Aquaculture system

Abstract

Observe the conditions inside the tanks where the aquaculture takes place. [Solution] The aquaculture system 100 includes a tank 114 for raising aquatic organisms, a camera 142 for imaging the tank, an image processing unit 162 for sharpening the images captured by the camera 142, an image determination unit 166 for determining the state of the Pacific white shrimp 108 and the amount of residual substances in the water based on the sharpened images, and an adjustment unit 168 for adjusting the amount of feed given to the Pacific white shrimp 108 based on the determination result.

Description

Technical Field

[0001] The present invention relates to a monitoring technology for an aquaculture tank.

Background Art

[0002] With the global increase in demand for fishery resources, the development of aquaculture technologies to achieve a stable supply of aquatic organisms such as shrimp and fish has been promoted. In recent years, with the problems of global warming and ocean pollution, land-based aquaculture, which is less affected by these factors, has attracted attention. Land-based aquaculture has advantages such as fewer location restrictions compared to offshore aquaculture, less susceptibility to weather and natural disasters by using indoor facilities, and the ability to reduce the environmental load caused by aquaculture. In a land-based aquaculture system, a circulation path is provided to circulate and purify the water in the tank (hereinafter referred to as "breeding water") for breeding aquatic organisms in order to keep the water clean (see Patent Document 1).

[0003] In the aquaculture of aquatic organisms, the feeding amount is an important growth factor. If there is too much feed, it may lead to pollution of the breeding water due to uneaten feed, an increase in the number of bacteria associated with it, and the generation of ammonia. Overfeeding not only results in wasted feed costs but also leads to an increase in the mortality rate of aquatic organisms due to water quality deterioration. On the other hand, if there is too little feed, the growth of aquatic organisms will be delayed. If the feeding is too little, there is also a possibility of nutritional disorders and cannibalism. Underfeeding also leads to an increase in the mortality rate of aquatic organisms.

[0004] Aquaculture farmers usually adjust the feeding amount after grasping the health status of aquatic organisms from their appearance and behavior. Also, if there is an abnormality in the aquatic organisms, it is necessary to detect it early and take immediate countermeasures. For this purpose as well, it is important to frequently and accurately confirm the health status of aquatic organisms.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

[0006] Aquarium water becomes cloudy due to leftover food and waste from aquatic organisms. When the water is cloudy, it becomes difficult to see the aquatic organisms. While increasing the circulation of the aquarium water can reduce turbidity, increasing the circulation too much is not economical. It is also said that aquatic organisms experience less stress in aquarium water that is somewhat cloudy than in perfectly clear water.

[0007] To achieve labor-saving and unmanned operation in land-based aquaculture, it is desirable to be able to photograph the tanks with cameras and adjust the amount of feed based on the information obtained from the captured images. However, if the images become unclear due to turbidity in the rearing water, it becomes difficult to understand the condition of the aquatic organisms.

[0008] This invention was completed in view of the above-mentioned problems, and its main purpose is to provide a technology for acquiring imaging images suitable for understanding the conditions inside a tank in which aquaculture takes place. [Means for solving the problem]

[0009] An aquaculture system in one aspect of the present invention comprises a tank for raising aquatic organisms, a camera for imaging the tank, an image processing unit for sharpening the images captured by the camera, an image determination unit for determining either or both the state of the aquatic organisms and the amount of residual substances in the water based on the sharpened images, and an adjustment unit for adjusting the cultivation environment for the aquatic organisms based on the determination results.

[0010] A different embodiment of the present invention provides an aquaculture system comprising: a tank for raising aquatic organisms; a camera for imaging the tank; an image determination unit for determining either or both the state of the aquatic organisms and the amount of residual substances in the water based on the captured image; and an adjustment unit for adjusting the cultivation environment for the aquatic organisms based on the determination result. In the aquarium, an observation area with a relatively shallow water depth is formed compared to other areas, and the camera is positioned so that the imaging range is within this observation area. [Effects of the Invention]

[0011] According to the present invention, it becomes easier to observe the conditions inside the aquarium. [Brief explanation of the drawing]

[0012] [Figure 1] This is a hardware configuration diagram of the aquaculture system. [Figure 2] This is a schematic diagram showing the configuration of an aquaculture farm and its management equipment. [Figure 3] This is a diagram showing the surrounding area of ​​the stage in the aquarium. [Figure 4] This is a functional block diagram of the aquaculture management system. [Figure 5] This is a screenshot of the original image. [Figure 6] This is a screenshot of a sharpened image. [Figure 7] This is a flowchart showing the observation process of Pacific white shrimp. [Figure 8] This is a schematic diagram showing the relationship between status data and the amount of feed given. [Figure 9] This is a diagram of the monitoring screen. [Figure 10] This is a diagram showing the configuration of the water tank in the modified example 1. [Figure 11] This is a diagram showing the configuration of the water tank in modified example 2. [Modes for carrying out the invention]

[0013] Hereinafter, an embodiment of the present invention will be described while referring to the drawings. In the following embodiments and their modifications, substantially the same components are denoted by the same reference numerals, and the description thereof will be omitted as appropriate. In this embodiment, the whiteleg shrimp, which is a kind of aquatic organism, will be described as an example.

[0014] FIG. 1 is a hardware configuration diagram of the aquaculture system 100. An aquaculture management device 110 is installed in the aquaculture farm 106. In the aquaculture system 100, the aquaculture management device 110, the server 104, and the communication terminal 112 are interconnected via the Internet 102. The aquaculture farm 106 is a facility for land-based aquaculture of whiteleg shrimps, but it may also be a facility other than land-based aquaculture such as sea-based aquaculture. The aquaculture management device 110 adjusts the growth environment of whiteleg shrimps by analyzing the state of whiteleg shrimps in the aquaculture farm 106 and feeding them accordingly.

[0015] The aquaculture management device 110 captures images of whiteleg shrimps and uploads the captured images to the server 104. The user can remotely observe the situation of the aquaculture farm 106 through the communication terminal 112 and issue various instructions to another user (aquaculture farmer) in the aquaculture farm 106 as needed. The communication terminal 112 may be a portable computer such as a PC, a smartphone, or a tablet computer installed with software for monitoring and control. The server 104 functions as a relay device and a data storage device for the aquaculture management device 110 and the communication terminal 112.

[0016] FIG. 2 is a schematic diagram showing the configuration of the aquaculture farm 106 and the aquaculture management device 110. The aquaculture facility 106 implements a closed-loop land-based aquaculture system and includes a tank 114 for raising Pacific white shrimp 108, a circulation path 116 for circulating the water (aquaculture water) in the tank 114, and a biological filtration tank 118 installed in the circulation path 116. A main pump 120 is installed in the circulation path 116. By driving the main pump 120, the aquaculture water circulates through the circulation path 116. The main pump 120 adjusts the amount of aquaculture water circulated and the direction of water supply (discharge direction). These aquaculture facilities are located in an indoor facility.

[0017] Land-based aquaculture methods are generally broadly classified into two types: the flow-through system and the closed-circuit system. The flow-through system involves pumping water from the sea or river and supplying it to tanks, and then draining the dirty water from the tanks, thus maintaining the cleanliness of the rearing water through water replacement. In contrast, the closed-circuit system has a circulation path for reusing the rearing water stored in the tanks, and maintains its cleanliness by passing the water through a filtration tank installed in that circulation path. There is also a semi-circular system, which is a hybrid of the flow-through and closed-circuit systems, in which the rearing water is circulated while a portion of it is replaced. In this embodiment, the closed-circuit system is adopted among land-based aquaculture methods to eliminate the introduction of pathogens from the outside and reduce seasonal factors that affect water temperature, etc. The rearing water temperature is set to a temperature suitable for the growth of Pacific white shrimp 108.

[0018] In this embodiment, the aquatic organism is Pacific white shrimp 108, but it may be a saltwater fish or a freshwater fish. It may also be shellfish, crustaceans, or other fish and shellfish. The composition of the rearing water is adjusted depending on whether the aquatic organism is a saltwater or freshwater organism.

[0019] A physical filter 122 is provided downstream of the tank 114 in the circulation path 116, and a biological filter tank 118 is provided downstream of that. Foreign matter such as molted shells, carcasses, and feces of the Pacific white shrimp 108 are captured by the physical filter 122.

[0020] In tank 114, toxic ammonia is generated due to the metabolic activity of the Pacific white shrimp 108 and the decomposition of organic matter such as leftover food. Therefore, the aquarium water is circulated and passed through the biological filter tank 118 to perform biological filtration, which decomposes the ammonia and converts it into less toxic nitrate. The biological filter tank 118 contains nitrifying bacteria (microorganisms) that oxidize ammonia in oxygen-containing water, converting it into nitrite and then nitrate.

[0021] The biological filtration tank 118 is connected to a foam separator 124 and a denitrification tank 126. Water discharged from the tank 114 and guided to the biological filtration tank 118 is driven by the pump 128 to the foam separator 124. The foam separator 124 separates the pollutants contained in the water by adsorbing them onto foam and causing them to float, and then returns the water to the biological filtration tank 118.

[0022] The nitrified water in the biological filter tank 118 is guided to the denitrification tank 126 by the drive of the pump 130 and returned to the biological filter tank 118. The denitrification tank 126 holds denitrifying bacteria and reduces the nitrates contained in the filtered water to nitrogen gas, which is released into the atmosphere. After being detoxified in this way, the water in the biological filter tank 118 is pumped up by the main pump 120 and supplied to the aquarium 114. The main pump 120 functions as a "water supply unit" that circulates the rearing water in the circulation path 116 and supplies the rearing water filtered by the biological filter tank 118 to the aquarium 114.

[0023] The aquaculture farm 106 is further equipped with an oxygen supply device 132 and a feeding device 134. The oxygen supply device 132 supplies oxygen to the water. The feeding device 134 supplies feed.

[0024] A raised area 136 is provided in a portion of the tank 114 to create an observation area for the Pacific white shrimp 108. The raised area 136 is composed of a shallow area of ​​water rising from the bottom of the tank 114. Details of the raised area 136 will be described in relation to Figure 3 below.

[0025] The aquaculture management device 110 controls each environmental device in the aquaculture farm 106 according to the environment in the tank 114. The aquaculture management device 110 maintains the dissolved oxygen level in the tank 114 within an appropriate range by controlling the oxygen supply device 132. If the ammonia concentration in the tank 114 exceeds the permissible range, the aquaculture management device 110 controls the operation of the main pump 120 to increase the circulating water volume and promote biological filtration of the rearing water. The aquaculture management device 110 ensures appropriate feeding of the Pacific white shrimp 108 by controlling the feeding device 134. Through such control, an appropriate rearing environment for the Pacific white shrimp 108 in the tank 114 can be maintained.

[0026] Figure 3 is a view of the area surrounding stage 140 in tank 114. As described above, a raised structure 136 is installed in a part of the tank 114 to create an observation area with a relatively shallow water depth. The raised structure 136 is a cylindrical object, but it does not need to be an object that rises up from the bottom surface 138 of the tank. In other words, the raised structure 136 does not need to be an object that forms a stage 140 as an observation area at a position at least higher than the bottom surface 138 of the tank.

[0027] A camera 142 is fixed to the side of the tank 114. The imaging range of the camera 142 includes the stage 140 of the raised object 136. A lighting device 144 is installed directly above the stage 140. The illumination range of the lighting device 144 includes the stage 140. In this embodiment, multiple observation times are set throughout the day, such as from 6:00 AM to 6:20 AM, from 11:00 AM to 11:20 AM, from 1:00 PM to 1:20 PM, from 3:00 PM to 3:20 PM, from 7:00 PM to 7:20 PM, and so on. During the observation time, the lighting device 144 is turned on, and the camera 142 images the illuminated stage 140 (observation area).

[0028] The aquaculture management device 110 grasps the condition of the Pacific white shrimp 108 from the captured images and adjusts the rearing environment of the Pacific white shrimp 108 by controlling the amount of feed based on the condition of the Pacific white shrimp 108. Due to the reasons mentioned above, the rearing water may become cloudy, which may result in unclear captured images. The following measures are taken to improve the clarity of the captured images.

[0029] <Countermeasure 1> Image sharpening through image processing By applying image sharpening processing to the captured images, the images are converted into ones that are easier to recognize, such as the Pacific white shrimp 108, leftover food, and feces. In the following, the captured image before sharpening processing will be called the "original image," and the captured image after sharpening processing will be called the "sharpened image." The sharpening processing will be described later in relation to Figures 5 and 6.

[0030] <Countermeasure 2> Imaging at low depth using raised object 136 By creating the raised surface 136, the stage 140 (observation area) is made to a lower depth. The shallower the area, the less turbidity there is. By creating a low-depth stage 140 with the raised surface 136, the clarity of the original image is improved.

[0031] <Measure 3> Enhanced clarity through lighting The lighting device 144 brightens the stage 140, improving the visibility of the Pacific white shrimp 108 and other organisms in the original image. Since the Pacific white shrimp 108 has a semi-transparent body, illuminating it makes it easier to see not only its external shape but also its internal organs. Because the lighting is only applied during the observation time, the stress that the lighting causes to the Pacific white shrimp 108 can be minimized.

[0032] <Countermeasure 4> Stage 140 Color The colors of the tank bottom 138 and the stage 140 (observation area) are made different. The tank bottom 138 is painted in a color close to the natural seabed, such as sand or soil. Natural materials such as sand and stones may be laid on the tank bottom 138. By making the tank bottom 138 closer to a natural color, a breeding environment as close to nature as possible is reproduced. On the other hand, since the body color of the Pacific white shrimp 108 is a translucent grayish-brown, if the background of the original image is a natural color, it becomes difficult to distinguish the Pacific white shrimp 108 from the background. In this embodiment, the stage 140 is painted in a brighter color than the tank bottom 138, such as white, cream, or yellow. By making the color of the stage 140 such a bright color, especially a color that is easy to distinguish from the body color of the Pacific white shrimp 108, it becomes easier to see the Pacific white shrimp 108 in the original image. The color of Stage 140 should be white, cream, yellow, green, or turquoise blue (the complementary color to brown), or any color that stands out from the body color of Vannamei shrimp 108.

[0033] The aquaculture management device 110 acquires data (hereinafter referred to as "condition data") that indicates various conditions of the rearing environment, such as the size, activity level, number of individuals, amount of leftover feed, water quality, and amount of leftover feces of the Pacific white shrimp 108, based on captured images.

[0034] Figure 4 is a functional block diagram of the aquaculture management device 110. Each component of the aquaculture management device 110 is realized by hardware including arithmetic units such as a CPU and various coprocessors, memory and storage devices, and wired or wireless communication lines connecting them, and software stored in the storage devices that supplies processing instructions to the arithmetic units. The computer program may consist of device drivers, an operating system, various application programs located at a higher layer, and libraries that provide common functions to these programs. The blocks described below represent functional units, not hardware units.

[0035] The aquaculture management device 110 includes a user interface processing unit 150, a communication unit 154, a data processing unit 152, and a data storage unit 156. The user interface processing unit 150 accepts user input via input devices such as touch panels and is responsible for processing related to the user interface, such as displaying images and outputting sound. The communication unit 154 is responsible for communication with various control equipment included in the aquaculture farm 106, as well as with the server 104 and the communication terminal 112. The data storage unit 156 stores various types of data. The data processing unit 152 executes various processes based on the data input from the user interface processing unit 150, the data received by the communication unit 154, and the data stored in the data storage unit 156. The data processing unit 152 also functions as an interface to the communication unit 154, the user interface processing unit 150, and the data storage unit 156.

[0036] The user interface processing unit 150 includes an input unit 158 ​​that receives input from the user and an output unit 160 that outputs images, etc.

[0037] The data processing unit 152 includes an image processing unit 162, an observation control unit 164, an image determination unit 166, and an adjustment unit 168. The image processing unit 162 performs sharpening processing on the captured image (original image). The observation control unit 164 illuminates the illumination device 144 during the observation time and instructs the camera 142 to take an image. The image determination unit 166 extracts various state data based on the sharpened image. The adjustment unit 168 controls the amount of feed and adjusts the rearing environment for the Pacific white shrimp 108 based on the state data.

[0038] Figure 5 is a screenshot of the original image 170. The image determination unit 166 extracts the region corresponding to the Pacific white shrimp 108 from the captured image and measures the body length and body width of the Pacific white shrimp 108. Due to turbidity in the rearing water, the appearance of the Pacific white shrimp 108 in the original image 170 tends to be unclear. The original image 170 shown in Figure 5 is an image captured when the above-mentioned countermeasures 2, 3, and 4 were not adopted. When the original image 170 is unclear, it becomes difficult to extract state data of the Pacific white shrimp 108 from the original image 170.

[0039] Figure 6 is a screen view of the enhanced image 180. The image processing unit 162 generates a sharpened image 180 by applying sharpening processing to the original image 170. The sharpening processing enhances the visibility of the Pacific white shrimp 108 by highlighting the slight color difference between the Pacific white shrimp 108 and the background (such as the rearing water). For example, if the background is dark brown and the Pacific white shrimp 108 appears light brown, the color at the boundary between the dark brown and light brown (hereinafter referred to as the "boundary color") is identified, and the brightness of the parts darker than this boundary color is reduced, while the brightness of the parts lighter than this boundary color is increased, thereby highlighting the outline of the Pacific white shrimp 108. The above is just one example of sharpening processing, and the sharpening processing by the image processing unit 162 can be achieved by applying existing image processing technologies.

[0040] In the enhanced image 180, the Pacific white shrimp 108 is clearly visible compared to the original image 170. The image determination unit 166 identifies the region corresponding to the Pacific white shrimp 108 and measures the body length and body width of the Pacific white shrimp 108. Since a stable correlation is known between the body length and body weight of the Pacific white shrimp 108, it is also possible to estimate the body weight of the Pacific white shrimp 108 from its body length.

[0041] Since the body of the Pacific white shrimp 108 is semi-transparent, the enhanced image 180 also shows the intestinal tract 182 of the Pacific white shrimp 108. The image determination unit 166 can estimate the amount of food present in the intestinal tract 182 from the proportion of the dark area within the intestinal tract 182. When the proportion of the dark area contained in the intestinal tract 182 exceeds a predetermined threshold, the image determination unit 166 determines that an obstruction (hereinafter referred to as "intestinal blockage") has occurred in the intestinal tract 182 of the Pacific white shrimp 108.

[0042] In addition to body length, body width, and intestinal blockage, the image determination unit 166 extracts various condition data from the Pacific white shrimp 108, such as body color, presence or absence of missing body parts, and swaying. The image determination unit 166 also extracts residual substances such as feces, leftover food, molted exoskeletons, and carcasses from the enhanced image 180. By using countermeasures 2, 3, and 4 in combination, it becomes even easier to extract condition data from the enhanced image 180.

[0043] Figure 7 is a flowchart showing the observation process of 108 Pacific white shrimp. The observation control unit 164 turns on the illumination device 144 to illuminate the stage 140 when the observation time arrives (S10). The observation control unit 164 instructs the camera 142 to start imaging (S20). The camera 142 captures a video of the stage 140 during the observation time. This video becomes the source image. The camera 142 may also acquire the source image 170 as multiple still images.

[0044] The image processing unit 162 generates a sharpened image 180 by applying sharpening processing to the original image 170, which has been captured as a still or moving image (S14). The image determination unit 166 extracts various state data by analyzing the sharpened image 180 (S16). The adjustment unit 168 adjusts the rearing environment based on the state data (S18). In this embodiment, the amount of feed is adjusted to adjust the rearing environment, but control parameters other than the amount of feed, such as the amount of oxygen supplied and the amount of rearing water circulated, may also be adjusted.

[0045] The image determination unit 166 further determines whether an abnormality has occurred in the enhanced image 180 (S18). An abnormality here can be arbitrarily set, such as the vannamei shrimp 108 being discolored, the number of vannamei shrimp 108 carcasses being greater than a predetermined number, the number of vannamei shrimp 108 being less than a predetermined number, the amount of leftover food being greater than a predetermined number, or abnormal behavior such as cannibalism occurring. The image determination unit 166 determines whether the predefined abnormality determination conditions have been met by image recognition.

[0046] If an abnormality occurs (Y in S18), the output unit 160 outputs warning information to the user indicating the abnormality (S20). The communication unit 154 may also send the warning information to the communication terminal 112 of a remote user. If there is no abnormality (N in S18), the process in S20 is skipped. When the observation time ends, the observation control unit 164 turns off the illumination device 144 (S24). At this time, imaging also ends.

[0047] Remote users may specify various control parameters from the communication terminal 112. Alternatively, the communication terminal 112 may include an environmental control unit. The environmental control unit of the communication terminal 112 may automatically adjust the control parameters for the rearing environment of the Pacific white shrimp 108 using various AI (artificial intelligence) such as neural networks and rule-based systems, in accordance with warning information received from the aquaculture management device 110.

[0048] Figure 8 is a schematic diagram showing the relationship between state data and the amount of feed given. From the enhanced image 180, various state data such as activity level, swimming population ratio, swaying, and number of individuals detected can be obtained. The following lists each state data and how to determine it. (1) Activity level The image determination unit 166 calculates the activity level of the Pacific white shrimp 108 based on how the shrimp moves. Activity level is an indicator of whether the Pacific white shrimp 108 is still or moving. For example, even if a Pacific white shrimp 108 is at the bottom of the tank 114 and not moving much, if it is moving its legs in small increments, it is judged to be highly active. The image determination unit 166 calculates the activity level as the amount of positional change of the Pacific white shrimp 108 per unit time. In addition, the image determination unit 166 calculates the overall activity level as the average activity level of multiple Pacific white shrimp 108 that were captured in the enhanced image 180 during the observation time. Hereinafter, activity level will refer to the average activity level of multiple observed Pacific white shrimp 108. The increase or decrease in activity level during the current observation compared to the activity level during the previous observation is called the "activity level increase / decrease rate." The same method is used to determine the increase / decrease rate for the swaying level and other parameters shown below.

[0049] (2) Swimming population The image determination unit 166 calculates the proportion of Pacific white shrimp 108 swimming in the tank 114 as the swimming population rate. More specifically, the image determination unit 166 calculates the swimming population rate as the ratio of the number of Pacific white shrimp 108 whose movement per unit time exceeds a predetermined threshold to the total number of Pacific white shrimp 108 observed during the observation period.

[0050] (3) Degree of swaying The image determination unit 166 calculates the number of direction changes per unit time of the Pacific white shrimp 108 as "swaying." The image determination unit 166 calculates the average value of the number of swayings of multiple Pacific white shrimp 108 as the "degree of swaying." If the proportion of Pacific white shrimp 108 whose number of swayings per unit time exceeds a predetermined number exceeds a predetermined threshold, the image determination unit 166 determines that it is abnormal.

[0051] (4) Number of individuals detected The image determination unit 166 calculates the number of Pacific white shrimp 108 detected during the observation period as the individual detection count. By observing the increase or decrease in the individual detection count for each of the multiple observation opportunities, it is possible to estimate the increase or decrease in the number of Pacific white shrimp 108 in the tank 114.

[0052] (5) Body length and body width The image determination unit 166 measures the body length and body width of the Pacific white shrimp 108. The image determination unit 166 estimates the surface area and body weight of the Pacific white shrimp 108 from the body length and body width. The image determination unit 166 calculates the average of the estimated body weights of multiple Pacific white shrimp 108 as the "average body weight".

[0053] (6) Body color When Vannamei shrimp 108 become ill, their body color may change. The image determination unit 166 determines the body color of Vannamei shrimp 108. By determining the body color, the image determination unit 166 determines whether there is an abnormality in Vannamei shrimp 108. The image determination unit 166 calculates the percentage of Vannamei shrimp 108 whose body color has changed as the "discoloration rate," and determines that there is an abnormality if the discoloration rate is above a predetermined threshold. Note that "body color" here includes not only the color of the Vannamei shrimp 108's carapace, but also the color of its internal organs and tail.

[0054] (7) Partial defect The image determination unit 166 determines whether there are any defects in parts of the Pacific white shrimp 108, such as their tails or legs. If there are defects in parts, cannibalism may be occurring. The image determination unit 166 calculates the percentage of Pacific white shrimp 108 with defects as the "defect rate," and determines that there is an abnormality if the defect rate is above a predetermined threshold.

[0055] (8) Intestinal blockage The image determination unit 166 determines whether the intestinal tract 182 of the Pacific white shrimp 108 is blocked. Pacific white shrimp 108 with blocked intestinal tracts can be estimated to be in good health because they are able to feed and digest properly. The image determination unit 166 calculates the percentage of Pacific white shrimp 108 with blocked intestinal tracts as the "intestinal blockage rate".

[0056] (9) Amount of leftover feed The image determination unit 166 detects the bait visible in the enhanced image 180. The image determination unit 166 calculates the amount of leftover bait based on the area of ​​the remaining bait visible in the enhanced image 180. Alternatively, the number of bait particles in the water may be considered as the amount of leftover bait.

[0057] The image determination unit 166 calculates the following input parameters from this state data. (1) Health level Health status is calculated using activity level, swimming individual rate, dizziness, site defects, and intestinal obstruction as variables. Specifically, the image determination unit 166 calculates health status using an arbitrary function that monotonically increases with respect to the increase rates of activity level, swimming individual rate, and intestinal obstruction, and monotonically decreases with respect to the increase rates of dizziness and defect rate. For example, the function for calculating health status is defined as f(x1,x2,x3,x4,x5)=A1·x1+A2·x2+A3·x3+B1·x4+B2·x5. Here, x1 is the increase rate of activity level, x2 is the increase rate of swimming individual rate, x3 is the increase rate of intestinal obstruction rate, x4 is the increase rate of dizziness, and x5 is the increase rate of defect rate. Also, A1, A2, and A3 are positive coefficients, and B1 and B2 are negative coefficients.

[0058] (2) Satiety The image determination unit 166 calculates satiety based on the rate of intestinal blockage and the amount of leftover food. The image determination unit 166 calculates satiety using an arbitrary function that increases monotonically with respect to the rate of increase of the rate of increase of the rate of intestinal blockage and the rate of increase of the amount of leftover food, respectively.

[0059] (3) Growth rate The image determination unit 166 calculates the growth rate based on the average weight increase.

[0060] The adjustment unit 168 determines the amount of feed based on three types of input parameters. The adjustment unit 168 reduces the amount of feed using a monotonically decreasing function with respect to satiety and growth. Alternatively, the adjustment unit 168 may increase the amount of feed when the biomass weight (total weight = assumed number of individuals × assumed body weight) of the Pacific white shrimp 108 is increasing, and decrease the amount of feed when the biomass weight is decreasing. It may also maintain or increase the amount of feed when health is improving, and decrease the amount of feed when health is declining. It is also possible to adjust the amount of feed using input parameters other than health, satiety, and growth.

[0061] Figure 9 is a screenshot of the monitoring screen 190. The monitoring screen 190 is displayed on the aquaculture management device 110. The communication unit 154 can also display the monitoring screen 190 on the communication terminal 112 by transmitting the screen data of the monitoring screen 190 to the communication terminal 112. The monitoring screen 190 includes an image area 192, a transition area 194, and a warning information area 196. In the image area 192, the original image 170 or the enhanced image 180 from the most recent observation time is displayed.

[0062] Camera 142 may be kept running at all times, and the aquaculture management device 110 may continuously display the captured images of the tank 114 in the image area 192. With this control method, the condition of the tank 114 can be checked remotely.

[0063] The transition region 194 shows the time evolution of various input parameters. As shown in Figure 9, it can be seen that the Pacific white shrimp 108 are growing well. The user can quantitatively understand the growth status of the Pacific white shrimp 108 by using the transition region 194.

[0064] The warning information area 196 displays a list of information that should be noted in the rearing environment. In Figure 9, the degree of instability is "somewhat high." This means that the degree of instability of the entire group of Pacific white shrimp 108 exceeds a predetermined standard. In addition, the image determination unit 166 displays warning information regarding the defect rate in the warning information area 196 when the defect rate is high.

[0065] The image determination unit 166 notifies the user if there are signs of disease in the Pacific white shrimp 108. For example, several inspection items such as body color, behavior, and deformities of the Pacific white shrimp 108 are set, and if there are Pacific white shrimp 108 that meet these inspection items, the image determination unit 166 notifies the user of the possibility that disease has occurred. In addition, if there is a possibility of disease occurring, the adjustment unit 168 may automatically control the oxygen supply amount, temperature, etc.

[0066] The user can also send control instructions for circulation rate and feed rate to the aquaculture management device 110 from the communication terminal 112. The communication terminal 112 can also issue various instructions to the user at the aquaculture farm 106. As described above, the communication terminal 112 may automatically control various control parameters according to the information transmitted from the aquaculture system 100.

[0067] [Summary] The aquaculture system 100 has been described above based on the embodiments. By adjusting the amount of feed based on the condition data, the survival rate of the Pacific white shrimp 108 can be increased, and their healthy growth can be promoted. The aquaculture management device 110 images the tank 114 and acquires various condition data from the captured images. Since there is no need to capture and inspect the Pacific white shrimp 108, the shrimp 108 are not harmed during observation.

[0068] A certain degree of turbidity in the rearing water is considered favorable for Vannamei shrimp 108. While turbidity in the rearing water can easily blur the captured image (original image 170), image enhancement processing makes it easier to see the underwater conditions. This method allows for maintaining a comfortable environment for Vannamei shrimp 108 while appropriately monitoring their condition.

[0069] In tank 114, the stage 140 (observation area) is placed in a relatively shallow area. By illuminating the stage 140 and making it a different color from the tank bottom surface 138, the visibility of the Pacific white shrimp 108 can be improved. In addition, limiting the observation time can reduce stress on the Pacific white shrimp 108 associated with observation.

[0070] Automatic adjustment of feed amounts based on captured images will contribute to labor saving or unmanned operation of the aquaculture farm 106. Traditionally, the cultivation of Pacific white shrimp 108 depended on the experience and skill of aquaculture farmers. It is believed that incorporating the knowledge of aquaculture farmers will enable stable and economical aquaculture in the future. According to data from the Fisheries Agency, Japan's seafood imports in 2023 amounted to approximately 2 trillion yen. In monetary terms, shrimp accounted for approximately 9.6% of this (approximately 14% if processed shrimp products are included). Realizing land-based aquaculture of various aquatic organisms, including Pacific white shrimp 108, and especially promoting labor saving to improve the price competitiveness of land-based aquaculture, is important from the perspective of food security.

[0071] It should be noted that the present invention is not limited to the embodiments and modifications described above, and the components can be modified and implemented without departing from the spirit of the invention. Various inventions may be formed by appropriately combining the multiple components disclosed in the embodiments and modifications described above. In addition, some components may be deleted from all the components shown in the embodiments and modifications described above.

[0072] [Differentiation] Figure 10 is a diagram showing the configuration of the water tank 114 in the modified example 1. In the first modification, multiple water jets 198 are provided on the bottom surface 138 of the tank so as to surround the raised structure 136. Upon instruction from the observation control unit 164, the main pump 120 sprays water from the water jets 198 toward the raised structure 136 during the observation period. This drives the Pacific white shrimp 108 near the bottom surface 138 toward the stage 140. This control method allows for the observation of a larger number of Pacific white shrimp 108 at once. Being able to observe many Pacific white shrimp 108 at once makes it easier to shorten the observation time.

[0073] Figure 11 is a diagram showing the configuration of the water tank 114 in modified example 2. In Modification 2, the water tank 114 does not have a raised structure 136. The bottom surface 138 of the water tank is provided with numerous spout holes 198. The main pump 120 sprays water upward from the spout holes 198 during the observation period. In Modification 2, the lighting device 144 is installed facing the camera 142, and the space between the camera 142 and the lighting device 144 becomes the observation area.

[0074] The upward jet of water from the fountain nozzle 198 propels the Pacific white shrimp 108 towards the water surface. At this time, the lighting device 144 illuminates the propelled Pacific white shrimp 108, and the camera 142 captures images of the Pacific white shrimp 108.

[0075] In tank 114, the water is less turbid in the shallower areas than in the deeper areas. The upward jet of water from the fountain nozzle 198 draws the Pacific white shrimp 108 closer to the water surface, allowing for the acquisition of relatively clear images (original images 170). By positioning the camera 142 and the lighting device 144 facing each other and illuminating the Pacific white shrimp 108 from behind, the silhouette of the Pacific white shrimp 108 can be clearly captured in the images.

[0076] The main pump 120 may change its water supply direction during observation. Normally, the main pump 120 supplies water to the tank 114 from a horizontal pipe, but during observation, this pipe may be directed upwards to supply water upwards. With this control method, the Pacific white shrimp 108 may be driven into the observation area only during observation.

[0077] During the observation period, the observation control unit 164 may use light to guide the Pacific white shrimp 108 into the observation area. Specifically, the Pacific white shrimp 108 may be guided into the observation area by irradiating an area other than the observation area with light of a color that the Pacific white shrimp 108 dislike. In addition to light, guidance can also be provided by sound, smell, electromagnetic waves, vibrations, ultrasound, or an underwater swimming robot that mimics a natural enemy.

[0078] The observation control unit 164 may guide the Pacific white shrimp 108 to the observation area by controlling the feeding device 134. For example, the feeding device 134 may guide the Pacific white shrimp 108 to the stage 140 by scattering food on the stage 140 during the observation time.

[0079] The observation control unit 164 may instruct the main pump 120 to increase the water flow rate during the observation period. Pacific white shrimp 108 have a tendency to swim in areas with current. By increasing the water flow (circulation force of the rearing water) during the observation period, the swimming of the Pacific white shrimp 108 can be encouraged, thereby gathering many Pacific white shrimp 108 into the stage 140 (observation area).

[0080] The aquaculture management device 110 uploads various data, such as captured images and status data, to the server 104. The communication terminal 112 can check the data stored on the server 104.

[0081] In this embodiment, the aquaculture management device 110 is described as performing all of the following: imaging, sharpening processing, identification of state data, and determination of feed amounts. However, all or part of the functions of the aquaculture management device 110, such as the image processing unit 162, image determination unit 166, observation control unit 164, and adjustment unit 168, may be implemented by the server 104 instead of the aquaculture management device 110.

[0082] Instead of limiting the observation time, the aquarium 114 may be continuously observed by keeping the camera 142 running at all times. Furthermore, instead of limiting the observation area to a portion of the aquarium 114, the position and field of view of the camera 142 may be set so that the entire aquarium 114 is within the imaging range.

[0083] The camera 142 and the lighting device 144 may be installed either underwater or above the water. It is desirable that the camera 142 be close to the observation area. For this reason, in this embodiment, the camera 142 is installed underwater. For similar reasons, it is desirable to use a camera 142 with a short focal length.

[0084] Camera 142 may acquire multiple images while changing the magnification. Image determination unit 166 may select an image with a suitable magnification for analysis from among the multiple images and extract state data from the selected image. For example, when measuring the amount of leftover food, a low-magnification image with a wide imaging range is desirable, while a high-magnification image is desirable when determining intestinal blockage. Similarly, illumination device 144 may acquire images corresponding to multiple illumination conditions by changing the light intensity, irradiation angle, and illumination color.

[0085] The sharpening process may be modified depending on the subject. The sharpening process for the Pacific white shrimp 108, the sharpening process for feces, the sharpening process for leftover food, and the sharpening process for molted exoskeletons do not need to have the same settings for image correction. The image processing unit 162 may use multiple sharpening processes depending on the purpose of analysis.

[0086] The appropriate background color for image clarity varies depending on the aquatic organism. The color of Stage 140 can be arbitrarily set according to the body color of the aquatic organism being cultivated. The same applies to the lighting color. For example, shrimp do not mind red, while flounder prefer green; different aquatic organisms have preferred and disliked colors. It is preferable to consider these points when determining the colors of Stage 140 and the tank bottom 138, as well as the lighting color.

[0087] The main pump 120 may temporarily reduce the turbidity of the rearing water by increasing its circulation volume during observation. This control method can improve the clarity of the original image.

[0088] The adjustment unit 168 may control the amount of food dispensed based on the amount of leftover food visible in the captured image. Specifically, the adjustment unit 168 may decrease the amount of food dispensed when the area of ​​leftover food in the captured image increases, and increase the amount of food dispensed when the area of ​​leftover food decreases.

[0089] The adjustment unit 168 may change not only the amount of feed but also the type of feed. For example, the feed may be changed when the health of the Pacific white shrimp 108 is declining. Alternatively, the adjustment unit 168 may change the feed when the growth rate of the Pacific white shrimp 108 exceeds a predetermined value.

[0090] Condition data may include various data related to the Pacific white shrimp 108, as well as residual substances such as the amount of leftover feed, the amount of molted shells, the amount of dead shrimp, and the amount of feces. In addition, control parameters for adjusting the rearing environment may include the amount of feed, oxygen supply, water temperature, circulation rate (circulation force), and pH value (hydrogen ion concentration). For example, when the amount of feces increases, the aquaculture management device 110 may instruct the main pump 120 to increase the circulation force to enhance the ability to remove foreign matter such as feces. When the health of the Pacific white shrimp 108 decreases, the aquaculture management device 110 may instruct the oxygen supply device 132 to increase the oxygen supply.

[0091] The input and control parameters may be adjusted according to a reinforcement learning (artificial intelligence) model. For example, when a set of input parameters (e.g., health level) and a set of control parameters (e.g., feeding amount) are set, the learning model may be adjusted by providing a "reward" for the preferred parameters of the Pacific white shrimp 108, such as when health level increases.

[0092] In this embodiment, when calculating input parameters such as health status, satiety status, and input status, the "increase rate" of state data such as activity level and swimming individual rate from the previous observation is used as a variable. As a variation, when calculating input parameters, the state data itself (raw data) such as activity level and swimming individual rate may be used as a variable, or the normalized value of the raw data may be used as a variable. [Explanation of Symbols]

[0093] 100 Aquaculture system, 102 Internet, 104 Server, 106 Aquaculture farm, 108 Pacific white shrimp, 110 Aquaculture management device, 112 Communication terminal, 114 Tank, 116 Circulation path, 118 Biological filtration tank, 120 Main pump, 122 Physical filter, 124 Foam separation device, 126 Denitrification tank, 128 Pump, 130 Pump, 132 Oxygen supply device, 134 Feeding device, 136 Raised structure, 138 Tank bottom, 140 Stage, 142 Camera, 144 Lighting device, 150 User interface processing unit, 152 Data processing unit, 154 Communication unit, 156 Data storage unit, 158 Input unit, 160 Output unit, 162 Image processing unit, 164 Observation control unit, 166 Image determination unit, 168 Adjustment unit, 170 Original image, 180 Sharpened image, 182 Intestinal tract, 190 Monitoring screen, 192 Image area, 194 Transition area, 196 Warning information area, 198 Fountain hole

Claims

1. A tank for keeping aquatic creatures, A camera for imaging the aforementioned water tank, An image processing unit that sharpens the image captured by the aforementioned camera, An image determination unit that determines, based on the enhanced image, the state of aquatic organisms and the amount of residual substances in the water, or both. An aquaculture system comprising: an adjustment unit that adjusts the cultivation environment for aquatic organisms based on the results of the aforementioned determination.

2. In the aforementioned tank, an observation area with a relatively shallow water depth is formed compared to other areas. The aquaculture system according to claim 1, wherein the camera is installed so as to have the observation area as its imaging range.

3. The aquaculture system according to claim 2, further comprising an illumination device that illuminates the aforementioned observation area.

4. The aquaculture system according to claim 2, wherein the observation area is colored with a different color from the other areas.

5. During the observation period, an observation control unit is provided to acquire captured images by controlling the camera, The system further comprises a water supply unit for supplying breeding water to the aforementioned tank, The aquaculture system according to claim 1, wherein the observation control unit moves aquatic organisms toward the water surface by changing both or either the direction and amount of water supplied by the water supply unit during the observation time.

6. The image determination unit determines the amount of leftover feed from the captured image, The aquaculture system according to claim 1, wherein the adjustment unit changes the amount of feed given based on the amount of leftover feed.

7. The aquatic organism in question is a type of shrimp, The image determination unit determines the degree of obstruction of the intestinal tract of the aquatic organism from the captured image, The aquaculture system according to claim 1, wherein the adjustment unit changes the amount of feed supplied based on the degree of blockage.

8. A tank for keeping aquatic creatures, A camera for imaging the aforementioned water tank, An image determination unit that determines either or both the state of aquatic organisms and the amount of residual substances in the water based on the captured image, Based on the results of the above determination, the system includes an adjustment unit that adjusts the environment for raising aquatic organisms, In the aforementioned tank, an observation area with a relatively shallow water depth is formed compared to other areas. The camera is installed in the aquaculture system so as to capture the observation area.

9. A function to enhance the clarity of images captured inside a tank where aquatic organisms are kept, Based on the enhanced image, the function determines either or both the state of aquatic organisms and the amount of residual substances in the water. An aquaculture support program that uses a computer to perform functions to adjust the cultivation environment for aquatic organisms based on the results of the aforementioned determination.

Images