System for determining activity of farmed fish and system for determining feeding amount of farmed fish

Abstract

To provide a system for determining the activity level of farmed fish that can determine the activity level of farmed fish in response to feeding with high accuracy, even when the imaging environment, such as the location where the determination is made.SOLUTION: The system for determining the activity level of farmed fish includes: imaging means for imaging the activity state of the farmed fish; and activity level determination means for receiving a signal from the imaging means and determining the activity level of the farmed fish in response to feeding. The imaging means captures a reference image of the farmed fish before it moves and a plurality of still images at predetermined time intervals within a predetermined time period, and the activity level determining means compares the still image with the reference image that serves as a reference for determining the activity level of the farmed fish in response to feeding, and determines the activity level of the farmed fish.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a system for determining the activity level of farmed fish and a system for determining the amount of feed given to farmed fish. [Background technology]

[0002] Traditionally, farmed fish raised in fish cages have been fed using feeders. However, this method of feeding does not take into account the level of sufficiency of the farmed fish, and the feeder is only controlled by an on / off switch or a timer to determine the timing of feeding.

[0003] Therefore, the sufficiency of the feed for the farmed fish is not taken into consideration, resulting in waste of feed. Alternatively, only the farmed fish near the feed input are fed with a high level of sufficiency, making it difficult to feed all the farmed fish in the cage evenly. Furthermore, excess feed becomes leftover feed, which directly leads to pollution of the water quality in the cage, posing a problem for the aquaculture industry in terms of water quality.

[0004] For example, Patent Document 1 discloses an invention relating to an automatic feeding method that can appropriately adjust and control the amount and time of feeding farmed fish. The invention in Patent Document 1 is an invention for an automatic feeding method for farmed fish that uses a network camera that captures the feeding status of farmed fish, and artificial intelligence that uses machine learning to determine the amount of feed per day and the optimal feeding time for that day, as well as to determine the activity of farmed fish when they are feeding. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2020-78278 Summary of the Invention [Problem to be solved by the invention]

[0006] However, in the automatic feeding method for farmed fish in Patent Document 1, when filming the feeding status of farmed fish with a network camera, if the filming is done outdoors, there is a wide variety of filming conditions at the evaluation site, such as differences in outdoor light, the condition of the fish pens such as whether or not there is bird netting on the top of the pens, and whether or not there are clouds reflected on the water surface, resulting in a problem of insufficient evaluation accuracy.

[0007] Therefore, the present invention has been made in consideration of the above-mentioned problems, and aims to provide a system for determining the activity level of farmed fish and a system for determining the amount of feed to be given to farmed fish, which can determine the activity level of farmed fish in response to feeding with high accuracy, even under a wide variety of conditions in the environment in which the determination is made and the farmed fish. [Means for solving the problem]

[0008] One aspect of the present invention is a system for determining the activity level of farmed fish, which determines the activity level of farmed fish in response to feeding, and is equipped with an imaging means for capturing images of the activity state of the farmed fish, and an activity level determination means for receiving a signal from the imaging means and determining the activity level of the farmed fish in response to feeding, wherein the imaging means captures a reference image of the farmed fish before it moves and a plurality of still images at predetermined time intervals within a predetermined period of time, and the activity level determination means compares the still images with the reference image, which serves as a basis for determining the activity level of the farmed fish in response to feeding, to determine the activity level of the farmed fish.

[0009] With this configuration, a reference image and multiple still images of the farmed fish are captured at specified time intervals within a specified time period, and the activity level of the farmed fish in response to feeding is determined based on the reference image before the farmed fish moves and the multiple still images.This means that the activity level of the farmed fish in response to feeding can be determined without the need for a person to constantly monitor the activity level of the farmed fish in response to feeding.

[0010] Furthermore, this system for determining the activity level of farmed fish is characterized in that the activity level determination means determines the activity level of the farmed fish in response to feeding based on changes in the shade of the reference image and the multiple still images.

[0011] According to this configuration, for example, outdoor imaging environments such as fluctuations in light quantity due to changes over time, differences in backgrounds around the imaging area, fluctuations in the color of the water surface due to weather, and the presence or absence of bird-proof nets are diverse. By making a determination based on the shading changes between the reference image and a plurality of still images, it is possible to accurately determine the activity level of cultured fish for feeding in any determination environment.

[0012] Further, this cultured fish activity determination system is characterized in that the activity discrimination means includes means A for dividing the reference image and one still image into a plurality of blocks respectively, means B for calculating the difference in shading for each corresponding block between the reference image and the one still image, means C for calculating the difference in shading between the reference image and the one still image based on the calculated difference in shading for each block, and means D for discriminating the level of shading change based on the calculated difference in shading between the reference image and the one still image by a plurality of predetermined thresholds. According to this configuration, the discrimination accuracy regarding the activity level of cultured fish for feeding is improved.

[0013] Further, this cultured fish activity determination system is characterized in that means B is means for calculating the difference in the additive average value of the shading for each corresponding block between the reference image and the one still image. According to this configuration, by inputting the difference in the additive average value calculated by means A into a neural network, it is possible to accurately determine the activity level of cultured fish for feeding.

[0014] Further, this cultured fish activity determination system is characterized in that means C uses the difference in shading for each block as an input value to a neural network and calculates a high-order discrimination curve for non-linear discrimination as the difference information in shading between the reference image and the one still image, and means D is means for discriminating the level of shading change based on the high-order discrimination curve by a plurality of predetermined thresholds. According to this configuration, it is possible to accurately determine the activity level of cultured fish for feeding. Also, it is possible to determine the activity level for feeding that is not affected by the determination environment, and to make a strict and flexible determination.

[0015] Furthermore, this farmed fish activity determination system is characterized in that the activity determination means includes means E for executing means A to D for each of the plurality of still images taken at predetermined time intervals within a predetermined time period. With this configuration, by having means E for executing means A to D for each of the plurality of still images, it is possible to accurately determine the activity of farmed fish in response to feeding.

[0016] Furthermore, this system for determining activity levels of farmed fish is characterized in that the activity determination means determines the activity levels of the farmed fish in response to feeding by majority voting of the determination results of the levels of change in shading obtained by executing means A to D for the reference image and each of the still images. With this configuration, activity levels in response to feeding can be determined with high accuracy.

[0017] Furthermore, in this farmed fish activity determination system, the activity determination means determines activity only when at least two of the determination results for the still images are the same, based on a majority vote, and invalidates the determination if all of the determination results are different. This configuration improves the reliability of the determination.

[0018] Furthermore, this farmed fish activity determination system is characterized in that the neural network is trained using learning data that pairs a reference image with a still image corresponding to the activity level for each activity level.

[0019] According to this configuration, the accuracy of judgment is improved by learning using training data consisting of a pair of a reference image and a still image. Conventional training data involves linking activity levels with a single image data, i.e., activity level 3 corresponds to this image data, and learning is performed by linking activity levels with a single image data. On the other hand, this invention links activity levels with a pair of image data, such as an activity level and its corresponding image data + a reference image, thereby significantly improving judgment accuracy compared to conventional methods. This is particularly effective when images fluctuate depending on the environment of the aquaculture farm.

[0020] The system for determining the activity of farmed fish is also characterized in that the activity determination means performs the discrimination between the plurality of still images and the reference image at predetermined time intervals multiple times within a predetermined time period. With this configuration, by discriminating between the still images and the reference image multiple times at different times, the activity of farmed fish in response to feeding can be determined with high accuracy.

[0021] This farmed fish activity determination system is also characterized in that the imaging means is located on land and the still image is an image of the water surface. With this configuration, the imaging means captures an image of the water surface, and the activity level of the farmed fish is determined based on the image corresponding to the state of the school of fish during feeding. This is advantageous because it allows the appetite activity of farmed fish to be determined in their natural state without installing a camera, which is a foreign object, in the water.

[0022] Furthermore, in this system for determining the activity level of farmed fish, the images obtained by the imaging means are visible images or near-infrared images. With this configuration, a system that can be used for a variety of purposes can be realized by signal processing using information from the visible images or near-infrared images.

[0023] In addition, this activity level determination system for cultured fish is provided with sensor irradiation means for irradiating sensor signals such as ultrasonic waves instead of the imaging means. The sensor irradiation means captures a reference reflection signal before the cultured fish moves and a plurality of reflection signals at predetermined time intervals within a predetermined time. The activity level discrimination means uses the reflection signals instead of the still images and uses the reference reflection signal instead of the reference image. It receives the reflection signal from the sensor irradiation means, compares the reflection signal with the reference reflection signal that serves as a reference for determining the activity level of the cultured fish with respect to feeding, and discriminates the activity level of the cultured fish with respect to feeding.

[0024] According to this configuration, even when the imaging environment is poor, the activity level of the cultured fish with respect to feeding can be accurately discriminated by the sensor irradiation means that irradiates sensor signals such as ultrasonic waves instead of the imaging means.

[0025] The activity level determination system for cultured fish according to one aspect of the present invention is characterized in that the feeding amount of the cultured fish is determined based on the determination result obtained by this activity level determination system. According to this configuration, fully automatic and efficient feeding becomes possible.

Effects of the Invention

[0026] According to the activity level determination system for cultured fish of the present invention, even in a variety of situations where the imaging environment such as the place to be determined is diverse, the activity level of the cultured fish with respect to feeding can be determined with high accuracy. Further, according to the automatic feeding system of the present invention, by efficiently feeding from the feeder based on the result determined by this activity level determination system, it is possible to reduce the cost of feeding by avoiding overfeeding and to avoid water quality pollution.

Brief Description of the Drawings

[0027]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Mode for Carrying Out the Invention

[0028] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited by this embodiment. For example, in the following embodiment, an example of use in a marine fish cage is used for explanation, but the present invention is not limited to marine fish cages, and can also be applied to, for example, land fish cages, aquariums for land farming, display aquariums in aquariums and zoos, and household aquariums.

[0029] <1. Configuration of the Activity Determination System> The activity determination system 1 for cultured fish according to this embodiment is an activity determination system premised on determining the activity of cultured fish in a fish cage with respect to feeding. As shown in FIG. 1, this activity determination system 1 includes an imaging means 2 and an activity discrimination means 3.

[0030] The imaging means 2 is an imaging device that images the activity state of cultured fish with respect to feeding in a fish cage. As the imaging means 2, any camera can be used, for example, a UVC camera that can obtain visible images.

[0031] In this embodiment, a UVC camera is used as the imaging means 2, but it may be a digital camera, an industrial camera, a network camera, a smartphone (mobile phone), a tablet terminal, etc. Further, the imaging means 2 may be a near-infrared camera that can obtain near-infrared images.

[0032] The imaging means 2 is installed on the aquaculture raft and is disposed at a position where it can image the water surface where bubbles are generated when the cultured fish are fed. In the present embodiment, as an example, the imaging means 2 is installed at a position where it can image the water surface from an obliquely upper direction. Further, the imaging means 2 can be suspended from the aquaculture raft via a cable and imaged from underwater the activity status of the cultured fish with respect to feeding, etc., and it is considered applicable as long as it is at a position where the activity state of the cultured fish with respect to feeding can be grasped.

[0033] The imaging means 2 preferably can adjust zoom in, zoom out, and the angle, and can change the position of the imaging area. Further, the number of the imaging means 2 is not limited to one, and a plurality of units may be used.

[0034] Then, using the UVC camera image which is the imaging means 2, imaging and determination are repeated at regular intervals for a certain period of time, the level determination of the activity degree of the cultured fish with respect to feeding is performed, and one final activity degree level determination of the cultured fish with respect to feeding is performed according to the majority principle of each determination result.

[0035] In the present embodiment, as an example, three images are acquired and determined at 15 - second intervals. And when the same determination result appears two or more times among the three determinations using the three images, it is output as the final determination result. In the case where all three determination results are different, it is determined as non - determinable.

[0036] Then, this series of determinations (from the three determinations to the calculation of the final determination result) is repeated a plurality of times, for example, within 5 minutes, and the activity degree is determined at any time. And based on the determination results performed at any time, the feeding amount is varied. Note that this time and interval are not limited to 15 seconds within 5 minutes, and can also be set, for example, to perform determination every 1 minute within 30 minutes.

[0037] Also, as shown in FIG. 1, the imaging means 2, the feeding device 4, and the tablet terminal 5 are connected to the activity degree discrimination means 3. And using this tablet terminal 5, the setting and operation of the activity degree determination means 3, as well as the setting and operation of the imaging means 2 and the feeding device 4, are possible.

[0038] These may be connected to the activity level determination means 3 by wired or wireless connection. The tablet terminal 5 is an example, and the operating terminal may be a PC or a remote terminal using an internet connection, etc. Instead of these terminals, the activity level determination means 3 may be directly equipped with a keyboard and a display.

[0039] The activity level determination means 3 is configured with a CPU, memory, SSD, etc. (not shown), and determines the activity level and controls the imaging means 2. The activity level determination means 3 also has means A to means E for determining the activity level. These means A to means E are also executed by the CPU, memory, SSD, etc.

[0040] The means A is a means for dividing a reference image and one still image into a plurality of blocks.

[0041] Means B is a means for calculating the difference between the average shading values of corresponding blocks between the reference image and one still image. By inputting the difference between the average shading values calculated by means B into a neural network, it is possible to accurately determine the activity of farmed fish in response to feeding.

[0042] Means C is a means for calculating a high-order discrimination curve for nonlinear discrimination using the difference in shading between each block as an input value of a neural network, as difference information in shading between the reference image and the one still image.

[0043] Means D is a means for discriminating the level of a change in shading by using a high-order discrimination curve with a plurality of predetermined thresholds.

[0044] The activity level of farmed fish can be determined with high accuracy by using the methods C and D. Furthermore, by using the above methods, the activity level of farmed fish can be determined without being affected by the environment of the farm, allowing for strict and flexible determination.

[0045] Further, means E is a means for executing means A to D for each of a plurality of still images at a predetermined time interval within a predetermined time. By having means E for executing each of the plurality of still images for each still image, the activity level of the cultured fish can be accurately discriminated.

[0046] Also, in this embodiment, the above determination result is calculated by being classified into activity level levels from 0 to 5, and the result is displayed on the tablet terminal 5. Level 0 is the state where there is no activity of the cultured fish and the activity level is the lowest, and level 5 is the state where the activity level of the cultured fish is the highest. Then, according to this activity level, a preset feeding amount is calculated.

[0047] <2. Operation of the Activity Determination System> Next, with reference to the flowchart of FIG. 2, the operation of the activity determination system 1 will be described in detail.

[0048] First, when the activity determination system 1 is activated, an activation signal interrupts the activity discrimination means 3, and the system starts. Then, the imaging means 2 disposed at a position on the culture raft where the water surface can be imaged images the water surface (activity level 0 level) in a state where no nabla has occurred before the cultured fish move, and the activity discrimination means 3 receives and stores the image (step S1). This image becomes the reference image.

[0049] Next, a predetermined time interval for imaging and determination is specified (step S2). Note that this time interval may be set in advance. FIG. 3 shows an example of the main menu screen displayed on the tablet terminal 5 connected to the activity discrimination means 3. In this embodiment, as an example, the time interval Δt = 15 seconds is set for three determinations, and an example is shown where imaging and determination are performed three times, the activity level 0 is determined three times, and the final determination is displayed as the activity level 0.

[0050] Next, when feeding starts, while extracting still images at specified time intervals (Δt = 15 seconds in this embodiment) from the video captured by the imaging means 2, the activity level is determined using each still image. Specifically, the activity level determination means 3 extracts the first still image 1 and the reference image, and performs the first motion analysis using a neural network with the first still image 1 and the reference image (step S3).

[0051] This analysis determines the appetite activity level based on the shading change between the first still image 1 and the reference image. Specifically, the in-block addition average values of the blocks obtained by block-dividing the still image 1 and the reference image are created for the still image 1 and the reference image respectively, and the difference between them is input into the neural network to determine the level of the shading change.

[0052] In this neural network, as a set of two still images paired with the reference image according to the activity level, multiple sets of the sets of each still image and the reference image according to each activity level are pre-learned as learning data.

[0053] Figs. 4(a) to (f) are diagrams showing examples of learning data from activity level 0 to activity level 5. The even-numbered (e.g., cutted-0) file names of the image data at each activity level are the reference images, and the nabla-occurring image data with the odd-numbered (e.g., cutted-1) adjacent to the right of the reference image is the still image according to the activity level.

[0054] Therefore, in Fig. 4, it is an example of being learned with three sets of the reference image and the still image for each activity level. Note that the activity level 0 has the lowest activity level, and the activity level 5 has the highest activity level.

[0055] Although not an essential configuration, in addition to the pre-learned learning data, a configuration may be adopted in which the actually determined data is accumulated and learned. It is considered that this can further improve the determination system in that environment.

[0056] Then, by inputting the data of the shade change obtained from the reference image and the still image 1 into the neural network, a high-order separation curve for non-linear discrimination can be obtained. By comparing this high-order separation curve with the high-order separation curve based on the shade change for each activity level obtained from the learning data shown in FIG. 5, the appetite activity level is determined. Note that the determination in this embodiment displays the result in six levels of activity levels 0 to 5.

[0057] In this way, by performing activity level determination using a neural network with learning data learned as a pair of image data, i.e., a still image + reference image, corresponding to the activity level, the determination system can be significantly improved compared to the conventional one.

[0058] That is, the conventional learning data is in the form of one image data for each activity level, i.e., for activity level 3, it is this image data corresponding to it. When the activity level and one image data are used, the determination is made based on thresholds for each point, etc. In the case where there is a shaded part protruding from a part of the image, the possibility of misjudgment increases. However, in this embodiment, the possibility of such misjudgment can be significantly suppressed, and the determination system can be greatly improved. In particular, it is effective when the variation of the image is large according to the environment of the aquaculture farm as in this embodiment.

[0059] After calculating the determination result of the first still image, next, the second still image 2 and the reference image Δt (15 seconds) after the still image 1 are used to perform the second motion analysis with the neural network (step S4).

[0060] Furthermore, after that, the third still image 3 and the reference image 2Δt (30 seconds) after the still image 1 are used to perform the third motion analysis with the neural network (step S5).

[0061] If two or more of the three judgment results are the same, this is the final judgment. If all three judgment results are different, the final judgment is invalid because it is impossible to judge. Specifically, in the case of a three-judgment judgment in this embodiment, the final judgment result is only when the same judgment result is obtained in two or more of the three judgment results, which is more than half of the total. If the judgment results are inconsistent and do not exceed the majority, the judgment is invalid, and a "?", for example, is displayed in the inspection result display area (step S6).

[0062] The activity level of the final determination result is then output to a monitor, USB port, file, etc. (Step S7). Steps S1 to S7 form an infinite loop, and are executed multiple times at predetermined time intervals within a predetermined time. When the predetermined time has elapsed and feeding has ended, an end signal is sent and the system is shut down.

[0063] <3. Feeding Operation> Once the activity level of the final assessment result of three sets of activity assessments has been calculated, the activity assessment means 3 determines the amount of feed to be fed based on that result. Specifically, first, the optimal feed amount according to the activity level of the final assessment result is selected from a correlation table with feed amounts determined for each activity level, which is preset according to the type of farmed fish. This optimal feed amount data is then sent to the feeding device 4, which then feeds the farmed fish.

[0064] This process is carried out for each set of S1 to S7 of the activity determination means 3 to determine the amount of feed to be fed, and the amount of feed to be fed is determined for each set of S1 to S7 which forms an infinite loop, and the farmed fish are fed the corresponding amount of feed from the feeding means 4. Furthermore, by repeating the above process throughout the day, or for two or three days, once the feed amount settings for each time of day have been determined, feeding can be carried out based on those settings thereafter.

[0065] <4. Other embodiments> As described above, the preferred embodiments of the present invention have been described with reference to the drawings. However, it goes without saying that the present invention is not limited to the above-described embodiments, and various modifications or variations within the scope described in the claims also belong to the technical scope of the present invention. For example, although a UVC camera is used as the imaging means in this embodiment, it is also possible to use a digital camera, an industrial camera, a network camera, a smartphone, a tablet terminal, or the like.

[0066] For example, in the above embodiment, the activity level is determined using an image of the water surface where the state of the nabla can be seen by a camera. However, it is also conceivable to install the camera underwater and determine the activity level using an image of a fish school or the state of fish taken by the camera.

[0067] Also, in the above embodiment, the activity level is determined using a video or a still image captured by a camera. However, instead of the imaging means 2 that captures the activity state of the cultured fish, a sensor irradiation means that irradiates a sensor signal such as ultrasonic waves is provided. The sensor irradiation means captures a reference reflection signal before the cultured fish moves and a plurality of reflection signals at a predetermined time interval within a predetermined time. The activity level discrimination means uses the reflection signal instead of the still image and uses the reference reflection signal instead of the reference image. It is also conceivable to apply it to a configuration in which the reflection signal from the sensor irradiation means is received and the activity level of the cultured fish with respect to feeding is discriminated by comparing the reflection signal with a reference reflection signal that serves as a reference for determining the activity level of the cultured fish with respect to feeding.

[0068] According to this, it is considered that even when the imaging environment is poor, the activity level of the cultured fish with respect to feeding can be discriminated by the sensor irradiation means that irradiates a sensor signal such as ultrasonic waves instead of the imaging means. The sensors to be used are composed of, for example, non-contact sensors such as ultrasonic sensors, electrostatic sensors, and vibrators.

Explanation of Reference Numerals

[0069] 1 Activity level determination system 2 Imaging means 3 Activity level discrimination means 4 Feeding device 5 Tablet terminals

Claims

1. A system for determining activity of farmed fish, which determines activity of farmed fish in response to feeding, comprising: an imaging means for imaging the activity state of the farmed fish; and an activity determination means for receiving a signal from the imaging means and determining the activity of the farmed fish in response to feeding, receiving a signal from the imaging means, capturing a reference image that serves as a reference for determining the appetite activity of the farmed fish, and a plurality of still images taken at predetermined time intervals within a predetermined time period; The still image is compared with the reference image to determine the appetite activity of the farmed fish. A system for determining the activity level of farmed fish.

2. The activity determination means The determination of the activity level of the farmed fish in response to feeding is performed based on a change in shading between the reference image and the plurality of still images. The system for determining the activity of farmed fish according to claim 1.

3. The activity determination means a means A for dividing the reference image and one still image into a plurality of blocks; a means B for calculating a difference in shading for each corresponding block between the reference image and the one still image; a means C for calculating a difference in shading between the reference image and the one still image based on the calculated difference in shading of each block; and means D for determining a level of change in shading of the calculated difference in shading between the reference image and the one still image based on a plurality of predetermined thresholds. The system for determining the activity of farmed fish according to claim 2.

4. The means B is a means for calculating a difference between an average value of shading for each corresponding block between the reference image and the one still image. The system for determining the activity of farmed fish according to claim 3.

5. the means C is means for calculating a high-order discrimination curve for nonlinear discrimination using the difference in shading between the reference image and the one still image as input values of a neural network, and The means D is a means for determining the level of change in shading based on a plurality of predetermined thresholds of the high-order discrimination curve. The system for determining the activity of farmed fish according to claim 4.

6. The activity determination means a means E for executing the means A to D for each of the plurality of still images at predetermined time intervals within a predetermined time period; The system for determining the activity of farmed fish according to claim 3.

7. The activity determination means For the reference image and each of the still images, the activity level of the cultured fish with respect to feeding is determined by a majority vote of the discrimination results of the density change levels obtained by executing the means A to D. The cultured fish activity determination system according to claim 6.

8. The activity discrimination means is When the majority vote of the discrimination results of each of the still images performs discrimination of the activity level only when at least two or more of the discrimination results are the same, when all of the discrimination results are different, the discrimination is invalidated. The cultured fish activity determination system according to claim 6.

9. The neural network is trained using learning data in which a reference image and a still image corresponding to the activity level are paired for each activity level. The cultured fish activity determination system according to claim 5.

10. The activity discrimination means is performs discrimination between the plurality of still images and the reference image at a predetermined time interval within a predetermined time a plurality of times at different times. The cultured fish activity determination system according to claim 1.

11. The imaging means is arranged on the ground, and the still image is an image obtained by imaging the water surface. The cultured fish activity determination system according to claim 1.

12. The image obtained by the imaging means is a visible image or a near-infrared image. The cultured fish activity determination system according to claim 1.

13. Instead of the imaging means, a sensor irradiation means for irradiating a sensor signal such as ultrasonic waves is provided, the sensor irradiation means captures a reference reflection signal before the cultured fish moves and a plurality of reflection signals at a predetermined time interval within a predetermined time, the activity discrimination means uses the reflection signal instead of the still image and uses the reference reflection signal instead of the reference image, receives the reflection signal from the sensor irradiation means, and compares the reflection signal with the reference reflection signal serving as a reference for determining the activity level of the cultured fish with respect to feeding, and discriminates the activity level of the cultured fish with respect to feeding. The cultured fish activity determination system according to any one of claims 1 to 11.

14. A cultured fish feeding amount determination system that determines the feeding amount of cultured fish based on the determination result obtained by the activity determination system according to any one of claims 1 to 12.

Images