CN110532926A - Pig neurogenic disease intelligence Forecasting Method based on deep learning - Google Patents
Pig neurogenic disease intelligence Forecasting Method based on deep learning Download PDFInfo
- Publication number
- CN110532926A CN110532926A CN201910783813.5A CN201910783813A CN110532926A CN 110532926 A CN110532926 A CN 110532926A CN 201910783813 A CN201910783813 A CN 201910783813A CN 110532926 A CN110532926 A CN 110532926A
- Authority
- CN
- China
- Prior art keywords
- pig
- gait
- point
- deep learning
- neurogenic disease
- 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
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 35
- 201000010099 disease Diseases 0.000 title claims abstract description 34
- 230000001272 neurogenic effect Effects 0.000 title claims abstract description 25
- 238000013135 deep learning Methods 0.000 title claims abstract description 15
- 238000013277 forecasting method Methods 0.000 title claims abstract description 9
- 230000005021 gait Effects 0.000 claims abstract description 23
- 238000012360 testing method Methods 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 14
- 238000000034 method Methods 0.000 claims abstract description 13
- 238000001514 detection method Methods 0.000 claims abstract description 10
- 238000005457 optimization Methods 0.000 claims abstract description 6
- 239000000284 extract Substances 0.000 claims abstract description 5
- 238000002790 cross-validation Methods 0.000 claims abstract description 4
- 230000001360 synchronised effect Effects 0.000 claims abstract description 3
- 230000008859 change Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 4
- 210000003194 forelimb Anatomy 0.000 claims description 2
- 210000003141 lower extremity Anatomy 0.000 claims description 2
- 230000008569 process Effects 0.000 claims description 2
- 238000003708 edge detection Methods 0.000 claims 1
- 206010044565 Tremor Diseases 0.000 abstract description 4
- 206010003591 Ataxia Diseases 0.000 abstract description 3
- 206010010947 Coordination abnormal Diseases 0.000 abstract description 3
- 238000003745 diagnosis Methods 0.000 abstract description 3
- 208000016290 incoordination Diseases 0.000 abstract description 3
- 238000000605 extraction Methods 0.000 abstract 1
- 241000282898 Sus scrofa Species 0.000 description 42
- 238000005516 engineering process Methods 0.000 description 6
- 238000010801 machine learning Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000011160 research Methods 0.000 description 4
- 238000013480 data collection Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 210000005036 nerve Anatomy 0.000 description 3
- 208000024891 symptom Diseases 0.000 description 3
- 238000009360 aquaculture Methods 0.000 description 2
- 244000144974 aquaculture Species 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 210000004556 brain Anatomy 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000009977 dual effect Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000001629 suppression Effects 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 208000003926 Myelitis Diseases 0.000 description 1
- 208000009525 Myocarditis Diseases 0.000 description 1
- 206010030113 Oedema Diseases 0.000 description 1
- 208000036142 Viral infection Diseases 0.000 description 1
- 241000607479 Yersinia pestis Species 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000006806 disease prevention Effects 0.000 description 1
- 238000013399 early diagnosis Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 235000013372 meat Nutrition 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000009304 pastoral farming Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 208000009305 pseudorabies Diseases 0.000 description 1
- 230000000384 rearing effect Effects 0.000 description 1
- 238000003307 slaughter Methods 0.000 description 1
- 230000009385 viral infection Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Radiology & Medical Imaging (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Public Health (AREA)
- Human Computer Interaction (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a kind of pig neurogenic disease intelligence Forecasting Method based on deep learning, pig normal walking and crippled video sample are acquired first, extract a series of single-frame images and complete objective contour, key point by extracting pig body profile constructs triangle skeleton model, according to model extraction gait angle information, realize the Synchronous fluorimetry of feature selecting and gait parameter, the gait information after optimization is allocated as training pattern and test model finally by test platform, final detection platform is obtained after cross validation.The present invention can fast and accurately determine that the incoordination of pig neurogenic disease early stage pig body, gait such as wave, tremble at the illnesss using the method for non-contact detecting and diagnosis.
Description
Technical field
The present invention relates to video image processing and nerual network technique method, in particular to a kind of pig based on deep learning
Neurogenic disease intelligence Forecasting Method.
Background technique
At this stage, the scale level of China's pig raising is greatly improved.But since technology and cost limit
The problems such as, infectiousness schweineseuche disease occurs often.If can infectiousness schweineseuche disease early stage it is carried out timely and effectively every
From treatment, it will greatly reduce the loss caused by pig aquaculture.5th (2018) large scale of pig farm new technology world discussion
Communication and discussion can have been carried out with regard to large-scale pig farm disease prevention technology, wherein Hua Zhong Agriculture University Chen Huanchun academician is in China pig
The thought of " pipe overweight feeding, support and overweight anti-, anti-overweight is controlled " is proposed in sick prevalence situation and prevention and control policy report.It is how effective
Raising scale livestock farming management, establish effective pig disease intelligent early-warning mechanism, it has also become China establishes large scale of pig farm
Field focus on research direction.Typically several causes of disease include: porcine pseudorabies, bowel oedema disease, pig transmissible brain to pig neurogenic disease
Myelitis, pig brain myocarditis, listerellosis etc..The clinical symptoms of the above pig neurogenic disease have following common ground: mutual aid
Imbalance, gait are waved, are trembled, and the death rate is high, viral infection.If epidemic disease early stage cannot control the isolation of sick pig as early as possible
Treatment processing, epidemic situation will rapidly spread expansion, cause biggish economic loss to pig aquaculture.In existing swine rearing condition
Under, there will be biggish difficulty in actual production using artificial observation and the method for Countable manual: firstly, observer needs
It is chronically in the environment of pigsty, is easy to generate the health of human body large effect, and staff's labour is strong
It spends larger.Secondly, influenced by certain subjective factors, the erroneous judgement of observed result and omit there is also it is larger a possibility that.
Main counter-measure is to slaughter sick pig, the movement of limitation animal and limitation meat listing at present, and this kind of solution is not only
The basic living of the mankind can be had an impact, but also will cause heavy losses economically.
Currently, the artificial intelligence technology of new generation with big data, cloud computing, machine learning etc. for representative enters acceleration hair
In the stage of exhibition, the new focus of international competition is had become, development will change human society life deeply.Now develop with artificial intelligence
Energy technology is that the reading intelligent agriculture of core has become the inevitable direction of China's agricultural innovation development.In recent years, pattern-recognition, expert system
System and artificial neural network (neural network, NN) are widely used in agricultural pest prediction, and achieve
Success.Wherein deep learning is research field more popular in machine learning.Compared with many machine learning methods, depth
Practising from complicated image and largely to divide with stronger data without learning effective characteristic of division in label complex data automatically
Class identify sum number it is predicted that ability, and it is many it is complicated, have and interior achieve successful application in terms of performance characteristic study.
Summary of the invention
Goal of the invention: the present invention provides a kind of methods using non-contact detecting and diagnosis, that is, may recognize that pig nerve
Property disease early stage pig body incoordination, gait waves, the illnesss such as tremble, and realizes the pig based on deep learning of pig neurogenic disease
Neurogenic disease intelligence measuring and reporting system.
Technical solution: steps are as follows for a kind of pig neurogenic disease intelligence measuring and reporting system based on deep learning:
(1) video sample required for acquisition is tested, including pig normal walking and crippled video;In target video image
It is middle to extract a series of single-frame images, and extract complete objective contour.
(2) by drawing profile center & periphery point distance Curve, pig body profile key point is extracted, by by key profile
Point is connected with profile central point constructs triangle skeleton model.
(3) according to the triangle skeleton model having had been built up, gait angle information is extracted, using packaged type feature selecting
Method realizes the Synchronous fluorimetry of feature selecting and gait parameter.
(4) test platform is designed using the RPN network structure under Open Framework caffe, by the gait after characteristic optimization
Information is allocated as training pattern and test model, and cross validation obtains final detection platform.
Further, the implementation method in the step 1 is as follows:
(1) a series of single-frame images is extracted to target video image, gray processing is carried out to target image pig.
(2) to image gaussian filtering, the gaussian coefficient put to each in template is carried out using two-dimensional Gaussian function formula
It calculates, final dimensional Gaussian template is obtained by normalization.
(3) difference Gx and Gy both horizontally and vertically are calculated using edge difference operator Sobel, obtains gradient-norm and side
To.
(4) non-maxima suppression finds pixel local maximum.
(5) selection of dual threshold, Canny algorithm distinguish edge pixel using a high threshold and a Low threshold.Such as
Fruit edge pixel point gradient value is greater than high threshold, then is considered as strong edge point.If edge gradient value is less than high threshold, it is greater than
Low threshold is then labeled as weak marginal point.Point less than Low threshold is then suppressed.
(6) hysteresis bounds track, accurate as a result, caused weak edge caused by removal noise or color change to obtain
Point.
Further, the two-dimensional Gaussian function formula are as follows:
Further, test platform is designed using the RPN network structure under Open Framework caffe, after characteristic optimization
Gait information 85% be used as data source, input the test platform for training pattern;Remaining 15% gait information is as survey
Die trial type recycles 5 times altogether, and cross validation obtains final detection platform, by the detection model having had built up, as observing and predicting mesh
Logo image instability of gait, incoordination, the symptoms such as tremble realize that pig neurogenic disease intelligently observes and predicts.
The utility model has the advantages that compared with prior art, the present invention has following remarkable result: the present invention is in deep learning application technology
On the basis of, pig gait parameter is extracted by building skeleton pattern, using the angle statistical nature of skeleton pattern, in deep learning frame
Pre-training model is designed on the basis of structure, obtains Optimized model after trained and test, realizes that pig neurogenic disease intelligently observes and predicts.
Detailed description of the invention
Fig. 1 is the Establishing process block diagram of test and training pattern;
Fig. 2 is that profile center & periphery point distance calculates schematic diagram;
Fig. 3 is profile center & periphery point distance Curve figure;
Fig. 4 is triangle skeleton model schematic;
The angle that Fig. 5 is ∠ AOB counts part exemplary diagram.
Specific embodiment
As shown in Figs. 1-5, pig neurogenic disease illness Image Acquisition, processing and big data set building.Acquisition experiment
Required video sample, including pig normal walking and crippled video;A series of single frames is extracted in target video image
Image carries out gray processing to target image, and then to image gaussian filtering, the realization of image gaussian filtering can directly pass through one
Convolution of a two-dimensional Gaussian kernel is realized, that is, two-dimensional convolution template, two-dimensional Gaussian function formula are as follows:The gaussian coefficient that each in template is put can be calculated by above formula, then again by returning
One mode changed obtains final dimensional Gaussian template.Difference both horizontally and vertically is calculated using edge difference operator Sobel
Divide Gx and Gy, obtains gradient-norm and direction.By non-maxima suppression, pixel local maximum is found.The selection of dual threshold,
Canny algorithm distinguishes edge pixel using a high threshold and a Low threshold.If edge pixel point gradient value is greater than height
Threshold value is then considered as strong edge point.If edge gradient value is less than high threshold, it is greater than Low threshold, then is labeled as weak marginal point,
Point less than Low threshold is then suppressed, hysteresis bounds tracking, accurate as a result, drawing caused by noise or color change to obtain
The weak marginal point risen should remove, and weak marginal point caused by usual true edge is connected to strong edge point, and is drawn by noise
Rise weak marginal point then will not, finally obtain objective contour.Seek the central point (x of objective contourc,yc) are as follows:
It is wherein marginal point sum, (xi,yi) it is a certain marginal point;Pass through profile up contour point marginal point (xi,yi) seek
Central point (xc,yc).The forelimb C hind leg D that obtains nose A, root of the tail B from center-marginal point distance Curve and can be detected,
To obtain triangle skeleton model.Extract gait angle character ∠ AOB.For the certainty for verifying data of the present invention, experiment is led to
The acquisition research to multiple groups different monitoring video data is crossed to differentiate the ∠ AOB of pig normal walking and crippled walking, the above acquisition
Angle statistical information as the present invention in training data set.Calculate the change of the ∠ AOB angle of successive objective frame
Change.
It is by the acquisition to multiple groups different monitoring video data that pig nerve disease, which early diagnoses deep learning model construction,
Study the ∠ AOB to differentiate pig normal walking and crippled walking.By obtaining pre- to the training of training dataset convolutional neural networks
It surveys disaggregated model and brings test data set into simultaneously, obtain prediction result.
The early diagnosis of pig neurogenic disease and intelligent measuring and reporting system build be with the pig neurogenic disease different onset cause of disease and
Image large data sets carry out model training as image sample data collection under different phase of falling ill, and building machine learning model is explored
The recognition methods of common pig neurogenic disease.Based on sample image data collection, image segmentation and mesh based on deep learning are studied
Detection algorithm is marked, pig neurogenic disease identification feature and rule are summarized, carries out pig nerve using depth convolutional neural networks model
Property disease identification, establish take into account accuracy rate and immediately detection demand pig neurogenic disease diagnosis prediction model.
By machine learning, model training is carried out using angle statistical picture large data sets as image sample data collection.Benefit
Test platform is designed with the RPN network structure under Open Framework caffe, regard the gait information 85% after characteristic optimization as number
According to source, the test platform is inputted for training pattern;Remaining 15% gait information recycles 5 times altogether as test model, intersects
Verifying obtains final detection platform, and by the detection model having had built up, as target image instability of gait is observed and predicted, mutual aid is lost
It adjusts, the symptoms such as tremble, realizes that pig neurogenic disease intelligently observes and predicts.Finally, being built on the basis of previous experiments research in pig house
Detection device builds the early warning of pig neurogenic disease and diagnostic system.
Claims (5)
1. a kind of pig neurogenic disease intelligence Forecasting Method based on deep learning, it is characterised in that: steps are as follows:
(1) video sample required for acquisition is tested, including pig normal walking and crippled video;It is mentioned in target video image
A series of single-frame images is taken, and extracts complete objective contour;
(2) by draw profile center & periphery point distance Curve, extract pig body profile key point, by by key profile point with
Profile central point, which is connected, constructs triangle skeleton model;
(3) according to the triangle skeleton model having had been built up, gait angle information is extracted, using packaged type feature selecting side
Method realizes the Synchronous fluorimetry of feature selecting and gait parameter;
(4) test platform is designed using the RPN network structure under Open Framework caffe, by the gait information after characteristic optimization
It is allocated as training pattern and test model, cross validation obtains final detection platform.
2. the pig neurogenic disease intelligence Forecasting Method according to claim 1 based on deep learning, it is characterised in that: institute
Implementation method in the step 1 stated are as follows: a series of single-frame images is extracted to target video image, gray scale is carried out to target image
Change, gaussian filtering obtains dimensional Gaussian template, calculates difference Gx both horizontally and vertically using edge difference operator Sobel
And Gy, gradient-norm and direction are obtained, by Canny edge detection algorithm, obtains objective contour.
3. the pig neurogenic disease intelligence Forecasting Method according to claim 1 based on deep learning, it is characterised in that: institute
Point (the x at the profile center of target in the step 2 statedc,yc) are as follows:
Wherein NbFor marginal point sum, (xi,yi) it is a certain marginal point;Draw center & periphery point distance Curve are as follows: in pig body wheel
Exterior feature, which is taken up an official post, takes a marginal point (xi,yi) it is used as starting point, and calculate marginal point (xi,yi) and center (xc,yc) distance, i.e., in
The heart-marginal point distance;Setting by the way that the upper cut-off frequency of low-pass filter is arranged can be from center-marginal point distance Curve
The forelimb C hind leg D that obtains nose A, root of the tail B and can be detected, to obtain triangle skeleton model.
4. the pig neurogenic disease intelligence Forecasting Method according to claim 1 based on deep learning, it is characterised in that: institute
Gait angle information obtains detailed process in the step 3 stated are as follows: the ∠ AOB of normal walking and crippled walking, with the angle of acquisition
Statistical information is training data set.
5. the pig neurogenic disease intelligence Forecasting Method according to claim 1 based on deep learning, it is characterised in that: institute
Gait information 85% in the step 4 stated after characteristic optimization is used as data source, inputs the test platform for training pattern;It is surplus
Under 15% gait information as test model, recycle 5 times altogether.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910783813.5A CN110532926A (en) | 2019-10-09 | 2019-10-09 | Pig neurogenic disease intelligence Forecasting Method based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910783813.5A CN110532926A (en) | 2019-10-09 | 2019-10-09 | Pig neurogenic disease intelligence Forecasting Method based on deep learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110532926A true CN110532926A (en) | 2019-12-03 |
Family
ID=68664068
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910783813.5A Pending CN110532926A (en) | 2019-10-09 | 2019-10-09 | Pig neurogenic disease intelligence Forecasting Method based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110532926A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112150506A (en) * | 2020-09-27 | 2020-12-29 | 成都睿畜电子科技有限公司 | Target state detection method, device, medium and electronic equipment |
CN112401834A (en) * | 2020-10-19 | 2021-02-26 | 南方科技大学 | Movement-obstructing disease diagnosis device |
CN112434577A (en) * | 2020-11-12 | 2021-03-02 | 中国农业大学 | Milk cow lameness detection method and milk cow lameness detection device |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105243675A (en) * | 2015-09-28 | 2016-01-13 | 江苏农林职业技术学院 | Star-shaped skeleton model based pig hobbling identification method |
CN107133604A (en) * | 2017-05-25 | 2017-09-05 | 江苏农林职业技术学院 | A kind of pig abnormal gait detection method based on ellipse fitting and predictive neutral net |
-
2019
- 2019-10-09 CN CN201910783813.5A patent/CN110532926A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105243675A (en) * | 2015-09-28 | 2016-01-13 | 江苏农林职业技术学院 | Star-shaped skeleton model based pig hobbling identification method |
CN107133604A (en) * | 2017-05-25 | 2017-09-05 | 江苏农林职业技术学院 | A kind of pig abnormal gait detection method based on ellipse fitting and predictive neutral net |
Non-Patent Citations (2)
Title |
---|
朱家骥等: "基于星状骨架模型的猪步态分析", 《江苏农业科学》 * |
王祥等: "基于关键轮廓点模型的猪的步态异常检测", 《科技视界》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112150506A (en) * | 2020-09-27 | 2020-12-29 | 成都睿畜电子科技有限公司 | Target state detection method, device, medium and electronic equipment |
CN112401834A (en) * | 2020-10-19 | 2021-02-26 | 南方科技大学 | Movement-obstructing disease diagnosis device |
CN112434577A (en) * | 2020-11-12 | 2021-03-02 | 中国农业大学 | Milk cow lameness detection method and milk cow lameness detection device |
CN112434577B (en) * | 2020-11-12 | 2024-03-26 | 中国农业大学 | Dairy cow lameness detection method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wu et al. | Lameness detection of dairy cows based on the YOLOv3 deep learning algorithm and a relative step size characteristic vector | |
Wang et al. | A deep learning approach incorporating YOLO v5 and attention mechanisms for field real-time detection of the invasive weed Solanum rostratum Dunal seedlings | |
Ozguven et al. | Automatic detection and classification of leaf spot disease in sugar beet using deep learning algorithms | |
CN108364006B (en) | Medical image classification device based on multi-mode deep learning and construction method thereof | |
Ran et al. | Cataract detection and grading based on combination of deep convolutional neural network and random forests | |
CN107895367A (en) | A kind of stone age recognition methods, system and electronic equipment | |
CN110532926A (en) | Pig neurogenic disease intelligence Forecasting Method based on deep learning | |
CN105513077A (en) | System for screening diabetic retinopathy | |
CN109635846A (en) | A kind of multiclass medical image judgment method and system | |
CN102096804A (en) | Method for recognizing image of carcinoma bone metastasis in bone scan | |
CN109461163A (en) | A kind of edge detection extraction algorithm for magnetic resonance standard water mould | |
CN104299242A (en) | Fluorescence angiography fundus image extraction method based on NGC-ACM | |
CN110246109A (en) | Merge analysis system, method, apparatus and the medium of CT images and customized information | |
CN114596448A (en) | Meat duck health management method and management system thereof | |
CN112750117A (en) | Blood cell image detection and counting method based on convolutional neural network | |
Kaldera et al. | MRI based glioma segmentation using deep learning algorithms | |
CN103839048B (en) | Stomach CT image lymph gland recognition system and method based on low-rank decomposition | |
Yang et al. | Unsupervised domain adaptation for cross-device OCT lesion detection via learning adaptive features | |
Cai et al. | A deep learning-based algorithm for crop Disease identification positioning using computer vision | |
CN110163103A (en) | A kind of live pig Activity recognition method and apparatus based on video image | |
CN111402231B (en) | Automatic evaluation system and method for lung CT image quality | |
CN116030063B (en) | Classification diagnosis system, method, electronic device and medium for MRI image | |
Hassan et al. | RRI-Net: classification of multi-class retinal diseases with deep recurrent residual inception network using OCT scans | |
Mali et al. | Evaluation and Segregation of Fruit Quality using Machine and Deep Learning Techniques | |
CN114176602B (en) | Method for simultaneously positioning electrocardiograph P wave, QRS wave and T wave based on deep learning multi-target detection |
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: 20191203 |
|
RJ01 | Rejection of invention patent application after publication |