CN114609952A - Fish culture monitoring method and system based on machine vision - Google Patents
Fish culture monitoring method and system based on machine vision Download PDFInfo
- Publication number
- CN114609952A CN114609952A CN202210361912.6A CN202210361912A CN114609952A CN 114609952 A CN114609952 A CN 114609952A CN 202210361912 A CN202210361912 A CN 202210361912A CN 114609952 A CN114609952 A CN 114609952A
- Authority
- CN
- China
- Prior art keywords
- fish
- module
- vision
- underwater
- fishes
- 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.)
- Granted
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 title claims abstract description 22
- 241000251468 Actinopterygii Species 0.000 claims abstract description 181
- 230000004438 eyesight Effects 0.000 claims abstract description 46
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 35
- 230000002159 abnormal effect Effects 0.000 claims abstract description 25
- 238000004891 communication Methods 0.000 claims abstract description 20
- 238000012545 processing Methods 0.000 claims abstract description 16
- 241000894007 species Species 0.000 claims abstract description 4
- 230000035945 sensitivity Effects 0.000 claims description 22
- 238000005516 engineering process Methods 0.000 claims description 18
- 230000000694 effects Effects 0.000 claims description 10
- 238000009372 pisciculture Methods 0.000 claims description 9
- 230000005856 abnormality Effects 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 6
- 230000006399 behavior Effects 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 5
- 230000003068 static effect Effects 0.000 claims description 5
- 230000000007 visual effect Effects 0.000 description 12
- 238000005265 energy consumption Methods 0.000 description 7
- 230000009182 swimming Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000037406 food intake Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 230000004297 night vision Effects 0.000 description 2
- 230000036544 posture Effects 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 206010038743 Restlessness Diseases 0.000 description 1
- 241000519995 Stachys sylvatica Species 0.000 description 1
- 241000212749 Zesius chrysomallus Species 0.000 description 1
- 210000001015 abdomen Anatomy 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 210000004690 animal fin Anatomy 0.000 description 1
- 230000036528 appetite Effects 0.000 description 1
- 235000019789 appetite Nutrition 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000740 bleeding effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 208000027744 congestion Diseases 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 210000004209 hair Anatomy 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000003097 mucus Anatomy 0.000 description 1
- 244000045947 parasite Species 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000028327 secretion Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
- 210000004916 vomit Anatomy 0.000 description 1
- 230000008673 vomiting Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
- G05B19/0423—Input/output
-
- 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
- A01K61/10—Culture of aquatic animals of fish
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/25—Pc structure of the system
- G05B2219/25257—Microcontroller
-
- 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
- Life Sciences & Earth Sciences (AREA)
- Environmental Sciences (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Marine Sciences & Fisheries (AREA)
- Zoology (AREA)
- Animal Husbandry (AREA)
- Biodiversity & Conservation Biology (AREA)
- Farming Of Fish And Shellfish (AREA)
Abstract
The invention relates to the technical field of fish digital culture, and particularly discloses a fish culture monitoring method and system based on machine vision, wherein the system comprises a land vision module, a communication module, a central computing module, a processing module, an auxiliary module and an underwater mobile vision module, the land vision module is deployed at the shore near a water area, a yolov4 algorithm is used for identifying and judging water flow, sea waves and the like, when targets such as water flow, sea waves and the like are identified, signals are transmitted to the communication module and transmitted to the central computing module, the underwater mobile vision module is carried into each underwater robot, and species of different fishes in cultured fishes are identified and distinguished by using the yolov4 algorithm. The health/abnormity models of different fishes are constructed by combining the underwater and overwater vision modules, and the reliability of decision making is increased; when the fish is judged to be abnormal, the subsequent processing is carried out on the fish, and the problems that in the prior art, the pertinence is poor and the like caused by the fact that a unified model is used by various fishes are solved.
Description
Technical Field
The invention relates to the technical field of digital fish culture, in particular to a fish culture monitoring method and system based on machine vision.
Background
In fish culture, a great deal of economic loss is brought by disease diffusion, water body pollution and the like caused by incapability of timely treating fishes due to various diseases or injuries, so that the health monitoring of the cultured fishes is necessary to ensure the yield and the quality of products. The prior art mainly monitors the health of fishes by acquiring images and videos in the activities of the fishes and applying methods such as a neural network and the like;
the existing fish health detection and tracking method has the following defects:
1. the underwater environment is complex, the visual identification effect is influenced, the prior art does not mention how to enhance the accuracy of underwater visual identification, and only roughly mentions that a machine vision and neural network method is used for detecting the fishes.
2. The prior art mostly tracks and detects fishes, the energy consumption of the device is large by the scheme, and due to underwater environment factors, energy supply and energy consumption of the robot are always to be improved.
The prior art only roughly mentions morphological characteristics of fishes and does not describe the morphological characteristics in detail, and the judgment accuracy is reduced by using the same decision basis when different fishes have different abnormal behaviors.
Disclosure of Invention
The invention aims to provide a fish culture monitoring method and system based on machine vision, which solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme: a fish culture monitoring method based on machine vision comprises the following steps:
s1: calculating the area of a water area, and dividing the responsible area according to the number of underwater robots;
s2: obtaining fish information in each area, matching the fish information with an original database, and identifying the fish type;
s3: constructing a corresponding sensitivity model according to the fish species;
s4: the underwater robot is released, the fish luring technology is used for drawing the distance between the underwater robot and the fish luring technology, and according to the sensitivity model, the light source is enhanced, so that two conditions can occur: 1): the fish are disturbed; 2) the fish are not disturbed;
s5: when the fish is disturbed, predicting the moving track of the fish, and communicating with other robots in the track;
s6: the other underwater robots are matched with each other to monitor the disturbed fish, and then S4 is repeated;
s7: when the fish is not disturbed, the underwater robot and the fish are relatively static to monitor the behaviors and the characteristics of the underwater robot and the fish;
s8: comparing with the fish abnormal model, and outputting whether a judgment result is: 1): judging health; 2) judging abnormality;
s9: when the health is judged, feeding back data, training the model for the next time, and waiting for the next operation;
s10: when the abnormality is judged, taking a timely fishing measure;
s11: and finally, further detection and treatment are carried out.
A fish culture monitoring system based on machine vision for implementing the method comprises a land vision module, a communication module, a central computing module, a processing module, an auxiliary module and an underwater mobile vision module, wherein the output end of the land vision module is in signal connection with the input end of the communication module, the output end of the communication module is in signal connection with the input end of the central computing module, the output end of the central computing module is in signal connection with the input end of the processing module, the output end of the auxiliary module is in signal connection with the input end of the land vision module, the output end of the auxiliary module is in signal connection with the input end of the underwater mobile vision module, and the output end of the underwater mobile vision module is in signal connection with the input end of a parallel module.
Preferably, the land vision module is deployed on the shore near the water area, the yolov4 algorithm is used for identifying and judging water flow, sea waves and the like, when targets such as water flow, sea waves and the like are identified, signals are transmitted to the communication module and then transmitted to the central computing module, so that misjudgment of the system caused by influences (such as exercise intensity, ingestion rate, growth rate and the like) on fish activities caused by factors such as water flow, sea waves and the like is prevented, and the accuracy and precision of judgment of the system are improved. And identifying fish above the water surface and calculating the time for the fish to float out of the water surface.
Preferably, the underwater mobile vision module is carried into each underwater robot, species of different fishes in cultured fishes are identified and distinguished by using a yolov4 algorithm, characteristic values of the currently identified activities of the fishes are extracted after various image processing technologies in an underwater complex environment, the characteristic values are matched with the database of the invention to judge the health state of the fishes, and a space coordinate system is established for the target and the device per se according to the relative position of the target per se and the target. The fish are marked using the prior art (using a non-contact electronic marking technique) when it is determined to be abnormal.
Preferably, the central computing module comprises a health and abnormal model feature training system, a fish sensitivity model building system and a multi-robot coordination operation system.
Preferably, the communication module comprises wireless and wired communication, and is used for signal and data transmission among the modules of the system and data transmission among the modules of the shore vision module, and particularly, the module is connected with an energy control system of a device carrying the system, so as to acquire energy consumption conditions and transmit the energy consumption conditions to the central computing module.
Preferably, the processing module is used for calibrating the fishes judged to be abnormal, fishing the fishes, and sending the fishes to further monitoring or treatment.
Preferably, the auxiliary module provides a light source for the vision module, fish are not disturbed as much as possible under the condition of taking the sensitivity model as a basis, the light source is provided, the fish attracting technology is used for attracting the fish to be close to the vision module and drawing the distance between the fish and the vision module, the identification precision and the judgment accuracy are improved, various common algorithms are carried in the vision module, such as a night vision algorithm for night, a defogging algorithm for foggy weather, a rain removing algorithm for a rainy environment and the like, and the operation of the vision module in various complex environments is guaranteed.
The fish culture monitoring method and system based on machine vision provided by the invention have the following beneficial effects:
(1) different sensitivity models are constructed for different fishes, the sensitivity models are constructed according to the sensitivity of the fishes to factors and the sensitivity intensity, the light is enhanced based on the models, the fishes cannot be disturbed while the visual identification is enhanced, the fish luring technology is used, the distance between the two models is shortened, the long-time monitoring can be realized in a relatively static state, and the problem that the visual identification effect of the underwater complex environment is poor, which is not mentioned in the prior art, is solved;
(2) the overwater and underwater vision module is combined to construct health/abnormal models of different fishes, so that the reliability of decision making is increased; when the abnormal condition is judged, performing subsequent processing on the abnormal condition; the defects of poor pertinence and the like caused by the fact that a plurality of fishes use a unified model in the prior art are overcome.
(3) And a multi-underwater robot cooperative operation system is constructed, tracking in the prior art is replaced, energy consumption of the device is reduced, and operation time of the system is prolonged.
Drawings
FIG. 1 is a schematic block diagram of a machine vision based fish farming monitoring system according to the present invention;
FIG. 2 is a schematic flow chart of a fish farming monitoring method based on machine vision according to the present invention.
Detailed Description
Referring to fig. 1-2, the method and system for monitoring fish culture based on machine vision according to the present embodiment use different sensitivity models constructed for different fishes to solve the problems in the background art.
In order to achieve the purpose, the invention provides the following technical scheme: a fish culture monitoring method based on machine vision comprises the following steps:
s1: calculating the area of a water area, and dividing the responsible area according to the number of underwater robots;
s2: obtaining fish information in each area, matching the fish information with an original database, and identifying the fish type;
s3: constructing a corresponding sensitivity model according to the fish species;
s4: the underwater robot is released, the fish luring technology is used for drawing the distance between the underwater robot and the fish luring technology, and according to the sensitivity model, the light source is enhanced, so that two conditions can occur: 1): the fish are disturbed; 2) the fish are not disturbed;
s5: when the fish is disturbed, predicting the moving track of the fish, and communicating with other robots in the track;
s6: the other underwater robots are matched with each other to monitor the disturbed fish, and then S4 is repeated;
s7: when the fish is not disturbed, the underwater robot and the fish are relatively static to monitor the behaviors and the characteristics of the underwater robot and the fish;
s8: comparing with the fish abnormal model, and outputting whether a judgment result is: 1): judging health; 2) judging abnormality;
s9: when the health is judged, feeding back data, training the model for the next time, and waiting for the next operation;
s10: when the abnormality is judged, taking a timely fishing measure;
s11: and finally, further detection and treatment are carried out.
Referring to fig. 2, a fish culture monitoring system based on machine vision for implementing the method comprises a land vision module, a communication module, a central computing module, a processing module, an auxiliary module and an underwater mobile vision module, wherein the output end of the land vision module is in signal connection with the input end of the communication module, the output end of the communication module is in signal connection with the input end of the central computing module, the output end of the central computing module is in signal connection with the input end of the processing module, the output end of the auxiliary module is in signal connection with the input end of the land vision module, the output end of the auxiliary module is in signal connection with the input end of the underwater mobile vision module, and the output end of the underwater mobile vision module is in signal connection with the input end of a peer module;
the land vision module is deployed on the shore near a water area, the yolov4 algorithm is used, water flow, sea waves and the like are identified and judged, when targets such as water flow, sea waves and the like are identified, signals are transmitted to the communication module and transmitted into the central computing module, so that misjudgment of the system caused by influences (such as movement intensity, ingestion rate, growth rate and the like) on fish activities caused by the factors such as water flow, sea waves and the like is prevented, and the accuracy and precision of judgment of the system are improved. Identifying fish above the water surface, and calculating the time for the fish to float out of the water surface;
the underwater mobile vision module is carried into each underwater robot, species of different fishes in cultured fishes are identified and distinguished by using a yolov4 algorithm, characteristic values of the currently identified fish activities are extracted after various image processing technologies in an underwater complex environment are carried out, the characteristic values are matched with the database of the invention, the health state of the fish is judged, and a space coordinate system is established for a target and a self device according to the relative position of the self device and the target. Marking the fish using existing techniques (using non-contact electronic marking techniques) when it is determined to be abnormal;
the central computing module comprises a health and abnormal model feature training module, a fish sensitivity model building module and a multi-robot coordination operation system, wherein the health and abnormal model feature training module comprises the following steps:
1) floating head characteristics: the fish sense is sensitive, and when being close to the fish, the fish can swim away far away, but when the fish is abnormal, the fish is close to the fish, and the fish still floats on the water surface, and when the vibration is generated, the fish slowly swims away and enters the water, but floats on the water surface after a while. (identification is carried out by using a land vision module, and a robot is controlled to approach);
2) speed of activity, body balance characteristics: when the fishes are abnormal, the fishes float on the water surface for a long time and can not sink, or sink on the bottom layer of the water tank for a long time and can not move upstream of the water surface, or stand still in the water in a special posture, lie on the side, stand upside down in the water, or swim irregularly and rotate;
3) keep away from shoal of fish characteristic: the novel swimming pool is characterized in that the swimming pool can not control the swimming to move forwards or backwards, when the fish is abnormal, the dorsal fins of the fish become not very hard, the pectoral fins look like to be powerless, the ventral fins do not open, the tail fins are very weak and can droop, and the fish vomits after eating;
4) spots or abnormalities in the surface: the fish body is generally not glossy, some white or red spots and plaques appear, scales sometimes fall off, hairs also grow on the fish body, or the fish fin is lack of a part, mucus on the fish body is increased, and the fish belly becomes larger in severe cases, so that saprolegniasis is possibly infected;
5) fish gill characteristics: when the gill of the fish has abnormal symptoms such as congestion, paleness, gray green and the like, the fish can be in an abnormal state;
6) behavior characteristics: the fishes are restless and swim frequently and rapidly, the bodies of the fishes continuously rub and rub the ramparts, the fish fins are shaken, the swimming postures are abnormal, the appetite is reduced, the fish bodies are locally inflamed, bleeding and ulcerated, and the secretion is increased. Possible infection with parasites;
constructing a fish sensitivity model: factors including but not limited to light intensity, light color, water flow, external moving objects, flashing light, etc. have an effect on fish's being stimulated to move away, and the following functions are constructed for different fish. For convenient calculation, the invention sets the illumination intensity as X1Coefficient of K1(ii) a The color of illumination is X2Coefficient of K2(ii) a The water flow is X3Coefficient of K3(ii) a The external moving object is X4Coefficient of K4(ii) a Flash of light is X5Coefficient of K5(ii) a The other factor being XnCoefficient of Kn。
Y＝K1X1+K2X2+K3X3+K4X4+K5X5+KnXn
When the fish category is visually identified, matching is performed in a database using different sensitivity models for different fish species. When a plurality of fish types are identified, the threshold value of the fish with the highest sensitivity factor is adopted to ensure that all the fish types are not disturbed, and in combination with the prior fish luring technology, the supplementary light source is used for improving the visual identification environment, and simultaneously, the fish is attracted to the vicinity of the visual module so as to be observed at a short distance for a long time. And after the abnormity is judged, calibrating by using a non-contact electronic marking method.
The multifunctional robot coordination operation system comprises: calculating the area S of the responsible water area, dividing the water area into n sub-areas, and calculating the area S of m areas which are responsible for the underwater robots carrying the systemm
S＝mSm
When the fish is disturbed, the moving track of the fish is predicted, and the fish is communicated with the other underwater robots in the moving track of the fish, so that the fish is identified and monitored.
The communication module comprises wireless and wired communication, performs signal and data transmission among all modules of the system and data transmission among the shore vision modules, and is particularly connected with an energy control system of a device carrying the system so as to acquire energy consumption conditions and transmit the energy consumption conditions to the central computing module;
the processing module is used for calibrating the fishes judged to be abnormal, salvaging the fishes and sending the fishes to further monitoring or treatment;
the auxiliary module provides a light source for the visual module, fish are not disturbed as much as possible under the condition based on the sensitivity model, the light source is provided, the fish attracting technology is used for attracting the fish to be close to the visual module and pulling the fish to be in the distance between the visual module and the visual module, the identification precision and the judgment accuracy are improved, and various common algorithms are carried in the visual module, such as a night vision algorithm for night, a defogging algorithm for heavy fog weather, a rain removing algorithm for a rainy environment and the like, so that the operation of the visual module in various complex environments is guaranteed.
When the fish culture monitoring method and the system based on the machine vision are used, the area of a water area is calculated, and the responsible area is divided according to the number of underwater robots; constructing a corresponding sensitivity model according to the type of the fish; the fish luring technology is used for drawing the distance between the fish luring technology and the fish luring technology, and according to the sensitivity model, the light source is enhanced, so that two conditions can occur: 1): the fish are disturbed; 2) the fish are not disturbed; when the fish is disturbed, predicting the moving track of the fish, and communicating with other robots in the track; the other robots cooperate with each other to monitor the disturbed fish, and then S4 is repeated; when the fish is not disturbed, the underwater robot and the fish are relatively static to monitor the behaviors and the characteristics of the underwater robot and the fish; comparing with the fish abnormal model, and outputting whether a judgment result is: 1): judging health; 2) judging abnormality; when the health is judged, feeding back data, training the model for the next time, and waiting for the next operation; when the abnormal condition is judged, taking a timely fishing measure; and finally, further detection and treatment are carried out.
The health/abnormity models of different fishes are constructed by combining the underwater and overwater vision modules, and the reliability of decision making is increased; when the fish is judged to be abnormal, the subsequent processing is carried out on the fish, and the problems that in the prior art, the pertinence is poor and the like caused by the fact that a unified model is used by various fishes are solved.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (8)
1. A fish culture monitoring method based on machine vision is characterized by comprising the following steps:
s1: calculating the area of a water area, and dividing the responsible area according to the number of underwater robots;
s2: obtaining fish information in each area, matching the fish information with an original database, and identifying the fish type;
s3: constructing a corresponding sensitivity model according to the fish species;
s4: the underwater robot is released, the fish luring technology is used for drawing the distance between the underwater robot and the fish luring technology, and according to the sensitivity model, the light source is enhanced, so that two conditions can occur: 1): the fish are disturbed; 2) the fish are not disturbed;
s5: when the fish is disturbed, predicting the moving track of the fish, and communicating with other robots in the track;
s6: the other underwater robots are matched with each other to monitor the disturbed fish, and then S4 is repeated;
s7: when the fish is not disturbed, the underwater robot and the fish are relatively static to monitor the behaviors and the characteristics of the underwater robot and the fish;
s8: comparing with the fish abnormal model, and outputting whether a judgment result is: 1): judging health; 2) judging abnormality;
s9: when the health is judged, feeding back data, training the model for the next time, and waiting for the next operation;
s10: when the abnormality is judged, taking a timely fishing measure;
s11: and finally, further detection and treatment are carried out.
2. A machine vision based fish farming monitoring system implementing the method of claim 1, characterized by: the underwater mobile vision module comprises a land vision module, a communication module, a central computing module, a processing module, an auxiliary module and an underwater mobile vision module, wherein the output end of the land vision module is in signal connection with the input end of the communication module, the output end of the communication module is in signal connection with the input end of the central computing module, the output end of the central computing module is in signal connection with the input end of the processing module, the output end of the auxiliary module is in signal connection with the input end of the land vision module, the output end of the auxiliary module is in signal connection with the input end of the underwater mobile vision module, and the output end of the underwater mobile vision module is in signal connection with the input end of the parallel module.
3. The machine vision-based fish farming monitoring system of claim 2, wherein: the land vision module is deployed on the shore near the water area, the yolov4 algorithm is used, the water flow, the sea wave and the like are identified and judged, and when targets such as the water flow, the sea wave and the like are identified, signals are transmitted to the communication module and transmitted to the central computing module.
4. The machine vision-based fish farming monitoring system of claim 2, wherein: the underwater mobile vision module is mounted in each underwater robot, the yolov4 algorithm is used for identifying and distinguishing species of different fishes in cultured fishes, and characteristic values of the currently identified fish activities are extracted after various image processing technologies in the underwater complex environment.
5. The machine-vision-based fish farming monitoring system of claim 2, wherein: the central computing module comprises a health and abnormal model feature training module, a fish sensitivity model building module and a multi-robot coordination operation system.
6. The machine vision-based fish farming monitoring system of claim 2, wherein: the communication module comprises wireless and wired communication, and is used for signal and data transmission among the modules of the system and data transmission among the shore vision modules.
7. The machine vision-based fish farming monitoring system of claim 2, wherein: the processing module is used for calibrating the fishes judged to be abnormal, fishing the fishes, and sending the fishes to further monitoring or treatment.
8. The machine vision-based fish farming monitoring system of claim 2, wherein: the auxiliary module provides a light source for the vision module, fish is not disturbed as much as possible under the condition of taking the sensitivity model as a basis, the light source is provided, the fish attracting technology is used for attracting the fish to be close to the vision module and pulling the fish to be in the distance between the vision module and the vision module, the recognition precision and the judgment accuracy are improved, and various common algorithms are carried in the vision module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210361912.6A CN114609952B (en) | 2022-04-07 | 2022-04-07 | Fish culture monitoring method and system based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210361912.6A CN114609952B (en) | 2022-04-07 | 2022-04-07 | Fish culture monitoring method and system based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114609952A true CN114609952A (en) | 2022-06-10 |
| CN114609952B CN114609952B (en) | 2024-04-26 |
Family
ID=81870046
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210361912.6A Active CN114609952B (en) | 2022-04-07 | 2022-04-07 | Fish culture monitoring method and system based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114609952B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105119656A (en) * | 2015-09-09 | 2015-12-02 | 惠州伟志电子有限公司 | LED fish luring illumination system with visible light communication function |
| KR20160029474A (en) * | 2014-09-05 | 2016-03-15 | 부산대학교 산학협력단 | Apparatus and method for monitoring the health of farmed fish |
| CN108132099A (en) * | 2017-12-20 | 2018-06-08 | 佛山市洛克威特科技有限公司 | A kind of fish monitoring system |
| CN108459059A (en) * | 2017-12-17 | 2018-08-28 | 江南大学 | The fish pond dissolved oxygen wireless detection device cleaned automatically can be achieved |
| 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 |
| CN109856138A (en) * | 2018-12-18 | 2019-06-07 | 杭州电子科技大学 | Deep sea net cage healthy fish identifying system and method based on deep learning |
| JP2020110138A (en) * | 2019-05-15 | 2020-07-27 | 株式会社ＦｕｌｌＤｅｐｔｈ | Fish monitoring system and camera unit |
| CN112418087A (en) * | 2020-11-23 | 2021-02-26 | 中山大学 | Underwater video fish identification method based on neural network |
-
2022
- 2022-04-07 CN CN202210361912.6A patent/CN114609952B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20160029474A (en) * | 2014-09-05 | 2016-03-15 | 부산대학교 산학협력단 | Apparatus and method for monitoring the health of farmed fish |
| CN105119656A (en) * | 2015-09-09 | 2015-12-02 | 惠州伟志电子有限公司 | LED fish luring illumination system with visible light communication function |
| CN108459059A (en) * | 2017-12-17 | 2018-08-28 | 江南大学 | The fish pond dissolved oxygen wireless detection device cleaned automatically can be achieved |
| CN108132099A (en) * | 2017-12-20 | 2018-06-08 | 佛山市洛克威特科技有限公司 | A kind of fish monitoring system |
| 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 |
| CN109856138A (en) * | 2018-12-18 | 2019-06-07 | 杭州电子科技大学 | Deep sea net cage healthy fish identifying system and method based on deep learning |
| JP2020110138A (en) * | 2019-05-15 | 2020-07-27 | 株式会社ＦｕｌｌＤｅｐｔｈ | Fish monitoring system and camera unit |
| CN112418087A (en) * | 2020-11-23 | 2021-02-26 | 中山大学 | Underwater video fish identification method based on neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114609952B (en) | 2024-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109591982B (en) | Unmanned monitoring underwater aquaculture robot | |
| Wang et al. | Intelligent fish farm—the future of aquaculture | |
| CN104008367B (en) | The automatism analysis system and method for growing and fattening pigs based on computer vision | |
| WO2019232247A1 (en) | Biomass estimation in an aquaculture environment | |
| Yu et al. | Segmentation and measurement scheme for fish morphological features based on Mask R-CNN | |
| WO2019245722A1 (en) | Sea lice detection and classification in an aquaculture environment | |
| Liu et al. | Measuring feeding activity of fish in RAS using computer vision | |
| Wu et al. | Application of intelligent and unmanned equipment in aquaculture: A review | |
| Qiao et al. | Cattle body detection based on YOLOv5-ASFF for precision livestock farming | |
| CN109472883A (en) | Patrol pool method and apparatus | |
| CN113158364A (en) | Circulating pump bearing fault detection method and system | |
| CN106370667A (en) | Visual detection apparatus and method for quality of corn kernel | |
| CN115879771B (en) | Intelligent control method and system for dissolved oxygen in aquaculture pond | |
| CN106259095A (en) | The behavior of a kind of pond cultured freshwater fish is to the determination methods in terms of aquaculture | |
| CN110999944A (en) | Automatic large-size fish slaughtering device and control method | |
| Zhang et al. | Intelligent fish feeding based on machine vision: A review | |
| CN111666897A (en) | Oplegnathus punctatus individual identification method based on convolutional neural network | |
| CN116778310A (en) | Acoustic-optical image fusion monitoring method and system for aquaculture | |
| Liu et al. | Research progress of computer vision technology in abnormal fish detection | |
| CN113869252B (en) | Crayfish head-tail separation control method and crayfish head-tail separation control system based on deep learning | |
| CN114609952A (en) | Fish culture monitoring method and system based on machine vision | |
| Zhang et al. | Fully automatic system for fish biomass estimation based on deep neural network | |
| CN116452967B (en) | Fish swimming speed identification method based on machine vision | |
| CN112965557A (en) | Intelligent lobster breeding monitoring device and monitoring method | |
| Wang et al. | A literature review of underwater image detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |