CN113920426A - CNN and LSTM based intelligent pest and disease identification method and system - Google Patents
CNN and LSTM based intelligent pest and disease identification method and system Download PDFInfo
- Publication number
- CN113920426A CN113920426A CN202111090780.XA CN202111090780A CN113920426A CN 113920426 A CN113920426 A CN 113920426A CN 202111090780 A CN202111090780 A CN 202111090780A CN 113920426 A CN113920426 A CN 113920426A
- Authority
- CN
- China
- Prior art keywords
- pest
- disease
- lstm
- picture
- cnn
- 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.)
- Withdrawn
Links
- 241000607479 Yersinia pestis Species 0.000 title claims abstract description 234
- 201000010099 disease Diseases 0.000 title claims abstract description 188
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 188
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 59
- 241000238631 Hexapoda Species 0.000 claims abstract description 50
- 238000012549 training Methods 0.000 claims abstract description 39
- 238000012545 processing Methods 0.000 claims abstract description 15
- 239000013598 vector Substances 0.000 claims description 43
- 238000012360 testing method Methods 0.000 claims description 21
- 230000006870 function Effects 0.000 claims description 17
- 238000006243 chemical reaction Methods 0.000 claims description 12
- 230000015654 memory Effects 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 238000005070 sampling Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 6
- 230000010365 information processing Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 4
- 230000014509 gene expression Effects 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 241000196324 Embryophyta Species 0.000 description 18
- 206010039509 Scab Diseases 0.000 description 7
- 241000209140 Triticum Species 0.000 description 7
- 235000021307 Triticum Nutrition 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003973 irrigation Methods 0.000 description 1
- 230000002262 irrigation Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- 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
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an intelligent pest and disease identification method and system based on CNN and LSTM, belonging to the technical field of artificial intelligent computer vision, aiming at solving the technical problem of how to utilize CNN and LSTM to identify crops and carry out corresponding treatment measures at the matching part after identification to achieve the purposes of finding and radically treating pests, and the adopted technical scheme is as follows: the method utilizes a convolutional neural network to judge the pest and disease damage condition on crops, and then realizes the purpose of discovering and solving the pest and disease damage through an LSTM matching aiming at a pest and disease damage condition processing scheme; the method comprises the following specific steps: constructing a crop disease and pest data set; constructing and training a disease and insect pest model; a pest model was deployed and used.
Description
Technical Field
The invention relates to the technical field of artificial intelligence computer vision, in particular to an intelligent pest and disease identification method and system based on CNN and LSTM.
Background
Neural networks (neural networks) are part of the field of artificial intelligence research, and the most popular neural network at present is deep Convolutional Neural Networks (CNNs), which are rarely used for reasons of accuracy, expressiveness, and the like, although they also have a shallow structure. Referring to CNNs and convolutional neural networks, there is no specific distinction between academic and industrial fields, and they generally refer to convolutional neural networks with deep structures, and the number of layers varies from "several layers" to "tens to hundreds". CNNs are currently enjoying tremendous success in many areas of research, such as speech recognition, image segmentation, natural language processing, etc. Although the problems addressed in these areas are not the same, these application methods can be generalized in that CNNs can automatically learn features from (usually large-scale) data and generalize the results to the same type of unknown data. The general steps for implementing the task of image recognition using CNN are: the input layer reads in a regularized (uniform size) image, each neuron of each layer takes a group of small local adjacent units of the previous layer as input, namely local receptive field and weight sharing, the neuron extracts some basic visual features such as edges, angular points and the like, and the features are used by the neurons of the higher layer. The convolutional neural network obtains a feature map by a convolution operation, and at each position, the units from different feature maps obtain different types of features respectively. A convolutional layer usually contains a plurality of feature maps with different weight vectors, so that richer features of the image can be retained. The back of the convolutional layer is connected with the pooling layer for down-sampling operation, so that the resolution of the image can be reduced, the parameter quantity can be reduced, and the robustness of translation and deformation can be obtained. The alternating distribution of the convolution layer and the pooling layer leads the number of the characteristic maps to be gradually increased and the resolution to be gradually reduced, thus the structure is a double pyramid.
LSTM (Long Short-Term Memory) is a Long Short-Term Memory network, a time recurrent neural network, suitable for processing and predicting important events with relatively Long intervals and delays in time series. LSTM has found many applications in the scientific field. LSTM based systems may learn tasks such as translating languages, controlling robots, image analysis, document summarization, speech recognition image recognition, handwriting recognition, controlling chat robots, predicting diseases, click rates and stocks, synthesizing music, and so forth.
The plant diseases and insect pests are one of the important factors causing the yield reduction of crops, and the traditional crop disease and insect pest identification method basically relies on the professional knowledge and the working experience of relevant experts for identification, so that the time and the labor are wasted, the efficiency is low, human errors are easy to generate, and the efficiency of the crop disease and insect pest control work is seriously influenced.
Therefore, how to identify crops by using CNN and LSTM and to carry out corresponding treatment measures at the matching position after identification is to achieve the purposes of finding and radically treating plant diseases and insect pests.
Disclosure of Invention
The invention aims to provide an intelligent pest and disease identification method and system based on CNN and LSTM, which solve the problems of how to identify crops by using CNN and LSTM and realize the discovery of pests and the radical treatment of pests by corresponding treatment measures at the matching part after identification.
The technical task of the invention is realized in the following way, an intelligent pest and disease damage identification method based on CNN and LSTM judges the pest and disease damage condition on crops by using a convolutional neural network, and then achieves the purpose of pest and disease damage discovery and solution by using an LSTM matching aiming at a pest and disease damage condition processing scheme; the method comprises the following specific steps:
constructing a crop disease and pest data set;
constructing and training a disease and insect pest model;
a pest model was deployed and used.
Preferably, the crop disease and insect pest data set comprises disease and insect pest pictures, the name of each disease and insect pest picture and a treatment scheme corresponding to the disease and insect pest;
the ratio of the training set to the testing set in the crop pest data set is 4: 1.
Preferably, the pest model is constructed and trained as follows:
extracting pest and disease picture characteristics through CNN and generating a name corresponding to each pest and disease picture characteristic;
the picture characteristics of the plant diseases and insect pests are converted into corresponding text descriptions through an LSTM;
combining the results of the CNN and the LSTM on the pest and disease picture processing to obtain a pest and disease model;
and (5) carrying out pest and disease model test and optimization by using the test set.
Preferably, the pest and disease picture features are extracted through the CNN and each pest and disease picture feature is generated and is named specifically as follows:
and (3) convolution operation: inputting the pest and disease picture with names in the training set into the convolutional layer, and outputting a feature vector of the pest and disease picture through feature operation of the pest and disease picture;
sampling operation: and (4) corresponding the characteristic vector of the pest and disease picture with the corresponding name of the pest and disease picture, and inputting the characteristic vector into a vector space.
Preferably, the picture characteristics of the plant diseases and insect pests are converted into corresponding text descriptions through the LSTM, and the text descriptions are as follows:
preprocessing a data text: deleting all symbols except letters, numbers and Chinese characters in the space vector text information by defining a function;
loading a stop word list: in the text information processing process, deleting words or expressions which have no influence on the text information according to the stop word list;
structuring treatment: converting the processed text information into structured data through word vectors; the method comprises the following specific steps:
generating a dictionary according to the text information of the Chinese characters;
converting the text data after word segmentation into a rectangular form through a dictionary;
the text data in the rectangular form is subjected to dimension unification, namely, a word number threshold value of one row of information in a matrix is set, and the following conditions are adopted:
when the number of characters of the text data is larger than a threshold value, deleting redundant text data;
when the number of characters of the text data is smaller than a threshold value, zero filling processing is carried out on insufficient text data;
model training: the vector space of the processed text data is input into an LSTM model for training.
Preferably, the pest model is deployed and used as follows:
deploying the pest and disease damage model into a server;
shooting crops by using an unmanned aerial vehicle;
uploading the photos to a server;
the server calls the pest and disease damage model to identify pest and disease damage types;
inputting the pest and disease model identification result into a server search function;
searching and the like to a specific treatment scheme aiming at the plant diseases and insect pests by a server;
the server inputs the specific treatment scheme of the plant diseases and insect pests into the unmanned aerial vehicle, and the unmanned aerial vehicle treats the crops according to the specific treatment scheme of the plant diseases and insect pests.
An intelligent pest and disease damage identification system based on CNN and LSTM, which comprises,
the construction unit is used for constructing a crop disease and pest data set; the crop disease and insect pest data set comprises disease and insect pest pictures, the name of each disease and insect pest picture and a treatment scheme corresponding to the disease and insect pest; the ratio of the training set to the testing set in the crop pest data set is 4: 1;
the building and training unit is used for building and training a pest model;
the deployment unit is used for deploying and using the pest and disease damage model; the method specifically comprises the following steps: deploying the pest and disease model into a server, taking a picture of crops by using an unmanned aerial vehicle, uploading the picture to the server, calling the pest and disease model by the server to identify the pest and disease types, and inputting the identification result of the pest and disease model into a server search function; finally, searching and the like by the server until a specific treatment scheme aiming at the plant diseases and insect pests is achieved; the server inputs the specific treatment scheme of the plant diseases and insect pests into the unmanned aerial vehicle, and the unmanned aerial vehicle treats the crops according to the specific treatment scheme of the plant diseases and insect pests.
Preferably, said building and training unit comprises,
the extraction module is used for extracting the pest and disease picture characteristics through CNN and generating a name corresponding to each pest and disease picture characteristic; the extraction module comprises a plurality of modules which are connected with each other,
the convolution submodule is used for inputting the pest and disease damage pictures with names in the training set into the convolution layer and outputting the feature vector of the pest and disease damage pictures through the feature operation of the pest and disease damage pictures;
the sampling submodule is used for corresponding the characteristic vector of the pest and disease picture with the corresponding name of the pest and disease picture and inputting the characteristic vector into a vector space;
the conversion module is used for converting the picture characteristics of the plant diseases and insect pests into corresponding text descriptions through the LSTM; the conversion module comprises a conversion module and a conversion module,
the preprocessing submodule is used for preprocessing the data text, namely deleting all symbols except letters, numbers and Chinese characters in the space vector text information through a defined function;
the loading submodule is used for deleting words or phrases which have no influence on the text information according to the stop word list in the text information processing process;
the structuring submodule is used for converting the processed text information into structured data through word vectors;
the training submodule is used for inputting the vector space of the processed text data into the LSTM model for training;
the combination module is used for combining the results of the CNN and the LSTM on the pest and disease picture processing to obtain a pest and disease model;
and the test module is used for carrying out pest and disease damage model test and optimization by using the test set.
An electronic device, comprising: a memory and at least one processor;
wherein the memory has stored thereon a computer program;
the at least one processor executes the computer program stored by the memory, so that the at least one processor executes the intelligent pest identification method based on CNN and LSTM as described above.
A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, the computer program being executable by a processor to implement the CNN and LSTM based intelligent pest identification method as described above.
The intelligent pest and disease damage identification method and system based on CNN and LSTM have the following advantages:
the method utilizes CNN + LSTM in deep learning to train a large number of crop pest and disease images, can learn the representation of each pest and disease characteristic, establishes a corresponding pest and disease treatment scheme, and implements the corresponding pest and disease treatment scheme by utilizing equipment such as an unmanned aerial vehicle and automatic irrigation, without manual execution, thereby greatly reducing the manual labor intensity and improving the treatment efficiency of crop pests;
secondly, the invention identifies crops based on convolutional neural network and LSTM in deep learning and carries out a matched pest and disease treatment scheme after identification, thereby achieving the purpose of finding and radically treating crop pests;
thirdly, the invention judges the pest and disease damage condition of the crop in the convolutional neural network, and then matches the scheme aiming at the pest and disease damage through the LSTM according to the pest and disease damage condition, so as to discover and solve the pest and disease damage.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a flow chart of an intelligent pest and disease identification method based on CNN and LSTM;
FIG. 2 is a block diagram of a process for constructing and training a pest model;
FIG. 3 is a flow chart diagram for converting pest and disease picture features into corresponding text descriptions through LSTM.
Detailed Description
The intelligent CNN and LSTM-based pest and disease identification method and system of the invention are described in detail below with reference to the drawings and specific examples.
Example 1:
as shown in the attached figure 1, the intelligent pest and disease damage identification method based on CNN and LSTM of the invention judges the pest and disease damage condition on crops by using a convolutional neural network, and then achieves the purpose of pest and disease damage discovery and solution by an LSTM matching aiming at a pest and disease damage condition processing scheme; the method comprises the following specific steps:
s1, constructing a crop disease and pest data set;
s2, constructing and training a pest model;
and S3, deploying and using a pest and disease damage model.
In this embodiment, the crop pest data set of step S1 includes pest pictures, names of each pest picture, and processing schemes corresponding to the pest;
for example: a picture of a plant with wheat scab was then named "wheat scab" and a treatment protocol for the relevant wheat scab was entered.
The ratio of the training set to the testing set in the crop pest data set is 4: 1.
In this embodiment, the construction and training of the pest model in step S2 are specifically as follows:
s201, extracting pest and disease picture features through CNN and generating names corresponding to the pest and disease picture features;
s202, converting the picture characteristics of the plant diseases and insect pests into corresponding text descriptions through LSTM;
s203, combining the results of the CNN and the LSTM on the pest and disease picture processing to obtain a pest and disease model;
and S204, carrying out pest and disease damage model test and optimization by using the test set.
In this embodiment, the step S201 of extracting pest and disease picture features through CNN and generating each pest and disease picture feature and the corresponding name are specifically as follows:
s20101, convolution operation: inputting the pest and disease picture with names in the training set into the convolutional layer, and outputting a feature vector of the pest and disease picture through feature operation of the pest and disease picture;
s20102, sampling operation: and (4) corresponding the characteristic vector of the pest and disease picture with the corresponding name of the pest and disease picture, and inputting the characteristic vector into a vector space.
In this embodiment, the step S202 of converting the pest and disease picture features into corresponding text descriptions through LSTM is specifically as follows:
s20201, preprocessing a data text: deleting all symbols except letters, numbers and Chinese characters in the space vector text information by defining a function;
s20202, loading a stop word list: in the text information processing process, deleting words or expressions which have no influence on the text information according to the stop word list;
s20203, structuring: converting the processed text information into structured data through word vectors;
s20204, model training: the vector space of the processed text data is input into an LSTM model for training.
For example, when a picture of wheat scab is input into a pest model, wheat scab is output.
And uploading the output result of the disease and pest model to a server, and searching out a solution through the searching function of the server.
For example, a picture of wheat scab is input into the system, and the server outputs a solution to wheat scab.
The structuring process in step S20203 in this embodiment is specifically as follows:
s2020301, generating a dictionary according to the text information of the Chinese characters;
s2020302, converting the text data after word segmentation into a rectangular form through a dictionary;
s2020303, dimension unification is performed on the text data in rectangular form, that is, a word count threshold of one line of information in the matrix is set, and the following cases are set:
firstly, when the number of characters of the text data is greater than a threshold value, deleting redundant text data;
and secondly, when the number of characters of the text data is smaller than a threshold value, zero filling processing is carried out on the insufficient text data.
For example: the threshold is set to 250, i.e. 250 words are used as one line of information, more erasures, and insufficient zero padding.
In this embodiment, the deployment and use of the pest model in step S3 are specifically as follows:
s301, deploying the pest and disease damage model into a server;
s302, shooting crops by using an unmanned aerial vehicle;
s303, uploading the photo to a server;
s304, calling a disease and insect pest model by the server to identify the disease and insect pest type;
s305, inputting a pest and disease damage model identification result into a server searching function;
s306, searching and the like by the server until a specific treatment scheme aiming at the plant diseases and insect pests is achieved;
s307, the server inputs the specific treatment scheme of the diseases and insect pests into the unmanned aerial vehicle, and the unmanned aerial vehicle treats the crops according to the specific treatment scheme of the diseases and insect pests.
Example 2:
the invention relates to an intelligent pest and disease identification system based on CNN and LSTM, which comprises,
the construction unit is used for constructing a crop disease and pest data set; the crop disease and insect pest data set comprises disease and insect pest pictures, the name of each disease and insect pest picture and a treatment scheme corresponding to the disease and insect pest; the ratio of the training set to the testing set in the crop pest data set is 4: 1;
the building and training unit is used for building and training a pest model;
the deployment unit is used for deploying and using the pest and disease damage model; the method specifically comprises the following steps: deploying the pest and disease model into a server, taking a picture of crops by using an unmanned aerial vehicle, uploading the picture to the server, calling the pest and disease model by the server to identify the pest and disease types, and inputting the identification result of the pest and disease model into a server search function; finally, searching and the like by the server until a specific treatment scheme aiming at the plant diseases and insect pests is achieved; the server inputs the specific treatment scheme of the plant diseases and insect pests into the unmanned aerial vehicle, and the unmanned aerial vehicle treats the crops according to the specific treatment scheme of the plant diseases and insect pests.
The building and training unit in this embodiment comprises,
the extraction module is used for extracting the pest and disease picture characteristics through CNN and generating a name corresponding to each pest and disease picture characteristic; the extraction module comprises a plurality of modules which are connected with each other,
the convolution submodule is used for inputting the pest and disease damage pictures with names in the training set into the convolution layer and outputting the feature vector of the pest and disease damage pictures through the feature operation of the pest and disease damage pictures;
the sampling submodule is used for corresponding the characteristic vector of the pest and disease picture with the corresponding name of the pest and disease picture and inputting the characteristic vector into a vector space;
the conversion module is used for converting the picture characteristics of the plant diseases and insect pests into corresponding text descriptions through the LSTM; the conversion module comprises a conversion module and a conversion module,
the preprocessing submodule is used for preprocessing the data text, namely deleting all symbols except letters, numbers and Chinese characters in the space vector text information through a defined function;
the loading submodule is used for deleting words or phrases which have no influence on the text information according to the stop word list in the text information processing process;
the structuring submodule is used for converting the processed text information into structured data through word vectors;
the training submodule is used for inputting the vector space of the processed text data into the LSTM model for training;
the combination module is used for combining the results of the CNN and the LSTM on the pest and disease picture processing to obtain a pest and disease model;
and the test module is used for carrying out pest and disease damage model test and optimization by using the test set.
Example 3:
an embodiment of the present invention further provides an electronic device, including: a memory and a processor;
wherein the memory stores computer execution instructions;
the processor executes the computer execution instructions stored in the memory, so that the processor executes the intelligent pest and disease identification method based on CNN and LSTM in any embodiment of the invention.
Example 4:
the embodiment of the invention also provides a computer-readable storage medium, wherein a plurality of instructions are stored, and the instructions are loaded by the processor, so that the processor executes the intelligent pest and disease damage identification method based on CNN and LSTM in any embodiment of the invention. Specifically, a system or an apparatus equipped with a storage medium on which software program codes that realize the functions of any of the above-described embodiments are stored may be provided, and a computer (or a CPU or MPU) of the system or the apparatus is caused to read out and execute the program codes stored in the storage medium.
In this case, the program code itself read from the storage medium can realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.
Examples of the storage medium for supplying the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (e.g., CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RYM, DVD-RW, DVD + RW), a magnetic tape, a nonvolatile memory card, and a ROM. Alternatively, the program code may be downloaded from a server computer via a communications network.
Further, it should be clear that the functions of any one of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like operating on the computer to perform a part or all of the actual operations based on instructions of the program code.
Further, it is to be understood that the program code read out from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion unit connected to the computer, and then causes a CPU or the like mounted on the expansion board or the expansion unit to perform part or all of the actual operations based on instructions of the program code, thereby realizing the functions of any of the above-described embodiments.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. An intelligent pest and disease damage identification method based on CNN and LSTM is characterized in that a convolutional neural network is used for judging pest and disease damage conditions on crops, and then a scheme for processing pest and disease damage conditions is adopted through LSTM matching, so that the purpose of discovering and solving pest and disease damage is achieved; the method comprises the following specific steps:
constructing a crop disease and pest data set;
constructing and training a disease and insect pest model;
a pest model was deployed and used.
2. An intelligent pest identification method based on CNN and LSTM according to claim 1, wherein the crop pest data set comprises pest pictures, the name of each pest picture and the treatment scheme of the corresponding pest;
the ratio of the training set to the testing set in the crop pest data set is 4: 1.
3. The intelligent pest identification method based on CNN and LSTM according to claim 1, wherein the pest model is constructed and trained as follows:
extracting pest and disease picture characteristics through CNN and generating a name corresponding to each pest and disease picture characteristic;
the picture characteristics of the plant diseases and insect pests are converted into corresponding text descriptions through an LSTM;
combining the results of the CNN and the LSTM on the pest and disease picture processing to obtain a pest and disease model;
and (5) carrying out pest and disease model test and optimization by using the test set.
4. An intelligent pest identification method based on CNN and LSTM according to claim 3, characterized in that, extracting pest picture features and generating each pest picture feature and its corresponding name by CNN are as follows:
and (3) convolution operation: inputting the pest and disease picture with names in the training set into the convolutional layer, and outputting a feature vector of the pest and disease picture through feature operation of the pest and disease picture;
sampling operation: and (4) corresponding the characteristic vector of the pest and disease picture with the corresponding name of the pest and disease picture, and inputting the characteristic vector into a vector space.
5. An intelligent pest identification method based on CNN and LSTM according to claim 3 or 4, wherein the image features of pest are converted into corresponding text descriptions by LSTM as follows:
preprocessing a data text: deleting all symbols except letters, numbers and Chinese characters in the space vector text information by defining a function;
loading a stop word list: in the text information processing process, deleting words or expressions which have no influence on the text information according to the stop word list;
structuring treatment: converting the processed text information into structured data through word vectors; the method comprises the following specific steps:
generating a dictionary according to the text information of the Chinese characters;
converting the text data after word segmentation into a rectangular form through a dictionary;
the text data in the rectangular form is subjected to dimension unification, namely, a word number threshold value of one row of information in a matrix is set, and the following conditions are adopted:
when the number of characters of the text data is larger than a threshold value, deleting redundant text data;
when the number of characters of the text data is smaller than a threshold value, zero filling processing is carried out on insufficient text data;
model training: the vector space of the processed text data is input into an LSTM model for training.
6. An intelligent pest identification method based on CNN and LSTM according to claim 1, wherein the pest model is deployed and used as follows:
deploying the pest and disease damage model into a server;
shooting crops by using an unmanned aerial vehicle;
uploading the photos to a server;
the server calls the pest and disease damage model to identify pest and disease damage types;
inputting the pest and disease model identification result into a server search function;
searching and the like to a specific treatment scheme aiming at the plant diseases and insect pests by a server;
the server inputs the specific treatment scheme of the plant diseases and insect pests into the unmanned aerial vehicle, and the unmanned aerial vehicle treats the crops according to the specific treatment scheme of the plant diseases and insect pests.
7. An intelligent pest and disease identification system based on CNN and LSTM is characterized in that the system comprises,
the construction unit is used for constructing a crop disease and pest data set; the crop disease and insect pest data set comprises disease and insect pest pictures, the name of each disease and insect pest picture and a treatment scheme corresponding to the disease and insect pest; the ratio of the training set to the testing set in the crop pest data set is 4: 1;
the building and training unit is used for building and training a pest model;
the deployment unit is used for deploying and using the pest and disease damage model; the method specifically comprises the following steps: deploying the pest and disease model into a server, taking a picture of crops by using an unmanned aerial vehicle, uploading the picture to the server, calling the pest and disease model by the server to identify the pest and disease types, and inputting the identification result of the pest and disease model into a server search function; finally, searching and the like by the server until a specific treatment scheme aiming at the plant diseases and insect pests is achieved; the server inputs the specific treatment scheme of the plant diseases and insect pests into the unmanned aerial vehicle, and the unmanned aerial vehicle treats the crops according to the specific treatment scheme of the plant diseases and insect pests.
8. An intelligent CNN and LSTM based pest identification system according to claim 7 wherein said building and training unit includes,
the extraction module is used for extracting the pest and disease picture characteristics through CNN and generating a name corresponding to each pest and disease picture characteristic; the extraction module comprises a plurality of modules which are connected with each other,
the convolution submodule is used for inputting the pest and disease damage pictures with names in the training set into the convolution layer and outputting the feature vector of the pest and disease damage pictures through the feature operation of the pest and disease damage pictures;
the sampling submodule is used for corresponding the characteristic vector of the pest and disease picture with the corresponding name of the pest and disease picture and inputting the characteristic vector into a vector space;
the conversion module is used for converting the picture characteristics of the plant diseases and insect pests into corresponding text descriptions through the LSTM; the conversion module comprises a conversion module and a conversion module,
the preprocessing submodule is used for preprocessing the data text, namely deleting all symbols except letters, numbers and Chinese characters in the space vector text information through a defined function;
the loading submodule is used for deleting words or phrases which have no influence on the text information according to the stop word list in the text information processing process;
the structuring submodule is used for converting the processed text information into structured data through word vectors;
the training submodule is used for inputting the vector space of the processed text data into the LSTM model for training;
the combination module is used for combining the results of the CNN and the LSTM on the pest and disease picture processing to obtain a pest and disease model;
and the test module is used for carrying out pest and disease damage model test and optimization by using the test set.
9. An electronic device, comprising: a memory and at least one processor;
wherein the memory has stored thereon a computer program;
the at least one processor executing the memory-stored computer program causing the at least one processor to perform the CNN and LSTM based intelligent pest identification method of any one of claims 1 to 6.
10. A computer-readable storage medium having stored thereon a computer program executable by a processor to implement the CNN and LSTM based intelligent pest identification method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111090780.XA CN113920426A (en) | 2021-09-17 | 2021-09-17 | CNN and LSTM based intelligent pest and disease identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111090780.XA CN113920426A (en) | 2021-09-17 | 2021-09-17 | CNN and LSTM based intelligent pest and disease identification method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113920426A true CN113920426A (en) | 2022-01-11 |
Family
ID=79235314
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111090780.XA Withdrawn CN113920426A (en) | 2021-09-17 | 2021-09-17 | CNN and LSTM based intelligent pest and disease identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113920426A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764268A (en) * | 2018-04-02 | 2018-11-06 | 华南理工大学 | A kind of multi-modal emotion identification method of picture and text based on deep learning |
| CN108874776A (en) * | 2018-06-11 | 2018-11-23 | 北京奇艺世纪科技有限公司 | A kind of recognition methods of rubbish text and device |
| CN108985237A (en) * | 2018-07-20 | 2018-12-11 | 安徽农业大学 | A kind of detection method and its system of the wheat scab based on depth mixing |
| CN109389059A (en) * | 2018-09-26 | 2019-02-26 | 华南理工大学 | A kind of P300 detection method based on CNN-LSTM network |
| CN109409392A (en) * | 2017-08-18 | 2019-03-01 | 广州极飞科技有限公司 | The method and device of picture recognition |
| CN110457982A (en) * | 2018-12-28 | 2019-11-15 | 中国科学院合肥物质科学研究院 | A kind of crop disease image-recognizing method based on feature transfer learning |
| CN113362088A (en) * | 2021-08-10 | 2021-09-07 | 中博信息技术研究院有限公司 | CRNN-based telecommunication industry intelligent customer service image identification method and system |
-
2021
- 2021-09-17 CN CN202111090780.XA patent/CN113920426A/en not_active Withdrawn
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409392A (en) * | 2017-08-18 | 2019-03-01 | 广州极飞科技有限公司 | The method and device of picture recognition |
| CN108764268A (en) * | 2018-04-02 | 2018-11-06 | 华南理工大学 | A kind of multi-modal emotion identification method of picture and text based on deep learning |
| CN108874776A (en) * | 2018-06-11 | 2018-11-23 | 北京奇艺世纪科技有限公司 | A kind of recognition methods of rubbish text and device |
| CN108985237A (en) * | 2018-07-20 | 2018-12-11 | 安徽农业大学 | A kind of detection method and its system of the wheat scab based on depth mixing |
| CN109389059A (en) * | 2018-09-26 | 2019-02-26 | 华南理工大学 | A kind of P300 detection method based on CNN-LSTM network |
| CN110457982A (en) * | 2018-12-28 | 2019-11-15 | 中国科学院合肥物质科学研究院 | A kind of crop disease image-recognizing method based on feature transfer learning |
| CN113362088A (en) * | 2021-08-10 | 2021-09-07 | 中博信息技术研究院有限公司 | CRNN-based telecommunication industry intelligent customer service image identification method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021190451A1 (en) | Method and apparatus for training image processing model | |
| CN110084281B (en) | Image generation method, neural network compression method, related device and equipment | |
| CN111241304B (en) | Answer generation method based on deep learning, electronic device and readable storage medium | |
| WO2023024412A1 (en) | Visual question answering method and apparatus based on deep learning model, and medium and device | |
| CN110532397B (en) | Question-answering method and device based on artificial intelligence, computer equipment and storage medium | |
| CN111275046B (en) | Character image recognition method and device, electronic equipment and storage medium | |
| CN111382868B (en) | Neural network structure searching method and device | |
| CN112183577A (en) | Training method of semi-supervised learning model, image processing method and equipment | |
| CN112765312B (en) | Knowledge graph question-answering method and system based on graph neural network embedded matching | |
| CN112070069A (en) | Method and device for identifying remote sensing image | |
| CN111753995B (en) | Local interpretable method based on gradient lifting tree | |
| CN112381079A (en) | Image processing method and information processing apparatus | |
| CN114090780A (en) | Prompt learning-based rapid picture classification method | |
| US20190228297A1 (en) | Artificial Intelligence Modelling Engine | |
| CN113936295A (en) | Character detection method and system based on transfer learning | |
| WO2023284716A1 (en) | Neural network searching method and related device | |
| CN113010683B (en) | Entity relationship identification method and system based on improved graph attention network | |
| CN113434699A (en) | Pre-training method of BERT model, computer device and storage medium | |
| CN112580627A (en) | Yoov 3 target detection method based on domestic intelligent chip K210 and electronic device | |
| CN114742224A (en) | Pedestrian re-identification method and device, computer equipment and storage medium | |
| CN112307048A (en) | Semantic matching model training method, matching device, equipment and storage medium | |
| CN110163206B (en) | License plate recognition method, system, storage medium and device | |
| CN116075820A (en) | Method, non-transitory computer readable storage medium and apparatus for searching image database | |
| CN113920426A (en) | CNN and LSTM based intelligent pest and disease identification method and system | |
| Kalliomäki | Real-time object detection for autonomous vehicles using deep learning |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20220111 |