CN109856138A - Deep sea net cage healthy fish identifying system and method based on deep learning - Google Patents
Deep sea net cage healthy fish identifying system and method based on deep learning Download PDFInfo
- Publication number
- CN109856138A CN109856138A CN201811547720.4A CN201811547720A CN109856138A CN 109856138 A CN109856138 A CN 109856138A CN 201811547720 A CN201811547720 A CN 201811547720A CN 109856138 A CN109856138 A CN 109856138A
- Authority
- CN
- China
- Prior art keywords
- module
- underwater
- real
- fish
- image
- 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
Landscapes
- Farming Of Fish And Shellfish (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of Deep sea net cage healthy fish identifying system and method based on deep learning, including image capture module, picture recognition module and health identification monitor terminal, wherein, image capture module includes underwater main control plate, real-time Transmission module, memory module, underwater lighting module, Underwater Camera, underwater battery cabin, power circuit；The health identification monitor terminal includes data center, remote monitoring terminal and real-time reception module.Intensive, highly efficient and productive, ecological, the safe growth requirement of aquiculture of the present invention, based on deep learning and wireless sensor technology exploitation, collection image information online acquisition, wireless transmission, image identifying and processing, warning information publication, decision support, remotely with automatic control etc. functions.
Description
Technical field
The invention belongs to marine field, in particular to a kind of Deep sea net cage healthy fish identifying system based on deep learning
And method.
Background technique
With the fast development of Deep sea net cage cultivation in recent years, the yield and quality of aquatic products has all obtained significantly mentioning
It rises.However some problems are also exposed, for example in net cage originally fish disease timely cannot understand and handle, and lead to fish in net cage
Mortality etc., tends to occur similar situation and requires fisherman by virtue of experience to judge whether to take measures, time-consuming and laborious.In order into
One step improves the culture efficiency of Deep sea net cage, saves cost of labor, and appeal event is avoided to occur, and is badly in need of the depth based on deep learning
Extra large net cage healthy fish identifying system is monitored the monitor video real-time Transmission in net cage to land using wireless communication technique
Center, and in monitor terminal real-time display, the picture recognition module of monitoring center builds deep learning mould based on using CNN
Type, selection is suitble to the softmax classifier of more classification tasks to judge the classification in picture, while being mentioned using convolution kernel
The relevant feature of biology is taken, the type and health status of fish are judged according to these features, so that poultry feeders are according to these letters
Breath, takes corresponding measure, avoids disaster, improves culture efficiency.
Summary of the invention
In view of this, the purpose of the present invention is to provide a kind of, the Deep sea net cage healthy fish based on deep learning identifies system
System, the growth requirement that aquiculture is intensive, highly efficient and productive, ecological, safe, based on deep learning and wireless sensor skill
Art exploitation, collection image information online acquisition, wireless transmission, image identifying and processing, warning information publication, decision support, far
The functions such as journey and automatic control.
In order to achieve the above objectives, the present invention provides a kind of, and the Deep sea net cage healthy fish based on deep learning identifies system
System, including image capture module, picture recognition module and health identification monitor terminal, wherein
Described image acquisition module includes underwater main control plate, real-time Transmission module, memory module, underwater lighting module, water
Lower video camera, underwater battery cabin, power circuit；It is described health identification monitor terminal include data center, remote monitoring terminal and
Real-time reception module；
Underwater battery cabin and power circuit are other parts power supply in image capture module；Underwater main control plate and real-time Transmission
Module, memory module, underwater lighting module and Underwater Camera are separately connected, and underwater lighting module is that Underwater Camera is being shot
When auxiliary light, the image that Underwater Camera takes is stored in memory module, and underwater main control plate handles acquired image
Real-time reception module is sent to by real-time Transmission module afterwards, real-time reception module is connect with data center, picture recognition module
It is connect with remote monitoring terminal respectively at data center, picture recognition module is to the image in data center according to convolutional Neural
The deep learning model built based on network selects to be sent to remote monitoring terminal after being suitble to classifier to carry out identification judgement；
Remote monitoring terminal receives and real-time display monitor video, and the recognition result sent to picture recognition module carries out analysis and answers
With, and alarm based on the analysis results.
Preferably, described image acquisition module further includes temperature sensor and pressure sensor, acquires underwater environment temperature
With hydraulic pressure.
Preferably, the memory module is mechanical hard disk.
Preferably, the real-time Transmission of the real-time Transmission module and real-time reception module uses wireless bridge.
Preferably, the wireless bridge uses LA-5839 bridge 08.
Based on above-mentioned purpose, the present invention also provides a kind of image-recognizing methods, comprising the following steps:
Data center's Video stream information is resolved into pictorial information by S10, is carried out by hand using LabelImg tool to picture
Calibration obtains class label, and target fish health status and the classification of fish in each calibration frame are determined with label；
S20 is trained and verifies to the data marked, using picture and picture tag as deep learning neural network
Training set and verifying collection, pass through the last full articulamentum of fine-tune VGG16 convolutional neural networks and using Adam optimization
Algorithm is trained data to obtain optimal healthy fish identification model；
S30 tests multiple test datas, calls optimal healthy fish identification model and test program to test chart
Piece carries out the detection of fish, by observing test result, analyzes the generalization ability of healthy fish identification model effect, subsequent monitoring
Video uses this identification model, carries out differentiating fish species and health status, and result is deposited to data center.
Preferably, the model is returned by five convolutional layers, five pond layers, three full articulamentums and a Softmax
Layer composition.
Compared with prior art, the Deep sea net cage healthy fish identifying system and side disclosed by the invention based on deep learning
Method, image capture module in net cage fish monitor, by front end monitor be responsible for net cage around and case in environment monitoring, so
Data are returned in such a way that nearest convergence and backbone transport combine afterwards, just finally by health identification monitor terminal
The monitoring comprehensive to culturing area may be implemented, in conjunction with picture recognition module, healthy fish situation identification function can be realized
Energy.
Picture recognition module is to the video and pictorial information of Image Acquisition, with building depth based on convolutional neural networks
Learning model selects that classifier is suitble to judge the classification in picture, while extracting the relevant spy of biology using convolution kernel
Sign, the type and health status of fish are judged according to these features.
Health identification monitor terminal by realtime image data collect with display, healthy early warning, data loading, permission control,
The functions such as system administration, apply recognition result.Intensive, highly efficient and productive, ecological, the safe development of aquiculture
Demand is developed based on deep learning and wireless sensor technology, collects image information online acquisition, wireless transmission, image recognition
With processing, warning information publication, decision support, remotely with automatically control etc. functions.
Detailed description of the invention
In order to keep the purpose of the present invention, technical scheme and beneficial effects clearer, the present invention provides following attached drawing and carries out
Illustrate:
Fig. 1 is the structural block diagram of Deep sea net cage healthy fish identifying system of the embodiment of the present invention based on deep learning；
Fig. 2 is Deep sea net cage healthy fish identifying system image capture module knot of the embodiment of the present invention based on deep learning
Structure schematic diagram；
Fig. 3 is Deep sea net cage healthy fish recognition methods flow chart of steps of the embodiment of the present invention based on deep learning.
Below in conjunction with attached drawing, a preferred embodiment of the present invention will be described in detail.
System embodiment is referring to Fig. 1,2, including image capture module 10, picture recognition module 20 and health identification monitoring are eventually
End, wherein
Image capture module 10 includes underwater main control plate 11, real-time Transmission module 12, memory module 13, underwater lighting module
14, Underwater Camera 15, underwater battery cabin 16, power circuit 17；Health identification monitor terminal includes data center 31, long-range prison
Control terminal 32 and real-time reception module 33；
Underwater battery cabin 16 and power circuit 17 are other parts power supply in image capture module 10；Underwater main control plate 11 with
Real-time Transmission module 12, memory module 13, underwater lighting module 14 and Underwater Camera 15 are separately connected, underwater lighting module 14
For Underwater Camera 15, auxiliary light, the image that Underwater Camera 15 takes are stored in memory module 13 when shooting, under water
Master control borad 11 passes through real-time Transmission module 12 after handling acquired image and is sent to real-time reception module 33, real-time reception mould
Block 33 is connect with data center 31, and picture recognition module 20 is connect with remote monitoring terminal 32 respectively at data center 31, image
Identification module 20 is to the image in data center 31 according to the deep learning model built based on convolutional neural networks, selection
Suitable classifier is sent to remote monitoring terminal 32 after carrying out identification judgement；Remote monitoring terminal 32 receives and real-time display monitoring
Video, and the recognition result sent to picture recognition module 20 carries out analysis application, and alarms based on the analysis results.
In specific embodiment, it is the underwater main control plate 11 of core and reality that the core circuit of image capture module 10, which is by ARM,
When transmission module 12 form, in entire deep water mesh cage healthy fish identifying system be in extremely important status, can by it
Switched with controlling Underwater Camera 15 and the switch of underwater lighting module 14 with 15 operating mode of Underwater Camera, and by video or
Image information is stored after the resume module is complete to locally or by cable technology remote monitoring terminal 32 is sent to, and image is adopted
Collection module 10 is based on ICP/IP protocol, and video is passed through video, audio, data acquisition system using H.264 coding techniques together
Wireless network transmissions equipment carries out point-to-point wireless transmission.The LA-5839 bridge 08 that real-time Transmission module 12 uses is a height
The remote outdoor wireless bridge of cost performance, 08 binary channels output power 800MW of LA-5839 bridge, antenna using built-in 5.8G,
18dbi dual polarization mimo antenna (16 degree of horizontal/vertical), 15 kilometers of signal stabilization transmission range, 10 km transmission rate 100M
Left and right.Device systems carry frequency detecting function, can detecte nearby corresponding 5G channel usage, the channel that avoids interference and mention
High spectrum utilization rate；Temperature sensor and pressure sensor provide image capture device operating temperature and water pressures, prevent from setting
It is standby to work long hours under extreme environment；Memory module 13 can be mechanical hard disk, the backup for image data.This module
Wireless bridge uses 802.11A/B/G/N and OFDM/QPSK modulation technique, has long transmission distance, receiving sensitivity height, non-view
Feature away from long transmission distance, image clearly smoothness uploads high definition real time monitoring biological in net cage with data to realize.
Health identification monitor terminal provides real-time monitor video for user and shows and remotely control image capture module
10 working conditions apply the recognition result of picture recognition module 20, realize the functions such as fish identification and early warning, alarm.
Other than it can monitor in real time, it can also configure intra-record slack byte and arbitrarily selected between 1-999 minutes, user is facilitated to play back monitoring view
Frequently.User can send instruction control underwater lighting module 14 by terminal and carry out auxiliary shooting, to obtain higher-quality
Video and pictorial information also can control the switch of the Underwater Camera 15 of high definition stabilization.When event occurs for image capture module 10
When barrier, system issues fault alarm automatically, informs that staff replaces related equipment in time.Remote monitoring terminal 32 receives
Recognition result from data center 31 is shown and is applied, and is facilitated user to understand fish growth and health status, is avoided calamity
Evil occurs, and improves the degree of automation of Deep sea net cage cultivation, improves benefit.
The step flow chart of embodiment of the method referring to Fig. 3 the following steps are included:
Data center's Video stream information is resolved into pictorial information by S10, is carried out by hand using LabelImg tool to picture
Calibration obtains class label, and target fish health status and the classification of fish in each calibration frame are determined with label；
S20 is trained and verifies to the data marked, using picture and picture tag as deep learning neural network
Training set and verifying collection, pass through the last full articulamentum of fine-tune VGG16 convolutional neural networks and using Adam optimization
Algorithm is trained data to obtain optimal healthy fish identification model；
S30 tests multiple test datas, calls optimal healthy fish identification model and test program to test chart
Piece carries out the detection of fish, by observing test result, analyzes the generalization ability of healthy fish identification model effect, subsequent monitoring
Video uses this identification model, carries out differentiating fish species and health status, and result is deposited to data center.
In specific embodiment, model is returned by five convolutional layers, five pond layers, three full articulamentums and a Softmax
Layer is returned to form.The work of convolutional layer is the feature to select each angle of preceding layer characteristic pattern from different angles.Getting volume
After product feature, convolution feature is divided on disjoint range, then obtains Chi Huahou's with the maximum feature in these regions
Convolution feature.First layer obtains 64 different characteristic patterns after the convolution twice of 64 different convolution kernels, and using primary
Pooling carries out down-sampling；The second layer is after 128 convolution nuclear convolution twice, using a pooling；Third layer warp
It crosses after the convolution of 256 convolution kernel three times, using pooling；4th layer repeats three 512 convolution nuclear convolutions with layer 5
Pooling again later carries out feature recombination using the full articulamentum of 3 FC, finally plus a Softmax return layer obtain it is each
The class probability of class.
The above is merely preferred embodiments of the present invention, be not intended to limit the invention, it is all in spirit of the invention and
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within principle.
Claims (7)
1. a kind of Deep sea net cage healthy fish identifying system based on deep learning, which is characterized in that including image capture module,
Picture recognition module and health identification monitor terminal, wherein
Described image acquisition module includes underwater main control plate, real-time Transmission module, memory module, underwater lighting module, takes the photograph under water
Camera, underwater battery cabin, power circuit；It is described health identification monitor terminal include data center, remote monitoring terminal and in real time
Receiving module；
Underwater battery cabin and power circuit are other parts power supply in image capture module；Underwater main control plate and real-time Transmission mould
Block, memory module, underwater lighting module and Underwater Camera are separately connected, underwater lighting module be Underwater Camera when shooting
Auxiliary light, the image that Underwater Camera takes is stored in memory module, after underwater main control plate handles acquired image
Be sent to real-time reception module by real-time Transmission module, real-time reception module is connect with data center, picture recognition module with
Remote monitoring terminal is connected respectively at data center, and picture recognition module is to the image in data center according to convolutional Neural net
The deep learning model built based on network selects to be sent to remote monitoring terminal after being suitble to classifier to carry out identification judgement；Far
Range monitoring terminal receives and real-time display monitor video, and the recognition result sent to picture recognition module carries out analysis application,
And it alarms based on the analysis results.
2. system according to claim 1, which is characterized in that described image acquisition module further includes temperature sensor and pressure
Force snesor acquires underwater environment temperature and hydraulic pressure.
3. system according to claim 1, which is characterized in that the memory module is mechanical hard disk.
4. system according to claim 1, which is characterized in that the real-time Transmission module is real-time with real-time reception module
Transmission uses wireless bridge.
5. system according to claim 4, which is characterized in that the wireless bridge uses LA-5839 bridge 08.
6. image-recognizing method in system described in a kind of one of claim 1-5, which comprises the following steps:
Data center's Video stream information is resolved into pictorial information by S10, carries out manual calibration to picture using LabelImg tool
Class label is obtained, target fish health status and the classification of fish in each calibration frame are determined with label；
S20 is trained and verifies to the data marked, using picture and picture tag as the instruction of deep learning neural network
Practice collection and verifying collection, pass through the last full articulamentum of fine-tune VGG16 convolutional neural networks and utilizes Adam optimization algorithm
Data are trained to obtain optimal healthy fish identification model；
S30 tests multiple test datas, call optimal healthy fish identification model and test program to test picture into
The detection of row fish analyzes the generalization ability of healthy fish identification model effect, subsequent monitor video by observing test result
With this identification model, carry out differentiating fish species and health status, and result is deposited to data center.
7. according to the method described in claim 6, it is characterized in that, the model is by five convolutional layers, five pond layers, three
Full articulamentum and a Softmax return layer composition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811547720.4A CN109856138A (en) | 2018-12-18 | 2018-12-18 | Deep sea net cage healthy fish identifying system and method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811547720.4A CN109856138A (en) | 2018-12-18 | 2018-12-18 | Deep sea net cage healthy fish identifying system and method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109856138A true CN109856138A (en) | 2019-06-07 |
Family
ID=66891491
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811547720.4A Pending CN109856138A (en) | 2018-12-18 | 2018-12-18 | Deep sea net cage healthy fish identifying system and method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109856138A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110738999A (en) * | 2019-10-30 | 2020-01-31 | 浙江大学宁波理工学院 | Fish swim bladder abnormality detection method and device based on audio recognition technology |
| CN110764093A (en) * | 2019-09-30 | 2020-02-07 | 苏州佳世达电通有限公司 | Underwater biological identification system and method thereof |
| CN111198549A (en) * | 2020-02-18 | 2020-05-26 | 陈文翔 | Poultry breeding monitoring management system based on big data |
| CN111406693A (en) * | 2020-04-23 | 2020-07-14 | 上海海洋大学 | Marine ranch fishery resource maintenance effect evaluation method based on bionic sea eels |
| CN111738139A (en) * | 2020-06-19 | 2020-10-02 | 中国水产科学研究院渔业机械仪器研究所 | Cultured fish monitoring method and system based on image recognition |
| NO20190760A1 (en) * | 2019-06-19 | 2020-12-21 | Subc3D As | System and procedure for imaging and counting external structures on a fish |
| CN112931309A (en) * | 2021-02-02 | 2021-06-11 | 中国水利水电科学研究院 | Method and system for monitoring fish proliferation and releasing direction |
| CN113159005A (en) * | 2021-05-28 | 2021-07-23 | 青海中水数易信息科技有限责任公司 | Machine learning-based water level and foreign matter identification integrated monitoring system and method |
| CN113569971A (en) * | 2021-08-02 | 2021-10-29 | 浙江索思科技有限公司 | Image recognition-based catch target classification detection method and system |
| CN114240686A (en) * | 2022-02-24 | 2022-03-25 | 深圳市旗扬特种装备技术工程有限公司 | Wisdom fishery monitoring system |
| CN114609952A (en) * | 2022-04-07 | 2022-06-10 | 广州大学 | Fish culture monitoring method and system based on machine vision |
| CN114626460A (en) * | 2022-03-15 | 2022-06-14 | 河海大学 | Underwater acquisition and in-situ recognition device and method for algae image data |
| CN114954863A (en) * | 2022-07-05 | 2022-08-30 | 中国农业大学 | Autonomous inspection early warning bionic robotic dolphin system and control method |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104992189A (en) * | 2015-07-22 | 2015-10-21 | 河海大学常州校区 | Deep learning network model based method for identifying abnormal behavior of fish shoal |
| CN106198909A (en) * | 2016-06-30 | 2016-12-07 | 中南大学 | A kind of aquaculture water quality Forecasting Methodology based on degree of depth study |
| CN108038459A (en) * | 2017-12-20 | 2018-05-15 | 深圳先进技术研究院 | A kind of detection recognition method of aquatic organism, terminal device and storage medium |
| CN108121968A (en) * | 2017-12-20 | 2018-06-05 | 佛山市洛克威特科技有限公司 | A kind of fish monitoring method |
| CN108235001A (en) * | 2018-01-29 | 2018-06-29 | 上海海洋大学 | A kind of deep-sea video quality objective assessment model based on space-time characteristic |
| CN108444447A (en) * | 2018-02-28 | 2018-08-24 | 哈尔滨工程大学 | A kind of fishing net in underwater obstacle avoidance system autonomous detection method in real time |
| CN207940236U (en) * | 2018-01-30 | 2018-10-09 | 上海理工大学 | Shoal of fish automatic health detects and archive management device |
| CN108805064A (en) * | 2018-05-31 | 2018-11-13 | 中国农业大学 | A kind of fish detection and localization and recognition methods and system based on deep learning |
| CN108875644A (en) * | 2018-06-21 | 2018-11-23 | 四川盈乾建设工程有限公司 | A kind of aquatic animal density survey method |
| CN108921058A (en) * | 2018-06-19 | 2018-11-30 | 厦门大学 | Fish identification method, medium, terminal device and device based on deep learning |
-
2018
- 2018-12-18 CN CN201811547720.4A patent/CN109856138A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104992189A (en) * | 2015-07-22 | 2015-10-21 | 河海大学常州校区 | Deep learning network model based method for identifying abnormal behavior of fish shoal |
| CN106198909A (en) * | 2016-06-30 | 2016-12-07 | 中南大学 | A kind of aquaculture water quality Forecasting Methodology based on degree of depth study |
| CN108038459A (en) * | 2017-12-20 | 2018-05-15 | 深圳先进技术研究院 | A kind of detection recognition method of aquatic organism, terminal device and storage medium |
| CN108121968A (en) * | 2017-12-20 | 2018-06-05 | 佛山市洛克威特科技有限公司 | A kind of fish monitoring method |
| CN108235001A (en) * | 2018-01-29 | 2018-06-29 | 上海海洋大学 | A kind of deep-sea video quality objective assessment model based on space-time characteristic |
| CN207940236U (en) * | 2018-01-30 | 2018-10-09 | 上海理工大学 | Shoal of fish automatic health detects and archive management device |
| CN108444447A (en) * | 2018-02-28 | 2018-08-24 | 哈尔滨工程大学 | A kind of fishing net in underwater obstacle avoidance system autonomous detection method in real time |
| CN108805064A (en) * | 2018-05-31 | 2018-11-13 | 中国农业大学 | A kind of fish detection and localization and recognition methods and system based on deep learning |
| CN108921058A (en) * | 2018-06-19 | 2018-11-30 | 厦门大学 | Fish identification method, medium, terminal device and device based on deep learning |
| CN108875644A (en) * | 2018-06-21 | 2018-11-23 | 四川盈乾建设工程有限公司 | A kind of aquatic animal density survey method |
Non-Patent Citations (3)
| Title |
|---|
| 刘敬彪等: "基于光缆的深海摄像***的设计与实现", 《嵌入式技术》 * |
| 胡昱等: "基于3G+VSaas技术的深水网箱养殖远程视频监控***", 《南方水产科学》 * |
| 顾郑平等: "基于深度学习的鱼类分类算法研究", 《计算机应用与软件》 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2599532A (en) * | 2019-06-19 | 2022-04-06 | Subc3D As | System and method for depiction and counting of external structures on a fish |
| NO20190760A1 (en) * | 2019-06-19 | 2020-12-21 | Subc3D As | System and procedure for imaging and counting external structures on a fish |
| WO2020256566A1 (en) * | 2019-06-19 | 2020-12-24 | Subc3D As | System and method for depiction and counting of external structures on a fish |
| GB2599532B (en) * | 2019-06-19 | 2022-10-26 | Subc3D As | System and method for depiction and counting of external structures on a fish |
| NO347348B1 (en) * | 2019-06-19 | 2023-09-25 | Subc3D As | System and procedure for imaging and counting external structures on a fish |
| CN110764093A (en) * | 2019-09-30 | 2020-02-07 | 苏州佳世达电通有限公司 | Underwater biological identification system and method thereof |
| CN110738999A (en) * | 2019-10-30 | 2020-01-31 | 浙江大学宁波理工学院 | Fish swim bladder abnormality detection method and device based on audio recognition technology |
| CN110738999B (en) * | 2019-10-30 | 2022-08-16 | 浙江大学宁波理工学院 | Fish swim bladder abnormality detection method and device based on audio recognition technology |
| CN111198549A (en) * | 2020-02-18 | 2020-05-26 | 陈文翔 | Poultry breeding monitoring management system based on big data |
| CN111198549B (en) * | 2020-02-18 | 2020-11-06 | 湖南伟业动物营养集团股份有限公司 | Poultry breeding monitoring management system based on big data |
| CN111406693A (en) * | 2020-04-23 | 2020-07-14 | 上海海洋大学 | Marine ranch fishery resource maintenance effect evaluation method based on bionic sea eels |
| CN111738139A (en) * | 2020-06-19 | 2020-10-02 | 中国水产科学研究院渔业机械仪器研究所 | Cultured fish monitoring method and system based on image recognition |
| CN112931309B (en) * | 2021-02-02 | 2021-11-09 | 中国水利水电科学研究院 | Method and system for monitoring fish proliferation and releasing direction |
| CN112931309A (en) * | 2021-02-02 | 2021-06-11 | 中国水利水电科学研究院 | Method and system for monitoring fish proliferation and releasing direction |
| CN113159005A (en) * | 2021-05-28 | 2021-07-23 | 青海中水数易信息科技有限责任公司 | Machine learning-based water level and foreign matter identification integrated monitoring system and method |
| CN113569971A (en) * | 2021-08-02 | 2021-10-29 | 浙江索思科技有限公司 | Image recognition-based catch target classification detection method and system |
| CN113569971B (en) * | 2021-08-02 | 2022-03-25 | 浙江索思科技有限公司 | Image recognition-based catch target classification detection method and system |
| CN114240686A (en) * | 2022-02-24 | 2022-03-25 | 深圳市旗扬特种装备技术工程有限公司 | Wisdom fishery monitoring system |
| CN114626460A (en) * | 2022-03-15 | 2022-06-14 | 河海大学 | Underwater acquisition and in-situ recognition device and method for algae image data |
| CN114626460B (en) * | 2022-03-15 | 2024-06-07 | 河海大学 | Underwater acquisition and in-situ identification device and method for algae image data |
| CN114609952A (en) * | 2022-04-07 | 2022-06-10 | 广州大学 | Fish culture monitoring method and system based on machine vision |
| CN114609952B (en) * | 2022-04-07 | 2024-04-26 | 广州大学 | Fish culture monitoring method and system based on machine vision |
| CN114954863A (en) * | 2022-07-05 | 2022-08-30 | 中国农业大学 | Autonomous inspection early warning bionic robotic dolphin system and control method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109856138A (en) | Deep sea net cage healthy fish identifying system and method based on deep learning | |
| CN107094683B (en) | Automatic feeding and water quality monitoring control system for aquaculture | |
| KR101598898B1 (en) | The system for precaution for paralichthys olivaceus disease using and an analysis of image and sound | |
| CN102509305B (en) | Animal behavior detection device based on omnidirectional vision | |
| CN201846435U (en) | Automatic monitoring and continuous shooting device for artificial fish reef trapping test | |
| KR102002632B1 (en) | Fruit monitoring system and method at the same | |
| CN109145794B (en) | Health monitoring method for marine ranch net cage | |
| CN112131927A (en) | Sow delivery time prediction system based on posture transformation characteristics in later gestation period | |
| CN114581816A (en) | Real-time detection and counting method for solanaceous vegetables and fruits in plant factory | |
| CN113762113A (en) | Livestock parturition behavior monitoring method and device | |
| CN114898405B (en) | Portable broiler chicken anomaly monitoring system based on edge calculation | |
| CN110771548B (en) | Biomass evaluation device and method for deep and open sea aquaculture net cage based on statistics | |
| CN113822198A (en) | Peanut growth monitoring method, system and medium based on UAV-RGB image and deep learning | |
| CN116778310A (en) | Acoustic-optical image fusion monitoring method and system for aquaculture | |
| CN103563808A (en) | Offshore type automatic system for deep sea aquaculture net cages | |
| CN114399664A (en) | Intelligent monitoring and control method and system for growth state of plant seedlings | |
| CN111027436A (en) | Northeast black fungus disease and pest image recognition system based on deep learning | |
| CN112036248A (en) | Intelligent fishpond management system based on scene recognition | |
| CN117315719A (en) | Safety helmet wearing identification method based on edge technology and deep learning | |
| CN112965557A (en) | Intelligent lobster breeding monitoring device and monitoring method | |
| CN112233353A (en) | Artificial intelligence-based anti-fishing monitoring system and monitoring method thereof | |
| CN104969885B (en) | A kind of marine cage fish culture system and method | |
| CN117636169A (en) | Forest patrol method and system | |
| CN112714411A (en) | Agricultural monitoring device with movable nodes | |
| CN102986604A (en) | Intelligent octopus catching device and use method thereof |
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: 20190607 |