Water tank and culture support system
The aquarium system with a sand-imitating design and integrated monitoring and purification features addresses the inefficiencies of land-based prawn farming by enabling stable and efficient prawn cultivation with reduced environmental impact.
Patent Information
- Application Number
- JP2023079500
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-12
- Publication Date
- 2024-11-22
AI Technical Summary
Conventional aquaculture systems face challenges in efficiently cultivating delicate aquatic animals like prawns on land due to their sand-burrowing habits, making observation and management difficult, and they are susceptible to environmental changes, leading to inefficient farming practices and environmental contamination.
An aquarium system with heat-insulating properties and an uneven, sand-imitating bottom surface, equipped with an imaging device and control system to monitor and manage aquatic animals without the need for bottom sand, along with a water purification system to maintain water quality.
The system allows stable and efficient cultivation of prawns by mimicking marine conditions, facilitating easy observation and management, reducing environmental impact, and enhancing growth and health through improved water quality.
Smart Images

Figure 2024163678000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to an aquarium and an aquaculture support system. [Background technology]
[0002] Conventionally, marine fish farms such as shrimp and other aquatic animals (fish and shellfish) have been established in marine cages or in farm ponds where mangroves have been cut down in areas where mangroves grow abundantly. When cultivating aquatic animals in such marine cages or farm ponds, the animals are raised while being influenced by external factors such as climate change and changes in water quality. In addition, from the viewpoint of efficiency, aquatic animals are cultivated in large numbers at high densities in the cages or farm ponds, and synthetic feed is given to promote the growth of the aquatic animals and antibiotics are administered to prevent diseases.
[0003] In aquaculture in marine cages and ponds, the success or failure of aquaculture is often affected by external factors, and the breeders have to spend a lot of time monitoring the water quality. In addition, the water quality of the cages and ponds can be polluted by the accumulation of leftover synthetic feed, antibiotics, and feces from large amounts of aquatic animals, and the polluted wastewater can have a negative impact on the environment around the cages and ponds. In particular, aquaculture ponds can accumulate water and soil pollution, and after about five years of use as aquaculture ponds, they can become unusable for aquaculture or for other purposes. When aquaculture ponds become unusable due to pollution, mangroves in other water areas are cut down to make room for the next aquaculture pond, which can cause a rapid decline in mangroves.
[0004] In order to carry out stable aquaculture, a growing environment that can suppress the influence of external factors is required. Furthermore, aquaculture of aquatic animals may cause mangrove cutting and water and soil pollution, which may contribute to environmental destruction. Therefore, there is a demand for a shift to land-based aquaculture, in which aquatic animals are cultivated in tanks installed on land. For example, there is an aquaculture device described in Patent Document 1 as an aquaculture device for land-based aquaculture. [Prior art documents] [Patent documents]
[0005] [Patent Document 1] JP 2021-193946 A Summary of the Invention [Problem to be solved by the invention]
[0006] However, in conventional land-based shrimp farming, whitenamei shrimp are often cultivated, which are highly resistant to disease and have high production efficiency because they swim in water.On the other hand, kuruma shrimp, which are delicate and have a tendency to burrow into sand (sand burrowing), are difficult to cultivate on land, so they are generally cultivated in the sea.
[0007] In the cultivation of kuruma shrimp, which have a tendency to burrow into sand, it is necessary to provide bottom sand in the cages or cultivation ponds, which makes it difficult to observe the kuruma shrimp and remaining food from above the water. Therefore, even in the case of marine cultivation, experienced workers at the farm must go underwater every day to observe the size of the kuruma shrimp and the amount of remaining food, and finely adjust the amount of food. Only strong individuals that have survived disease and environmental changes can grow to be kuruma shrimp over 15 cm and be shipped, so it takes an enormous amount of time and effort before they can be shipped.
[0008] Even in the case of land-based aquaculture, in order to observe the prawns burrowing into the sand, the water must be drained from the tank to lower the water level and the sand must be dug up to observe the prawns and any remaining food, making aquaculture an extremely time-consuming and labor-intensive process.
[0009] Therefore, efficient farming systems are required for aquatic animals such as kuruma shrimp, which are delicate, vulnerable to environmental changes, and prone to burrowing. In particular, it is difficult to farm kuruma shrimp on land using conventional farming systems.
[0010] In consideration of the above circumstances, an object of the present invention is to provide an aquarium and an aquaculture support system that enable efficient aquaculture. [Means for solving the problem]
[0011] In order to solve the above problems, the present invention proposes the following means. The aquarium of the first aspect of the present invention is an aquarium capable of housing aquatic animals that have a sand-burrowing tendency in an aquaculture area, facing the aquaculture area, having insulating properties, and having a bottom with an uneven surface on the side facing the aquaculture area that resembles sand.
[0012] In a second aspect of the present invention, in the aquarium according to the first aspect, the uneven surface is black.
[0013] A third aspect of the present invention is the aquarium according to the second aspect, wherein the aquatic animals are kuruma shrimp.
[0014] An aquaculture support system according to a fourth aspect of the present invention comprises an aquarium according to any one of the first to third aspects, an imaging device capable of photographing the aquaculture area, and a control device capable of detecting an identification area for identifying the aquatic animal in the image captured by the imaging device.
[0015] A fifth aspect of the present invention is the aquaculture support system according to the fourth aspect, wherein the control device estimates a size of the aquatic animal based on the identification area.
[0016] In a sixth aspect of the present invention, in the aquaculture support system according to the fourth aspect, the control device estimates the quantity of the aquatic animals based on the identification area.
[0017] A seventh aspect of the present invention is an aquaculture support system relating to the fourth aspect, wherein the control device detects feature points indicating the position of the aquatic animal in a first image and a second image taken at a different time than the first image, selected from the plurality of images, and estimates the activity of the aquatic animal based on the amount of movement per unit time of the feature points in the first image and the second image.
[0018] An eighth aspect of the present invention is an aquaculture support system relating to the fourth aspect, in which the activity of the aquatic animal is estimated based on the difference between a first image selected from the plurality of images and a second image taken at a different time than the first image.
[0019] A ninth aspect of the present invention is the aquaculture support system according to the fourth aspect, further comprising a water purification device capable of removing impurities from the aquaculture water contained in the aquaculture area. Effect of the Invention
[0020] The aquarium of the first embodiment faces the aquaculture area in which sand-burrowing aquatic animals are housed, has insulating properties, and has a bottom with an uneven shape that imitates sand formed on the surface facing the aquaculture area, so that aquatic animals that are sensitive to temperature changes and have a sand-burrowing tendency can be cultivated stably and efficiently without using bottom sand.
[0021] In the aquarium of the second embodiment, the uneven shape formed on the surface of the bottom facing the aquaculture area is black, so that aquatic animals can be cultured in an environment closer to marine surface aquaculture with bottom sand, and aquatic animals that have a sand-burrowing tendency can be cultured more stably and efficiently.
[0022] The tank according to the third aspect makes it possible to stably and efficiently cultivate kuruma shrimp, which are delicate, vulnerable to environmental changes, and have a tendency to burrow into sand, making them difficult to cultivate on land using conventional farming systems.
[0023] According to the fourth aspect, the aquaculture support system includes an aquarium according to any one of the first to third aspects, an imaging device capable of photographing the aquaculture area in the aquarium, and a control device capable of detecting an identification area for identifying the aquatic animal in the image captured by the imaging device. Since the tank has an uneven shape that mimics sand, there is no need to use bottom sand, and there is no risk of the aquatic animals hiding in the sand, so the aquatic animals and remaining food can be easily photographed with the imaging device. In addition, since the control device can detect the identification area that identifies the aquatic animals, the farm worker can observe the aquatic animals and the amount of remaining food without having to perform steps such as draining the water to lower the water level.
[0024] According to the aquaculture support system of the fifth aspect, the control device estimates the size of the aquatic animals based on the identification area, so that the aquaculture workers can grasp the size of the aquatic animals without performing processes such as draining the water and lowering the water level.
[0025] According to the aquaculture support system of the sixth aspect, the control device estimates the quantity of aquatic animals based on the identification area, so that aquaculture workers can grasp the quantity of aquatic animals without performing processes such as draining water and lowering the water level.
[0026] According to the seventh aspect of the aquaculture support system, the control device detects feature points indicating the position of the aquatic animals in a first image and a second image taken at a different time than the first image, selected from a plurality of images captured by the imaging device, and estimates the activity level of the aquatic animals based on the movement per unit time of the feature points in the first image and the second image, so that the aquaculture workers can grasp the physical condition of the aquatic animals without having to perform processes such as draining the water and lowering the water level.
[0027] According to the eighth aspect of the aquaculture support system, the control device estimates the activity level of the aquatic animals based on the difference between a first image selected from multiple images captured by the imaging device and a second image captured at a different time from the first image. This makes it easy to grasp the physical condition of aquatic animals such as baby shrimp, which swim freely in all directions within the water of the aquarium and whose movement amount is difficult to measure using characteristic points.
[0028] According to the ninth aspect of the aquaculture support system, since the aquaculture support system further includes a water purification device capable of removing impurities from the aquaculture water contained in the aquaculture area, it is possible to prevent diseases of aquatic animals caused by contamination of the aquaculture water, and also to prevent the surrounding environment from being polluted by contaminated wastewater. [Brief description of the drawings]
[0029] [Figure 1] FIG. 1 is a perspective view showing a schematic diagram of an aquaculture support system according to an embodiment of the present invention. [Diagram 2]FIG. 2 is a cross-sectional view showing a schematic diagram of an aquaculture area in the aquaculture support system. [Diagram 3] 13 is an example of a first detection image output by a control device of the aquaculture support system. [Figure 4] 4 is a flowchart showing a method for estimating the size and quantity of aquatic animals in the control device. [Diagram 5] 5 is a flowchart showing details of an estimation method in the flowchart shown in FIG. 4. [Figure 6] 4 is a flowchart showing an example of a method for estimating the activity level of an aquatic animal in the control device. [Figure 7] 11 is an example of a second detection image output by the control device. [Figure 8] 4 is a flowchart showing an example of a method for estimating the activity level of an aquatic animal in the control device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0030] An embodiment of the present invention will be described with reference to the drawings. Fig. 1 is a perspective view showing a model of the aquaculture support system according to the present embodiment, and Fig. 2 is a cross-sectional view showing a model of the aquaculture area in the aquaculture support system according to the present embodiment.
[0031] As shown in FIGS. 1 and 2, the aquaculture support system 1 includes an aquarium 10, a water purification device 20, a control device 30, and an imaging device 40.
[0032] 2, the aquarium 10 includes a bottom 10b that forms the bottom surface of the aquarium 10, and a sidewall 10s that forms the side surface of the aquarium 10. A culture area 10w is formed in the area surrounded by the bottom 10b and the sidewall 10s. The bottom 10b and the sidewall 10s are provided facing the culture area 10w.
[0033] In this embodiment, the aquarium 10 has an opening on the vertically upper side and is a rectangular box shape when viewed from above. The shape of the aquarium 10 is not limited to this, and may be, for example, a polygonal shape such as a hexagon or octagon, or a circular shape when viewed from above.
[0034] As shown in Fig. 1, the culture area 10w of the aquarium 10 contains culture water W and aquatic animals S. The culture water W is water necessary for the aquatic animals S to live in, and is, for example, seawater. For the culture water W, water suited to the habitat of the aquatic animals S to be cultured can be used.
[0035] The aquatic animals S are, for example, crustaceans such as shrimp, fish, shellfish, etc. In this embodiment, the aquatic animals S are mainly kuruma shrimp. The aquatic animals S are not limited to kuruma shrimp, and may be other fish and shellfish.
[0036] The size and thickness of the bottom 10b and the side walls 10s can be appropriately determined depending on, for example, the number of aquatic animals S to be cultured and the environment in which the aquarium 10 is installed. The bottom 10b and the side walls 10s are formed of, for example, reinforced polystyrene foam having sufficient strength for cultivating the aquatic animals S and have thermal insulation properties.
[0037] The bottom 10b and the side walls 10s have thermal insulation properties, which can prevent the temperature of the culture water W from changing suddenly due to temperature changes around the aquarium tank 10. It is desirable that the bottom 10b and the side walls 10s have a thickness that can maintain sufficient thermal insulation properties.
[0038] In particular, in climates unsuitable for aquaculture, such as the cold winter months, the risk of the aquatic animals S becoming ill due to a drop in the temperature of the aquaculture water W and stress on the aquatic animals S due to temperature changes can be suppressed.
[0039] The bottom 10b and the sidewall 10s have a sand-like uneven surface on the surface facing the culture area 10w. The sand-like uneven surface is formed, for example, by painting the surface of reinforced polystyrene foam with polyurea, which is a resin compound formed by a chemical reaction between isocyanate and polyamine.
[0040] The bottom 10b and the side walls 10s are covered with a water-resistant paint such as polyurea, so that the culture water W contained in the culture area 10w does not seep into the bottom 10b and the side walls 10s. Moreover, the culture water W does not leak out of the tank 10, except for the circulation between the tank 10 and the water purification device 20 described later.
[0041] The bottom 10b and the surface of the side wall 10s facing the culture area 10w are formed with an uneven shape resembling sand, so that the aquatic animals S can be given the illusion that sand is laid inside the aquarium 10. Therefore, even if the aquatic animals S are shrimp or the like that have a sand-burrowing nature and live in an environment with bottom sand, the stress on the aquatic animals S can be reduced during culture.
[0042] In addition, since the side wall 10s has an uneven shape that imitates sand, the aquatic animals S tend to settle not only on the bottom portion 10b but also on the side wall 10s. Therefore, more aquatic animals S can be cultured compared to a case in which the aquatic animals S settle only on the bottom portion 10b.
[0043] In the aquarium 10, the bottom 10b and the surface of the side wall 10s facing the culture area 10w are formed with an uneven shape resembling sand, so that sand-burrowing aquatic animals S can be cultured without using bottom sand. Compared to culturing with bottom sand, the aquaculture worker can observe the aquatic animals S without having to retrieve them from the sand, and since no residual food is mixed into the sand, the amount of residual food can be easily determined. Furthermore, the accumulation of anaerobic bacteria in the sand and the generation of harmful substances can be suppressed, so contamination of the culture water W can be suppressed.
[0044] The uneven shape formed on the surface of the bottom 10b and the side wall 10s facing the culture area 10w is preferably black. By making the surface black, the uneven shape looks more similar to sand, so that the stress on the aquatic animals S, which have a tendency to burrow into sand, can be further reduced. The surface of the uneven shape is not limited to black, and may be a color close to black (a color with low brightness) or a color that easily absorbs sunlight like sand deposited on the seabed. The surface of the uneven shape may be any color that the aquatic animals S can mistake for sand, and may be, for example, brown.
[0045] The water purification device 20 is a device that purifies the culture water W contaminated by excrement and leftover food of the aquatic animals S, and is provided near the aquarium 10. As shown in FIG. 1, the water purification device 20 includes a filtration section 21, a purification section 22, and a water pipe 23.
[0046] In the aquaculture support system 1, the water tank 10, the filtration unit 21, and the purification unit 22 are connected via a water pipe 23. The culture water W contained in the water tank 10 circulates through the water pipe 23 inside the water tank 10, the filtration unit 21, and the purification unit 22.
[0047] The filtration unit 21 is a filtering device that filters the culture water W flowing therein and removes impurities from the culture water W. Here, the impurities are, for example, leftover food left over by the aquatic animals S and feces excreted by the aquatic animals S. The culture water W flows from the aquarium 10 through a water pipe 23 into the filtration unit 21, and the impurities are removed by the filtration unit 21.
[0048] By removing impurities, the culture water W becomes more likely to absorb oxygen. Increasing the oxygen concentration in the culture water W can activate the nutrient absorption and immune function of the aquatic animals S, promoting the growth of the aquatic animals S. It can also make the aquatic animals S more resistant to disease. Furthermore, it can prevent the aquatic animals S from suffering from oxygen deficiency.
[0049] The purifier 22 is a bio-purifier that uses microorganisms to decompose ammonia, nitric acid, and the like contained in the culture water W flowing inside, and purifies the culture water W. The culture water W flows from the filtration unit 21 through the water pipe 23 into the purifier 22 and is purified by the purifier 22.
[0050] By decomposing ammonia, nitrate, and the like contained in the culture water W using microorganisms, it is possible to suppress diseases of the aquatic animals S. In addition, it is possible to reduce odors generated from the culture water W and the aquatic animals S.
[0051] The culture water W purified in the purification section 22 flows again into the culture area 10w of the aquarium 10 through the water pipe 23.
[0052] The water pipe 23 is, for example, a cylindrical water pipe. The shape and length of the water pipe 23 can be appropriately selected depending on the environment in which the aquaculture support system 1 is installed and the amount of the circulating aquaculture water W. In the example shown in Fig. 1, the water pipe 23 connecting the purification unit 22 and the water tank 10 is one, but the water pipe 23 may be branched into multiple pipes to allow the aquaculture water W to flow into the water tank 10.
[0053] The water pipe 23 may be provided so that the culture water W flowing out from the purification section 22 flows into the opening of the water tank 10 from above.
[0054] By providing the aquaculture support system 1 with the water purification device 20, contamination of the culture water W can be prevented even in a closed circulation type aquaculture system, and the aquatic animals S can be cultured efficiently.
[0055] The control device 30 controls the entire aquaculture support system 1. The control device 30 is, for example, a program-executable device (computer) equipped with a processor, a memory, a storage unit, and the like. Each function of the control device 30 is realized by, for example, one or more processors such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit) executing a program stored in a program memory. However, all or part of these functions may be realized by hardware (e.g., a circuit unit; circuitry) such as an LSI (Large Scale Integration; large-scale integrated circuit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a PLD (Programmable Logic Device). In addition, all or part of the above functions may be realized by a combination of software and hardware. The storage unit is realized by a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), a ROM (Read-Only Memory), a RAM (Random Access Memory; readable and writable memory), or the like.
[0056] The imaging device 40 is, for example, a camera capable of capturing images of the surroundings. The imaging device 40 is provided so as to be capable of capturing images of the aquaculture area 10w. In the example shown in Fig. 2, the aquaculture support system 1 includes a plurality of imaging devices 40, but the number of imaging devices 40 included in the aquaculture support system 1 can be appropriately determined depending on the size of the aquarium 10, the performance of the imaging devices 40, and the like.
[0057] The imaging device 40 is controlled by the control device 30 and captures images of the culture area 10w. The imaging device 40 may be provided above the aquarium 10 and capture images of the entire culture area 10w from an overhead view of the aquarium 10, or may be provided in the culture area 10w and capture images of the entire or part of the culture area 10w in the culture water W. When the imaging device 40 is provided in the culture water W, a waterproof underwater camera can be used as the imaging device 40.
[0058] The imaging device 40 is preferably an underwater camera provided in the aquaculture area 10w in order to accurately capture images of the aquatic animals S in the aquaculture water W. Furthermore, in order to accurately capture images of the aquatic animals S, the aquaculture support system 1 is preferably equipped with a lighting device (not shown) that illuminates the aquatic animals S. The control device 30 turns on the lighting device, and the aquatic animals S illuminated by the lighting device are photographed by the imaging device 40, thereby enabling the aquatic animals S to be photographed accurately. The lighting device that illuminates the aquatic animals S may be, for example, a lighting device attached to the imaging device 40, such as a camera flash.
[0059] The image captured by the imaging device 40 is transmitted to the control device 30. The control device 30 performs image processing on the received (acquired) image. Fig. 3 shows an example of a first detected image PP1 output by the control device 30 after performing image processing.
[0060] The control device 30 performs image processing on the image acquired from the imaging device 40 to detect the discrimination area P1 and the information area P2, and outputs a first detection image PP1 including the discrimination area P1 and the information area P2. The first detection image PP1 output from the control device 30 is transmitted to a display device (not shown) such as a display or a terminal device (not shown) such as a smartphone or tablet, and the aquaculture worker can check the first detection image PP1 displayed on the display device or terminal device. A method for detecting the discrimination area P1 will be described later.
[0061] The classification area P1 is a classifier that classifies the aquatic animal S. In the example shown in FIG. 3, the aquatic animal S is a shrimp (prawn). The classification area P1 is a rectangular area that surrounds the aquatic animal S as shown in FIG. 3, and one classification area P1 is detected for one aquatic animal S. That is, the same number of classification areas P1 as the number of individual aquatic animals S captured in the first detected image PP1 are detected. However, depending on the posture of the aquatic animal S, the aquatic animal S may not be accurately identified, and there may be cases where the classification area P1 is not detected for some of the aquatic animals S captured in the first detected image PP1. The posture of the aquatic animal S will be described later.
[0062] The information area P2 is an area for displaying desired information about the aquatic animal S photographed in each identification area P1. Information displayed in the information area P2 may be, for example, information indicating the growth level of the aquatic animal S, such as "adult (adult shrimp)" or "juvenile fish (juvenile shrimp)." Even if a farm worker does not have sufficient knowledge about the farmed aquatic animals S, he or she can easily grasp information about the aquatic animal S (for example, the growth level) by checking the information displayed in the information area P2.
[0063] 3, the information region P2 is a rectangular region provided inside the identification region P1. If there is no need to display information about the aquatic animal S in the information region P2, the control device 30 does not detect the information region P2, and the information region P2 does not need to be included in the first detection image PP1.
[0064] Next, a method for estimating the size and quantity of the aquatic animals S in the control method of the aquaculture support system 1 by the control device 30 will be described. FIG. 4 is a flow chart showing a method for estimating the size and quantity of aquatic animals S.
[0065] (Step S100) First, the control device 30 turns on the lighting device to illuminate the aquatic animals S in the culture area 10w (step S100). At this time, if the aquatic animals S are shrimp, it is desirable to minimize the time and amount of light that the lighting device is turned on, since shrimp are nocturnal and tend to avoid light. Also, if the aquatic animals S can be photographed accurately without turning on the lighting device, step S100 may be omitted.
[0066] (Step S200) Next, the control device 30 controls the imaging device 40 to capture an image of the aquatic animal S (step S200). The imaging device 40 transmits the captured image to the control device 30.
[0067] At this time, the imaging device 40 may transmit images of multiple areas in the aquaculture area 10w to the control device 30. For example, when the aquaculture support system 1 includes multiple imaging devices 40, the control device 30 acquires images captured by each imaging device 40.
[0068] The imaging device 40 may also have a mechanism for changing the imaging position or angle so that images of multiple areas in the aquaculture area 10w can be captured. In this case, the control device 30 changes the imaging position or angle of the imaging device 40, and the imaging device 40 captures multiple images at different imaging positions or angles and transmits each image to the control device 30.
[0069] (Step S300) The control device 30 processes the image acquired from the imaging device 40 in step S200 and outputs a first detected image PP1 (step S300). At this time, it is sufficient that the output first detected image PP1 has at least an identification area P1, and does not have to have an information area P2. In step S300, the control device 30 detects an identification area P1 for identifying an aquatic animal S in the image acquired from the imaging device 40, and determines the aquatic animal S.
[0070] The control device 30 uses, for example, a trained model that has been machine-learned to determine the aquatic animal S. For example, the trained model is generated by machine learning for object detection using images of the aquatic animal S labeled with position coordinates as training data or validation data.
[0071] Furthermore, the control device 30 may detect the identification area P1 and determine whether the aquatic animal S is a fish or not by using a pattern matching method such as template matching.
[0072] (Step S400) Next, the control device 30 judges whether or not to complete the photographing of the aquatic animal S (step S400). If it is determined in step S400 that the photographing of the aquatic animal S is not to be completed, the control device 30 proceeds to step S500.
[0073] Here, since the photographing of the aquatic animal S has not been completed, the aquaculture support system 1 needs to photograph the aquatic animal S again. For example, when the aquatic animal S moves quickly and the control device 30 cannot properly detect the identification area P1 from the moving image photographed by the imaging device 40, the control device 30 determines that the photographing of the aquatic animal S has not been completed.
[0074] (Step S500) In step S500, the control device 30 waits for a predetermined time (for example, 10 minutes). If the aquatic animal S cannot be determined in step S400 because the aquatic animal S moves too quickly, the control device 30 waits for the aquatic animal S to calm down by waiting for the predetermined time.
[0075] After waiting for a predetermined time, the control device 30 returns to step S200, again captures an image of the aquatic animal S with the imaging device 40, and executes the processes from step S200 onwards.
[0076] (Step S600) When the control device 30 determines in step S400 that the photographing of the aquatic animal S is completed, the process proceeds to step S600. At this time, the aquatic animal S has been appropriately determined by the identification area P1 detected by the control device 30. The control device 30 executes a size and quantity estimation process (step S600) based on the identification area P1 detected in step S300.
[0077] (Step S601) FIG. 5 is a flowchart showing the details of step S600 in FIG. First, the quantity estimation step (step S601) will be described.
[0078] In step S601, the control device 30 estimates the quantity (number of individuals) of the aquatic animals S captured in the first detected image PP1. The control device 30 calculates the quantity of the aquatic animals S by counting the number of identification regions P1 detected in step S300.
[0079] In step S601, for example, the control device 30 2 The number of aquatic animals S per unit is calculated. In this case, the area of the area captured in one first detection image PP1 is 1 m 2 If the difference is less than 1 m, the control device 30 calculates the difference based on a plurality of first detected images PP1 having different captured areas. 2 The quantity of aquatic animal S per unit may be calculated.
[0080] (Step S602) Furthermore, in step S600, the control device 30 executes a posture estimation step (step S602). In step S602, the control device 30 estimates the posture of the aquatic animal S based on the identification region P1 detected in step S300. Here, the posture of the aquatic animal S refers to the posture of the aquatic animal S, such as "the aquatic animal S has a straight body posture" or "the aquatic animal S has a bent body posture".
[0081] The control device 30 estimates the posture of the aquatic animal S, for example, by using a trained model that has been machine-learned. For example, the trained model is generated by machine learning for object detection and posture estimation using images with position coordinate labels and posture labels of the aquatic animal S as training data or validation data.
[0082] Furthermore, the control device 30 may estimate the posture of the aquatic animal S using a pattern matching method such as template matching.
[0083] The control device 30 calculates the coordinates of the identification region P1 in the first detected image PP1 and estimates the posture of the aquatic animal S. After estimating the posture of the aquatic animal S, the control device 30 proceeds to step S603.
[0084] (Step S603) In step S603, the control device 30 judges whether the posture of the aquatic animal S displayed in the identification area P1 is straight or curved based on the result of estimation in step S602. The control device 30 also judges whether the entire length of the aquatic animal S is captured within the identification area P1.
[0085] For an aquatic animal S that is in a straight position and whose entire length is captured in the image, the control device 30 can estimate the size of the aquatic animal S in step S604, which will be described later. On the other hand, it is difficult to estimate the size of an aquatic animal S that is in a crooked position or an aquatic animal S that is in a straight position but whose entire length is not captured in the image. Hereinafter, an aquatic animal S whose posture is straight and whose entire length is photographed will be referred to as an "individual whose size can be estimated." An aquatic animal S whose posture is bent or whose posture is straight but whose entire length is not photographed will be referred to as an "individual whose size is difficult to estimate."
[0086] For the aquatic animal S that is a "difficult-to-size-estimate individual" as a result of estimation in step S602, the control device 30 does not proceed to step S604 and ends the control of the aquaculture support system 1. At this time, the control device 30 outputs the quantity of the aquatic animal S estimated in step S601.
[0087] (Step S604) The control device 30 executes a size estimation step (step S604) for the aquatic animal S in the identification region P1 that has been determined to be a "size-estimable individual" in step S603.
[0088] In step S604, the control device 30 estimates the size of the aquatic animal S, for example, using markers. For example, a marker is placed on the bottom 10b of the culture area 10w, and in step S200, the imaging device 40 captures an image of the marker and the aquatic animal S. The control device 30 detects the marker from the captured image using a method such as template matching. Also, the actual size (dimension) of the marker is stored in advance in a memory unit (not shown) of the control device 30.
[0089] The control device 30 compares the size of the detected marker with the size of the identification region P1 surrounding the aquatic animal S in the first detection image PP1, and estimates the size of the aquatic animal S. The size of the aquatic animal S may be estimated using a trained model that has been machine-learned.
[0090] After estimating the size of the aquatic animals S in step S604, the control device 30 ends the control of the aquaculture support system 1. At this time, the control device 30 outputs the quantity of the aquatic animals S estimated in step S601 and the size of the aquatic animals S estimated in step S604.
[0091] The estimated results of the size and quantity of the aquatic animals S output by the control device 30 are transmitted to and displayed on the above-mentioned display device or terminal device. By checking the displayed estimated results, the farm worker can easily grasp the size and quantity of the farmed aquatic animals S.
[0092] By knowing the size of the aquatic animals S, the farm worker can understand the growth rate of the aquatic animals S and determine the amount of feed according to the growth and quantity of the aquatic animals S. In addition, by knowing the quantity of the aquatic animals S, the farm worker can observe the survival rate and growth environment condition of the aquatic animals S in the farming area 10w.
[0093] In the above-described method for estimating the size and quantity of aquatic animals S, the imaging device 40 may capture a video and transmit it to the control device 30, but it is preferable to use images that are easier to process by the control device 30. When the control device 30 processes the video, for example, the control device 30 can output a first detection image PP1 by performing image processing on a predetermined frame (image) extracted from the video captured by the imaging device 40.
[0094] Next, a method for estimating the activity of the aquatic animal S will be described. 6 is a flowchart showing an example of a method for estimating the activity level of the aquatic animal S in the control device 30. FIG. 7 is an example of a second detection image PP2 output by the control device 30 in the flowchart shown in FIG.
[0095] Here, activity level refers to the degree of activity of the aquatic animal S after feeding (feeding activity). By knowing the activity level of the aquatic animal S, the farm worker can understand the physical condition of the aquatic animal S and determine the amount of food to be fed. Furthermore, for example, if the aquatic animal S is a shrimp, the time at which the aquatic animal S is active (activity time) changes every day, and feeding can be done efficiently by feeding the aquatic animal S at an appropriate time relative to the activity time. Therefore, by estimating the activity level of the aquatic animal S, it can be farmed more efficiently.
[0096] 6 and 7, a method for estimating the activity level of an aquatic animal S by tracking feature points will be described. In this embodiment, feature point tracking, which will be described later, is effective for estimating the activity level of an aquatic animal S that is of a certain size, such as an adult shrimp, and that is active on the bottom 10b. The method for estimating activity shown in FIG. 6 is carried out after feeding the aquatic animal S.
[0097] (Step S110) For example, after feeding the aquatic animals S, the aquaculture worker starts control of the aquaculture support system 1 by the control device 30 using an input device such as a keyboard connected to the control device 30. When control of the aquaculture support system 1 is started, first, the control device 30 controls the imaging device 40 to capture an image of the aquaculture area 10w (step S110).
[0098] At this time, similarly to step S100 shown in FIG. 4, the image may be taken while illuminating the culture area 10w with a lighting device.
[0099] Furthermore, in step S110, the imaging device 40 captures a plurality of images and transmits them to the control device 30.
[0100] (Step S210) Next, the control device 30 executes a feature point tracking step (step S210). In step S210, the control device 30 processes the multiple images acquired from the imaging device 40 in step S110, and outputs a second detected image PP2.
[0101] 7, the control device 30 extracts feature points P3 from a plurality of images acquired from the imaging device 40. The feature points P3 are detected (extracted) by applying a feature point extraction algorithm such as SIFT (Scale-Invariant Feature Transform) to the images acquired from the imaging device 40.
[0102] The characteristic point P3 is a point (characteristic point) detected as a characteristic point of the aquatic animal S in the image captured by the imaging device 40. The control device 30 detects the characteristic point P3 in the multiple images acquired from the imaging device 40.
[0103] The control device 30 detects characteristic points P3 of the aquatic animal S in a first image and a second image selected from the plurality of images. Here, the first image is an image selected from a plurality of images captured by the imaging device 40, and is an image captured at a predetermined shooting time. The second image is an image selected from a plurality of images captured by the imaging device 40, and is an image captured at a different shooting time from the first image.
[0104] The control device 30 may obtain a video captured by the imaging device 40 and detect the feature point P3 by applying a feature point extraction algorithm to each frame of the obtained video. The control device 30 may also detect the feature point P3 by using a trained model that has been machine-learned.
[0105] (Step S310) Next, the control device 30 executes an activity measurement step (step S310). In step S310, the control device 30 calculates the movement amount per unit time of the characteristic point P3 detected in the first image and the second image in step S210, and outputs the calculated movement amount of the characteristic point P3 as the movement amount of the aquatic animal S.
[0106] The images for which the control device 30 detects the feature point P3 are not limited to the first image and the second image; the control device 30 may, for example, select three images (first to third images) from the multiple images acquired from the imaging device 40 and detect the feature point P3 for each image.
[0107] For example, the first to third images are assumed to have been captured at earlier times in this order. In this case, the control device 30 calculates the movement distance (first movement distance) of the feature point P3 between the first image and the second image. Also, the control device 30 calculates the movement distance (second movement distance) of the feature point P3 between the second image and the third image.
[0108] The control device 30 determines the sum of the first movement distance and the second movement distance as the total movement distance of the aquatic animal S, and calculates the movement amount of the characteristic point P3 per unit time. In other words, the control device 30 outputs the sum of the lengths of the line segments of the characteristic point P3 last detected among multiple images captured at different times as the total movement distance of the aquatic animal S.
[0109] The number of images in which the control device 30 detects the feature point P3 may be more than three. In addition, the control device 30 may calculate the amount of movement of the feature point P3 using a trained model that has been machine-learned.
[0110] The outputted amount of movement of the aquatic animals S is transmitted to and displayed on the above-mentioned display device or terminal device. The farm worker can use this amount of movement as the activity level to grasp the physical condition of the aquatic animals S and determine the amount of feed to be given.
[0111] 3. Furthermore, the second detected image PP2 generated in step S210 may have an identification region P1 or an information region P2, as shown in FIG. 7, similar to the first detected image PP1 shown in FIG.
[0112] Next, a method of estimating the activity of the aquatic animal S that differs from the estimation method based on feature point tracking in FIG. 6 will be described.
[0113] Fig. 8 is a flow chart showing an example of a method for estimating the activity level of the aquatic animal S in the control device 30. The method for estimating the activity level of the aquatic animal S by frame difference calculation will be described with reference to Fig. 8.
[0114] In this embodiment, the frame difference calculation is effective for estimating the activity of aquatic animals S that have small body lengths, such as juvenile shrimp, and that float evenly throughout the culture water W. The method for estimating activity shown in FIG. 8 is carried out after feeding the aquatic animal S.
[0115] (Step S120) For example, after feeding the aquatic animals S, the aquaculture worker starts control of the aquaculture support system 1 by the control device 30 using an input device such as a keyboard connected to the control device 30. When control of the aquaculture support system 1 is started, first, the control device 30 controls the imaging device 40 to capture an image of the aquaculture area 10w (step S120).
[0116] At this time, similarly to step S100 shown in FIG. 4, the image may be taken in a state where the culture area 10w is illuminated by a lighting device.
[0117] Furthermore, in step S120, the imaging device 40 captures a plurality of images and transmits them to the control device 30.
[0118] (Step S220) Next, the control device 30 executes a frame difference calculation step (step S220). In step S220, the control device 30 first converts the first image and the second image selected from the multiple images captured by the imaging device 40 into grayscale images. Next, the control device 30 applies a smoothing filter to the grayscaled first image and the second image to reduce noise.
[0119] Here, the first image and the second image are images captured at different times, similarly to the above-described method of estimating activity level by tracking feature points.
[0120] (Step S320) Next, the control device 30 executes an activity measurement step (step S320). In step S320, the control device 30 calculates the difference in brightness between the first image and the second image that have been smoothed in step S220, and outputs the calculated difference as a score (point) of the feeding activity.
[0121] Here, similarly to the above-described method of estimating activity level by tracking feature points, the control device 30 may execute steps S220 and S320 for three or more images.
[0122] For example, the control device 30 acquires the first to third images from the imaging device 40. The first to third images are assumed to have been captured at earlier times in this order. In this case, the control device 30 calculates the difference in luminance between the first and second images (first difference) and the difference in luminance between the second and third images (second difference), and outputs the sum of the first and second differences as the feeding activity score.
[0123] The number of images for which the control device 30 calculates the difference may be more than three. Furthermore, the control device 30 may calculate the difference using a trained model that has been machine-learned.
[0124] The output feeding activity score is transmitted to and displayed on the display device or terminal device. The farm worker can use the feeding activity score as the activity level to understand the physical condition of the aquatic animal S and to determine the feeding amount.
[0125] It is difficult to estimate the activity of aquatic animals S, such as juvenile shrimp, which have small body lengths and float evenly in the culture water W, using the above-mentioned feature point tracking. By using the activity estimation method using frame difference calculation shown in Figure 8, it is possible to estimate the activity of juvenile shrimp, for which feature point tracking is difficult to apply.
[0126] According to the aquarium 10 and the aquaculture support system 1 of this embodiment, the aquarium 10 capable of housing sand-burrowing aquatic animals S in the aquaculture area 10w has a bottom 10b that faces the aquaculture area 10w, has thermal insulation properties, and has an uneven shape simulating sand formed on the surface on the aquaculture area 10w side. This allows the aquatic animals S to have the illusion that there is bottom sand in the aquaculture area 10w of the aquarium 10, and allows stable and efficient aquaculture of aquatic animals S (e.g., kuruma shrimp) that are vulnerable to temperature changes and have sand-burrowing properties, even without providing bottom sand.
[0127] Furthermore, the aquaculture support system 1 of this embodiment includes the aquarium 10 having an uneven shape resembling sand, the imaging device 40 capable of photographing the aquaculture area 10w, and the control device 30 capable of detecting the identification area P1 for identifying the aquatic animals S in the image photographed by the imaging device 40, so there is no need to provide bottom sand in the aquarium 10 and there is no risk of the aquatic animals S (e.g., kuruma shrimp) hiding in the sand, and the aquatic animals S and remaining feed can be easily photographed by the imaging device 40. Therefore, the aquaculture worker can easily observe the aquatic animals S and the amount of remaining feed without having to perform a process such as draining the water to lower the water level.
[0128] Furthermore, the control device 30 estimates the size of the aquatic animal S based on the identification region P1. Therefore, the farm worker can grasp the size (growth level) of the aquatic animal S and determine the amount of feed according to the growth level without carrying out a process such as draining the water to lower the water level, and the sand-burrowing aquatic animal S can be farmed stably and efficiently.
[0129] The control device 30 also estimates the quantity of the aquatic animals S based on the identification area P1. Therefore, the farming worker can grasp the quantity of the aquatic animals and determine the amount of feed according to the quantity without carrying out a process such as draining the water and lowering the water level. Furthermore, the farming worker can observe the survival rate and growth environment of the aquatic animals S in the farming area 10w based on the quantity of the aquatic animals S. As a result, the aquatic animals S that have a sand-burrowing tendency can be farmed stably and efficiently.
[0130] Furthermore, the control device 30 can estimate the activity of the aquatic animals S based on the images captured by the imaging device 40. Therefore, the farm worker can grasp the physical condition of the aquatic animals S without carrying out a process such as draining the water to lower the water level, and can stably and efficiently farm the sand-burrowing aquatic animals S.
[0131] Although one embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and design modifications and the like within the scope of the present invention are also included. In addition, the components shown in the above-mentioned embodiment and the modified examples shown below can be appropriately combined to form a configuration.
[0132] (Variation 1) Although the aquaculture support system 1 of the above embodiment includes one water tank 10, the aspect of the aquaculture support system is not limited to this. The aquaculture support system may include a plurality of water tanks 10. For example, the aquaculture support system can be provided with an appropriate number of aquariums 10 depending on the size of the installation site, the number of aquatic animals S to be cultured, etc. When the aquaculture support system includes a plurality of aquariums 10, the aquaculture water W in each aquarium 10 can be prevented from being polluted by arranging the water pipes 23 so that the aquaculture water W in each aquarium 10 circulates inside the filtration section 21 and the purification section 22, as in the aquaculture support system 1 of the above embodiment.
[0133] (Variation 2) In the above embodiment, the bottom 10b and the sidewall 10s of the aquarium 10 have an uneven shape that imitates sand, but the embodiment of the aquarium 10 is not limited to this. It is sufficient that at least the bottom 10b of the aquarium has an uneven shape that imitates sand. For example, when the aquatic animal S to be cultured is kuruma shrimp, adult kuruma shrimp, which have a tendency to burrow into sand, mainly live on the bottom portion 10b. Therefore, even if the side walls do not have an uneven shape that imitates sand, as long as at least the bottom portion 10b has an uneven shape that imitates sand, the kuruma shrimp can be made to mistake the uneven shape of the bottom portion 10b for sand, reducing stress, and stable culture is possible. [Explanation of symbols]
[0134] 1. Aquaculture support system 10. Aquarium 10b bottom 10s side wall 10w aquaculture area 20 Water purification equipment 30 Control device 40 Imaging device S. Aquatic Animals PP1 First detected image (first image, second image) PP2 Second detection image (first image, second image) P1 identification area P3 feature points
Claims
1. An aquarium capable of housing sand-burrowing aquatic animals in an aquaculture area, The bottom faces the culture area, has thermal insulation properties, and has a bottom on the surface facing the culture area on which an uneven shape simulating sand is formed. Aquarium.
2. The uneven surface is black.
2. The aquarium of claim 1.
3. The aquatic animal is a prawn.
3. The aquarium according to claim 2.
4. The water tank according to any one of claims 1 to 3, An imaging device capable of photographing the aquaculture area; a control device capable of detecting an identification area for identifying the aquatic animal in the image captured by the imaging device; Equipped with Aquaculture support system.
5. The control device estimates the size of the aquatic animal based on the identification area. The aquaculture support system according to claim 4.
6. The control device estimates the quantity of the aquatic animals based on the identification area. The aquaculture support system according to claim 4.
7. The control device includes: detecting feature points indicating the position of the aquatic animal for a first image and a second image captured at a different time from the first image, the first image and the second image being selected from the plurality of images; estimating an activity level of the aquatic animal based on a movement amount per unit time of the feature point in the first image and the second image; The aquaculture support system according to claim 4.
8. the control device estimates the activity level of the aquatic animal based on a difference between a first image selected from the plurality of images and a second image captured at a different time from the first image; The aquaculture support system according to claim 4.
9. Further comprising a water purification device capable of removing impurities from the culture water contained in the culture area; The aquaculture support system according to claim 4.
Citation Information
Patent Citations
Culture apparatus and culture method
JP2021193946A
Cited By
Seawater crab attack ability testing device based on living body target
CN121003173A
Living target-based seawater crab attack ability testing device
CN121003173B
