CN116740562A - Artificial forest pest and disease damage identification method based on snake group optimization algorithm and CNN algorithm - Google Patents
Artificial forest pest and disease damage identification method based on snake group optimization algorithm and CNN algorithm Download PDFInfo
- Publication number
- CN116740562A CN116740562A CN202310591625.9A CN202310591625A CN116740562A CN 116740562 A CN116740562 A CN 116740562A CN 202310591625 A CN202310591625 A CN 202310591625A CN 116740562 A CN116740562 A CN 116740562A
- Authority
- CN
- China
- Prior art keywords
- artificial forest
- snake
- convolution layer
- pest
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 241000270295 Serpentes Species 0.000 title claims abstract description 91
- 241000607479 Yersinia pestis Species 0.000 title claims abstract description 66
- 201000010099 disease Diseases 0.000 title claims abstract description 36
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 36
- 238000005457 optimization Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 title claims abstract description 18
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 53
- 238000013528 artificial neural network Methods 0.000 claims abstract description 34
- 238000012549 training Methods 0.000 claims abstract description 6
- 235000013305 food Nutrition 0.000 claims description 13
- 241000238631 Hexapoda Species 0.000 claims description 11
- 238000011176 pooling Methods 0.000 claims description 9
- 238000011161 development Methods 0.000 claims description 7
- 230000018109 developmental process Effects 0.000 claims description 7
- 230000013011 mating Effects 0.000 claims description 4
- 241000282461 Canis lupus Species 0.000 claims description 3
- 235000013601 eggs Nutrition 0.000 claims description 3
- 230000012447 hatching Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/082—Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an artificial forest pest and disease damage identification method based on a snake group optimization algorithm and a CNN algorithm, which comprises the following steps: acquiring an image dataset of the artificial forest tree; extracting leaf image features from the tree image; constructing an artificial forest pest and disease damage identification CNN neural network; obtaining optimal structural parameters of the artificial forest pest and disease damage identification CNN neural network based on a snake group optimization algorithm; taking leaf image characteristics in the artificial forest tree image data set as input for training an artificial forest pest and disease damage identification CNN neural network with optimal structural parameters; the artificial forest pest and disease damage identification CNN neural network after training is used for judging whether the tree image to be identified is a healthy tree or the degree of pest and disease damage of leaves in the image. According to the invention, the optimal structural parameters of the artificial forest pest identification CNN neural network are obtained based on the snake group optimization algorithm, so that whether the artificial forest is affected by the pest and the extent of the affected pest can be accurately identified.
Description
Technical Field
The invention relates to an artificial forest pest identification method, in particular to an artificial forest pest identification method based on a snake group optimization algorithm and a CNN algorithm.
Background
The characteristics of a single tree species of the artificial forest lead the leaves of the artificial forest to be easy to be attacked by diseases and insect pests and the spreading speed of the diseases and insect pests is extremely high. The plant diseases and insect pests cause serious economic loss to the artificial forests and also cause great influence to the ecological environment. Traditional artificial forest tree leaf pest and disease damage identification often relies on field investigation and empirical analysis, space limitation is very large, and occurrence and degree of pest and disease damage are difficult to judge in time. With the development and rising of deep learning, the method is increasingly applied to the field of pest and disease identification. One method is to train a convolutional neural network by taking an image data set of an artificial forest tree as input, so that the convolutional neural network can accurately identify the occurrence degree of leaf diseases and insect pests in the image. The structural parameters of the convolutional neural network are important factors influencing the identification performance, but are often difficult to determine, and are a difficult problem for building the neural network.
Disclosure of Invention
The invention aims to solve the technical problem of providing an artificial forest pest identification method based on a snake group optimization algorithm and a CNN algorithm aiming at the defects of the prior art, wherein the artificial forest pest identification method based on the snake group optimization algorithm and the CNN algorithm adopts the snake group optimization algorithm to acquire the optimal structural parameters of an artificial forest pest identification CNN neural network, so that whether the leaves in the artificial forest tree are damaged by the pest and the degree of damage by the pest can be accurately identified.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
an artificial forest pest and disease damage identification method based on a snake group optimization algorithm and a CNN algorithm comprises the following steps:
(1) Acquiring an image dataset of an artificial forest tree, wherein the image dataset comprises healthy tree images and tree images with different degrees of diseases and insect pests;
(2) Respectively extracting leaf image features from all tree images;
(3) Constructing an artificial forest pest and disease damage identification CNN neural network;
(4) Obtaining optimal structural parameters of the artificial forest pest and disease damage identification CNN neural network based on a snake group optimization algorithm; taking leaf image characteristics in the artificial forest tree image data set as input for training an artificial forest pest and disease damage identification CNN neural network with optimal structural parameters;
(5) Collecting a tree image to be identified, extracting leaf image features in the tree image to be identified, and inputting the leaf image features in the tree image to be identified into a trained artificial forest pest identification CNN neural network, so as to judge the pest damage degree of the leaves in the tree image to be identified.
As a further improved technical scheme of the invention, the leaf image features comprise color features, texture features, shape features and spatial relationship features.
As a further improved technical scheme of the invention, the output of the artificial forest pest identification CNN neural network is the percentage of the total leaf area occupied by the pest.
As a further improved technical scheme of the invention, the artificial forest pest identification CNN neural network sequentially comprises an input layer, a first convolution layer, a pooling layer, a second convolution layer, a pooling layer, a third convolution layer, a pooling layer, a fourth convolution layer, a full connection layer and an output layer.
As a further improved technical scheme of the invention, the obtaining of the optimal structural parameters of the CNN based on the snake group optimization algorithm comprises the following steps:
(4.1), setting the number of parameters and the maximum number of iterations:
(4.2) initializing a snake group, calculating a value of individual fitness, and equally dividing the initial snake group into male and female snakes;
(4.3), calculating the food quantity Q and the temperature Temp, and judging the current stage:
(4.3.1), exploration phase: when Q <0.25, the male and female snakes search for food and move the positions, respectively; executing the step (4.4);
(4.3.2), development stage: when Q >0.25, temp >0.6, both male and female snakes move toward food; executing the step (4.4);
(4.3.3), development stage: when Q is more than 0.25 and temp is less than 0.6, the snake can perform a combat mode or mating mode; hatching snake eggs, and substituting new individuals for individuals with the worst applicability in the original population by referring to a wolf algorithm; executing the step (4.4);
and (4.4) judging whether the maximum iteration times are reached, if not, adding 1 to the iteration times and returning to the step (4.3), otherwise, outputting the position of the individual with the optimal applicability of the current snake group as the optimal structural parameter of the artificial forest pest identification CNN neural network.
As a further improved technical solution of the present invention, the position x= (a 1, b1, c1, a2, b2, c2, a3, b3, c3, a4, b4, c4, η) of the current individual with optimal snake group, wherein a1 represents the number of channels of the first convolution layer, b1 represents the size of the first convolution layer, c1 represents the step size of the first convolution layer, a2 represents the number of channels of the second convolution layer, b2 represents the size of the second convolution layer, c2 represents the step size of the second convolution layer, a3 represents the number of channels of the third convolution layer, b3 represents the size of the third convolution layer, c3 represents the step size of the third convolution layer, a4 represents the number of channels of the fourth convolution layer, b4 represents the size of the fourth convolution layer, c4 represents the step size of the fourth convolution layer, η represents the learning rate.
The beneficial effects of the invention are as follows:
the convolution kernel parameter of the CNN convolution neural network is obtained by using the snake group optimization algorithm, so that the adjustment time of the CNN network parameter is reduced, manpower and material resources are saved, the identification efficiency and accuracy are improved, and the problem that the structural parameter of the neural network is difficult to determine in the prior art is solved. The CNN network can accurately identify whether the artificial forest tree leaves are damaged by diseases and insect pests or not and the damage degree of the artificial forest tree leaves.
Drawings
FIG. 1 is a schematic flow chart of the present invention.
FIG. 2 is a flow chart of the snake group optimization algorithm of the invention.
Fig. 3 is a schematic flow chart of the CNN algorithm of the present invention.
Detailed Description
The following is a further description of embodiments of the invention, with reference to the accompanying drawings:
as shown in fig. 1, the artificial forest pest and disease damage identification method based on the snake group optimization algorithm and the CNN algorithm comprises the following steps:
(1) An image dataset of an artificial forest tree is obtained, including images of healthy trees and images of trees that are affected by different degrees of disease and insect pests.
(2) Extracting leaf image features from the tree image; the leaf image features comprise color features, texture features, shape features, spatial relationship features and the like.
(3) And (5) building a structure of an artificial forest pest and disease damage identification CNN neural network.
(4) Obtaining optimal structural parameters of the artificial forest pest and disease damage identification CNN neural network based on a snake group optimization algorithm; the result of the snake group optimization algorithm is taken as the optimal structural parameter of the artificial forest pest identification CNN neural network algorithm, the leaf image characteristics of the tree images in the artificial forest tree image data set are taken as input, the percentage of the leaf area occupied by the pest in the tree images to the total leaf area is taken as output, and then the artificial forest pest identification CNN neural network with the optimal structural parameter is trained.
(5) Collecting a tree image to be identified, extracting leaf image features in the tree image to be identified, and inputting the leaf image features in the tree image to be identified into a trained artificial forest pest identification CNN neural network, so as to judge the pest damage degree of the leaves in the tree image to be identified. The degree of the tree image that the tree leaves are damaged by diseases is the percentage of the total area of the tree leaves, which is damaged by diseases, in the tree image. The output of the artificial forest pest identification CNN neural network is that the leaf area of the pest is a percentage y of the total leaf area, and y is a number greater than or equal to 0 and less than or equal to 1. When y is equal to 0, the tree image is a healthy tree image.
As shown in fig. 2, obtaining the optimal structural parameters of the CNN neural network based on the snake swarm optimization algorithm includes:
(4.1), setting the parameter number and the maximum iteration number, and initializing the snake group:
X i ＝X min +r×(X max -X min )；
wherein X is max And X min Respectively represent the upper and lower boundaries of the position of the snake group, r represents one [0,1 ]]Is a random number of (a) in the memory.
(4.2) calculating individual fitness f by using the initialized snake group position as an initial structural parameter of the CNN network i And the initial group of snakes was equally divided into male and female snakes.
(4.3) calculating food quantity Q and temperature Temp, judging current belonging stage, in whichWherein Q represents food amount, c1 is food amount parameter, and 0.5 is taken; />Wherein Temp is the temperature, and T respectively represent the current iteration number and the maximum iteration number.
(4.3.1), exploration phase: when Q is<0.25, the male and female snakes search for food by random position and move the position, updating male snake position X i,m And female snake position X i,f ：
X i,m (t+1)＝X rand,m (t)±0.05×A m ×((X max -X min )×rand+X min )
X i,f (t+1)＝X rand,f (t)±0.05×A f ×((X max -X min )×rand+X min )
Wherein, subscripts m and f respectively represent male snake and female snake, subscript i represents ith snake, rand represents random number, A m Indicating the ability of the male snake to find food, A f Indicating the ability of male snakes to find food, X max And X min Respectively representing the upper boundary and the lower boundary of the position of the snake group, wherein t represents the current iteration times; x is X i,m (t+1) represents the position of the ith male snake in the t+1st cycle; x is X i,f (t+1) represents the position of the ith female snake in the t+1st cycle; x is X rand,m (t) represents the t th timeRandom position of circulation (male snake); x is X rand,f (t) represents the random position of the t cycle (female snake);
thereafter, step (4.4) is performed.
(4.3.2), development stage: when Q >0.25, temp >0.6, both male and female snakes move toward food;
X i,m(f) (t+1)＝X best,m(f) ±2×Temp×rand×(X best,m(f) -X i,m(f) (t))；
X i,m(f) (t+1) represents the position of the ith male snake (female snake) in the t+1st cycle; x is X best,m(f) Indicating the optimal position of a male snake (female snake); x is X i,m(f) (t) represents the position of the ith male snake (female snake) of the t-th cycle;
thereafter, step (4.4) is performed.
(4.3.3), development stage: when Q >0.25, temp <0.6, the snake will go into combat mode or mating mode:
F i,m(f) ＝exp(-f best,f(m) /f i )
M i,m(f) ＝exp(-f i,f(m) /f i,m(f) )
X i,m(f) (t+1)＝X i,m(f) (t)+2×F i,m(f) ×rand×(Q×X best,f(m) -X i,m(f) (t))
X i,m(f) (t+1)＝X i,m(f) (t)+2×M i,m(f) ×rand×(Q×X i,f(m) -X i,m(f) (t))
F i,m(f) represents the combat power of the ith male snake (female snake); f (f) best,f(m) Indicating the optimal fitness of the female (male) snake; f (f) i Indicating the fitness of the ith snake; m is M i,m(f) Represents mating ability of the ith male snake (female snake); f (f) i,f(m) Indicating the fitness of the ith female snake (male snake); f (f) i,m(f) Indicating the fitness of the ith male snake (female snake); x is X i,m(f) (t+1) represents the position of the ith male snake (female snake) in the t+1st cycle; x is X i,m(f) (t) represents the position of the ith male snake (female snake) of the t-th cycle; x is X best,f(m) Indicating the optimal position of the female (male) snake; x is X i,f(m) Representing the ith female snake (male snake)A location;
hatching snake eggs, and substituting new individuals for individuals with the worst applicability in the original population by referring to a wolf algorithm; thereafter, step (4.4) is performed.
And (4.4) judging whether the maximum iteration times are reached, if not, adding 1 to the current iteration times t and returning to (4.3), otherwise, outputting the position of the individual with the optimal applicability of the current snake group as the optimal structural parameter of the artificial forest pest identification CNN neural network.
As shown in fig. 3, the CNN neural network for identifying plant diseases and insect pests in artificial forests sequentially includes an input layer, a convolutional layer C1 (i.e., a first convolutional layer), a pooling layer P1, a convolutional layer C2 (i.e., a second convolutional layer), a pooling layer P2, a convolutional layer C3 (i.e., a third convolutional layer), a pooling layer P3, a convolutional layer C4 (i.e., a fourth convolutional layer), a full connection layer H, and an output layer. Each convolution layer uses a softsign function as an activation function.
Definition of the softsign function is
The position x= (a 1, b1, C1, a2, b2, C2, a3, b3, C3, a4, b4, C4, η) of the current individual with optimal snake group applicability, wherein a1 represents the number of channels of the convolution layer C1, b1 represents the size of the convolution layer C1, C1 represents the step size of the convolution layer C1, a2 represents the number of channels of the convolution layer C2, b2 represents the size of the convolution layer C2, C2 represents the step size of the convolution layer C2, a3 represents the number of channels of the convolution layer C3, b3 represents the size of the convolution layer C3, C3 represents the step size of the convolution layer C3, a4 represents the number of channels of the convolution layer C4, b4 represents the size of the convolution layer C4, and η represents the learning rate.
The CNN super parameters are set in this embodiment as follows: by adopting the AdaDelta method, the maximum iteration number is set to 1000 times, the learning rate reduction factor is set to 0.1, the learning rate becomes 0.1 eta after training for 800 times, the data set is disturbed during each training, and the batch size is set to 128.
According to the embodiment, the convolution kernel parameters of the convolution neural network are obtained by using the snake group optimization algorithm, so that the adjustment time of the network parameters is shortened, manpower and material resources are saved, and the identification efficiency and accuracy are improved.
The scope of the present invention includes, but is not limited to, the above embodiments, and any alterations, modifications, and improvements made by those skilled in the art are intended to fall within the scope of the invention.
Claims (6)
1. The artificial forest pest and disease damage identification method based on the snake group optimization algorithm and the CNN algorithm is characterized by comprising the following steps:
(1) Acquiring an image dataset of an artificial forest tree, wherein the image dataset comprises healthy tree images and tree images with different degrees of diseases and insect pests;
(2) Respectively extracting leaf image features from all tree images;
(3) Constructing an artificial forest pest and disease damage identification CNN neural network;
(4) Obtaining optimal structural parameters of the artificial forest pest and disease damage identification CNN neural network based on a snake group optimization algorithm; taking leaf image characteristics in the artificial forest tree image data set as input for training an artificial forest pest and disease damage identification CNN neural network with optimal structural parameters;
(5) Collecting a tree image to be identified, extracting leaf image features in the tree image to be identified, and inputting the leaf image features in the tree image to be identified into a trained artificial forest pest identification CNN neural network, so as to judge the pest damage degree of the leaves in the tree image to be identified.
2. The method for identifying artificial forest diseases and insect pests based on the snake group optimization algorithm and the CNN algorithm according to claim 1, wherein the leaf image features comprise color features, texture features, shape features and spatial relationship features.
3. The artificial forest pest identification method based on the snake group optimization algorithm and the CNN algorithm according to claim 1, wherein the output of the artificial forest pest identification CNN neural network is the percentage of the total leaf area occupied by the pest.
4. The artificial forest pest identification method based on the snake group optimization algorithm and the CNN algorithm according to claim 1, wherein the artificial forest pest identification CNN neural network sequentially comprises an input layer, a first convolution layer, a pooling layer, a second convolution layer, a pooling layer, a third convolution layer, a pooling layer, a fourth convolution layer, a full connection layer and an output layer.
5. The method for identifying plant diseases and insect pests of artificial forests based on the snake swarm optimization algorithm and the CNN algorithm according to claim 4, wherein the obtaining the optimal structural parameters of the CNN neural network based on the snake swarm optimization algorithm comprises the following steps:
(4.1), setting the number of parameters and the maximum number of iterations:
(4.2) initializing a snake group, calculating a value of individual fitness, and equally dividing the initial snake group into male and female snakes;
(4.3), calculating the food quantity Q and the temperature Temp, and judging the current stage:
(4.3.1), exploration phase: when Q <0.25, the male and female snakes search for food and move the positions, respectively; executing the step (4.4);
(4.3.2), development stage: when Q >0.25, temp >0.6, both male and female snakes move toward food; executing the step (4.4);
(4.3.3), development stage: when Q is more than 0.25 and temp is less than 0.6, the snake can perform a combat mode or mating mode; hatching snake eggs, and substituting new individuals for individuals with the worst applicability in the original population by referring to a wolf algorithm; executing the step (4.4);
and (4.4) judging whether the maximum iteration times are reached, if not, adding 1 to the iteration times and returning to the step (4.3), otherwise, outputting the position of the individual with the optimal applicability of the current snake group as the optimal structural parameter of the artificial forest pest identification CNN neural network.
6. The method for identifying the artificial forest pest and disease damage based on the snake group optimization algorithm and the CNN algorithm according to claim 5, wherein the position X= (a 1, b1, c1, a2, b2, c2, a3, b3, c3, a4, b4, c4, eta) of the individual with the optimal applicability of the current snake group is characterized in that a1 represents the number of channels of a first convolution layer, b1 represents the size of the first convolution layer, c1 represents the step size of the first convolution layer, a2 represents the number of channels of a second convolution layer, b2 represents the size of the second convolution layer, c2 represents the step size of the second convolution layer, a3 represents the number of channels of a third convolution layer, b3 represents the size of the third convolution layer, c3 represents the step size of the fourth convolution layer, a4 represents the number of channels of the fourth convolution layer, b4 represents the step size of the fourth convolution layer, eta represents the learning rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310591625.9A CN116740562A (en) | 2023-05-24 | 2023-05-24 | Artificial forest pest and disease damage identification method based on snake group optimization algorithm and CNN algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310591625.9A CN116740562A (en) | 2023-05-24 | 2023-05-24 | Artificial forest pest and disease damage identification method based on snake group optimization algorithm and CNN algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116740562A true CN116740562A (en) | 2023-09-12 |
Family
ID=87912422
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310591625.9A Pending CN116740562A (en) | 2023-05-24 | 2023-05-24 | Artificial forest pest and disease damage identification method based on snake group optimization algorithm and CNN algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116740562A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117708697A (en) * | 2023-11-29 | 2024-03-15 | 国网湖北省电力有限公司超高压公司 | Transformer DGA fault diagnosis method based on PNN network |
| CN118130072A (en) * | 2024-05-07 | 2024-06-04 | 石家庄铁道大学 | Bridge girder erection machine braking system fault diagnosis method based on improved snake optimization network model |
-
2023
- 2023-05-24 CN CN202310591625.9A patent/CN116740562A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117708697A (en) * | 2023-11-29 | 2024-03-15 | 国网湖北省电力有限公司超高压公司 | Transformer DGA fault diagnosis method based on PNN network |
| CN118130072A (en) * | 2024-05-07 | 2024-06-04 | 石家庄铁道大学 | Bridge girder erection machine braking system fault diagnosis method based on improved snake optimization network model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116740562A (en) | Artificial forest pest and disease damage identification method based on snake group optimization algorithm and CNN algorithm | |
| CN111723738B (en) | Coal rock chitin group microscopic image classification method and system based on transfer learning | |
| CN109711471A (en) | A kind of rice disease image-recognizing method based on depth convolutional neural networks | |
| CN112036512A (en) | Image classification neural network architecture searching method and device based on network clipping | |
| CN115331732B (en) | Gene phenotype training and predicting method and device based on graph neural network | |
| CN111127423B (en) | Rice pest and disease identification method based on CNN-BP neural network algorithm | |
| Kamrul et al. | Machine vision based rice disease recognition by deep learning | |
| CN111340133A (en) | Image classification processing method based on deep convolutional neural network | |
| Nikith et al. | Leaf disease detection and classification | |
| CN110289987A (en) | Multi-agent system network resilience appraisal procedure based on representative learning | |
| CN111625992B (en) | Mechanical fault prediction method based on self-optimal deep learning | |
| Veeragandham et al. | Effectiveness of convolutional layers in pre-trained models for classifying common weeds in groundnut and corn crops | |
| Zhang et al. | Robust image segmentation method for cotton leaf under natural conditions based on immune algorithm and PCNN algorithm | |
| CN113076873A (en) | Crop disease long-tail image identification method based on multi-stage training | |
| CN118447297A (en) | Knowledge distillation-based light crop disease identification method | |
| Rehana et al. | Plant disease detection using region-based convolutional neural network | |
| CN112907503B (en) | Penaeus vannamei Boone quality detection method based on self-adaptive convolutional neural network | |
| CN117557948A (en) | Corn seedling identification method based on residual selective nuclear attention mechanism | |
| Chauhan et al. | Deep residual neural network for plant seedling image classification | |
| Singh et al. | Tomato crop disease classification using convolution neural network and transfer learning | |
| Jin et al. | An improved mask r-cnn method for weed segmentation | |
| CN115731464A (en) | Litchi pest and disease identification method based on whale optimization algorithm and CNN model | |
| Hu | A rice pest identification method based on a convolutional neural network and migration learning | |
| Costales et al. | Development of a Prototype Application for Rice Disease Detection Using Convolutional Neural Networks | |
| Gao et al. | Classification Method of Rape Root Swelling Disease Based on Convolution Neural Network |
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 |