CN104680524B - A kind of leafy vegetable disease screening method - Google Patents
A kind of leafy vegetable disease screening method Download PDFInfo
- Publication number
- CN104680524B CN104680524B CN201510073648.6A CN201510073648A CN104680524B CN 104680524 B CN104680524 B CN 104680524B CN 201510073648 A CN201510073648 A CN 201510073648A CN 104680524 B CN104680524 B CN 104680524B
- Authority
- CN
- China
- Prior art keywords
- disease
- leafy vegetable
- image
- information
- texture
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/095—Traffic lights
Abstract
The present invention discloses a kind of leafy vegetable disease screening method, and method includes:S1:Noise reduction process is carried out to leafy vegetable blade face image, obtains the first tone images of leafy vegetable blade face image;S2:Color feature extracted is carried out to the first tone images, obtains the characteristic information value and the second tone images of leafy vegetable blade face image;S3:Texture feature extraction is carried out to the second tone images, obtains the texture eigenvalue of the second tone images;S4:Calculate the mean value of the texture eigenvalue of all pictures in preset leafy vegetable disease picture library;S5:According to the mean value of texture eigenvalue, disease threshold point is obtained;S6:According to preset discriminant function and disease threshold point, the characteristic information value of leafy vegetable blade face image is diagnosed as disease leafy vegetable for linearly related leafy vegetable.The leafy vegetable disease screening method of the present invention is merged by disease professional knowledge with the more preferable of computer technology, and realization more rapidly, accurately diagnoses leafy vegetable disease by image processing and pattern recognition.
Description
Technical field
The present invention relates to leafy vegetable disease screening technical fields, and in particular to a kind of leaf class based on blade face image symptom
Vegetable disease diagnostic method.
Background technology
Leafy vegetable is in process of production often since disease occurs for the factors such as weather, environment, and previous peasant household is due to lacking
The reasons such as disease knowledge often occur failing to pinpoint a disease in diagnosis and mistaken diagnosis, affect disease treatment adversely, the extensive underproduction of leafy vegetable, product will be caused when serious
Matter declines, and restricts leafy vegetable productivity effect.And improve crop disease control, increase economic well-being of workers and staff depend on it is timely to disease
With correct diagnosis.According to incompletely statistics, the first visit accuracy of some common diseases is less than 60%, as relatively rare disease
Disease, the accuracy of first visit are lower.Thus, disease recognition is leafy vegetable diagnosis prevention as the important step of disease screening
Key with basis.
In the past, peasant relied primarily on that eye is seen, experience identifies, judges crop disease feelings in the actual production process
Condition, thus Activities of Some Plants expert also went out many relevant diagnosis catalogues.But this method is limited to peasant itself to disease
Recognition capability, and many diseases are often just diagnosed when occurred frequently as a result, affect treatment adversely, largely effect on the receipts of crop
Into.In recent years, with image processing techniques and the gradual perfection of Development of IT Application in Agriculture, domestic and foreign scholars start to get down to extensively
Application image treatment technology has carried out the research of plant disease identifying and diagnosing.Many correlation scholars have carried out many researchs, such as
Okamoto, T etc. are using the processing of image with identification come disease (Sasaki Y, Okamoto T, the Imou K, et of diagnosis of plant
al.Automatic diagnosis of plant disease:Recognition between healthy and
diseased leaf[J].Journal of the Japanese society of agricultural machinery,
1999,61), Cao Liying etc. has carried out the identification of maize diseases with image processing techniques with BP neural network algorithm and diagnosis is ground
Study carefully (Cao Liying, Zhang Xiaoxian, umbrella dawn brightness, wait maize diseases diagnostic methods of the based on image processing techniques and BP neural network algorithm
Research [J] computer science, 2012,39 (10):300-302.).It can be seen that use image procossing and pattern-recognition skill
Art can effectively diagnose disease, and many experts have carried out the verification of many methods with scholar also by experiment.
However, in the accuracy and agility of leafy vegetable disease screening, presently used method still has limitation, many
Method fails to be combined well with domain knowledge so that also with failing to pinpoint a disease in diagnosis and mistaken diagnosis during leafy vegetable disease screening
Situation.
Invention content
The technical problems to be solved by the invention are the institutes at present in the accuracy and agility of leafy vegetable disease screening
Method still has limitation, and many methods fail to be combined well with domain knowledge so that in leafy vegetable disease screening
In the process also with failing to pinpoint a disease in diagnosis and the problem of mistaken diagnosis.
For this purpose, the present invention proposes a kind of leafy vegetable disease screening method, the method includes:
S1:Noise reduction process is carried out to leafy vegetable blade face image, obtains the first tone of leafy vegetable blade face image
Image;
S2:Color feature extracted is carried out to first tone images, obtains the feature of leafy vegetable blade face image
The value of information and the second tone images;
S3:Texture feature extraction is carried out to second tone images, obtains the textural characteristics of second tone images
Value;
S4:Calculate the mean value of the texture eigenvalue of all pictures in preset leafy vegetable disease picture library;
S5:According to the mean value of the texture eigenvalue, disease threshold point is obtained;
S6:According to preset discriminant function and the disease threshold point, by the feature of leafy vegetable blade face image
The value of information is diagnosed as disease leafy vegetable for linearly related leafy vegetable.
Optionally, before the step S1, the method further includes following preprocessing process:
According to preset leafy vegetable disease screening knowledge base and preset leafy vegetable disease picture library, obtained leaf class
Vegetable disease mark sheet, the leafy vegetable Disease Characters table include disease title and Damage Types, the Damage Types packet
It includes:Disease shape, disease color, disease size, Disease Characters and disease sprawling mode, the Disease Characters are according to the leaf
Class vegetable disease diagnostic knowledge base determines.
Optionally, the step S1 is specifically included:
According to preset additive mixture RGB models, color feature extracted is carried out to leafy vegetable image, obtains the leaf
Red R information, green G information, the blue B information of class vegetables image;
It, will be described red according to preset hexagonal pyramid HSV models and the conversion formula of preset additive mixture RGB models
Color R, green G, blue B information are converted into tone H information, saturation degree S information and brightness V information;
The tone H information is averagely divided into 16 gradients, the leaf class vegetable is drawn according to 16 gradients display
First tone images of dish image.
Optionally, in the step S3, the texture eigenvalue of second tone images is by solving gray scale symbiosis
Matrix obtains, and the angular second moments of the gray level co-occurrence matrixes is evenly distributed degree and texture fineness degree for gradation of image, the ash
The moment of inertia of degree co-occurrence matrix be local gray level correlation in image, the gray level co-occurrence matrixes correlation COR be, it is described
The unfavourable balance of gray level co-occurrence matrixes is believed away from the homogeney that L is image texture, the entropy of the gray level co-occurrence matrixes possessed by image
The measurement of breath amount.
Optionally, in the step S3, the method further includes the screening steps to the texture eigenvalue:
According to the correlation of the texture eigenvalue, the texture eigenvalue of extraction is selected.If Data Representation outlet
Property can divide, and select;It is abandoned if the Data Representation is cutting edge aligned inseparable or linear separability is not strong;
The image texture characteristic value obtained input decision function will be selected to carry out corresponding parameter calculating.
Optionally, the step S5 is specifically included:
S5.1 calculates the projection of sample in each training set
S5.2 calculates the positive definite matrix L of sampleμ
S5.3 is according to core FISHER method of discrimination, computational discrimination demarcation threshold point y0。
Optionally, the step S6 is specifically included:
Decision Classfication is carried out according to the discriminant function that S5 is solved and threshold point, if numerical value meets linear correlation, is made a definite diagnosis
For corresponding disease;If conditions are not met, it then needs that judgement is identified again according to the comparison of calculated value and illness specific features value.
Compared with the prior art, leafy vegetable disease screening method of the invention passes through disease professional knowledge and computer skill
The more preferable fusion of art, realization more rapidly, accurately examine leafy vegetable disease by image processing and pattern recognition
It is disconnected.
Description of the drawings
Fig. 1 shows a kind of leafy vegetable disease screening method flow diagram;
Fig. 2 shows a kind of leafy vegetable disease geo-radar image characteristic processing flow charts;
Fig. 3 shows a kind of flow chart for differentiating leafy vegetable disease.
Specific embodiment
Purpose, technical scheme and advantage to make the embodiment of the present invention are clearer, below in conjunction with the embodiment of the present invention
In attached drawing, the technical solution in the embodiment of the present invention is explicitly described, it is clear that described embodiment be the present invention
Part of the embodiment, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not having
All other embodiments obtained under the premise of creative work are made, shall fall within the protection scope of the present invention.
As shown in Figure 1, the present embodiment discloses a kind of leafy vegetable disease screening method.Fig. 2 shows a kind of leafy vegetables
Disease geo-radar image characteristic processing flow chart;Fig. 3 shows a kind of flow chart for differentiating leafy vegetable disease.
The present embodiment has continued to use the related data of plant disease expert system first, adopts plant disease expert and is published
Leafy vegetable disease knowledge base and picture library, and leafy vegetable disease geo-radar image information is classified, concluded and is arranged.This
The relevant information of the knowledge base of leafy vegetable disease and picture library middle period class vegetable disease is converted into the letter of materialization by embodiment
Breath value, including picture color, texture, illness pathology etc..As machine occurs for scab color, Lesion size, spot pattern and scab
The illnesss performance information such as reason and state.The occurrence of leafy vegetable foliage disease information can be used as leafy vegetable disease screening according to
According to, meanwhile, when the value when diagnostic image in subsequent step is unsatisfactory for the correlation of default training sample, leafy vegetable can be passed through
The occurrence analysis of foliage disease information is as reference.By taking celery as an example, disease geo-radar image is stored according to classification.
Its Common Diseases, which embodies mark sheet, to be obtained by the collection of data with arranging, corresponding as shown in table 2.
2 Adult plant celery Common Diseases illness of table embodies mark sheet
The key of the present embodiment using leafy vegetable blade face symptom as leafy vegetable disease screening is started with, by leaf class vegetable
The processing of dish leaf face symptom diagnoses leafy vegetable disease with analysis.It is normal by the leafy vegetable blade face image acquired on the spot
There are the problems such as noise is big, quality, therefore need to carry out gaussian filtering noise reduction process to image first.The process of gaussian filtering noise reduction
Including being weighted average, the value of each pixel to image, all by other pixel values in itself and neighborhood by adding
It is obtained after weight average.Concrete operations are:With each pixel in a template (or convolution, mask) scan image, mould is used
The weighted average gray value of pixel removes alternate template center gray value in the neighborhood that plate determines, picture noise is reduced so as to reach
Effect.
Since disease geo-radar image is often obtained with RGB models, however RGB models carry out matched discrimination not by machine vision
Height, is difficult to obtain in processing procedure larger difference value, thus the present embodiment has selected that be converted to machine vision more easy to identify
HSV models.
Wherein, three parameters of HSV models are respectively tone (H), saturation degree (S) and brightness (V), HSV models and RGB
The formula of model conversion is as follows:
V=max (R, G, B);
H=h × 60 °;
H component informations are obtained after processing, the image H information for converting obtained averagely is divided into 16 gradients, is then pressed
The H component maps of image are shown according to 16 gradients of H components.The component map obtained can clearly retouch the illness information of image
It states.In the present embodiment by image H information be averagely divided into 16 gradients can effective translated image H component values, avoid point
Carefully lead to very much the wasting of resources and that divides slightly causes very much to be difficult to.
Texture feature extraction is carried out again to gained disease geo-radar image H component maps.The present embodiment is extracted using gray level co-occurrence matrixes
Texture eigenvalue, wherein solving its angular second moment (energy) ASM, the moment of inertia (contrast) CON, correlation COR, local homogeneity
(unfavourable balance away from) L and entropy H.Specific formula is as follows:
Local homogeneity (unfavourable balance away from)
Wherein, angular second moment for gray level co-occurrence matrixes element quadratic sum, also referred to as energy, reflection gradation of image distribution
Uniformity coefficient and texture fineness degree.The moment of inertia, that is, contrast reflects the clarity of image and the degree of the texture rill depth.Ash
For the i.e. big pixel of contrast of degree difference to more, CON is bigger.Entropy is the measurement of information content possessed by image, and texture information also belongs to
It is the measurement of a randomness in the information of image.Unfavourable balance is away from the homogeney for reflecting image texture, measurement image texture part
Transformation number, value then illustrates to lack variation between the different zones of image texture greatly, local uniform.
The feature space that classification for disease is j, the dispersion of training sampleFor
If projection vector is non-vanishing vector w, then the FISHER discriminant functions in feature space are
FISHER differentiations based on kernel function are exactly by above-mentioned maximization identification function, find best discriminant vector structure
Into projection matrix, projection of the sample on projection matrix w is as the foundation differentiated.
Sample characteristics can be by function phi (x1),φ(x2),…,φ(xn) linear expression, that is, have
Then
In above formula,
Lμ=L+ μ I
The projective transformation of final feature space w is
Then for the FISHER method of discrimination based on kernel function, demarcation threshold point is
WhereinFor the mean value of all categories after projection.
Decision Classfication is carried out according to discriminant function and threshold point, if five groups of texture eigenvalues of sample image to be tested point
Not Man Zu with to have made a definite diagnosis the image texture characteristic value of disease geo-radar image (training sample) linearly related, then make a definite diagnosis disease, otherwise basis
Calculated value compares with illness specific features value and judgement is identified again, and e.g., Disease Characters show as brown ellipse scab, then
Scab color, shape, size are calculated by observation and computer to diagnose.Illness specific features value is according to leafy vegetable blade face disease
Evil knowledge base in disease Knowledge Acquirement, scab color, spot pattern, Lesion size, scab feature including each disease with
And there is a situation where sprawling modes for disease.
Although being described in conjunction with the accompanying embodiments of the present invention, those skilled in the art can not depart from this hair
Various modifications and variations are made in the case of bright spirit and scope, such modifications and variations are each fallen within by appended claims
Within limited range.
Claims (3)
- A kind of 1. leafy vegetable disease screening method, which is characterized in that the method includes:S1:Noise reduction process is carried out to leafy vegetable blade face image, obtains the first tone images of leafy vegetable blade face image;S2:Color feature extracted is carried out to first tone images, obtains the characteristic information of leafy vegetable blade face image Value and the second tone images;S3:Texture feature extraction is carried out to second tone images, obtains the texture eigenvalue of second tone images;S4:Calculate the mean value of the texture eigenvalue of all pictures in preset leafy vegetable disease picture library;S5:According to the mean value of the texture eigenvalue, disease threshold point is obtained;S6:According to preset discriminant function and the disease threshold point, by the characteristic information of leafy vegetable blade face image It is worth and is diagnosed as disease leafy vegetable for linearly related leafy vegetable;Before the step S1, the method further includes following preprocessing process:According to preset leafy vegetable disease screening knowledge base and preset leafy vegetable disease picture library, obtained leafy vegetable Disease Characters table, the leafy vegetable Disease Characters table include disease title and Damage Types, and the Damage Types include:Disease Evil shape, disease color, disease size, Disease Characters and disease sprawling mode, the Disease Characters are according to the leaf class vegetable Dish disease screening knowledge base determines;In the step S3, the texture eigenvalue of second tone images is obtains by solving gray level co-occurrence matrixes, institute The angular second moments of gray level co-occurrence matrixes is stated to be evenly distributed degree and texture fineness degree for gradation of image;The gray level co-occurrence matrixes The moment of inertia is the local gray level correlation in image;The correlation COR of the gray level co-occurrence matrixes is the consistency of image, is spent The similarity degree on row or column direction is measured;The unfavourable balance of the gray level co-occurrence matrixes is measured away from the homogeney that L is image texture The number of image texture localized variation;The entropy of the gray level co-occurrence matrixes is the measurement of information content possessed by image;The step S6 is specifically included:Decision Classfication is carried out according to discriminant function and the threshold point that S5 is solved, if numerical value meets linearly related, is diagnosed as pair The disease answered;If conditions are not met, it then needs that judgement is identified again according to the comparison of calculated value and illness specific features value.
- 2. according to the method described in claim 1, it is characterized in that, the step S1 is specifically included:According to preset additive mixture RGB models, color feature extracted is carried out to leafy vegetable image, obtains the leaf class vegetable Red R information, green G information, the blue B information of dish image;According to the conversion formula of preset hexagonal pyramid HSV models and preset additive mixture RGB models, by the red R, Green G, blue B information are converted into tone H information, saturation degree S information and brightness V information;The tone H information is averagely divided into 16 gradients, the leafy vegetable figure is drawn according to 16 gradients display First tone images of picture.
- 3. according to the method described in claim 1, it is characterized in that, the step S5 is specifically included:S5.1 calculates the projection of sample in each training setS5.2 calculates the positive definite matrix L of sampleμS5.3 is according to core FISHER method of discrimination, computational discrimination demarcation threshold point y0。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510073648.6A CN104680524B (en) | 2015-02-11 | 2015-02-11 | A kind of leafy vegetable disease screening method |
PCT/CN2016/103928 WO2017157022A1 (en) | 2015-02-11 | 2016-10-30 | Light array drive control system for passing control system of smart electronic police |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510073648.6A CN104680524B (en) | 2015-02-11 | 2015-02-11 | A kind of leafy vegetable disease screening method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104680524A CN104680524A (en) | 2015-06-03 |
CN104680524B true CN104680524B (en) | 2018-06-22 |
Family
ID=53315517
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510073648.6A Active CN104680524B (en) | 2015-02-11 | 2015-02-11 | A kind of leafy vegetable disease screening method |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN104680524B (en) |
WO (1) | WO2017157022A1 (en) |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105181914A (en) * | 2015-09-06 | 2015-12-23 | 中国农业大学 | Evaluation and tracing method for nutrition quality information of leafy vegetables |
CN105261045B (en) * | 2015-09-12 | 2020-04-03 | 黑龙江八一农垦大学 | Digital method for rapidly evaluating loss degree of alfalfa diseases and insect pests |
CN106295654B (en) * | 2016-08-01 | 2019-11-12 | 杭州励飞软件技术有限公司 | A kind of image-recognizing method and device |
CN106570898A (en) * | 2016-10-11 | 2017-04-19 | 浙江大学 | Method for quantitatively measuring green leaf rate of herbaceous plant quickly based HSV color space |
CA3039557A1 (en) | 2016-10-13 | 2018-04-19 | Mccain Foods Limited | Method, medium, and system for detecting potato virus in a crop image |
CN108921849A (en) * | 2018-09-30 | 2018-11-30 | 靖西海越农业有限公司 | For preventing and treating the wisdom Agricultural Monitoring early warning system of fertile mandarin orange pest and disease damage |
CN109035057A (en) * | 2018-09-30 | 2018-12-18 | 田东县文设芒果专业合作社 | Mango insect pest intelligent monitor system with sightseeing corridor |
CN109342421A (en) * | 2018-09-30 | 2019-02-15 | 靖西海越农业有限公司 | A kind of wisdom agricultural monitoring platform based on Internet of Things |
CN109658409A (en) * | 2018-12-29 | 2019-04-19 | 佛山科学技术学院 | A kind of Leaf of banana leaf spot detection method |
CN110414537A (en) * | 2019-01-08 | 2019-11-05 | 电子科技大学 | A kind of recognition methods of Highland Barley In Xizang lesion |
CN110390359A (en) * | 2019-01-08 | 2019-10-29 | 电子科技大学 | A kind of recognition methods of Highland Barley In Xizang blade lesion |
CN114694382B (en) * | 2022-04-01 | 2023-01-17 | 桂林电子科技大学 | Dynamic one-way traffic control system based on Internet of vehicles environment |
CN115376032B8 (en) * | 2022-10-25 | 2023-05-26 | 金乡县林业保护和发展服务中心(金乡县湿地保护中心、金乡县野生动植物保护中心、金乡县国有白洼林场) | Forestry disease and insect intelligent identification method and system based on image understanding |
CN116258968B (en) * | 2023-05-10 | 2023-07-18 | 蒙阴县国有林场总场(蒙阴县国有岱崮林场) | Method and system for managing fruit diseases and insects |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102214306A (en) * | 2011-06-16 | 2011-10-12 | 中国农业大学 | Leaf disease spot identification method and device |
CN102752479A (en) * | 2012-05-30 | 2012-10-24 | 中国农业大学 | Scene detection method of vegetable diseases |
CN103559511A (en) * | 2013-11-20 | 2014-02-05 | 天津农学院 | Automatic identification method of foliar disease image of greenhouse vegetable |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2910024B2 (en) * | 1994-05-23 | 1999-06-23 | 株式会社タツノ・メカトロニクス | Travel guide member |
CN1308306A (en) * | 2001-02-16 | 2001-08-15 | 李亚林 | Double setup method of traffic signal light and double traffic signal light device |
CN100397435C (en) * | 2005-04-11 | 2008-06-25 | 云霄 | Crossroads traffic management control system |
JP2009187340A (en) * | 2008-02-07 | 2009-08-20 | Yoko Naruiwa | Curbstone partition side auxiliary traffic light |
CN101719318B (en) * | 2009-03-19 | 2011-11-30 | 王苏华 | Method for guiding vehicles at road intersection |
CN201918020U (en) * | 2011-01-31 | 2011-08-03 | 韩景福 | Traffic signal control device capable of improving traffic capacity on level-crossing roads |
CN102592462A (en) * | 2012-03-06 | 2012-07-18 | 昆明理工大学 | Whole-journey independent phase green wave induction control method and system of traffic flow |
-
2015
- 2015-02-11 CN CN201510073648.6A patent/CN104680524B/en active Active
-
2016
- 2016-10-30 WO PCT/CN2016/103928 patent/WO2017157022A1/en active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102214306A (en) * | 2011-06-16 | 2011-10-12 | 中国农业大学 | Leaf disease spot identification method and device |
CN102752479A (en) * | 2012-05-30 | 2012-10-24 | 中国农业大学 | Scene detection method of vegetable diseases |
CN103559511A (en) * | 2013-11-20 | 2014-02-05 | 天津农学院 | Automatic identification method of foliar disease image of greenhouse vegetable |
Non-Patent Citations (1)
Title |
---|
"基于Fisher判别分析的玉米叶部病害图像识别";王娜等;《中国农业科学》;20091110;第42卷(第11期);摘要,第1.2、1.3、1.4、2.3节 * |
Also Published As
Publication number | Publication date |
---|---|
WO2017157022A1 (en) | 2017-09-21 |
CN104680524A (en) | 2015-06-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104680524B (en) | A kind of leafy vegetable disease screening method | |
CN104598908B (en) | A kind of crops leaf diseases recognition methods | |
Yun et al. | PNN based crop disease recognition with leaf image features and meteorological data | |
Yadav et al. | Classification of glaucoma based on texture features using neural networks | |
CN106845497B (en) | Corn early-stage image drought identification method based on multi-feature fusion | |
CN107330449A (en) | A kind of BDR sign detection method and device | |
CN109767440A (en) | A kind of imaged image data extending method towards deep learning model training and study | |
CN107844797A (en) | A kind of method of the milking sow posture automatic identification based on depth image | |
DE102008060789A1 (en) | System and method for unmonitored detection and Gleason grading for a prostate cancer preparation (whole-mount) using NIR fluorescence | |
CN106340000A (en) | Bone age assessment method | |
CN105320970B (en) | A kind of potato disease diagnostic device, diagnostic system and diagnostic method | |
CN107563996B (en) | A kind of novel discus nervi optici dividing method and system | |
CN110910377B (en) | Cerebral infarction MRI image identification method based on neural network | |
Jadhav et al. | Soybean leaf disease detection and severity measurement using multiclass SVM and KNN classifier | |
CN110288574A (en) | A kind of adjuvant Ultrasonographic Diagnosis hepatoncus system and method | |
CN110400293A (en) | A kind of non-reference picture quality appraisement method based on depth forest classified | |
CN110163839A (en) | The recognition methods of leopard line shape eye fundus image, model training method and equipment | |
CN107392251A (en) | A kind of method that target detection network performance is lifted using category images | |
CN112270667A (en) | TI-RADS-based integrated deep learning multi-tag identification method | |
CN114240874A (en) | Bone age assessment method and device based on deep convolutional neural network and feature fusion and computer readable storage medium | |
Lyu et al. | Deep tessellated retinal image detection using Convolutional Neural Networks | |
CN111028230A (en) | Fundus image optic disc and macula lutea positioning detection algorithm based on YOLO-V3 | |
CN104751461B (en) | A kind of leukocyte cell core dividing method based on histogram thresholding and low-rank representation | |
Huang et al. | HEp-2 cell images classification based on textural and statistic features using self-organizing map | |
CN106446805A (en) | Segmentation method and system for optic cup in eye ground photo |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |