CN110210434A - Pest and disease damage recognition methods and device - Google Patents
Pest and disease damage recognition methods and device Download PDFInfo
- Publication number
- CN110210434A CN110210434A CN201910494898.5A CN201910494898A CN110210434A CN 110210434 A CN110210434 A CN 110210434A CN 201910494898 A CN201910494898 A CN 201910494898A CN 110210434 A CN110210434 A CN 110210434A
- Authority
- CN
- China
- Prior art keywords
- plant
- image
- identified
- pest
- network model
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
Pest and disease damage recognition methods provided by the present application and device are applied to data processing equipment.The data processing equipment is pre-configured with trained neural network model, wherein the neural network model is trained acquisition by the leaf image and plant image of multiple illness plants.The data processing equipment obtains the leaf image and plant image of plant to be identified, is identified by feature of the neural network model to plant to be identified, and then obtains the type and pest and disease damage type of the plant to be identified.In this way, improving the discrimination of floristic discrimination and different plant illness types by combining the plant image of the leaf image of plant and the plant.
Description
Technical field
This application involves field of image recognition, in particular to a kind of pest and disease damage recognition methods and device.
Background technique
Plant pest refers to that plant is infected by other biological, as germ is infected, fungal infection or pest are infected；
Perhaps the destruction of the normal physiological function due to caused by unfavorable environmental condition such as arid, freeze or inflate pollution.I
State usually causes great loss as large agricultural country, diseases and pests of agronomic crop disease to national economy.
In existing pest and disease damage detection technique, most methods are directed to single class plant or pest and disease damage is identified, due to identification
Means extract feature dependent on artificial, therefore, accurate recognition methods are able to carry out on certain class crops and is generally difficult to again
It uses on another kind of plant.Accordingly, there exist poor universality, the unstable problem of discrimination between different plants.
Summary of the invention
In order to overcome at least one deficiency in the prior art, the first purpose of the application is to provide a kind of pest and disease damage knowledge
Other method, is applied to data processing equipment, and the data processing equipment is configured with trained neural network model, the training
Good neural network model is trained acquisition, the method packet by the leaf image and plant image of multiple illness plants
It includes:
The image of plant to be identified is obtained, the image of the plant to be identified includes leaf image and plant image；
The image of the plant to be identified is inputted into the neural network model, so that the neural network model is to described
The feature of plant to be identified is identified, the type of the plant to be identified and the pest and disease damage class of the plant to be identified are obtained
Type.
Optionally, the neural network model includes that floristics identification network and multiple pest species identify network,
The image by the plant to be identified inputs the neural network model, so that the neural network model is to described wait know
The feature of other plant is identified, the step of the type of the plant to be identified and the pest and disease damage type of the plant to be identified is obtained
Suddenly include:
Identify that network knows the leaf image and plant image of the plant to be identified by the floristics
Not, the floristics to be identified is obtained；
Corresponding target pest species identification network is determined according to the floristics to be identified；
Identify network to the leaf image and plant image of the plant to be identified by the target pest species
It is identified, obtains the pest species of the plant to be identified.
Optionally, the method also includes the steps of the training to the neural network model:
Obtain the plant sample image for being marked with floristics label and pest species label, the plant sample figure
As including leaf image and plant image；
The plant sample image is inputted to the neural network model to be trained；
Based on default loss function, the trained neural network model weight is treated by back-propagation algorithm and is changed
Generation adjustment, until the loss function output valve be less than preset threshold, obtain the trained neural network model.
Optionally, the neural network model includes that floristics identification network and multiple pest species identify network,
It is described to be based on default loss function, the trained neural network model weight is treated by back-propagation algorithm and is iterated tune
It is whole, until the step of output valve of the loss function is less than preset threshold, obtains trained neural network model packet
It includes:
The plant sample image is inputted into the floristics and identifies network, so that the floristics identifies network root
It is trained according to the floristics label of the object sample image；
According to the type label of the plant sample image, different types of plant sample image is inputted respectively different
Pest species identify network, so that pest species identification network is trained according to the pest species label；
Identify that network and multiple pest species identify that network constitutes the instruction by the trained floristics
The neural network model perfected.
Optionally, the plant sample image is divided into the training sample and test sample of preset ratio, the test
Sample is used to detect the accuracy rate by the trained neural network model of the training sample.
Optionally, the plant image is the plant panorama sketch obtained from multiple shooting angle.
The another object of the embodiment of the present application is to provide a kind of pest and disease damage identification device, is applied to data processing equipment,
The data processing equipment is configured with trained neural network model, and the trained neural network model is suffered from by multiple
The leaf image and plant image of sick plant are trained acquisition, and the pest and disease damage identification device includes obtaining module and identification
Module；
The image for obtaining module and being used to obtain plant to be identified, the image of the plant to be identified includes leaf image
And plant image；
The identification module is used to the leaf image and plant image inputting the neural network model, so that institute
Neural network model is stated to identify the feature of the plant to be identified, obtain the plant to be identified type and should be to
Identify the pest and disease damage type of plant.
Optionally, the neural network model includes that floristics identification network and multiple pest species identify network,
The identification module in the following manner identifies the plant to be identified:
Identify that network knows the leaf image and plant image of the plant to be identified by the floristics
Not, the floristics to be identified is obtained；
Corresponding target pest species identification network is determined according to the floristics to be identified；
Identify network to the leaf image and plant image of the plant to be identified by the target pest species
It is identified, obtains the pest species of the plant to be identified.
Optionally, the pest and disease damage identification device further includes adjustment module；
The module that obtains is also used to obtain the plant sample for being marked with floristics label and pest species label
Image, the plant sample image include leaf image and plant image, and the plant sample image is inputted to institute to be trained
State neural network model；
The adjustment module is used to treat the trained nerve net by back-propagation algorithm based on default loss function
Network model weight is iterated adjustment, until the loss function output valve be less than preset threshold, obtain described trained
Neural network model.
Optionally, the plant sample image is divided into the training sample and test sample of preset ratio, the test
Sample is used to detect the accuracy rate by the trained neural network model of the training sample.
In terms of existing technologies, the application has the advantages that
Pest and disease damage recognition methods provided by the embodiments of the present application and device are applied to data processing equipment.The data processing
Equipment is pre-configured with trained neural network model, wherein the neural network model passes through the blade figure of multiple illness plants
Picture and plant image are trained acquisition.The data processing equipment obtains the leaf image and plant figure of plant to be identified
Picture identified by feature of the neural network model to plant to be identified, so obtain the type of the plant to be identified with
And pest and disease damage type.In this way, improving plant species by combining the plant image of the leaf image of plant and the plant
The discrimination of the discrimination of class and different plant illness types.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only some embodiments of the application, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 is the hardware structure diagram of data processing equipment provided by the embodiments of the present application；
Fig. 2 is the step flow chart of pest and disease damage recognition methods provided by the embodiments of the present application；
Fig. 3 is that the structural diagrams of neural network model provided by the embodiments of the present application are intended to；
Fig. 4 is one of the structural schematic diagram of pest and disease damage identification device provided by the embodiments of the present application；
Fig. 5 is the second structural representation of pest and disease damage identification device provided by the embodiments of the present application.
Icon: 100- data processing equipment；120- memory；110- pest and disease damage identification device；130- processor；1101-
Obtain module；1102- identification module；1103- adjusts module.
Specific embodiment
To keep the purposes, technical schemes and advantages of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application
In attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is
Some embodiments of the present application, instead of all the embodiments.The application being usually described and illustrated herein in the accompanying drawings is implemented
The component of example can be arranged and be designed with a variety of different configurations.
Therefore, the detailed description of the embodiments herein provided in the accompanying drawings is not intended to limit below claimed
Scope of the present application, but be merely representative of the selected embodiment of the application.Based on the embodiment in the application, this field is common
Technical staff's every other embodiment obtained without creative efforts belongs to the model of the application protection
It encloses.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.
Fig. 1, the block diagram of data processing equipment 100 shown in FIG. 1 are please referred to, which includes
Memory 120, processor 130 and pest and disease damage identification device 110.Wherein, the memory 120, processor 130 and mutually it
Between be directly or indirectly electrically connected, to realize the transmission or interaction of data.For example, these elements can pass through one between each other
Or a plurality of communication bus or signal wire are realized and are electrically connected.The pest and disease damage identification device 110 includes at least one can be with software
Or the form of firmware (firmware) is stored in the memory 120 or is solidificated in the operation of the data processing equipment 100
Software function module in system (operating system, OS).The processor 130 is for executing the memory 120
The executable module of middle storage, such as software function module and computer program included by the pest and disease damage identification device 110
Deng.
In the present embodiment, the data processing equipment 100 be may be, but not limited to, smart phone, PC
(personal computer, PC), tablet computer, personal digital assistant (personal digital assistant, PDA),
Mobile internet surfing equipment (mobile Internet device, MID) etc..
Wherein, the memory 120 may be, but not limited to, random access memory (Random Access
Memory, RAM), read-only memory (Read Only Memory, ROM), programmable read only memory (Programmable
Read-Only Memory, PROM), erasable read-only memory (Erasable Programmable Read-Only
Memory, EPROM), electricallyerasable ROM (EEROM) (Electric Erasable Programmable Read-Only
Memory, EEPROM) etc..Wherein, memory 120 is for storing program, the processor 130 after receiving and executing instruction,
Execute described program.
The processor 130 may be a kind of IC chip, the processing capacity with signal.Above-mentioned processor can
To be general processor, including central processing unit (Central Processing Unit, abbreviation CPU), network processing unit
(Network Processor, abbreviation NP) etc.；Can also be digital signal processor (DSP), specific integrated circuit (ASIC),
Field programmable gate array (FPGA) either other programmable logic device, discrete gate or transistor logic, discrete hard
Part component.It may be implemented or execute disclosed each method, step and the logic diagram in the embodiment of the present application.General processor
It can be microprocessor or the processor be also possible to any conventional processor etc..
Referring to figure 2., Fig. 2 is the process of the pest and disease damage recognition methods applied to data processing equipment 100 shown in FIG. 1
Figure, the data processing equipment 100 are configured with trained neural network model, and the trained neural network model passes through
The leaf image and plant image of multiple illness plants are trained acquisition.Below will to the method includes each step into
Row elaborates.
Step S100, obtains the image of plant to be identified, and the image of the plant to be identified includes leaf image and plant
Strain image.
The leaf image and plant image are inputted the neural network model, so that the nerve by step S200
Network model identifies the feature of the plant to be identified, obtains the type and the plant to be identified of the plant to be identified
The pest and disease damage type of object.
Optionally, since floristics is various, there are different types of plant its blade shape, surface texture are extremely similar
The case where.However different types of plant, the type for being easy illness is also different, brings to pest and disease damage identification greatly dry
It disturbs.For these reasons, which obtains the image of plant to be identified, the image packet of the plant to be identified
Include leaf image and plant image.In this way, increasing the distinguishing characteristics between different plants by the plant forms of plant.
Optionally, the structure of the neural network model is as shown in figure 3, the neural network model includes floristics identification net
Network and multiple pest species identify network.After the data processing equipment 100 gets the image of plant to be identified, first input
Into floristics identification network；Identify that network handles the plant characteristics to be identified by the floristics, Jin Ershi
Not Chu the plant to be identified floristics.The data processing equipment 100 is according to the floristics of the plant to be identified from multiple
Target pest species identification network is selected in pest species identification network, to the illness type to the plant to be identified
It is identified.
For example, in a kind of possible example, which is known by floristics referring once again to Fig. 3
Other network gets the leaf image and plant image of plant to be identified, and does further to the feature of the plant to be identified
Processing, and then identify that the type of the plant to be identified is plant B.The data processing equipment 100 determines disease according to the plant B
Insect pest category identification network B.Further, which identifies that network waits knowing to this by the pest species
The image of other plant identified, determine the plant to be identified whether illness.
Optionally, the embodiment of the present application is also provided to the training step to the neural network model.The data processing equipment
100, which obtain multiple, is marked with the plant sample image of floristics and pest species label.In order to improve to floristics
Recognition effect, which includes leaf image and plant image.The data processing equipment 100 is by the plant sample
Picture inputs in the neural network model, preset loss function is based on, by back-propagation algorithm to the power of the neural network
Value is iterated adjustment, until the output for presetting loss function is less than preset threshold.
Optionally, since the neural network model includes floristics identification network and multiple pest species identification net
Network.The data processing equipment 100 identifies that network and multiple pest species identification network are trained to the floristics respectively.
Wherein, identify that network, the data processing equipment 100 input plant sample image wherein, so that the plant for the floristics
Species identification network is trained according to the type label of the plant sample；Network, the number are identified for the pest species
It is respectively that each plant distributes pest species identification network according to processing equipment 100, so that the pest species identify net
Network is served only for identifying a kind of pest species of plant.The data processing equipment 100 is carried out to pest species identification network
When training, using the pest species label of the plant sample as judgment basis.In this way, referring once again to Fig. 3, by training
Floristics identification network and multiple pest species identification network constitute the trained neural network model.
By the way that the neural network model is divided into floristics identification model and pest species identification network, list is simplified
The parameter of a network.When newly-increased plant or pest and disease damage classification, the network of repetition training is less.For example, in order to enable the mind
New plant can be identified through network model, it is only necessary to which network and newly-increased pest and disease damage, which identify, to be identified to floristics
Network is trained.In this way, reducing the network of repetition training；Since the neural network model is using two identification network levels
The mode of connection, each network only do individual thing, reduce the parameter of single network.
Optionally, in a kind of possible example, which is ResNet, and ResNet is a kind of depth convolution
The neural network framework of neural network has very high recognition accuracy to image recognition.
Optionally, which is divided into the training sample and test sample of preset ratio.For example, in one kind
In possible example, by the plant sample image according to 4:1 ratio it is random be divided into training sample set and test sample collection.Its
In, the test sample collection is for testing by the trained neural network model of the training sample set, to obtain the instruction
The accuracy of identification for the neural network model perfected.
Optionally, when shooting the leaf image in plant sample image, blade is tiled, shoots the position in blade.Together
The blade of illness can acquire multiple to be shot on one plant, and the blade of non-illness is avoided to impact sample.
Optionally, it when shooting the plant image in plant sample picture, is shot from multiple angles, obtains the complete of plant
Scape image.
Optionally, right using the mode of rotation, translation or cutting for the plant sample of parts of images lazy weight
Category plant sample data carry out augmentation operation, to increase the quantity of training sample and the robustness of model.
Referring to figure 4., the embodiment of the present application also provides a kind of pest and disease damage identification device 110, is applied to data processing equipment
100, the data processing equipment 100 is configured with trained neural network model, and the trained neural network model is logical
The leaf image and plant image for crossing multiple illness plants are trained acquisition, are functionally divided, the pest and disease damage
Identification device 110 includes obtaining module 1101 and identification module 1102.
The image for obtaining module 1101 and being used to obtain plant to be identified, the image of the plant to be identified includes blade
Image and plant image.
In the present embodiment, which is used to execute the step S100 in Fig. 2, about the acquisition module 1101
Detailed description, can refer to step S100 detailed description.
The identification module 1102 is used to the leaf image and plant image inputting the neural network model, makes
The neural network model identifies the feature of the plant to be identified, obtain the plant to be identified type and
The pest and disease damage type of the plant to be identified.
In the embodiment of the present application, which is used to execute the step S200 in Fig. 2, about the identification module
1102 detailed description can refer to the detailed description of step S200.
Optionally, the neural network model includes that floristics identification network and multiple pest species identify network,
The identification module 1102 in the following manner identifies the plant to be identified:
Identify that network knows the leaf image and plant image of the plant to be identified by the floristics
Not, the floristics to be identified is obtained；
Corresponding target pest species identification network is determined according to the floristics to be identified；
Identify network to the leaf image and plant image of the plant to be identified by the target pest species
It is identified, obtains the pest species of the plant to be identified.
Optionally, referring to figure 5., the pest and disease damage identification device 110 further includes adjustment module 1103；
The module 1101 that obtains is also used to obtain the plant for being marked with floristics label and pest species label
Sample image, the plant sample image include leaf image and plant image, and the plant sample image is inputted wait train
The neural network model be trained；
The adjustment module 1103 is used to treat the trained mind by back-propagation algorithm based on default loss function
Be iterated adjustment through network model weight, until the loss function output valve be less than preset threshold, obtain the training
Good neural network model.
The plant sample image is divided into the training sample and test sample of preset ratio, and the test sample is used for
The accuracy rate that detection passes through the trained neural network model of the training sample.
In conclusion pest and disease damage recognition methods provided by the embodiments of the present application and device, are applied to data processing equipment.It should
Data processing equipment is pre-configured with trained neural network model, wherein the neural network model passes through multiple illness plants
Leaf image and plant image be trained acquisition.The data processing equipment obtain plant to be identified leaf image and
Plant image is identified by feature of the neural network model to plant to be identified, and then obtains the plant to be identified
Type and pest and disease damage type.In this way, being improved by combining the plant image of the leaf image of plant and the plant
The discrimination of floristic discrimination and different plant illness types.
In embodiment provided herein, it should be understood that disclosed device and method, it can also be by other
Mode realize.The apparatus embodiments described above are merely exemplary, for example, the flow chart and block diagram in attached drawing are shown
According to device, the architectural framework in the cards of method and computer program product, function of multiple embodiments of the application
And operation.In this regard, each box in flowchart or block diagram can represent one of a module, section or code
Point, a part of the module, section or code includes one or more for implementing the specified logical function executable
Instruction.It should also be noted that function marked in the box can also be attached to be different from some implementations as replacement
The sequence marked in figure occurs.For example, two continuous boxes can actually be basically executed in parallel, they sometimes may be used
To execute in the opposite order, this depends on the function involved.It is also noted that each of block diagram and or flow chart
The combination of box in box and block diagram and or flow chart can be based on the defined function of execution or the dedicated of movement
The system of hardware is realized, or can be realized using a combination of dedicated hardware and computer instructions.
In addition, each functional module in each embodiment of the application can integrate one independent portion of formation together
Point, it is also possible to modules individualism, an independent part can also be integrated to form with two or more modules.
It, can be with if the function is realized and when sold or used as an independent product in the form of software function module
It is stored in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) execute each embodiment the method for the application all or part of the steps.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
The above, the only various embodiments of the application, but the protection scope of the application is not limited thereto, it is any
Those familiar with the art within the technical scope of the present application, can easily think of the change or the replacement, and should all contain
Lid is within the scope of protection of this application.Therefore, the protection scope of the application shall be subject to the protection scope of the claim.
Claims (10)
1. a kind of pest and disease damage recognition methods, which is characterized in that be applied to data processing equipment, the data processing equipment is configured with
Trained neural network model, leaf image and plant of the trained neural network model by multiple illness plants
Strain image is trained acquisition, which comprises
The image of plant to be identified is obtained, the image of the plant to be identified includes leaf image and plant image；
The image of the plant to be identified is inputted into the neural network model, so that the neural network model is to described wait know
The feature of other plant is identified, the type of the plant to be identified and the pest and disease damage type of the plant to be identified are obtained.
2. pest and disease damage recognition methods according to claim 1, which is characterized in that the neural network model includes plant species
Class identifies network and multiple pest species identify that network, the image by the plant to be identified input the neural network
Model obtains the plant to be identified so that the neural network model identifies the feature of the plant to be identified
The step of type and the pest and disease damage type of the plant to be identified includes:
It identifies that network identifies the leaf image and plant image of the plant to be identified by the floristics, obtains
Obtain the floristics to be identified；
Corresponding target pest species identification network is determined according to the floristics to be identified；
Identify that network carries out the leaf image and plant image of the plant to be identified by the target pest species
Identification, obtains the pest species of the plant to be identified.
3. pest and disease damage recognition methods according to claim 1, which is characterized in that the method also includes to the nerve net
The training step of network model:
Obtain the plant sample image for being marked with floristics label and pest species label, the plant sample image packet
Include leaf image and plant image；
The plant sample image is inputted to the neural network model to be trained；
Based on default loss function, the trained neural network model weight is treated by back-propagation algorithm and is iterated tune
It is whole, until the loss function output valve be less than preset threshold, obtain the trained neural network model.
4. pest and disease damage recognition methods according to claim 3, which is characterized in that the neural network model includes plant species
Class identifies network and multiple pest species identify network, described based on default loss function, is treated by back-propagation algorithm
The trained neural network model weight is iterated adjustment, until the loss function output valve be less than preset threshold,
The step of obtaining the trained neural network model include:
The plant sample image is inputted into the floristics and identifies network, so that the floristics identifies network according to institute
The floristics label for stating object sample image is trained；
According to the type label of the plant sample image, different types of plant sample image is inputted into different disease pests respectively
Evil category identification network, so that pest species identification network is trained according to the pest species label；
Identify that network and multiple pest species identify that network trains described in constituting by the trained floristics
Neural network model.
5. pest and disease damage recognition methods according to claim 3, which is characterized in that the plant sample image is divided into pre-
If the training sample and test sample of ratio, the test sample passes through the trained nerve net of the training sample for detecting
The accuracy rate of network model.
6. pest and disease damage recognition methods according to claim 3, which is characterized in that the plant image is from multiple shooting angles
Spend the plant panorama sketch obtained.
7. a kind of pest and disease damage identification device, which is characterized in that be applied to data processing equipment, the data processing equipment is configured with
Trained neural network model, leaf image and plant of the trained neural network model by multiple illness plants
Strain image is trained acquisition, and the pest and disease damage identification device includes obtaining module and identification module；
The module that obtains is used to obtain the image of plant to be identified, the image of the plant to be identified include leaf image and
Plant image；
The identification module is used to the leaf image and plant image inputting the neural network model, so that the mind
It is identified through feature of the network model to the plant to be identified, obtains the type of the plant to be identified and this is to be identified
The pest and disease damage type of plant.
8. pest and disease damage identification device according to claim 7, which is characterized in that the neural network model includes floristics
Identify that network and multiple pest species identify that network, the identification module in the following manner carry out the plant to be identified
Identification:
It identifies that network identifies the leaf image and plant image of the plant to be identified by the floristics, obtains
Obtain the floristics to be identified；
Corresponding target pest species identification network is determined according to the floristics to be identified；
Identify that network carries out the leaf image and plant image of the plant to be identified by the target pest species
Identification, obtains the pest species of the plant to be identified.
9. pest and disease damage identification device according to claim 7, which is characterized in that the pest and disease damage identification device further includes adjustment
Module；
The module that obtains is also used to obtain the plant sample image for being marked with floristics label and pest species label,
The plant sample image includes leaf image and plant image, and the plant sample image is inputted to the nerve to be trained
Network model；
The adjustment module is used to treat the trained neural network mould by back-propagation algorithm based on default loss function
Type weight is iterated adjustment, until the loss function output valve be less than preset threshold, obtain the trained nerve
Network model.
10. pest and disease damage identification device according to claim 9, which is characterized in that the plant sample image is divided into
The training sample and test sample of preset ratio, the test sample pass through the trained nerve of the training sample for detecting
The accuracy rate of network model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910494898.5A CN110210434A (en) | 2019-06-10 | 2019-06-10 | Pest and disease damage recognition methods and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910494898.5A CN110210434A (en) | 2019-06-10 | 2019-06-10 | Pest and disease damage recognition methods and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110210434A true CN110210434A (en) | 2019-09-06 |
Family
ID=67791522
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910494898.5A Pending CN110210434A (en) | 2019-06-10 | 2019-06-10 | Pest and disease damage recognition methods and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110210434A (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110851638A (en) * | 2019-11-06 | 2020-02-28 | 杭州睿琪软件有限公司 | Method and device for acquiring species identification name |
| CN110991454A (en) * | 2019-12-23 | 2020-04-10 | 云南大学 | Blade image recognition method and device, electronic equipment and storage medium |
| CN111340070A (en) * | 2020-02-11 | 2020-06-26 | 杭州睿琪软件有限公司 | Plant disease and insect pest diagnosis method and system |
| CN111461337A (en) * | 2020-03-05 | 2020-07-28 | 深圳追一科技有限公司 | Data processing method and device, terminal equipment and storage medium |
| CN111523457A (en) * | 2020-04-22 | 2020-08-11 | 七海行(深圳)科技有限公司 | Weed identification method and weed treatment equipment |
| CN111563439A (en) * | 2020-04-28 | 2020-08-21 | 北京海益同展信息科技有限公司 | Aquatic organism disease detection method, device and equipment |
| CN111568195A (en) * | 2020-02-29 | 2020-08-25 | 佛山市云米电器科技有限公司 | Brewed beverage identification method, device and computer-readable storage medium |
| CN111767849A (en) * | 2020-06-29 | 2020-10-13 | 京东数字科技控股有限公司 | Crop pest and disease identification method and device and storage medium |
| CN111914814A (en) * | 2020-09-01 | 2020-11-10 | 平安国际智慧城市科技股份有限公司 | Wheat rust detection method and device and computer equipment |
| CN112001370A (en) * | 2020-09-29 | 2020-11-27 | 中国农业科学院农业信息研究所 | Crop pest and disease identification method and system |
| CN112434579A (en) * | 2020-11-13 | 2021-03-02 | 深圳园林股份有限公司 | Method and device for extracting fallen leaf accommodating capacity of ground cover plant based on neural network |
| CN112465038A (en) * | 2020-11-30 | 2021-03-09 | 深圳市识农智能科技有限公司 | Method and system for identifying disease and insect pest types of fruit trees |
| CN112699805A (en) * | 2020-12-31 | 2021-04-23 | 上海市崇明区蔬菜科学技术推广站 | Intelligent recognition system for vegetable pest control |
| CN112699260A (en) * | 2019-10-22 | 2021-04-23 | 杭州睿琪软件有限公司 | Species identification method and device |
| CN112700260A (en) * | 2020-12-31 | 2021-04-23 | 上海市崇明区蔬菜科学技术推广站 | Visual traceability system for full life cycle of vegetables |
| CN112766178A (en) * | 2021-01-22 | 2021-05-07 | 广州大学 | Method, device, equipment and medium for positioning pests based on intelligent pest control system |
| CN112883873A (en) * | 2021-02-22 | 2021-06-01 | 暨南大学 | Leaf disease identification method and device, electronic device and storage medium |
| CN113095109A (en) * | 2019-12-23 | 2021-07-09 | 中移(成都)信息通信科技有限公司 | Crop leaf surface recognition model training method, recognition method and device |
| CN113627216A (en) * | 2020-05-07 | 2021-11-09 | 杭州睿琪软件有限公司 | Plant state evaluation method, system and computer readable storage medium |
| CN113780048A (en) * | 2021-01-15 | 2021-12-10 | 北京京东拓先科技有限公司 | Plant control method and device |
| CN114545833A (en) * | 2022-03-02 | 2022-05-27 | 腾圣福(广州)农业科技有限公司 | Facility agriculture intelligent interactive processing system based on internet of things technology |
| CN115294555A (en) * | 2022-09-27 | 2022-11-04 | 江苏景瑞农业科技发展有限公司 | Plant disease intelligent diagnosis method and system based on neural network |
| CN115618046A (en) * | 2022-09-05 | 2023-01-17 | 中国科学院昆明植物研究所 | Gridding scientific investigation processing method, system, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101646068A (en) * | 2009-09-07 | 2010-02-10 | 中国农业大学 | Plant diseases and insect pest information acquisition system and method |
| CN107392091A (en) * | 2017-06-09 | 2017-11-24 | 河北威远生物化工有限公司 | A kind of agriculture artificial intelligence makees object detecting method, mobile terminal and computer-readable medium |
| CN107563389A (en) * | 2017-09-11 | 2018-01-09 | 合肥工业大学 | A kind of corps diseases recognition methods based on deep learning |
| KR20180043082A (en) * | 2016-10-19 | 2018-04-27 | 한국전자통신연구원 | Plant disease detection system and method |
| CN109800806A (en) * | 2019-01-14 | 2019-05-24 | 中山大学 | A kind of corps diseases detection algorithm based on deep learning |
-
2019
- 2019-06-10 CN CN201910494898.5A patent/CN110210434A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101646068A (en) * | 2009-09-07 | 2010-02-10 | 中国农业大学 | Plant diseases and insect pest information acquisition system and method |
| KR20180043082A (en) * | 2016-10-19 | 2018-04-27 | 한국전자통신연구원 | Plant disease detection system and method |
| CN107392091A (en) * | 2017-06-09 | 2017-11-24 | 河北威远生物化工有限公司 | A kind of agriculture artificial intelligence makees object detecting method, mobile terminal and computer-readable medium |
| CN107563389A (en) * | 2017-09-11 | 2018-01-09 | 合肥工业大学 | A kind of corps diseases recognition methods based on deep learning |
| CN109800806A (en) * | 2019-01-14 | 2019-05-24 | 中山大学 | A kind of corps diseases detection algorithm based on deep learning |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112699260A (en) * | 2019-10-22 | 2021-04-23 | 杭州睿琪软件有限公司 | Species identification method and device |
| CN110851638A (en) * | 2019-11-06 | 2020-02-28 | 杭州睿琪软件有限公司 | Method and device for acquiring species identification name |
| CN113095109A (en) * | 2019-12-23 | 2021-07-09 | 中移(成都)信息通信科技有限公司 | Crop leaf surface recognition model training method, recognition method and device |
| CN110991454A (en) * | 2019-12-23 | 2020-04-10 | 云南大学 | Blade image recognition method and device, electronic equipment and storage medium |
| CN111340070A (en) * | 2020-02-11 | 2020-06-26 | 杭州睿琪软件有限公司 | Plant disease and insect pest diagnosis method and system |
| CN111340070B (en) * | 2020-02-11 | 2024-03-26 | 杭州睿琪软件有限公司 | Plant pest diagnosis method and system |
| CN111568195A (en) * | 2020-02-29 | 2020-08-25 | 佛山市云米电器科技有限公司 | Brewed beverage identification method, device and computer-readable storage medium |
| CN111461337A (en) * | 2020-03-05 | 2020-07-28 | 深圳追一科技有限公司 | Data processing method and device, terminal equipment and storage medium |
| CN111461337B (en) * | 2020-03-05 | 2023-08-18 | 深圳追一科技有限公司 | Data processing method, device, terminal equipment and storage medium |
| CN111523457B (en) * | 2020-04-22 | 2023-09-12 | 七海行(深圳)科技有限公司 | Weed identification method and weed treatment equipment |
| CN111523457A (en) * | 2020-04-22 | 2020-08-11 | 七海行(深圳)科技有限公司 | Weed identification method and weed treatment equipment |
| CN111563439B (en) * | 2020-04-28 | 2023-08-08 | 京东科技信息技术有限公司 | Aquatic organism disease detection method, device and equipment |
| CN111563439A (en) * | 2020-04-28 | 2020-08-21 | 北京海益同展信息科技有限公司 | Aquatic organism disease detection method, device and equipment |
| CN113627216B (en) * | 2020-05-07 | 2024-02-27 | 杭州睿琪软件有限公司 | Plant state evaluation method, system and computer readable storage medium |
| CN113627216A (en) * | 2020-05-07 | 2021-11-09 | 杭州睿琪软件有限公司 | Plant state evaluation method, system and computer readable storage medium |
| CN111767849A (en) * | 2020-06-29 | 2020-10-13 | 京东数字科技控股有限公司 | Crop pest and disease identification method and device and storage medium |
| CN111914814A (en) * | 2020-09-01 | 2020-11-10 | 平安国际智慧城市科技股份有限公司 | Wheat rust detection method and device and computer equipment |
| CN112001370A (en) * | 2020-09-29 | 2020-11-27 | 中国农业科学院农业信息研究所 | Crop pest and disease identification method and system |
| CN112434579A (en) * | 2020-11-13 | 2021-03-02 | 深圳园林股份有限公司 | Method and device for extracting fallen leaf accommodating capacity of ground cover plant based on neural network |
| CN112465038A (en) * | 2020-11-30 | 2021-03-09 | 深圳市识农智能科技有限公司 | Method and system for identifying disease and insect pest types of fruit trees |
| CN112700260A (en) * | 2020-12-31 | 2021-04-23 | 上海市崇明区蔬菜科学技术推广站 | Visual traceability system for full life cycle of vegetables |
| CN112699805A (en) * | 2020-12-31 | 2021-04-23 | 上海市崇明区蔬菜科学技术推广站 | Intelligent recognition system for vegetable pest control |
| CN113780048A (en) * | 2021-01-15 | 2021-12-10 | 北京京东拓先科技有限公司 | Plant control method and device |
| CN112766178B (en) * | 2021-01-22 | 2023-06-23 | 广州大学 | Disease and pest positioning method, device, equipment and medium based on intelligent deinsectization system |
| CN112766178A (en) * | 2021-01-22 | 2021-05-07 | 广州大学 | Method, device, equipment and medium for positioning pests based on intelligent pest control system |
| CN112883873B (en) * | 2021-02-22 | 2023-08-01 | 暨南大学 | Leaf disease identification method, device, electronic device and storage medium |
| CN112883873A (en) * | 2021-02-22 | 2021-06-01 | 暨南大学 | Leaf disease identification method and device, electronic device and storage medium |
| CN114545833B (en) * | 2022-03-02 | 2023-05-02 | 深圳市丰农数智农业科技有限公司 | Intelligent interactive processing system of facility agriculture based on internet of things |
| CN114545833A (en) * | 2022-03-02 | 2022-05-27 | 腾圣福(广州)农业科技有限公司 | Facility agriculture intelligent interactive processing system based on internet of things technology |
| CN115618046A (en) * | 2022-09-05 | 2023-01-17 | 中国科学院昆明植物研究所 | Gridding scientific investigation processing method, system, equipment and storage medium |
| CN115294555A (en) * | 2022-09-27 | 2022-11-04 | 江苏景瑞农业科技发展有限公司 | Plant disease intelligent diagnosis method and system based on neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110210434A (en) | Pest and disease damage recognition methods and device | |
| Mishra et al. | A Deep Learning-Based Novel Approach for Weed Growth Estimation. | |
| Murad et al. | Weed detection using deep learning: A systematic literature review | |
| Pujari et al. | Automatic fungal disease detection based on wavelet feature extraction and PCA analysis in commercial crops | |
| Bhagat et al. | A comprehensive survey on leaf disease identification & classification | |
| Sharma et al. | A New Dawn for Tomato-spotted wilt virus Detection and Intensity Classification: A CNN and LSTM Ensemble Model | |
| Prakash et al. | Design and development of Android based Plant disease detection using Arduino | |
| Liu et al. | Development of a mobile application for identification of grapevine (Vitis vinifera L.) cultivars via deep learning | |
| Shafik et al. | Using a novel convolutional neural network for plant pests detection and disease classification | |
| Sachan et al. | Paddy leaf disease detection using thermal images and convolutional neural networks | |
| Kaur et al. | Wheat leaf disease classification using EfficientNet B3 pre-trained architecture | |
| Sharma et al. | A Survey on the State of Art Approaches for Disease Detection in Plants | |
| Ahmed et al. | Detection and classification of plant diseases in leaves through machine learning | |
| Mishra et al. | A robust pest identification system using morphological analysis in neural networks | |
| YASIN et al. | Deep learning-based classification of black gram plant leaf diseases: A comparative study | |
| Singh et al. | Tomato crop disease classification using convolution neural network and transfer learning | |
| Vignesh et al. | Identification of unhealthy leaves in paddy by using computer vision based deep learning model | |
| Jadhav et al. | Comprehensive review on machine learning for plant disease identification and classification with image processing | |
| Qayyum et al. | Optimal feature extraction technique for crop classification using aerial imagery | |
| Li et al. | Identification of Crop Diseases Based on Improved Genetic Algorithm and Extreme Learning Machine. | |
| Banerjee et al. | Enhancing Snake Plant Disease Classification through CNN-Random Forest Integration | |
| Roy et al. | Disease Detection in Paddy Crop using Machine Learning Techniques | |
| Fida et al. | Leaf image recognition based identification of plants: Supportive framework for plant systematics | |
| Karthik et al. | Application for Plant’s Leaf Disease Detection using Deep Learning Techniques | |
| CN109444360B (en) | Fruit juice storage period detection algorithm based on cellular neural network and electronic nose feature extraction |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190906 |