robot and its control method are administered to target based on walking certainly for fruit tree identification positioning |
| CN111627031A (en) * | 2020-05-29 | 2020-09-04 | 郑州轻工业大学 | Tile-house-shaped polygon-based crop root system phenotype analysis device and method |
Non-Patent Citations (2)
| Title |
|---|
| 基于LS-SVM的浮选药剂量优化设定;张新林;《计测技术》;20150630;全文 * |
| 藜麦碾磨加工与营养分布研究进展;吴立根;《食品研究与开发》;20200831;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113100207A (en) | 2021-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | A review on weed detection using ground-based machine vision and image processing techniques | |
| CN105718945B (en) | Apple picking robot night image recognition method based on watershed and neural network | |
| CN107610114B (en) | optical satellite remote sensing image cloud and snow fog detection method based on support vector machine | |
| CN104268583B (en) | Pedestrian re-recognition method and system based on color area features | |
| CN116092013B (en) | Dangerous road condition identification method for intelligent monitoring | |
| CN103646249B (en) | A kind of greenhouse intelligent mobile robot vision navigation path identification method | |
| CN104463877B (en) | A kind of water front method for registering based on radar image Yu electronic chart information | |
| Qing et al. | Automated detection and identification of white-backed planthoppers in paddy fields using image processing | |
| CN108846831B (en) | Band steel surface defect classification method based on combination of statistical characteristics and image characteristics | |
| CN109858480A (en) | Digital instrument identification method | |
| CN109255336A (en) | Arrester recognition methods based on crusing robot | |
| CN113947570B (en) | Crack identification method based on machine learning algorithm and computer vision | |
| CN113100207B (en) | Accurate formula pesticide applying robot system based on wheat disease information and pesticide applying method | |
| CN114511770A (en) | Road sign plate identification method | |
| CN109766850B (en) | Fingerprint image matching method based on feature fusion | |
| CN113221881B (en) | Multi-level smart phone screen defect detection method | |
| CN109598200B (en) | Intelligent image identification system and method for molten iron tank number | |
| CN105975906B (en) | A kind of PCA static gesture identification methods based on area features | |
| CN115082776A (en) | Electric energy meter automatic detection system and method based on image recognition | |
| Mei et al. | A novel framework for container code-character recognition based on deep learning and template matching | |
| CN110188646B (en) | Human ear identification method based on fusion of gradient direction histogram and local binary pattern | |
| CN115311689A (en) | Cattle face identification feature extraction model construction method and cattle face identification method | |
| CN116524224A (en) | Machine vision-based method and system for detecting type of cured tobacco leaves | |
| CN111626150B (en) | Commodity identification method | |
| CN111950409A (en) | Intelligent identification method and system for road marking line |
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 |