CN107886077A - A kind of crop pests recognition methods and its system based on wechat public number - Google Patents
A kind of crop pests recognition methods and its system based on wechat public number Download PDFInfo
- Publication number
- CN107886077A CN107886077A CN201711145172.8A CN201711145172A CN107886077A CN 107886077 A CN107886077 A CN 107886077A CN 201711145172 A CN201711145172 A CN 201711145172A CN 107886077 A CN107886077 A CN 107886077A
- Authority
- CN
- China
- Prior art keywords
- image
- public number
- wechat public
- crop pests
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/35—Categorising the entire scene, e.g. birthday party or wedding scene
- G06V20/38—Outdoor scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/04—Real-time or near real-time messaging, e.g. instant messaging [IM]
- H04L51/046—Interoperability with other network applications or services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Biology (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Catching Or Destruction (AREA)
Abstract
The invention discloses a kind of method of work of the crop pests identification based on wechat public number, specific works method are as follows:S1. image uploads, and peasant household's concern wechat public number, adds pest and disease damage photo, image is uploaded into wechat public number;S2. image information pre-processes, and server receives image information and to the pretreatment of crop pests image, takes out the great quantities of spare interference information of original image, retains or strengthen the characteristic information of identification main body, and then obtains intermediate image;S3. image characteristics extraction, each point of the image to receiving carry out computing, detect the point and belong to any in color characteristic, textural characteristics and morphological feature;S4. image recognition is classified, and is matched according to the characteristics of image extracted with the pest and disease damage image of server end;S5. the image for the disease and insect that wechat platform end will identify that passes to user.Do not limited by time place, thoroughly solve a variety of deficiencies of plant protection expert limited amount and manual identified, the preventing and treating of diseases and pests of agronomic crop has critically important theory significance and practical value.
Description
Technical field
The present invention relates to agricultural technology field, and in particular to a kind of crop pests recognition methods based on wechat public number
And its system.
Background technology
The artificial stage is rested on mostly for the diagnosis of current China's crop pests, and there is objectivity is poor, efficiency is low, labor
Many weak points, this patent be present in the problems such as fatigue resistance is big, and the existing crop pests diagnosis based on image recognition
Systematic research has been carried out to the identifying pest system core theoretical algorithm based on image recognition and practical application.
The content of the invention
The purpose of the present invention is to overcome above mentioned problem, proposes a kind of crop pests identification side based on wechat public number
Method and its system.When peasant household runs into unknown insect, the on-the-spot instruction of plant disease pest diagnostician is not required to, is taken with a mobile phone
Pest and disease damage image, wechat public number is sent the image to, is transferred to server end, insect identification is carried out in server end, " to disappear
The form of breath " is supplied to user, and provides corresponding agricultural pests essential information and prophylactico-therapeutic measures.
Technical scheme:A kind of crop pests recognition methods based on wechat public number described in technical scheme,
Specific works method is as follows:
S1. image uploads, and peasant household's concern wechat public number, adds pest and disease damage photo, image is uploaded into wechat public number;
S2. image information pre-processes, and server receives image information and to the pretreatment of crop pests image, takes out
The great quantities of spare interference information of original image, retains or strengthens the characteristic information of identification main body, and then obtains intermediate image;
S3. image characteristics extraction, computing is carried out to each point of image received, detect the point belong to color characteristic,
It is any in textural characteristics and morphological feature;
S4. image recognition is classified, and is matched according to the characteristics of image extracted with the pest and disease damage image of server end;
S5. the image for the disease and insect that wechat platform end will identify that passes to user.
Further, preprocessing process is smoothing denoising, gray processing processing, image enhaucament and segmentation portion in the step S2
Reason.
Further, the method for image characteristics extraction is that color characteristic, texture first to image are special in the step S3
Morphological feature of seeking peace is extracted, and three feature vectors extracted are carried out into Fusion Features.Three feature vectors merge, and take
Long benefit is short, the respective advantage of competence exertion, evades defect, so as to truely and accurately express the feature of image, is optimized
Image recognition rate.
Further, the principle that crop pests image data set is built in the step S4 is as follows:According to farming species
Class, insect metamorphosis, insect difference posture, insect different shooting angles are built, and set unified background and big
Small-format.
Further, the image recognition classification in the step S4 includes two stages of sample training and image recognition.
Further, convolutional neural networks model is passed through in the pest and disease damage images match of server end in the step S4
Realize image classification identification mission, convolutional neural networks model specific works method is as follows:
(1) image of unified size is read in first, using image as input data;
(2) view data is sent into convolutional layer, view data and 6 filter convolution is formed into 6 characteristic patterns, obtain C1
Layer;
(3) down-sampled layer is sent into C1 layers, the down-sampled operation that 2*2 to 1 is carried out to C1 layers obtains S2 layers,;
(4) process of repetition above-mentioned (2)-(3), proper C5 layers, 120 1*1 characteristic pattern is included, then by 120
Characteristic pattern travels to 10*1 output layer by way of connecting entirely.
Further, the convolutional layer and down-sampled layer are alternately distributed connection, form " double pyramids " structure.
A kind of such as crop pests identifying system based on wechat public number, including wechat public number front end, wechat backstage
Public number is added by wechat client with wechat public number server user, establishes and communicates to connect with wechat public number front end,
Pest and disease damage photo is added by wechat public number front end, the pest and disease damage photo that user asks to differentiate is forwarded to wechat by wechat backstage
Public number server, wechat public number server carry out image recognition, and result is fed back to wechat backstage in the form of a message,
Wechat backstage relays to wechat public number front end.
Above-mentioned technical proposal can be seen that beneficial effects of the present invention:
(1) a kind of crop pests recognition methods based on wechat public number of the present invention, peasant household can pay close attention to micro-
Believe public number, add pest and disease damage photo, the picture to be added can be added from wechat photograph album and can also carried by mobile terminal
Camera shooting field picture.Image can carry out image recognition, in the form of " message " through wechat background transfer to server end
User is supplied to, and corresponding agricultural pests essential information and prophylactico-therapeutic measures are provided.
(2) a kind of crop pests identifying system based on wechat public number of the present invention realizes the reality of pest and disease damage
When automatic identification, this pest and disease damage identifying system can be widely applied without being limited by time place, thoroughly solved
A variety of deficiencies of plant protection expert limited amount and manual identified, the preventing and treating of diseases and pests of agronomic crop have critically important theory significance
And practical value, also there is very wide application prospect.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the present invention;
Fig. 2 is that the image of the crop pests identifying system of the present invention interacts schematic diagram.
Embodiment
In order that the technical means, the inventive features, the objects and the advantages of the present invention are easy to understand, tie below
Diagram is closed, above-mentioned technical proposal is expanded on further.
A kind of crop pests recognition methods based on wechat public number as shown in Figure 1-2, specific works method are as follows:
S1. image uploads, and peasant household's concern wechat public number, adds pest and disease damage photo, image is uploaded into wechat public number;
S2. image information pre-processes, and server receives image information and to the pretreatment of crop pests image, takes out
The great quantities of spare interference information of original image, retains or strengthens the characteristic information of identification main body, and then obtains intermediate image;
S3. image characteristics extraction, computing is carried out to each point of image received, detect the point belong to color characteristic,
It is any in textural characteristics and morphological feature;
S4. image recognition is classified, and is matched according to the characteristics of image extracted with the pest and disease damage image of server end;
S5. the image for the disease and insect that wechat platform end will identify that passes to user.
Preprocessing process is smoothing denoising, gray processing processing, image enhaucament and segmentation in step S2 described in the present embodiment
Processing.
The method of image characteristics extraction is that color characteristic, texture first to image are special in step S3 described in the present embodiment
Morphological feature of seeking peace is extracted, and three feature vectors extracted are carried out into Fusion Features.The global feature right and wrong of image
It is often complicated and changeable, the color of image, texture, morphological feature while having its own advantage, and inevitably because from
The speciality of the attribute of body feature, there is respective limitation.In image recognition, target is entered if only using single features
Row identification, will necessarily have certain error, it is impossible to obtain preferable recognition result, only melted three feature vectors
Close, learn from other's strong points to offset one's weaknesses, the respective advantage of competence exertion, evade the defects of respective, so as to truely and accurately express the feature of image, obtain
The image recognition rate of optimization.It will be carried out in the research of this project using the color, texture, the morphological feature parameter that consider image
Multiple features fusion.
Using a key issue of above-mentioned Feature-level fusion method, it is how to determine the weights of these features, this
In embodiment weights are sought using Fisher linear discriminant methods.Fisher linear discriminants are a kind of linear classification methods, and it is in mould
The fields such as formula classification, data analysis, dimensionality reduction are widely used.
If set pattern x has c pattern class, S in n-dimensional spacebThe scatter matrix between training sample class, SwFor training sample
Scatter matrix in class, S are training sample total population scatter matrix, are nonnegative definite matrix, and S=Sb+Sw, then Fisher criterion
Function is defined as
Jf(x) value is the bigger the better, that is, inter _ class relationship and within-cluster variance change value it is maximum when.
The principle that crop pests image data set is built in step S4 described in the present embodiment is as follows:According to farming species
Class, insect metamorphosis, insect difference posture, insect different shooting angles are built, and set unified background and big
Small-format.
Image recognition classification in step S4 described in the present embodiment includes two stages of sample training and image recognition.
In step S4 described in the present embodiment in the pest and disease damage images match of server end, pass through convolutional neural networks model
Realize image classification identification mission, convolutional neural networks model specific works method is as follows:
(1) image of unified size is read in first, using image as input data;
(2) view data is sent into convolutional layer, view data and 6 filter convolution is formed into 6 characteristic patterns, obtain C1
Layer;
(3) down-sampled layer is sent into C1 layers, the down-sampled operation that 2*2 to 1 is carried out to C1 layers obtains S2 layers,;
(4) process of repetition above-mentioned (2)-(3), proper C5 layers, 120 1*1 characteristic pattern is included, then by 120
Characteristic pattern travels to 10*1 output layer by way of connecting entirely.
The operation principle of convolutional neural networks is as follows:Input layer reads in the picture by simple rule (same size),
Unit in each layer passes through local sensing field, neuron using the small local neighbor unit of one group in preceding layer as input
Extract basic visual signature;
By the visual signature of extraction used in the neuron of higher, by using these features, convolutional neural networks profit
Image diverse location is distributed in a component but there is the unit of identical weight vector, to obtain the feature of image and form width spy
Figure is levied in each position, the respective different types of feature that the unit from different characteristic figure obtains, in a secondary characteristic pattern
Different units are restricted to carry out same operation to the local data of each diverse location in input figure, and being equal to input
Image carries out convolution for a small core, forms convolutional layer.Generally comprised in one convolutional layer it is multiple have different weights to
The characteristic pattern of amount so that can be obtained in same position a variety of different features to after convolution image carry out local average and
Down-sampled operation forms down-sampled layer, and the resolution ratio and network for reducing characteristic pattern are exported for position and the sensitivity deformed.
Convolutional layer described in the present embodiment and down-sampled layer are alternately distributed connection, form " double pyramids " structure.
A kind of crop pests identifying system based on wechat public number in the present embodiment, including wechat public number front end 1,
Wechat backstage 2 and wechat public number server 3, user add public number by wechat client, built with wechat public number front end 1
Vertical communication connection, pest and disease damage photo is added by wechat public number front end, the pest and disease damage that user asks to differentiate is shone on wechat backstage 2
Piece is forwarded to wechat public number server 3, and wechat public number server 3 carries out image recognition, by result in the form of a message
Wechat backstage 2 is fed back to, wechat backstage 2 relays to wechat public number front end 1.
The present invention combines plant pathology knowledge, computer networking technology, image processing techniques and mode identification technology, utilizes
Wechat platform, the image recognition technology of disease pest and weed is worked as by peasant household using on the mobile apparatus by the exploitation of wechat public number
When running into unknown insect, the on-the-spot instruction of plant disease pest diagnostician is not required to, pest and disease damage image is taken with a mobile phone, sends figure
As giving wechat public number, server end is transferred to, insect identification is carried out in server end, use is supplied in the form of " message "
Family, and corresponding agricultural pests essential information and prophylactico-therapeutic measures are provided.I.e. peasant household need to only be diagnosed by way of mobile phone " taking pictures "
Go out agricultural pest, it is simple to operate, real-time, cost is low.
Embodiment is only illustrative of the invention and is not intended to limit the scope of the invention, after the present invention has been read, this
Art personnel fall within the application appended claims limited range to the various equivalences of the present invention.
Claims (8)
- A kind of 1. crop pests recognition methods based on wechat public number, it is characterised in that:Specific works method is as follows:S1. image uploads, and peasant household's concern wechat public number, adds pest and disease damage photo, image is uploaded into wechat public number;S2. image information pre-processes, and server receives image information and to the pretreatment of crop pests image, takes out original The great quantities of spare interference information of image, retains or strengthens the characteristic information of identification main body, and then obtains intermediate image;S3. image characteristics extraction, each point of the image to receiving carry out computing, detect the point and belong to color characteristic, texture It is any in feature and morphological feature;S4. image recognition is classified, and is matched according to the characteristics of image extracted with the pest and disease damage image of server end;S5. the image for the disease and insect that wechat platform end will identify that passes to user.
- A kind of 2. crop pests recognition methods based on wechat public number according to claim 1, it is characterised in that:Institute It is smoothing denoising, gray processing processing, image enhaucament and dividing processing to state preprocessing process in step S2.
- A kind of 3. crop pests recognition methods based on wechat public number according to claim 2, it is characterised in that:Institute The method for stating image characteristics extraction in step S3 is that the color characteristic, textural characteristics and morphological feature of image are carried first Take, three feature vectors extracted are subjected to Fusion Features.
- A kind of 4. crop pests recognition methods based on wechat public number according to claim 3, it is characterised in that:Institute It is as follows to state the principle that crop pests image data set is built in step S4:According to crop specie, insect metamorphosis, insect Different postures, insect different shooting angles are built, and set unified background and format size.
- A kind of 5. crop pests recognition methods based on wechat public number according to claim 1, it is characterised in that:Institute Stating the classification of the image recognition in step S4 includes two stages of sample training and image recognition.
- A kind of 6. crop pests recognition methods based on wechat public number according to claim 1, it is characterised in that:Institute In the pest and disease damage images match for stating server end in step S4, appointed by the identification of convolutional neural networks model realization image classification Business, convolutional neural networks model specific works method are as follows:(1)The image of unified size is read in first, using image as input data;(2)View data is sent into convolutional layer, view data and 6 filter convolution are formed into 6 characteristic patterns, obtain C1 layers;(3)Down-sampled layer is sent into C1 layers, the down-sampled operation that 2*2 to 1 is carried out to C1 layers obtains S2 layers,;(4)Repeat above-mentioned(2)-(3)Process, proper C5 layers, 120 1*1 characteristic pattern is included, then by the feature of 120 Figure travels to 10*1 output layer by way of connecting entirely.
- A kind of 7. crop pests recognition methods based on wechat public number according to claim 6, it is characterised in that:Institute State convolutional layer and down-sampled layer is alternately distributed connection, form " double pyramids " structure.
- A kind of 8. crop pests identifying system based on wechat public number, including wechat public number as claimed in claim 1 Front end(1), wechat backstage(2)With wechat public number server(3), user adds public number by wechat client, with wechat Public number front end(1)Communication connection is established, pest and disease damage photo, wechat backstage are added by wechat public number front end(2)By user The pest and disease damage photo that request differentiates is forwarded to wechat public number server(3), wechat public number server(3)Carry out image knowledge Not, result is fed back to wechat backstage in the form of a message(2), wechat backstage(2)Relay to wechat public number front end (1).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711145172.8A CN107886077A (en) | 2017-11-17 | 2017-11-17 | A kind of crop pests recognition methods and its system based on wechat public number |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711145172.8A CN107886077A (en) | 2017-11-17 | 2017-11-17 | A kind of crop pests recognition methods and its system based on wechat public number |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107886077A true CN107886077A (en) | 2018-04-06 |
Family
ID=61777868
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711145172.8A Pending CN107886077A (en) | 2017-11-17 | 2017-11-17 | A kind of crop pests recognition methods and its system based on wechat public number |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107886077A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108921814A (en) * | 2018-05-16 | 2018-11-30 | 华南农业大学 | A kind of Citrus Huanglongbing pathogen on-line quick detection system and method based on deep learning |
CN109325499A (en) * | 2018-08-02 | 2019-02-12 | 浙江中农在线电子商务有限公司 | Pest and disease damage recognition methods and device |
CN110070071A (en) * | 2019-04-30 | 2019-07-30 | 重庆文理学院 | A kind of ginger cauline leaf pest and disease damage recognition methods and system based on image recognition |
CN112036470A (en) * | 2020-08-28 | 2020-12-04 | 扬州大学 | Cloud transmission-based multi-sensor fusion cucumber bemisia tabaci identification method |
CN112237087A (en) * | 2019-07-19 | 2021-01-19 | 迪尔公司 | Crop residue based field work adjustment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103903006A (en) * | 2014-03-05 | 2014-07-02 | 中国科学院合肥物质科学研究院 | Crop pest identification method and system based on Android platform |
CN105069622A (en) * | 2015-08-03 | 2015-11-18 | 福州海景科技开发有限公司 | Face identification payment system facing mobile terminal and face identification payment method facing mobile terminal |
WO2016026447A1 (en) * | 2014-08-19 | 2016-02-25 | 北京奇虎科技有限公司 | Alarm method of intelligent camera system, intelligent camera system, and network camera |
CN107153840A (en) * | 2017-04-21 | 2017-09-12 | 防城港市丰禾曲辰农业科技有限公司 | A kind of crop pests image-recognizing method based on convolutional Neural |
-
2017
- 2017-11-17 CN CN201711145172.8A patent/CN107886077A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103903006A (en) * | 2014-03-05 | 2014-07-02 | 中国科学院合肥物质科学研究院 | Crop pest identification method and system based on Android platform |
WO2016026447A1 (en) * | 2014-08-19 | 2016-02-25 | 北京奇虎科技有限公司 | Alarm method of intelligent camera system, intelligent camera system, and network camera |
CN105069622A (en) * | 2015-08-03 | 2015-11-18 | 福州海景科技开发有限公司 | Face identification payment system facing mobile terminal and face identification payment method facing mobile terminal |
CN107153840A (en) * | 2017-04-21 | 2017-09-12 | 防城港市丰禾曲辰农业科技有限公司 | A kind of crop pests image-recognizing method based on convolutional Neural |
Non-Patent Citations (2)
Title |
---|
JIANHUA XU ET AL.: "Applications of mobile social media:wechat among academic libraries in china", 《THE JOURNAL OF ACADEMIC LIBRARIANSHIP》 * |
张洁: "基于图像识别的农作物害虫诊断技术研究与应用", 《中国博士学位论文全文数据库 农业科技辑》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108921814A (en) * | 2018-05-16 | 2018-11-30 | 华南农业大学 | A kind of Citrus Huanglongbing pathogen on-line quick detection system and method based on deep learning |
CN109325499A (en) * | 2018-08-02 | 2019-02-12 | 浙江中农在线电子商务有限公司 | Pest and disease damage recognition methods and device |
CN110070071A (en) * | 2019-04-30 | 2019-07-30 | 重庆文理学院 | A kind of ginger cauline leaf pest and disease damage recognition methods and system based on image recognition |
CN112237087A (en) * | 2019-07-19 | 2021-01-19 | 迪尔公司 | Crop residue based field work adjustment |
CN112036470A (en) * | 2020-08-28 | 2020-12-04 | 扬州大学 | Cloud transmission-based multi-sensor fusion cucumber bemisia tabaci identification method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107886077A (en) | A kind of crop pests recognition methods and its system based on wechat public number | |
Pan et al. | A smart mobile diagnosis system for citrus diseases based on densely connected convolutional networks | |
CN106971152B (en) | Method for detecting bird nest in power transmission line based on aerial images | |
CN104992223B (en) | Intensive population estimation method based on deep learning | |
Khan et al. | Deep learning for apple diseases: classification and identification | |
CN106844614A (en) | A kind of floor plan functional area system for rapidly identifying | |
CN106780466A (en) | A kind of cervical cell image-recognizing method based on convolutional neural networks | |
CN106548159A (en) | Reticulate pattern facial image recognition method and device based on full convolutional neural networks | |
CN109145759A (en) | Vehicle attribute recognition methods, device, server and storage medium | |
CN107239514A (en) | A kind of plants identification method and system based on convolutional neural networks | |
CN110046631A (en) | System and method for inferring the variation of time-space image automatically | |
Shi et al. | Recent advances in plant disease severity assessment using convolutional neural networks | |
CN110929687B (en) | Multi-user behavior recognition system based on key point detection and working method | |
CN111127423B (en) | Rice pest and disease identification method based on CNN-BP neural network algorithm | |
CN114359727A (en) | Tea disease identification method and system based on lightweight optimization Yolo v4 | |
Alkhudaydi et al. | Spikeletfcn: Counting spikelets from infield wheat crop images using fully convolutional networks | |
CN108764018A (en) | A kind of multitask vehicle based on convolutional neural networks recognition methods and device again | |
Öztürk et al. | Transfer learning and fine‐tuned transfer learning methods' effectiveness analyse in the CNN‐based deep learning models | |
CN104063686A (en) | System and method for performing interactive diagnosis on crop leaf segment disease images | |
CN108228684A (en) | Training method, device, electronic equipment and the computer storage media of Clustering Model | |
CN111046213B (en) | Knowledge base construction method based on image recognition | |
CN104899604B (en) | A kind of feature level fusing method merged based on data set | |
Rai et al. | Classification of diseased cotton leaves and plants using improved deep convolutional neural network | |
CN113361928B (en) | Crowd-sourced task recommendation method based on heterogram attention network | |
Praveen et al. | To detect plant disease identification on leaf using machine learning algorithms |
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 |
Application publication date: 20180406 |
|
RJ01 | Rejection of invention patent application after publication |