CN111738140A - Image recognition-based cultured shrimp real-time monitoring method - Google Patents
Image recognition-based cultured shrimp real-time monitoring method Download PDFInfo
- Publication number
- CN111738140A CN111738140A CN202010565214.9A CN202010565214A CN111738140A CN 111738140 A CN111738140 A CN 111738140A CN 202010565214 A CN202010565214 A CN 202010565214A CN 111738140 A CN111738140 A CN 111738140A
- Authority
- CN
- China
- Prior art keywords
- shrimps
- shrimp
- real
- cultured
- training set
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 24
- 238000000034 method Methods 0.000 title claims abstract description 20
- 241000238557 Decapoda Species 0.000 title claims description 38
- 241000143060 Americamysis bahia Species 0.000 claims abstract description 47
- 238000012549 training Methods 0.000 claims abstract description 22
- 230000000384 rearing effect Effects 0.000 claims abstract description 11
- 238000002372 labelling Methods 0.000 claims description 6
- 238000009313 farming Methods 0.000 claims description 2
- 239000012855 volatile organic compound Substances 0.000 abstract 3
- 238000013528 artificial neural network Methods 0.000 abstract 1
- 230000004083 survival effect Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000009395 breeding Methods 0.000 description 2
- 230000001488 breeding effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 206010000349 Acanthosis Diseases 0.000 description 1
- 230000003187 abdominal effect Effects 0.000 description 1
- 238000005273 aeration Methods 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 210000000080 chela (arthropods) Anatomy 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 235000003642 hunger Nutrition 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 235000018343 nutrient deficiency Nutrition 0.000 description 1
- 230000017448 oviposition Effects 0.000 description 1
- 238000006213 oxygenation reaction Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01K—ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
- A01K61/00—Culture of aquatic animals
- A01K61/50—Culture of aquatic animals of shellfish
- A01K61/59—Culture of aquatic animals of shellfish of crustaceans, e.g. lobsters or shrimps
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/80—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in fisheries management
- Y02A40/81—Aquaculture, e.g. of fish
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Environmental Sciences (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Multimedia (AREA)
- Marine Sciences & Fisheries (AREA)
- Zoology (AREA)
- Animal Husbandry (AREA)
- Biodiversity & Conservation Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The invention discloses a method for monitoring cultured shrimps in real time based on image recognition, which comprises the following steps: the method comprises the steps of constructing a sample training set, marking the cultured shrimps in the sample training set, then exporting a VOC (volatile organic Compounds) data set, calling a residual convolution network in a Faster-RCNN (fast-remote storage neural network) network model to train the VOC data set to obtain the feature weight of the shrimps, shooting image data of all the cultured shrimps in a temporary rearing box, and importing the image data into the feature weight of the shrimps for real-time monitoring.
Description
Technical Field
The invention relates to the technical field of computer image recognition, in particular to a method for monitoring characteristics of shrimps cultured in a temporary culture environment in real time.
Background
The quality of the famous special shrimps is closely related to the profits of the famous special shrimps, and the health of the parent shrimps is the key for successful breeding and breeding of the shrimps. The existing temporary rearing box is simple and crude, the survival of shrimps is kept only through aeration and oxygenation, and the characteristic states of the shrimps are observed through manual inspection, but the characteristic states of the shrimps cannot be accurately known through manual observation in a dark environment. The shrimps are struggled, eaten mutually, molted and killed in the temporary rearing process, and the parent shrimps also have the characteristics of spawning, which are vain and difficult through manual statistics. The shrimps in the temporary rearing box can fight due to over-high density, and can also cause difficult shelling due to mutual rearing by hunger or nutrient deficiency, and some shrimps can also die due to stress reaction. Therefore, how to design a system capable of monitoring the living state of the shrimps in the temporary rearing box in real time for real-time monitoring is a problem to be solved at present.
Disclosure of Invention
The invention provides a method and a system for monitoring the survival state of shrimps cultured in a temporary culture box based on image recognition, and aims to solve the technical problem of monitoring the survival state of the shrimps cultured in the temporary culture box in real time.
The invention solves the technical problems through the following technical scheme:
a method for monitoring cultured shrimps in real time based on image recognition comprises the following steps:
constructing a sample training set;
marking the cultured shrimps in the sample training set and then deriving a VOC data set;
training the VOC data set by calling a residual convolution network in a Faster-RCNN network model to obtain shrimp characteristic weight;
and shooting image data of all the cultured shrimps in the temporary culture box, and importing the shrimp characteristic weight for real-time monitoring.
Preferably, before the constructing the sample training set, the method further comprises the step of using an image transmission module to control a camera to shoot a part of the images of the cultured shrimps in the temporary rearing box.
Preferably, labeling the farmed shrimp in the sample training set comprises:
setting category names according to the states of the cultured shrimps;
manually labeling the cultured shrimps in the sample training set;
the VOC data set is examined and exported.
Preferably, the obtaining of the shrimp feature weight includes:
arranging partial image position paths of the cultured shrimps;
calling the residual convolution network in the Faster-RCNN network model to train the VOC data set to obtain shrimp characteristic weight;
setting a class value and a learning rate.
Preferably, the real-time monitoring result is displayed on a screen and/or uploaded to a cloud database.
On the basis of the common knowledge in the field, the above preferred conditions can be combined randomly to obtain the preferred embodiments of the invention.
The positive progress effects of the invention are as follows: the survival state of the shrimps in the temporary rearing box is monitored in real time, and the method has the advantages of high automation degree, accurate small target detection and the like.
Drawings
FIG. 1 is a flowchart of a method of an embodiment of the present invention for monitoring shrimp cultivation based on image recognition;
FIG. 2 is a flow chart of a method for monitoring cultured shrimps in a training set by using labeled samples according to an embodiment of the invention;
fig. 3 is a flow chart of shrimp feature weight acquisition in an embodiment of the shrimp farming monitoring method based on image recognition.
Detailed Description
To facilitate an understanding of the present application, the present application will now be described more fully with reference to the accompanying drawings. Preferred embodiments of the present application are shown in the drawings. This application may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
It will be understood that when an element is referred to as being "connected" to another element, it can be directly connected to the other element and be integral therewith, or intervening elements may also be present. The terms "mounted," "one end," "the other end," and the like are used herein for illustrative purposes only.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the description of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
Fig. 1 shows a flow chart of a method for monitoring cultivated shrimps in an embodiment of the present invention:
s01: using an image transmission module to control a camera to shoot images of part of the cultured shrimps in the temporary culture box;
in one example, 1500-2500 pictures of shrimps in the temporary rearing box are taken by controlling the high power camera through the 5G wireless image transmission module, and specifically 1500, 2000, 2400 or the like.
S02: constructing a sample training set;
in one example, a sample training set is constructed by using a camera to take images of a portion of the cultured shrimps in the temporary rearing box.
S03: marking the cultured shrimps in the sample training set and then deriving a VOC data set;
in an alternative example, as shown in fig. 2, the steps of labeling the farmed shrimp in the sample training set are as follows:
s031: setting category names according to the states of the cultured shrimps;
in an alternative example, the set categories are called dead shrimp, live shrimp, acanthosis, oviposition shrimp, molting shrimp, wherein the dead shrimp is characterized by a reddish body, a recumbent posture, and a severely mutilated body. The live shrimp features normal body color, normal posture, sound shrimp body, the shrimp with pincers deficiency, the egg-holding shrimp features its abdominal egg mass and the shell-molting shrimp features its shell separated from shrimp body.
S032: manually labeling the cultured shrimps in the sample training set;
s033: the VOC data set is examined and exported.
In an alternative example, the marked sample is tested to check whether a fault occurs, and the VOC data set is derived after the test.
S04: training the VOC data set by calling a residual convolution network in a fast-RCNN network model to obtain shrimp characteristic weight
In an alternative example, as shown in fig. 3, the steps of obtaining the shrimp feature weights are as follows:
s041: arranging partial image position paths of the cultured shrimps;
in an alternative example, a train.txt file is made, specifically, a Python processing data set picture file and a target position file are integrated into a document with a training set picture path and a target position.
S042: calling the residual convolution network in the Faster-RCNN network model to train the VOC data set to obtain shrimp characteristic weight;
in an optional example, a reset 50 network is used as a Back bone network of the fast-RCNN network model, and after the reset 50 network trains the VOC data set, weights are saved every 1500 steps to 2500 steps, specifically 1500 steps, 2000 steps and 2500 steps, so as to generate a shrimp model weight VOC _ weights _ renewnet.
S043: setting a class value and a learning rate.
In an alternative example, the category value NUM _ CLASSES ═ 5 may be set to represent dead shrimp, live shrimp, delectated shrimp, oviferous shrimp, and decapsulated shrimp, respectively, and the learing _ RATE may be 0.001 to 0.01, specifically, 0.001, 0.005, 0.01, etc., INIT _ EPOCH ═ 0, and fresh _ EPOCH ═ 25.
S05: shooting image data of all the cultured shrimps in the temporary culture box, and importing the shrimp characteristic weight for real-time monitoring;
in one example, a high power camera view is received as video data for detection in a 5G wireless map, and each shrimp in the foster box is framed and displays the category name and detection accuracy.
S06: and displaying the real-time monitoring result on a screen and/or uploading the real-time monitoring result to a cloud database.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that this is by way of example only, and that the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the spirit and scope of the invention, and these changes and modifications are within the scope of the invention.
Claims (5)
1. A shrimp farming real-time monitoring method based on image recognition is characterized by comprising the following steps:
constructing a sample training set;
marking the cultured shrimps in the sample training set and then deriving a VOC data set;
training the VOC data set by calling a residual convolution network in a Faster-RCNN network model to obtain shrimp characteristic weight;
and shooting image data of all the cultured shrimps in the temporary culture box, and importing the shrimp characteristic weight for real-time monitoring.
2. The method as claimed in claim 1, wherein before the constructing of the sample training set, the method further comprises using a graph-passing module to control a camera to capture an image of a part of the cultured shrimps in the temporary rearing box.
3. The method of claim 1, wherein labeling the cultured shrimps in the sample training set comprises:
setting category names according to the states of the cultured shrimps;
manually labeling the cultured shrimps in the sample training set;
the VOC data set is examined and exported.
4. The method for monitoring the cultivated shrimps in real time based on the image recognition as claimed in claim 1, wherein the obtaining of the shrimp feature weight comprises:
arranging partial image position paths of the cultured shrimps;
calling the residual convolution network in the Faster-RCNN network model to train the VOC data set to obtain shrimp characteristic weight;
setting a class value and a learning rate.
5. The method for real-time monitoring of the cultivated shrimps based on the image recognition as claimed in any one of claims 1-4, wherein the real-time monitoring result is displayed on a screen and/or uploaded to a cloud database.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010565214.9A CN111738140A (en) | 2020-06-19 | 2020-06-19 | Image recognition-based cultured shrimp real-time monitoring method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010565214.9A CN111738140A (en) | 2020-06-19 | 2020-06-19 | Image recognition-based cultured shrimp real-time monitoring method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111738140A true CN111738140A (en) | 2020-10-02 |
Family
ID=72650722
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010565214.9A Pending CN111738140A (en) | 2020-06-19 | 2020-06-19 | Image recognition-based cultured shrimp real-time monitoring method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111738140A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109543679A (en) * | 2018-11-16 | 2019-03-29 | 南京师范大学 | A kind of dead fish recognition methods and early warning system based on depth convolutional neural networks |
CN109598224A (en) * | 2018-11-27 | 2019-04-09 | 微医云(杭州)控股有限公司 | Recommend white blood cell detection method in the Sections of Bone Marrow of convolutional neural networks based on region |
CN110211173A (en) * | 2019-04-03 | 2019-09-06 | 中国地质调查局发展研究中心 | A kind of paleontological fossil positioning and recognition methods based on deep learning |
CN110705630A (en) * | 2019-09-27 | 2020-01-17 | 聚时科技(上海)有限公司 | Semi-supervised learning type target detection neural network training method, device and application |
-
2020
- 2020-06-19 CN CN202010565214.9A patent/CN111738140A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109543679A (en) * | 2018-11-16 | 2019-03-29 | 南京师范大学 | A kind of dead fish recognition methods and early warning system based on depth convolutional neural networks |
CN109598224A (en) * | 2018-11-27 | 2019-04-09 | 微医云(杭州)控股有限公司 | Recommend white blood cell detection method in the Sections of Bone Marrow of convolutional neural networks based on region |
CN110211173A (en) * | 2019-04-03 | 2019-09-06 | 中国地质调查局发展研究中心 | A kind of paleontological fossil positioning and recognition methods based on deep learning |
CN110705630A (en) * | 2019-09-27 | 2020-01-17 | 聚时科技(上海)有限公司 | Semi-supervised learning type target detection neural network training method, device and application |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
MARTI et al. | Food Habits 8 | |
Barreto et al. | Emerging indicators of fish welfare in aquaculture | |
CN111738139A (en) | Cultured fish monitoring method and system based on image recognition | |
García-Salgado et al. | Evaluation of trail-cameras for analyzing the diet of nesting raptors using the Northern Goshawk as a model | |
CN114341951A (en) | Method and system for diagnosing diseases, physiological changes or other internal conditions of crustaceans by non-invasive means | |
CN112232977A (en) | Aquatic product cultivation evaluation method, terminal device and storage medium | |
Stockton et al. | Female mate preference in an invasive phytopathogen vector: how learning may influence mate choice and fecundity in D iaphorina citri | |
Irigoyen et al. | Habitat use, seasonality and demography of an apex predator: sevengill shark Notorynchus cepedianus in northern Patagonia | |
Arthur et al. | Feeding ecology of green turtles (Chelonia mydas) from Shoalwater Bay, Australia | |
Postma et al. | Age-related reproductive variation in a wild marine mammal population | |
Cooke et al. | On Improved Care of Black Bass During Live‐Release Competitive Angling Events–Recent Innovations and Associated Research Needs | |
Parachu Marco et al. | New evidence characterizing temperature-dependent sex determination in broad-snouted caiman, Caiman latirostris | |
CN111738140A (en) | Image recognition-based cultured shrimp real-time monitoring method | |
Guzmán et al. | Seasonal variation in reproductive parameters of the squat lobster Pleuroncodes monodon from a South Pacific population | |
Buchholz | Older males have bigger knobs: correlates of ornamentation in two species of curassow | |
Seymoure | Heliconius in a new light: The effects of light environments on mimetic coloration, behavior, and visual systems | |
CN115379150B (en) | System and method for automatically generating dynamic video of rice growth process in remote way | |
Bollen | Fruit-frugivore interactions in a Malagasy littoral forest: a community-wide approach of seed dispersal | |
Taparhudee et al. | Weight estimation of Nile tilapia (Oreochromis niloticus Linn.) using image analysis with and without fins and tail | |
Kozakai et al. | Intake rate of captive masked palm civets on large-sized cultivated fruit, Japanese persimmon | |
Mulvenna et al. | The ability of magnetic field sensors to monitor feeding in three domestic herbivores | |
Zhao et al. | Isotopic (δ15N and δ13C) profiles in dentine indicate sex differences and individual variability in resource use among narwhals (Monodon monoceros) | |
Baras et al. | Ontogenetic variation of food intake and gut evacuation rate in larvae of the doncella P seudoplatystoma punctifer, as measured using a non‐destructive method | |
Waldner et al. | Disease outbreak investigation in food animal practice | |
Flahauw et al. | Gonad volume assessment in the oyster Crassostrea gigas: Comparison between a histological method and a magnetic resonance imaging (MRI) method |
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: 20201002 |