CN110348349A - A kind of method and system collected, analyze pig behavior video data - Google Patents
A kind of method and system collected, analyze pig behavior video data Download PDFInfo
- Publication number
- CN110348349A CN110348349A CN201910583728.4A CN201910583728A CN110348349A CN 110348349 A CN110348349 A CN 110348349A CN 201910583728 A CN201910583728 A CN 201910583728A CN 110348349 A CN110348349 A CN 110348349A
- Authority
- CN
- China
- Prior art keywords
- pig
- image
- video
- module
- training
- 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
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 51
- 230000006835 compression Effects 0.000 claims abstract description 25
- 238000007906 compression Methods 0.000 claims abstract description 25
- 238000004458 analytical method Methods 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 22
- 238000012360 testing method Methods 0.000 claims abstract description 21
- 239000000284 extract Substances 0.000 claims abstract description 15
- 238000002372 labelling Methods 0.000 claims abstract description 10
- 230000006399 behavior Effects 0.000 claims description 58
- 238000011156 evaluation Methods 0.000 claims description 16
- 238000013527 convolutional neural network Methods 0.000 claims description 13
- 230000007717 exclusion Effects 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000009434 installation Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims 1
- 230000003542 behavioural effect Effects 0.000 abstract description 9
- 230000000694 effects Effects 0.000 abstract description 5
- 238000013145 classification model Methods 0.000 abstract description 3
- 238000009395 breeding Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000009360 aquaculture Methods 0.000 description 2
- 244000144974 aquaculture Species 0.000 description 2
- 230000001488 breeding effect Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000001537 neural effect Effects 0.000 description 2
- 244000144977 poultry Species 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000005194 fractionation Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 244000144972 livestock Species 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 239000002985 plastic film Substances 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Human Computer Interaction (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Image Analysis (AREA)
Abstract
The present invention is more particularly directed to a kind of method and systems collected, analyze pig behavior video data to be lain down, and to the place of pig observation during the break using pig bed body module for pig；Video acquisition module, the acquisition for observation and video data index to pig behavior state；Video image processing and analysis module, for being collected storage to video image, frame extracts and preserving module extracts real-time video or is stored in the frame of the video of storage hard disk, the behavior classification for the pig that image labeling module reflects image is labeled, image compression module compresses the image marked, image data base has all been marked to compression image and has been split as training set and test set at random, training set training image disaggregated model, the image classification model of training is evaluated using test set, the invention collects the behavioural characteristic of the intrinsic pig of pig bed and is labeled storage and disaggregated model training, technical support is provided for the pig Activity recognition based on video.
Description
Technical field
The invention belongs to livestock and poultry cultivation intelligence Processes and apparatus technical fields, in particular to a kind of to collect, analyze pig row
For the method and system of video data.
Background technique
With the scale of pig-breeding, intensivization development, manual type be increasingly difficult to meet pig raising it is low at
The requirement such as sheet, high efficiency, safety.Intellectualized technology is expected to can be improved the efficiency of pig raising, the timely typical behaviour for finding pig,
Various exceptions improve the efficiency of pig raising in time, automatically, pointedly take appropriate measures, and reduce aquaculture cost, do
To preventing trouble before it happens, to improve the competitiveness of pig raising enterprise, artificial intelligence will become the strong of boosting pig breeding industry transition and upgrade
Technological means.
It realizes intelligent pig raising, first has to collect a large amount of pig behavioral data, such as image, sound etc..Then, using pig
Behavioral data train classification models, by trained model be applied to pig behavior real-time prediction.Currently, there are many researchs
Person is dedicated to the Activity recognition research of pig.
Pig bed is a kind of important, widely used pig breeding instrument.Currently, people have had devised with heating, cooling, temperature
The pig bed of the automation functions such as control, cleaning up excrement, improves the intelligent level of pig raising to a certain extent.But it is existing at present
Pig bed is also lacking in the behavior perceptible aspect of pig.Pig bed is the important space of pig life, when pig had most of in one day
Between will stay in pig bed on spend.If being able to use the behavioral data that pig bed collects pig, and then intelligence is carried out to the behavioural characteristic of pig
It can analyze, important technology support will be provided for the intelligent recognition of pig behavior and intelligent pig raising technology, therefore, for above-mentioned phenomenon sheet
Invention proposes a kind of method and system collected, analyze pig behavior video data, passes through video camera and collects row of the pig on pig bed
It for data, extracts picture frame and is labeled, compresses, establish the image data base of pig behavior, and the image classification of training pig behavior
Model provides technical support for the pig Activity recognition based on video.
Summary of the invention
It is an object of the present invention to provide a kind of method and systems collected, analyze pig behavior video data, are received by pig bed
Collect the behavioral data of pig in pig bed, and then intellectual analysis is carried out to the behavioural characteristic of pig, is the intelligent recognition and intelligence of pig behavior
Pig raising technology provides technical support.
The present invention provides a kind of systems collected, analyze pig behavior video data to solve above-mentioned technical problem, including
Sequentially connected: pig bed body module lies down for pig, and during the break to the place of pig observation；Video acquisition mould
Block, the acquisition for observation and video data index to pig behavior state；Video image processing and analysis module, are used for
Storage is collected to video image, and processing analysis is carried out to video data.
The pig bed body module is made of bottom plate, top plate, side plate, bed curtain, and bottom plate left and right sides and rear side are provided with
Side plate, side plate upper end are connected with top plate, and the front end of top plate is provided with PVC bar shaped and hangs curtain, and pig can hang curtain disengaging pig by PVC bar shaped
Bed.
The video acquisition module includes an optics or thermal camera, and video camera acquires the work of pig in pig bed ontology
Dynamic video, and send vision signal to video image processing and analysis module.
The video image processing and analysis module is made of computer hardware and software systems, and computer hardware includes
Signal receiving module, storage hard disk, software systems are saved by video, video is shown, frame extracts and is saved, image labeling, image
The functional modules such as compression, model training and evaluation are constituted.
The video that video preserving module in the software systems receives computer is saved to local hard drive, video
Received video is exported to display and is shown by display module, and frame extracts and preserving module is responsible for extracting figure from received video
It as frame, and saves to local hard drive, the pig behavior that image labeling provides image marks man-machine interface, for professional to image
Pig behavior is labeled, and the compression of images of mark is specified height and width by image compression module, and saves as picture number
According to library, the data set of image data base is split as the training set of mutual exclusion by model training and evaluation module at random according to a certain percentage
And test set uses the performance of test set evaluation neural network using training set training deep learning convolutional neural networks.
The present invention also provides a kind of methods collected, analyze pig behavior video data, comprising:
S1: installing camera in pig bed, and installation site is preferably able to covering pig bed whole region using the visual field and flexibly selects as principle
It selects, camera can clearly take the crawler behavior of pig, complete information collection to pig by camera, and by camera
The vision signal taken passes to computer hardware；
S2: signal receiving module receives vision signal, and is stored in storage hard disk, frame in software systems extract and
Preserving module extracts real-time video or is stored in the frame of the video of storage hard disk, and the frame of extraction is color image or gray level image,
Image is stored in storage hard disk again, passes through the behavior classification for the pig that the image labeling module of software systems reflects image
It is labeled, the image marked is compressed by image compression module, and compression image is saved to image data base；
S3: image data base has all been marked the training set that compression image is split as mutual exclusion at random by model training and evaluation module
And test set, using training set training image disaggregated model, disaggregated model is made of convolutional neural networks, then reuses training
The image pattern classification of good disaggregated model prediction test set, prediction result and legitimate reading are compared, classification of assessment model
Performance；
S4: changing the type, structure, parameter of disaggregated model, repeats step S3, until model performance is met the requirements.
The step S3 includes:
S31: image data base has all been marked the training that compression image is split as mutual exclusion at random by model training and evaluation module
Collection and test set；
S32: the volume with convolutional layer Conv1, pond layer Pool1, convolutional layer Conv2, pond layer Pool2, full articulamentum is constructed
Product neural network；
S33: training set training convolutional neural networks are used；
S34: the image pattern classification of trained convolutional neural networks prediction test set is used；
S35: prediction result and legitimate reading are compared, and investigate classification accuracy, recall rate, f1 score, classification of assessment model
Performance.
The method and system of a kind of collection, analysis pig behavior video data of the invention, using what is be arranged in pig bed ontology
Camera acquires the motion video of pig, and by collected video collect into storage hard disk, then sends video to video figure
As processing and analysis module, establish the image data base of pig behavior, by video image processing and analysis module to video information into
Row processing analysis, and according to the behavior sophisticated image disaggregated model of pig, which can be according to the video data pair of offer
The behavior of pig is labeled identification, provides technical support for the pig Activity recognition based on video.
Detailed description of the invention
Fig. 1 is a kind of collection of the invention, the system module composition schematic diagram for analyzing pig behavior video data；
Fig. 2 is a kind of collection of the invention, the system structure diagram for analyzing pig behavior video data；
Fig. 3 is that a kind of collection of the invention, video acquisition, display, preservation, the frame analyzed in the method for pig behavior video data mention
Take, image analysis, disaggregated model training, model evaluation module between workflow schematic diagram；
Fig. 4 is a kind of collection of the invention, analyzes the frame image in the method for pig behavior video data from video extraction；
Fig. 5 is a kind of collection of the invention, analyzes the compressed image of Fig. 4 in the method for pig behavior video data；
Fig. 6 is a kind of collection of the invention, analyzes the image after Fig. 5 convolution in the method for pig behavior video data；
Fig. 7 is a kind of collection of the invention, analyzes the image of Fig. 6 Chi Huahou in the method for pig behavior video data；
Fig. 8 is a kind of collection of the invention, analyzes the image after Fig. 7 convolution in the method for pig behavior video data；
Fig. 9 is a kind of collection of the invention, analyzes the image of Fig. 8 Chi Huahou in the method for pig behavior video data；
Figure 10 is a kind of collection of the invention, analyzes the convolutional neural networks structure that constructs in the method for pig behavior video data and show
It is intended to.
Specific embodiment
Embodiment 1, as shown in Figure 1 and Figure 2, a kind of system collected, analyze pig behavior video data, including be sequentially connected
: pig bed body module 1 lies down for pig, and during the break to the place of pig observation；Video acquisition module 2, is used for
The acquisition of observation and video data index to pig behavior state；Video image processing and analysis module 3, for video
Image is collected storage, and carries out processing analysis to video data, wherein pig bed body module 1 by bottom plate, top plate, side plate,
Bed curtain is constituted, and bottom plate uses 20mm thickness density board, and left and right, rear 3 side plates use 12mm thickness density board, and top plate is thick using 12mm
Hollow plastic sheet is spent, bottom plate left and right sides and rear side are provided with side plate, and side plate upper end is connected with top plate, and the front end of top plate is provided with
Curtain is hung, hanging curtain is preferably that curtain is hung in PVC bar shaped, and the length size of pig bed ontology can customize according to actual needs, and pig can pass through
Hang curtain disengaging pig bed ontology；Video acquisition module 2 includes a video camera, and video camera is preferably thermal camera, in pig bed sheet
1 thermal camera is fixed in the roof center position of body, is continuously shot within thermal camera 24 hours the behavior of pig in pig bed ontology
Motion video, thermal camera connect with computer and vision signal are real-time transmitted to video image processing and analysis module 3；
Video image processing is made of with analysis module 3 computer hardware and software systems, computer hardware include signal receiving module,
Hard disk, display are stored, software systems use language and the API such as Python, OpenCV, Scikit-Image, Tensorflow
Exploitation, by video saves, video is shown, frame extracts and saves, the functions such as image labeling, compression of images, model training and evaluation
Module composition；Video preserving module in software systems saves the video that computer receives to storage hard disk with mp4 format,
Received video is exported to display and is shown by video display module, and frame extracts and preserving module is responsible for mentioning from received video
Picture frame is taken, and is saved to storage hard disk, image labeling module provides pig behavior and marks man-machine interface, for professional to image
Pig behavior be labeled, the compression of images of mark is specified height and width by image compression module, and saves as image
The data set of image data base is split as the training of mutual exclusion by database, model training and evaluation module at random according to a certain percentage
Collection and test set use the performance of test set evaluation neural network using training set training deep learning convolutional neural networks.
Embodiment 2, a kind of method collected, analyze pig behavior video data as shown in Fig. 3-10, comprising:
S1: installing camera in pig bed ontology, and installation site is preferably able to covering pig bed whole region as principle spirit using the visual field
Selection living, camera can clearly take the crawler behavior of pig, complete the information collection to pig, video camera by camera
It is continuously shot within 24 hours the behavioral activity video of pig in pig bed ontology, video camera connect with computer and passes vision signal in real time
Give video image processing and analysis module 3；
S2: signal receiving module receives vision signal, and is stored in storage hard disk, frame in software systems extract and
Preserving module extracts real-time video or is stored in the frame of the video of storage hard disk, and the frame of extraction is color image or gray level image,
The frame image extracted is as shown in figure 4, its size is width × height=960 × 544 pixels, format jpg, the processing of gray level image
Only one gray color channel, the processing of color image have 3 Color Channels: red (R), green (G), blue (B), different colours
Channel keeps image information more abundant from the feature of different dimension reaction images；Image is stored in storage hard disk again, is led to
The behavior classification (such as normal, sick, fight) for crossing the pig that the image labeling modules of software systems reflects image is labeled,
The image marked is compressed by image compression module, compression ratio can with flexible choice, such as: after the compression of images of Fig. 3
Image as shown in figure 5, its size be width × height=100 × 57 pixels, format jpg；And compression image is saved to picture number
According to library, the pixel value of each image is shown laid flat in 1 row, width × height × column of port number+1 record, wherein last 1 is classified as mark
As a result, 1 row, 100 × 57 × 3+1 column are saved as, wherein last 1 is classified as example, the pixel in 3 channels of Fig. 5 is successively flattened
The image mark (such as 0, indicate normal；1, it indicates sick；2, expression is fought), other pixel values for being classified as image will be every
Width image saves as a record, i.e. 1 row；
S3: image data base has all been marked the training set that compression image is split as mutual exclusion at random by model training and evaluation module
And test set, the ratio of fractionation can with sets itself, (such as: training set: test set=7: 3)；Use training set training image point
Class model, disaggregated model are made of convolutional neural networks, then reuse the image of trained disaggregated model prediction test set
Sample class compares prediction result and legitimate reading, the performance of classification of assessment model；
S4: changing the type, structure, parameter of disaggregated model, repeats step S3, until model performance is met the requirements.
The step S3 includes:
S31: image data base has all been marked the training that compression image is split as mutual exclusion at random by model training and evaluation module
Collection and test set；
S32: the volume with convolutional layer Conv1, pond layer Pool1, convolutional layer Conv2, pond layer Pool2, full articulamentum is constructed
Product neural network has 1 input wherein full articulamentum, there are two hidden layer (H1, H2), the neuronal quantity of each hidden layer is 50
In and 1 output layer Output of layer, as shown in Figure 10.All neurons use Relu excitation function, are rolled up using training set training
Product neural network, Fig. 6-9 are respectively that (6 convolution kernels, window size 3*3, step-length 1, obtaining 6 width sizes is by Fig. 5 convolution
The convolution results image of 100*57), pond (pond window size 2*2, step-length 2, obtain 6 width sizes be 50*29 Chi Huajie
Fruit image), again convolution (8 convolution kernels, window size 3*3, step-length 1, obtain 8 width sizes be 50*29 convolution results
Image), again pond (pond window size 2*2, step-length 2, obtain 8 width sizes be 25*15 pond result images) behaviour
Make result images, after these operations, pixel quantity becomes 25 × 15 × 8=3000, they are full articulamentum input layer
Neuronal quantity；
S33: training set training convolutional neural networks are used；
S34: the image pattern classification of trained convolutional neural networks prediction test set is used；
S35: prediction result and legitimate reading are compared, and investigate classification accuracy, recall rate, f1 score, classification of assessment model
Performance (such as f1 score is greater than 0.9).
The method and system of a kind of collection, analysis pig behavior video data of the invention, using what is be arranged in pig bed ontology
Camera acquires the motion video of pig, and by collected video collect into storage hard disk, then sends video to video figure
As processing and analysis module, establish the image data base of pig behavior, by video image processing and analysis module to video information into
Row processing analysis, and the behavior that pig in a large amount of pig bed of collection is passed through according to the Behavioral training image classification model of pig, the invention
Data carry out analysis mark to the various actions to pig, disaggregated model are enable more quickly and accurately to identify the behavior of pig,
The disaggregated model can be labeled identification to the behavior of pig according to the video data of offer, so that poultry feeders find pig in time
Typical behaviour, various exceptions in time, automatically, pointedly take appropriate measures improve the efficiency of pig raising, reduce
Aquaculture cost improves the competitiveness of pig raising enterprise, provides technical support for the pig Activity recognition based on video.
The foregoing is only a preferred embodiment of the present invention, is not intended to limit the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied, all within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (7)
1. a kind of system collected, analyze pig behavior video data, which is characterized in that including sequentially connected:
Pig bed body module, lies down for pig, and during the break to the place of pig observation；
Video acquisition module, the acquisition for observation and video data index to pig behavior state；
Video image processing and analysis module for being collected storage to video image, and carry out processing point to video data
Analysis.
2. a kind of system collected, analyze pig behavior video data according to claim 1, it is characterised in that: described
Pig bed body module is made of bottom plate, top plate, side plate, bed curtain, and bottom plate left and right sides and rear side are provided with side plate, and side plate upper end connects
It is connected to top plate, the front end of top plate, which is provided with, hangs curtain, and pig can pass in and out pig bed ontology by hanging curtain.
3. a kind of system collected, analyze pig behavior video data according to claim 1, it is characterised in that: described
Video acquisition module includes a video camera, and video camera acquires the motion video of pig in pig bed ontology, and vision signal is transmitted
To video image processing and analysis module.
4. a kind of system collected, analyze pig behavior video data according to claim 1, it is characterised in that: described
Video image processing and analysis module are made of computer hardware and software systems, computer hardware include signal receiving module,
Hard disk, display are stored, software systems are saved by video, video is shown, frame extracts and is saved, image labeling, compression of images, mould
The functional modules such as type training and evaluation are constituted.
5. a kind of system collected, analyze pig behavior video data according to claim 4, it is characterised in that: described
The video that video preserving module in software systems receives computer is saved to local hard drive, and video display module will receive
Video export to display and show, frame extracts and preserving module is responsible for extracting picture frame from received video, and saves extremely
Hard disk is stored, image labeling module provides pig behavior and marks man-machine interface, is labeled for professional to the pig behavior of image,
The compression of images of mark is specified height and width by image compression module, and saves as image data base, model training and
The data set of image data base is split as the training set and test set of mutual exclusion by evaluation module at random according to a certain percentage, uses instruction
Practice and collect training deep learning convolutional neural networks, uses the performance of test set evaluation neural network.
6. a kind of method collected, analyze pig behavior video data characterized by comprising
S1: installing camera in pig bed ontology, and installation site is preferably able to covering pig bed whole region as principle spirit using the visual field
Selection living, camera can clearly take the crawler behavior of pig, complete the information collection to pig by camera, and will take the photograph
The vision signal taken as head passes to computer hardware；
S2: signal receiving module receives vision signal, and is stored in storage hard disk, frame in software systems extract and
Preserving module extracts real-time video or is stored in the frame of the video of storage hard disk, and the frame of extraction is color image or gray level image,
Image is stored in storage hard disk again, passes through the behavior classification for the pig that the image labeling module of software systems reflects image
It is labeled, the image marked is compressed by image compression module, and compression image is saved to image data base；
S3: image data base has all been marked the training set that compression image is split as mutual exclusion at random by model training and evaluation module
And test set, using training set training image disaggregated model, disaggregated model is made of convolutional neural networks, then reuses training
The image pattern classification of good disaggregated model prediction test set, prediction result and legitimate reading are compared, classification of assessment model
Performance；
S4: changing the type, structure, parameter of disaggregated model, repeats step S3, until model performance is met the requirements.
7. a kind of method collected, analyze pig behavior video data as claimed in claim 6, it is characterised in that: the step
Suddenly S3 includes:
S31: image data base has all been marked the training that compression image is split as mutual exclusion at random by model training and evaluation module
Collection and test set；
S32: the volume with convolutional layer Conv1, pond layer Pool1, convolutional layer Conv2, pond layer Pool2, full articulamentum is constructed
Product neural network；
S33: training set training convolutional neural networks are used；
S34: the image pattern classification of trained convolutional neural networks prediction test set is used；
S35: prediction result and legitimate reading are compared, and investigate classification accuracy, recall rate, f1 score, classification of assessment model
Performance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910583728.4A CN110348349A (en) | 2019-07-01 | 2019-07-01 | A kind of method and system collected, analyze pig behavior video data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910583728.4A CN110348349A (en) | 2019-07-01 | 2019-07-01 | A kind of method and system collected, analyze pig behavior video data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110348349A true CN110348349A (en) | 2019-10-18 |
Family
ID=68177767
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910583728.4A Pending CN110348349A (en) | 2019-07-01 | 2019-07-01 | A kind of method and system collected, analyze pig behavior video data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110348349A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111242007A (en) * | 2020-01-10 | 2020-06-05 | 上海市崇明区生态农业科创中心 | Farming behavior supervision method |
| CN112329870A (en) * | 2020-11-11 | 2021-02-05 | 国网山东省电力公司威海供电公司 | Method for identifying state of transformer substation pressure plate based on YOLO3 algorithm |
| CN112415009A (en) * | 2020-12-14 | 2021-02-26 | 河南牧业经济学院 | Wheat grain non-adhesion image acquisition method and system based on sticky dot matrix |
| CN112710664A (en) * | 2020-12-14 | 2021-04-27 | 河南牧业经济学院 | Soybean grain non-adhesion image acquisition method and system based on special-shaped tube array |
| CN113219964A (en) * | 2021-03-30 | 2021-08-06 | 广州朗国电子科技有限公司 | Poultry house environment inspection and regulation method, equipment and medium |
| CN113269109A (en) * | 2021-06-03 | 2021-08-17 | 重庆市畜牧科学院 | Pig state analysis system and method based on visual AI |
| CN113283363A (en) * | 2021-06-04 | 2021-08-20 | 重庆市畜牧科学院 | Pig video data acquisition and pig behavior evaluation method based on visual AI |
| CN113467552A (en) * | 2021-07-05 | 2021-10-01 | 长沙瑞和数码科技有限公司 | Pigsty environment temperature control method, device, equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017177661A1 (en) * | 2016-04-15 | 2017-10-19 | 乐视控股（北京）有限公司 | Convolutional neural network-based video retrieval method and system |
| CN107844797A (en) * | 2017-09-27 | 2018-03-27 | 华南农业大学 | A kind of method of the milking sow posture automatic identification based on depth image |
| CN108830144A (en) * | 2018-05-03 | 2018-11-16 | 华南农业大学 | A kind of milking sow gesture recognition method based on improvement Faster-R-CNN |
-
2019
- 2019-07-01 CN CN201910583728.4A patent/CN110348349A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017177661A1 (en) * | 2016-04-15 | 2017-10-19 | 乐视控股（北京）有限公司 | Convolutional neural network-based video retrieval method and system |
| CN107844797A (en) * | 2017-09-27 | 2018-03-27 | 华南农业大学 | A kind of method of the milking sow posture automatic identification based on depth image |
| CN108830144A (en) * | 2018-05-03 | 2018-11-16 | 华南农业大学 | A kind of milking sow gesture recognition method based on improvement Faster-R-CNN |
Non-Patent Citations (2)
| Title |
|---|
| 闫丽等: "猪行为自动监测技术研究现状与展望", 《江苏农业科学》 * |
| 高云等: "基于深度学习的群猪图像实例分割方法", 《农业机械学报》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111242007A (en) * | 2020-01-10 | 2020-06-05 | 上海市崇明区生态农业科创中心 | Farming behavior supervision method |
| CN112329870A (en) * | 2020-11-11 | 2021-02-05 | 国网山东省电力公司威海供电公司 | Method for identifying state of transformer substation pressure plate based on YOLO3 algorithm |
| CN112415009A (en) * | 2020-12-14 | 2021-02-26 | 河南牧业经济学院 | Wheat grain non-adhesion image acquisition method and system based on sticky dot matrix |
| CN112710664A (en) * | 2020-12-14 | 2021-04-27 | 河南牧业经济学院 | Soybean grain non-adhesion image acquisition method and system based on special-shaped tube array |
| CN112415009B (en) * | 2020-12-14 | 2024-06-11 | 河南牧业经济学院 | Wheat grain non-adhesion image acquisition method and system based on viscous lattice |
| CN113219964A (en) * | 2021-03-30 | 2021-08-06 | 广州朗国电子科技有限公司 | Poultry house environment inspection and regulation method, equipment and medium |
| CN113269109A (en) * | 2021-06-03 | 2021-08-17 | 重庆市畜牧科学院 | Pig state analysis system and method based on visual AI |
| CN113269109B (en) * | 2021-06-03 | 2023-12-05 | 重庆市畜牧科学院 | Pig status analysis system and method based on visual AI |
| CN113283363A (en) * | 2021-06-04 | 2021-08-20 | 重庆市畜牧科学院 | Pig video data acquisition and pig behavior evaluation method based on visual AI |
| CN113467552A (en) * | 2021-07-05 | 2021-10-01 | 长沙瑞和数码科技有限公司 | Pigsty environment temperature control method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110348349A (en) | A kind of method and system collected, analyze pig behavior video data | |
| CN111178197B (en) | Mass R-CNN and Soft-NMS fusion based group-fed adherent pig example segmentation method | |
| Li et al. | Deep cascaded convolutional models for cattle pose estimation | |
| CN103517042B (en) | A kind of nursing house old man's hazardous act monitoring method | |
| CN102509305B (en) | Animal behavior detection device based on omnidirectional vision | |
| CN109431523B (en) | Autism primary screening device based on non-social voice stimulation behavior paradigm | |
| CN107844797A (en) | A kind of method of the milking sow posture automatic identification based on depth image | |
| CN111294565A (en) | Intelligent pig raising monitoring method and management terminal | |
| CN113411542A (en) | Intelligent working condition monitoring equipment | |
| CN107330403A (en) | A kind of yak method of counting based on video data | |
| CN112528823B (en) | Method and system for analyzing batcharybus movement behavior based on key frame detection and semantic component segmentation | |
| CN108596256B (en) | Object recognition classifier construction method based on RGB-D | |
| CN113762113A (en) | Livestock parturition behavior monitoring method and device | |
| CN112580552A (en) | Method and device for analyzing behavior of rats | |
| Wang et al. | A PCA-based frame selection method for applying CNN and LSTM to classify postural behaviour in sows | |
| CN110427894A (en) | A kind of crowd's abnormal motion monitoring method, system, equipment and medium | |
| CN113068657B (en) | Intelligent efficient pig raising method and system | |
| CN114898405A (en) | Portable broiler chicken abnormity monitoring system based on edge calculation | |
| CN110334642A (en) | The machine vision recognition method and system of one boar behavior | |
| CN117523610A (en) | Goose group individual gesture recognition method based on improved YOLO v7 | |
| CN115760523A (en) | Animal management method and system based on cloud platform | |
| CN115829234A (en) | Automatic supervision system based on classroom detection and working method thereof | |
| CN114943929A (en) | Real-time detection method for abnormal behaviors of fishes based on image fusion technology | |
| CN115049879A (en) | Swin transform plant leaf disease degree and type identification and classification method based on convolution | |
| CN113591672A (en) | Detection method for identifying fish state based on Mask-Rcnn |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191018 |