Barnacle larva detection device and agent concentration management system

Abstract

The invention provides a barnacle-type larva detection device which can easily detect barnacle-type larvae in an adhesion period and can continuously monitor the barnacle-type larvae. This barnacle larva detection device is provided with: an image data acquisition unit for acquiring image data; and a detection unit that detects barnacle-type larvae having a predetermined shape from the image data, and that detects and counts barnacle-type larvae having a predetermined shape on the basis of a learned model that is constructed using image data in which at least barnacle-type larvae having a predetermined shape are annotated as learning data.

Description

Technical Field

[0001] The invention relates to a barnacle larvae detection device and a drug concentration management system. Background Art

[0002] In thermal power plants and nuclear power plants that use seawater as cooling water, marine organisms such as barnacles sometimes adhere to the water intake channel that draws seawater from the sea and supplies it to the condenser, and the drainage channel that releases the seawater that has passed through the condenser to the sea. When the number of attached marine organisms increases, the cooling water flow channel may be blocked, resulting in failures such as reduced cooling performance. Therefore, in order to inhibit the attachment of marine organisms, chlorine-based agents such as agents for the purpose of inhibiting biological attachment have been injected into the cooling water in the past.

[0003] However, from the perspective of environmental protection and the cost of adding chemicals, it is not a good idea to add excessive amounts of chlorine-based chemicals to cooling water, and it is required to add an appropriate amount of chlorine-based chemicals. For example, if the number of attached organisms such as barnacles in seawater can be known, the amount of chlorine-based chemicals added can be appropriately adjusted according to the number.

[0004] As a method for surveying barnacles, a method of surveying the number of barnacles attached to a test plate immersed in seawater can be considered, but continuous monitoring is difficult, and there is also a problem that measures to inhibit attachment cannot be taken quickly because the survey is conducted after attachment. As a method for detecting attached juveniles of barnacles, a method using a detection kit using a monoclonal antibody specific to attached juveniles of barnacles is disclosed (see, for example, Patent Document 1).

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Publication No. 2006-246821.

[0008] The technique disclosed in Patent Document 1 can detect barnacle larvae in the settlement stage with high detection accuracy. However, since the reagent cartridge is disposable, continuous monitoring is not possible and the cost required for detection is high. Summary of the invention

[0009] Problem that the invention aims to solve

[0010] The present invention has been made in view of the above-mentioned circumstances, and an object of the present invention is to provide a barnacle larvae detection device that can easily detect barnacle larvae in the settlement stage and can perform continuous monitoring.

[0011] Solutions for solving problems

[0012] (1) The present invention is a barnacle larvae detection device comprising: an image data acquisition unit that acquires image data; and a detection unit that detects barnacle larvae having a specified shape from the image data, wherein the detection unit detects and counts the barnacle larvae having the specified shape based on a learned model, wherein the learned model is constructed using the image data that at least annotates the barnacle larvae having the specified shape as learning data.

[0013] (2) The barnacle larvae detection device according to (1), wherein the barnacle larvae having a predetermined shape are at least one of cyprids and nauplii.

[0014] (3) A drug concentration management system, comprising the barnacle larvae detection device, the drug concentration control device, and the microorganism photographing device described in (1) or (2), wherein the drug concentration control device comprises: a control unit that controls the amount of drug injected into the seawater from the drug injection unit of the seawater utilization facility, and the microorganism photographing device comprises: a sample acquisition unit that collects seawater near the seawater utilization facility; and a camera unit that photographs the seawater acquired by the sample acquisition unit, the image data acquisition unit acquires the image data photographed by the camera unit, and the control unit controls the amount of drug injected into the seawater based on the detection result of the detection unit.

[0015] Effects of the Invention

[0016] The invention provides a barnacle larvae detection device which can easily detect barnacle larvae in the attachment period and can perform continuous monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram showing the appearance and arrangement of organs of the larvae of the glandular membrane, which are the detection targets of the barnacle larvae detection device according to the present embodiment.

[0018] Figure 2 It is a block diagram showing the structure of the drug concentration management system according to the present embodiment.

[0019] Figure 3 This is a diagram showing learning data of the barnacle larvae detection device according to the present embodiment. DETAILED DESCRIPTION

[0020] <Barnacle larvae detection device>

[0021] The barnacle larvae detection device of the present embodiment includes a detection unit that detects barnacle larvae having a predetermined shape based on image data that may include barnacle larvae.

[0022] Barnacles are a general term for organisms classified into the phylum Arthropoda, class Crustacea, subclass Cirripedia, order Thoracica, including organisms belonging to the suborder Balanomorpha, such as striped barnacles, ivory barnacles, red giant barnacles, triangular barnacles, thorny giant barnacles, reticulated barnacles, oriental small barnacles, white-ridged barnacles, and densely striped barnacles.

[0023] Barnacles swim in the sea when they are larvae, and when they become settled larvae, they attach to appropriate settlements and metamorphose into adults. The following is an overview of the stages related to reproduction in the life history of barnacles. That is, after mating and fertilization between settled adults, planktonic nauplii hatch, and after repeated molting, the nauplii become settled cyprids (settled larvae), and then the cyprids attach to the substrate and metamorphose into young barnacles. This reproductive period is unique to various barnacles.

[0024] The cyprid larvae are attached to structures in the water, such as seawater intake equipment or seawater drainage equipment of power plants, coastal aquaculture equipment, and fishery equipment, and their subsequent growth may have adverse effects on these equipment. Therefore, by understanding the number of cyprid larvae and nauplii inhabiting the sea areas where these equipment are located, necessary measures such as injecting chlorine-based agents can be taken. Cyprid larvae and nauplii have characteristic prescribed shapes, so they can be image-detected using the barnacle larvae detection device 10.

[0025] Glandular larvae 1 have Figure 1 The prescribed shape shown. The glandular larvae 1 are larvae having a transparent carapace (shell) 9 that is flat and long on the left and right, and have a pair of first antennae 2 in front of the ventral surface, and six pairs of thoracic limbs 3 extending from the inside of the carapace 9 in the posterior half of the ventral surface. The first antennae 2 are organs with a suction cup-shaped tip for attachment, and the adhesive substance (quinone cross-linked binding protein) secreted from the mucus gland 4 through the mucus duct 5 is secreted on the surface of the attachment organ and adheres to the matrix. In addition, the glandular larvae repeatedly approach and detach from the matrix while investigating the adaptability to the matrix, and after temporarily attaching in a detachable manner, they are permanently attached to the fixed position finally determined. In addition, the reference numerals 6 in the figure represent oil cells, 7 represent compound eyes, and 8 represent thorax.

[0026] The barnacle larvae detection device of this embodiment is composed of, for example, Figure 2 The illustrated embodiment is a barnacle larvae detection device 10 of one structure of the drug concentration management system 100 . Figure 2 This is a block diagram showing an example of the main functions that the barnacle larvae detection device 10 may have.

[0027] The barnacle larvae detection device 10 includes an image data acquisition unit 14 that acquires image data and a detection unit 17 that detects barnacle larvae having a predetermined shape from the image data. In addition to the above, the barnacle larvae detection device 10 may include a storage unit 11, an input unit 12, an output unit 13, an image processing unit 15, a machine learning unit 16, and a communication unit 18.

[0028] The storage unit 11 is constituted by a storage device such as a hard disk drive (HDD) or a solid state drive (SSD), and stores images processed by the image processing unit 15 , learned models generated by the machine learning unit 16 , and the like.

[0029] The input unit 12 is constituted by, for example, a keyboard, a mouse, etc., and receives input from a user of the barnacle larvae detection device 10 .

[0030] The output unit 13 is composed of a display that displays an image processed by the image processing unit 15 and a detection result of the detection unit 17. The input unit 12 and the output unit 13 may be a structure that integrates display function and input function like a touch panel.

[0031] The image data acquisition unit 14 acquires image data. The above-mentioned image data is image data that may include barnacle larvae. In the present embodiment, the image data acquisition unit 14 acquires image data including barnacle larvae taken by the camera unit 32 of the microorganism photographing device 30. There is no limitation on the form of the image data acquired by the image data acquisition unit 14, and it may be a still image or a video. In addition, the image acquired by the image data acquisition unit 14 may be an image taken under a fluorescence microscope. This is because the position of the fluorescence of the barnacle larvae varies from larvae to larvae, and the type of the larvae can be identified by the position of the fluorescence.

[0032] The image processing unit 15 performs image processing on the image data acquired by the image data acquisition unit 14. Examples of image processing include size conversion, rotation, color conversion, noise removal, etc. The functions of the image data acquisition unit 14 and the image processing unit 15 are implemented by, for example, using a processor to execute a computer program stored in a memory.

[0033] The machine learning unit 16 constructs a model such as a convolutional neural network (CNN). CNN has an input layer, an intermediate layer, and an output layer. In the input layer, learning data (true value data) as a learning object is input. The intermediate layer is a layer that detects features based on the image input to the input layer. The output layer is a layer that outputs the recognition result of whether there are abnormal parts in the image based on the features extracted by the intermediate layer. CNN has a multi-layer structure and maintains multiple weight parameters. By updating the above-mentioned weight parameters from the initial value to the optimal value, the pre-learning model changes to the learned model.

[0034] The machine learning unit 16 outputs a learned model constructed using the image data as learning data, in which at least a barnacle larva having a predetermined shape is annotated for the image processed by the image processing unit 15. The output learned model may be a learned model for detecting and counting only the cyprid larvae, or a learned model for detecting and counting both the cyprid larvae and the nauplii.

[0035] Figure 3 is an example of annotated learning data. Figure 3 In the example of FIG. 1 , three types of barnacle larvae, namely, nauplii 1b, small cyprinid larvae 1a, and large cyprinid larvae 1c, are enclosed by bounding boxes and given annotations. The annotations are given according to the operation of the user of the barnacle larvae detection device 10 .

[0036] The detection unit 17 detects and counts barnacle larvae having a predetermined shape based on the image data processed by the image processing unit 15. The barnacle larvae having a predetermined shape are preferably glandular larvae, and it is more preferred that nauplii and glandular larvae can be detected and counted separately. The function of the detection unit 17 is realized based on the learned model output by the machine learning unit 16.

[0037] An example of detecting a barnacle larva by the detection unit 17 using the learned model output by the machine learning unit 16 is shown below.

[0038] (Machine Learning Condition 1)

[0039] Using the phototropism of barnacle larvae, only barnacle larvae were lured and recovered from the sample collected with a plankton net to obtain a recovered sample. Three times the amount of tap water was added to the recovered sample to stop the larvae from swimming and to cause them to settle. The settled larvae were then collected in a petri dish and enlarged photographs were taken from the bottom of the petri dish using a digital camera (manufactured by Olympus Corporation). The photographed images were analyzed as follows: Figure 3 The annotations shown in the figure are assigned to the learning data. In addition, according to the occurrence of adults at the sample collection site, the annotations are implemented as follows: among the larvae, the individuals assumed to be red giant barnacles and triangular barnacles are classified as large larvae, and the larvae assumed to be net-shaped barnacles, mud tube barnacles, and ivory barnacles are classified as small larvae. For nauplii, the species is not distinguished.

[0040] 1080 pieces of the above learning data were prepared, and a learned model was generated by machine learning. Using the generated learned model, 180 test images were used to detect barnacle larvae. The results were: small cyprid larvae AP (average accuracy) = 0.84, large cyprid larvae AP (average accuracy) = 0.39, nauplii AP (average accuracy) = 0.39, and the average AP mAP = 0.54.

[0041] (Machine Learning Condition 2)

[0042] The above learning data was rotated 90° and the images were copied to prepare 4320 learning data. The learned model was generated under the same conditions as machine learning condition 1. Using the generated learned model, 180 test images were used to detect barnacle larvae. The results were small cyprid larvae AP (average accuracy) = 0.85, large cyprid larvae AP (average accuracy) = 0.39, nauplii AP (average accuracy) = 0.36, and the average AP mAP = 0.53.

[0043] (Machine Learning Condition 3)

[0044] The learned model was generated under the same conditions as in machine learning condition 1. Using the generated learned model, the detection of cyprids was performed using 180 test images without distinguishing between large and small sizes. The average AP (mean average precision) value was mAP = 0.87, confirming that the results can be actually applied to the detection of cyprids of barnacles in images.

[0045] The communication unit 18 is a device for the barnacle larvae detection device 10 to communicate with another device via a communication means such as a communication line 41. The communication line 41 is a wired or wireless communication means, and is configured to be connectable to the Internet 40, for example.

[0046] The barnacle larvae detection device 10 focuses on the shapes of glandular larvae and nauplii having a specified shape, and detects barnacle larvae based on a learned model constructed using the image data obtained by the image data acquisition unit 14 as learning data, thereby being able to easily detect barnacle larvae and continuously monitor the presence and number of barnacle larvae.

[0047] Figure 2 The structure shown is only an example of the barnacle larvae detection device 10 and is not particularly limited to Figure 2 For example, the barnacle larvae detection device 10 may not include the storage unit 11, and the storage unit 11 may be provided in an external device that can communicate with the barnacle larvae detection device 10. Alternatively, the barnacle larvae detection device 10 may be configured not to include the communication unit 18.

[0048] <Drug concentration management system>

[0049] Next, a chemical concentration management system 100 having the barnacle detection device 10 is described. The chemical concentration management system 100 is a system for managing the concentration of chemical injection into seawater utilization facilities, such as Figure 2 As shown, there is a barnacle larvae detection device 10, a drug concentration control device 20, and a microorganism photographing device 30. The drug concentration management system 100 detects and counts barnacle larvae in the barnacle detection device 10 based on the microorganism image photographed by the microorganism photographing device 30, and manages the drug injection concentration in the drug concentration control device 20 based on the above detection and counting results, so that the drug can be injected at an effective time and with an appropriate dosage.

[0050] The reagent whose injection concentration is managed by the reagent concentration management system 100 is a reagent for the purpose of inhibiting the attachment of organisms to structures (seawater utilization facilities) in water. There is no particular limitation on such a reagent, and known reagents can be used. As the above-mentioned reagents, chlorine-based reagents such as sodium hypochlorite and chlorine dioxide, oxidants such as hydrogen peroxide, ozone, and peracetic acid can be cited. In addition to the above, organic compounds such as quaternary ammonium compounds or polyquaternary ammonium compounds, aromatic hydrocarbons, alkyl diamines, etc. as non-oxidizing reagents, metal salts as sources of copper ions, ammonium nitrates, pyrosulfites, antioxidants, catalytic enzymes, etc. can be cited. In addition, the reagent concentration management system of this embodiment can be controlled in a manner that the residual reagent concentration does not exceed a specified threshold value. Therefore, as the reagent, it is preferred that the residual concentration of the chlorine-based reagent is limited by agreements with the region from the perspective of water resource protection.

[0051] The seawater utilization facility to which the chemical concentration management system can be applied is not particularly limited, and examples thereof include power plants such as thermal power plants and nuclear power plants.

[0052] (Drug concentration control device)

[0053] For example Figure 2 As shown, the drug concentration control device 20 includes a storage unit 21 , an input unit 22 , an output unit 23 , a control unit 24 , a residual drug concentration measuring unit 25 , a drug injection unit 26 , and a communication unit 27 .

[0054] The storage unit 21, the input unit 22, and the output unit 23 can be configured similarly to the storage unit 11, the input unit 12, and the output unit 13. The storage unit 21 stores the measured value of the residual drug concentration measured by the residual drug concentration measuring unit 25 and the detection and counting results of the barnacle larvae received from the barnacle larvae detection device 10.

[0055] The control unit 24 controls the injection amount of the agent injected from the specified position of the seawater system of the seawater utilization facility through the agent injection unit 26. The control unit 24 controls the injection amount based on the detection and counting results of the barnacle larvae received from the barnacle larvae detection device 10. Specifically, the detection and counting results of the barnacle larvae received from the barnacle larvae detection device 10 can also be obtained in advance as the number of barnacle larvae detected in the specified volume of seawater. When the above-mentioned detection number exceeds a specified threshold, the injection amount of the agent can be increased compared with the normal situation. Alternatively, when the above-mentioned detection number is lower than the specified threshold, the injection amount of the agent can be reduced compared with the normal situation. In addition to the above, the control unit 24 can also control the injection amount in a manner that the residual agent concentration measured by the residual agent concentration measuring unit 25 does not exceed the specified threshold. The above-mentioned functions of the control unit 24 can be implemented manually at least in part based on the above-mentioned detection number, etc., or can be implemented by executing a program stored in the storage unit 21.

[0056] The residual reagent concentration measuring unit 25 measures the residual reagent concentration at a predetermined position (e.g., the inlet of the condenser) of the seawater system of the seawater utilization facility. The residual reagent concentration measuring unit 25 can control the amount of injected reagents by the control unit 24 so that it meets the reference value such as the environmental protection agreement value of the outlet of the seawater system of the seawater utilization facility. The structure of the residual reagent concentration measuring unit 25 is not particularly limited, and a concentration meter such as a known residual chlorine concentration meter can be used.

[0057] The reagent injector 26 injects reagents (sodium hypochlorite, etc.) from a predetermined position of the seawater system of the seawater utilization facility (for example, any position upstream of the condenser). The reagent injection means is not particularly limited, and the reagent may be injected or seawater may be electrolyzed.

[0058] The communication unit 27 is a device used by the drug concentration control device 20 to communicate with another device via a communication means such as a communication line 42. The communication line 42 is a wired or wireless communication means, and is configured to be connectable to the Internet 40, for example.

[0059] (Microbiological photography device)

[0060] The microorganism imaging device 30 acquires image data that may contain barnacle larvae based on seawater flowing through the seawater system of the seawater utilization facility or seawater near the seawater utilization facility. Figure 2 As shown, the microorganism imaging device 30 includes a sample acquisition unit 31 , an imaging unit 32 , and a communication unit 33 .

[0061] The sample acquisition unit 31 extracts seawater flowing in the seawater system of the seawater utilization facility or the seawater near the seawater utilization facility, and obtains a sample of a predetermined volume of seawater filtered by a plankton net, etc. In the sample acquisition unit 31, in order to assist the imaging of the camera unit 32, the sample can be arbitrarily diluted, separated, and the larvae of the glandular cyprids can be stopped from swimming by adding tap water.

[0062] The camera unit 32 captures the sample obtained by the sample acquisition unit 31 and acquires image data. The image data captured by the camera unit 32 is sent to the barnacle larvae detection device 10 via the communication unit 33. The camera unit 32 may capture the sample a plurality of times. The specific structure of the camera unit 32 is not particularly limited, and for example, a camera device having a function of capturing an enlarged image may be used.

[0063] The communication unit 33 is a device used by the microorganism imaging device 30 to communicate with another device via a communication means such as a communication line 43. The communication line 43 is a wired or wireless communication means, and is configured to be connectable to the Internet 40, for example.

[0064] Figure 2 The structures of the drug concentration control device 20 and the microorganism imaging device 30 shown are examples and are not limited to the above. The devices that realize the functions of the above devices do not necessarily need to be integrated, and the devices can also be independently provided in a communicable manner.

[0065] The preferred embodiments of the present invention have been described above. The present invention is not limited to the above-described embodiments, and modifications and improvements within the scope of achieving the object of the present invention are also included in the present invention.

[0066] Description of Reference Numerals

[0067] 1: Glandular larvae

[0068] 10: Barnacle larvae detection device

[0069] 14: Image data acquisition unit

[0070] 17: Inspection Department

[0071] 20: Drug concentration control device

[0072] 24: Control Department

[0073] 26: Drug injection department

[0074] 30: Microbiological photography device

[0075] 32: Camera Department

Claims

1. A barnacle larvae detection device, comprising: an image data acquisition unit that acquires image data; and a detection unit configured to detect a barnacle larva having a predetermined shape from the image data, The detection unit detects and counts the barnacle larvae having the predetermined shape based on a learned model constructed using the image data to which annotations are given at least to the barnacle larvae having the predetermined shape as learning data.

2. The barnacle larvae detection device according to claim 1, wherein: The barnacle larvae having a predetermined shape are at least one of cyprid larvae and nauplii.

3. A drug concentration management system, comprising the barnacle larvae detection device, the drug concentration control device, and the microorganism imaging device according to claim 1 or 2, wherein: The chemical concentration control device comprises: a control unit for controlling the amount of chemical injected into seawater from a chemical injection unit of a seawater utilization facility; The microorganism imaging device comprises: a sample acquisition unit for collecting seawater near the seawater utilization facility; and an imaging unit for imaging the seawater acquired by the sample acquisition unit. The image data acquisition unit acquires the image data captured by the imaging unit. The control unit controls the amount of the chemical injected into the seawater based on the detection result of the detection unit.

Images