Computer program, information processing method, and information processing device

Abstract

Provided is a computer program which enables accurate recognition of microorganisms.　The computer program causes a computer to execute processes of: acquiring an image in which a plurality of microorganisms are captured; applying, to the acquired image, a plurality of methods according to kinds or sizes of the microorganisms; and recognizing the microorganisms in the image.

Description

Computer program, information processing method, and information processing device

[0001] The present invention relates to a computer program, an information processing method, and an information processing device.

[0002] Wastewater treatment systems using the activated sludge process are widely used to treat wastewater discharged from food factories, chemical factories, etc., and public sewage. In wastewater treatment systems using the activated sludge process, wastewater to be treated is introduced into a treatment tank and subjected to an aeration process in which oxygen is supplied to the various types of aerobic microorganisms (activated sludge) present in the treatment tank. The organic pollutants contained in the wastewater in the treatment tank are decomposed by the action of the aerobic microorganisms, thereby purifying the wastewater.

[0003] The operational status of a wastewater treatment system is significantly affected by the state of microorganisms in the wastewater. Therefore, technologies for understanding the state of microorganisms have been proposed. For example, Patent Document 1 discloses a wastewater treatment operational status evaluation device that analyzes input images of microorganisms living in a water treatment tank and identifies which classification category the microorganisms belong to.

[0004] Patent No. 7172302

[0005] However, the technology described in Patent Document 1 has a problem in that the accuracy of recognizing microorganisms is insufficient. Such a problem can exist not only in wastewater treatment systems but also in any field that handles microorganisms.

[0006] An object of the present disclosure is to provide a computer program or the like that can accurately recognize microorganisms.

[0007] A computer program according to one aspect of the present disclosure acquires an image of a plurality of microorganisms, and applies a plurality of techniques to the acquired image according to the type or size of the microorganisms, causing the computer to perform a process of recognizing the microorganisms in the image.

[0008] An information processing method according to one aspect of the present disclosure involves a computer acquiring an image of a plurality of microorganisms, applying a plurality of techniques to the acquired image according to the type or size of the microorganisms, and performing a process of recognizing the microorganisms in the image.

[0009] An information processing device according to one aspect of the present disclosure includes a control unit that acquires an image of a plurality of microorganisms, and performs a process of recognizing the microorganisms in the image by applying a plurality of techniques to the acquired image according to the type or size of the microorganisms.

[0010] According to the present disclosure, microorganisms can be recognized with high accuracy.

[0011] FIG. 1 is a schematic diagram of a management system according to the present embodiment. FIG. 2 is a block diagram showing the configuration of an information processing device. FIG. 3 is a block diagram showing the configuration of a terminal device. FIG. 4 is a conceptual diagram showing a method for estimating the health state of activated sludge using a first method. FIG. 5 is a conceptual diagram showing a method for estimating the health state of activated sludge using a second method. FIG. 6 is a conceptual diagram showing a method for estimating the health state of activated sludge using a third method. FIG. 7 is a conceptual diagram showing a method for estimating the health state of activated sludge using a fourth method. FIG. 8 is a flowchart showing an example of a processing procedure executed by an information processing device. FIG. 9 is a schematic diagram showing an example of a display screen showing estimation results and proposed information.

[0012] The present disclosure will be specifically described with reference to the drawings showing embodiments thereof.

[0013] 1 is a schematic diagram of a management system 100 according to this embodiment. In this embodiment, an object to be managed by the management system 100 is a wastewater treatment system using an activated sludge method, and a case where microorganisms in wastewater are recognized will be described as an example.

[0014] The management system 100 includes an information processing device 1 as a main device. The information processing device 1 is communicably connected to a terminal device 2 and a plurality of wastewater treatment systems 3 via a network N such as the Internet. The wastewater treatment systems 3 are systems for performing wastewater treatment by an activated sludge method. The number of terminal devices 2 may be two or more. The number of wastewater treatment systems 3 may be one or three or more. The information processing device 1 may be integrated into any of the terminal devices 2 or wastewater treatment systems 3.

[0015] The information processing device 1 is a device capable of various information processing and sending and receiving information, and is, for example, a server computer, a personal computer, a quantum computer, etc. The information processing device 1 is managed, for example, by a service provider of a management service provided by the management system 100. The information processing device 1 acquires sludge images showing the state of activated sludge in each wastewater treatment system 3, and estimates the state of activated sludge in the wastewater treatment system 3 based on the acquired sludge images. The information processing device 1 provides various information including the estimated state of activated sludge to a terminal device 2. The information processing device 1 may be communicatively connected to a plurality of terminal devices 2 corresponding to a plurality of users who use the management service.

[0016] The terminal device 2 is an information processing terminal used by a user of the management service, and is, for example, a personal computer, a smartphone, a tablet terminal, etc. The user of the management service includes, for example, an organization such as a wastewater treatment business operator that owns the wastewater treatment system 3, an operator belonging to the organization, etc. The terminal device 2 may be a local computer provided in a factory that includes the wastewater treatment system 3.

[0017] The wastewater treatment system 3 includes a treatment tank 31 and a settling tank 32 as a wastewater treatment device that treats wastewater using activated sludge, and a control device 33 that controls the operation of the treatment tank 31 and the settling tank 32. The activated sludge used in the wastewater treatment system 3 is a complex microbial system and includes microorganisms such as bacteria, protozoa, and metazoa. The wastewater treated by the wastewater treatment system 3 contains organic pollutants. The wastewater is, for example, industrial wastewater, public sewage, etc.

[0018] Wastewater (water to be treated) flows into the treatment tank 31 through a water supply line 30a. A constant amount of the wastewater may be continuously supplied to the treatment tank 31.

[0019] The treatment tank 31 is equipped with an aeration means (not shown), and aerobically treats the water to be treated using activated sludge present in the treatment tank 31. The treatment tank 31 aerates the activated sludge stored therein by diffusing gases such as air and oxygen supplied by a blower (not shown). Aerobic microorganisms in the activated sludge take up dissolved oxygen from the water and oxidize and decompose organic pollutants in the water that they have taken in as food. The water to be treated that has been treated in the treatment tank 31 is sent to the settling tank 32 through the water supply line 30b.

[0020] The settling tank 32 allows the water to be treated, delivered from the treatment tank 31, to stand for a predetermined time, allowing the activated sludge to settle naturally. This separates the water into a supernatant and activated sludge. The supernatant is discharged as treated water via the water supply line 30c. A portion of the settled activated sludge is returned to the treatment tank 31 via line 30d and recycled. The remainder of the activated sludge is extracted from the system as excess sludge via line 30e. The excess sludge is dehydrated, dried, etc., and then incinerated or recycled as a soil conditioner, civil engineering or construction material, etc. Note that the settling tank 32 is not limited to a settling tank type and may be, for example, a membrane separation device.

[0021] The treatment tank 31 is provided with a camera 34 that captures images of the state of the activated sludge in the tank. The camera 34 is an example of an imaging device. The camera 34 may be a high-magnification camera that can capture images of various microorganisms contained in the activated sludge at high magnification. The camera 34 generates a sludge image that represents the state of the activated sludge. The sludge image is a still image or a moving image composed of a plurality of still images. The camera 34 is connected to the control device 33 by wire or wirelessly, and outputs the captured sludge image to the control device 33. The camera 34 may be provided in the settling tank 32. Multiple cameras 34 may be provided.

[0022] The wastewater treatment system 3 is also provided with a tank 35 for storing a nutrient solution containing microorganisms as added bacteria. The nutrient solution stored in the tank 35 is continuously added in predetermined amounts to the treatment tank 31, for example, using a pump (not shown). The form of the nutrient solution is not limited to a formulation, and may be, for example, a liquid or powder. The nutrient solution may be added in the settling tank 32 or the water supply lines 30a, 30b, or line 30d. The nutrient solution may also be added manually.

[0023] The wastewater treatment system 3 may be provided with various detectors (not shown) for detecting the temperature, flow rate, pH, DO (Dissolved Oxygen) concentration, MLSS (Mixed Liquor Suspended Solids), TOC (Total Organic Carbon), metal element concentration, etc. within the system or related to the wastewater. The detectors output the detected values ​​to the control device 33.

[0024] The configuration of the wastewater treatment system 3 is an example and is not limited thereto. The wastewater treatment system 3 may include, for example, a pretreatment tank and a pre-treatment settling tank provided before the treatment tank 31. The treatment tank 31 may be configured with multiple stages.

[0025] The control device 33 is a computer and includes a control unit, a storage unit, a communication unit, etc., which are not shown. The control device 33 can send and receive information to and from the information processing device 1 via the communication unit. The control device 33 generates control signals for controlling the operation of the treatment tank 31 and the settling tank 32, and sends them to the treatment tank 31 and the settling tank 32. The control device 33 also sends sludge images received from the camera 34 to the information processing device 1. The control device 33 may be communicatively connected to the information processing device 1 via a relay device (not shown), such as a router or a gateway.

[0026] The quality and amount of wastewater in the wastewater treatment system 3 vary greatly depending on the season and the day, and the types and amounts of organic pollutants contained in the wastewater also change daily. Therefore, the ecosystem of the microorganisms that make up the activated sludge also changes depending on the condition of the wastewater it comes into contact with.

[0027] The wastewater treatment performance of the wastewater treatment system 3 depends heavily on the activated sludge. The quality and volume of wastewater fluctuate significantly depending on the season and the day. Because activated sludge is an aggregate of many types of microorganisms, changes in the quality of the wastewater it comes into contact with also cause changes in the ecosystem of the microorganisms that make up the activated sludge. In order to stabilize and optimize the operation of the wastewater treatment system 3, it is considered important to accurately grasp the state of the activated sludge, including the state of the microorganisms, and in particular the health state of the activated sludge. However, it is not easy to grasp the health state of activated sludge, which is an aggregate of many types of microorganisms. The management system 100 of this embodiment supports the operation management of the wastewater treatment system 3 by estimating the health state of the activated sludge in the wastewater treatment system 3 and presenting it to the user.

[0028] 2 is a block diagram showing the configuration of the information processing device 1. The information processing device 1 is a computer and includes a control unit 11, a storage unit 12, and a communication unit 13. The information processing device 1 may be a multi-computer consisting of multiple computers, or may be a virtual machine virtually constructed by software.

[0029] The control unit 11 includes one or more arithmetic processing devices such as a central processing unit (CPU) or a graphics processing unit (GPU). The control unit 11 controls each component and executes processing using built-in memories such as a read-only memory (ROM) or a random access memory (RAM), a clock, a counter, etc. The functions of the information processing device 1 may be realized by software, or some or all of them may be realized by hardware such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

[0030] The storage unit 12 includes a nonvolatile memory such as a hard disk, a flash memory, or an SSD (Solid State Drive). The storage unit 12 may be an external storage device connected to the information processing device 1. The storage unit 12 stores various computer programs and data referenced by the control unit 11. The storage unit 12 stores a program 1P for causing a computer to execute processing related to estimation of the health state of activated sludge, and estimation data 121 required for executing the program 1P. The estimation data 121 includes, for example, sludge images received from each wastewater treatment system 3, detection models and formulas for performing the microbial state analysis described below, a proposed information table, various analysis results, and the like.

[0031] A computer program (computer program product) including program 1P may be provided by a non-transitory recording medium 1A on which the computer program is readably recorded. The storage unit 12 stores the computer program read from the recording medium 1A by a reading device (not shown). The recording medium 1A may be, for example, a magnetic disk, an optical disk, or a semiconductor memory. Alternatively, the computer program may be downloaded from an external server connected to a communications network and stored in the storage unit 12. Program 1P may be a single computer program or may be composed of multiple computer programs, and may be executed on a single computer or on multiple computers interconnected by a communications network.

[0032] The communication unit 13 includes a communication device that realizes communication via the network N. The control unit 11 transmits and receives data between the terminal device 2 and the control device 33 via the communication unit 13.

[0033] The configuration of the information processing device 1 is not limited to the above example, and may include, for example, an operation unit for accepting user operations, a display unit for displaying images, and the like.

[0034] 3 is a block diagram showing the configuration of the terminal device 2. The terminal device 2 is a computer, and includes a control unit 21, a storage unit 22, a communication unit 23, a display unit 24, and an operation unit 25.

[0035] The control unit 21 includes one or more arithmetic processing units such as a CPU, an MPU, a GPU, etc. The control unit 21 controls each component and executes processing using built-in memories such as a ROM or a RAM, a clock, a counter, etc.

[0036] The storage unit 22 includes a non-volatile memory such as a hard disk, a flash memory, or an SSD. The storage unit 22 stores various computer programs and data referenced by the control unit 21. The storage unit 22 stores a computer program for causing a computer to execute processing related to receiving the estimation result of the health state of activated sludge.

[0037] The communication unit 23 includes a communication device that realizes communication via the network N. The control unit 21 transmits and receives data to and from the information processing device 1 via the communication unit 23.

[0038] The display unit 24 includes a display device such as a liquid crystal display, an organic EL (Electro Luminescence) display, etc. The display unit 24 displays the estimation results received from the information processing device 1 in accordance with instructions from the control unit 21.

[0039] The operation unit 25 is an interface that accepts user operations. The operation unit 25 includes, for example, a keyboard, a mouse, a touch panel device with a built-in display, a speaker, a microphone, etc. The operation unit 25 accepts operation input from the user and sends a control signal according to the operation content to the control unit 21.

[0040] The information processing device 1 of this embodiment automatically executes a process of recognizing the state of microorganisms and estimating the health state of activated sludge based on sludge images acquired from the wastewater treatment system 3. The health state of activated sludge is estimated by extracting various microorganisms from the sludge image through image analysis of the sludge image and recognizing the state of the extracted microorganisms, such as the number, size, and type.

[0041] The state of microorganisms is analyzed by classifying the microorganisms constituting activated sludge into multiple groups and using different algorithms for each group. The classification of microorganisms is based on one or both of the type and size of the microorganisms. In this specification, the microorganisms to be recognized include not only microorganisms themselves, such as bacteria and protozoa, but also aggregates of microorganisms, such as flocs.

[0042] In this embodiment, four types of methods, Method 1 to Method 4, are used. Method 1 is a method for analyzing the state of protozoa having a size equal to or greater than a predetermined value. Method 2 is a method for analyzing the state of protozoa having a size less than a predetermined value. Method 3 is a method for analyzing the state of bacteria, particularly filamentous bacteria. Method 4 is a method for analyzing the state of flocs. Methods 1 to 4 will be described in detail below.

[0043] 4 is a conceptual diagram showing a method for estimating the health state of activated sludge using the first method. The information processing device 1 applies the first method to analyze the state of protozoa having a size equal to or larger than a predetermined value. Protozoa that are the subject of the first method are, for example, protozoa having a size equal to or larger than 100 μm. Hereinafter, protozoa having a size equal to or larger than a predetermined value will also be referred to as large-sized protozoa. Examples of large-sized protozoa include Colpoda, Paramecium, Rotifer, and Archera.

[0044] In the first method, a detection model 1M is used to detect (recognize) large-sized protozoan object regions in a sludge image. The detection model 1M is a learning model generated by machine learning. The detection model 1M is expected to be used as a program module that constitutes part of artificial intelligence software.

[0045] The detection model 1M receives a sludge image as input and outputs information indicating an object region in the sludge image. The detection model 1M is, for example, Mask R-CNN (Regions with Convolutional Neural Networks). The detection model 1M uses image recognition technology using instance segmentation to recognize, on a pixel-by-pixel basis, whether each pixel in the input image corresponds to an object region. When multiple objects of the detection target class (large-sized protozoa) are captured, the detection model 1M recognizes them separately.

[0046] Mask R-CNN is a variant of Faster R-CNN, which is primarily used for object detection, and is a model that adds a deconvolution layer to Faster R-CNN. Detection model 1M is equipped with a CNN and a Region Proposal Network (RPN), and inputs the features of the input data extracted by the CNN and the coordinate range of the target object extracted by the RPN into the deconvolution layer to identify the object class and extract the region. Detection model 1M ultimately generates a mask image that masks the image region corresponding to the object in the image.

[0047] FIG. 4 shows an example of a mask image in which multiple large-sized protozoa have been detected as objects. A mask indicating the detection result is applied to the image area corresponding to each object. In the example of FIG. 4, only the boundaries of the image area corresponding to the object are indicated by thick lines. The mask image allows the type, position, and shape of the protozoa to be visually recognized. The mask image may be configured to display the type of detected protozoa in an identifiable manner, for example, by applying a mask of a different color to each object class, i.e., each type of protozoa. The detection model 1M may be configured to output the position (area) of the protozoa using a bounding box.

[0048] The detection model 1M can be generated by preparing training data in which sludge images containing objects are associated with labels indicating the position and type of each object, and then using the training data to train an untrained neural network.

[0049] Specifically, the information processing device 1 inputs sludge images included in the training data to the input layer of the detection model 1M, undergoes arithmetic processing in the intermediate layer, and acquires the position and type of the object output from the output layer. The information processing device 1 compares the position and type of the object output from the output layer with the position and type of the object included in the training data, and optimizes parameters such as the weights (coupling coefficients) between neurons so that the position and type of the object output from the output layer approach the correct values. There are no particular limitations on the method for optimizing the parameters, and the information processing device 1 optimizes various parameters using, for example, an error backpropagation method.

[0050] The detection model 1M may be a 3D-CNN that handles three-dimensional input data. In this case, the detection model 1M can input an image set including multiple consecutive frames of sludge images over a predetermined unit time, and simultaneously output images in which image regions corresponding to objects are labeled for each of the multiple frames of sludge images. With the above configuration, objects can be detected taking into account the chronologically consecutive frame images, thereby improving detection accuracy.

[0051] Upon completion of the learning, a detection model 1M is constructed that has been trained to be able to appropriately recognize objects in sludge images. The detection model 1M provides detection results for large-sized protozoa in sludge images.

[0052] The configuration of the detection model 1M is not limited to the example shown in FIG. 4. The detection model 1M only needs to be able to identify large-sized protozoa contained in a sludge image. The detection model 1M may be a model based on other learning algorithms, such as U-Net, a recurrent neural network (RNN), a support vector machine (SVM), or a decision tree. The detection model 1M may also be a model that detects large-sized protozoa contained in a sludge image without distinguishing between types of protozoa.

[0053] The information processing device 1 counts the large-sized protozoa detected by the detection model 1M and determines whether the number of large-sized protozoa is appropriate. The determination of whether the number of large-sized protozoa is appropriate is performed, for example, by determining whether the increase in the number of protozoa calculated based on a comparison between the previous detection number and the current detection number is within a predetermined allowable range of increase. If the increase in the number of protozoa is within the predetermined allowable range, it is determined that the number of protozoa is appropriate. If the increase in the number of protozoa is not within the predetermined allowable range, it is determined that the number of protozoa is not appropriate and that the number of protozoa has increased. The determination of whether the number of protozoa is appropriate may be performed by determining whether the current detection number is within a predetermined allowable range of detection numbers. The determination of whether the number of large-sized protozoa is appropriate may be performed for each type of large-sized protozoa.

[0054] When the numbers of all the large-sized protozoa are appropriate, it can be determined that the health of the activated sludge is good, i.e., the activated sludge is healthy. When the number of at least one species of large-sized protozoa is not appropriate, it can be determined that the health of the activated sludge is poor, i.e., the activated sludge is not healthy.

[0055] If the health condition of activated sludge is poor, further details of the poor condition may be determined. The detailed determination of the poor condition can be made based on the detection results for each type of protozoa. The cause of an increase in the number of protozoa varies depending on the type of protozoa. For example, an increase in colpoda is likely to be caused by an increase in the ammonia concentration in the wastewater. Therefore, if the number of colpoda is inappropriate, the health condition can be determined to be poor (increased ammonia concentration). Alternatively, an increase in paramecium is likely to be caused by a decrease in the dissolved oxygen concentration in the wastewater. Therefore, if the number of paramecium is inappropriate, the health condition can be determined to be poor (decreased dissolved oxygen concentration).

[0056] 5 is a conceptual diagram showing a method for estimating the health state of activated sludge using the second method. The information processing device 1 applies the second method to analyze the state of protozoa having a size less than a predetermined value. Protozoa that are the subject of the second method are, for example, protozoa having a size less than 100 μm. Hereinafter, protozoa having a size less than the predetermined value will also be referred to as small-sized protozoa. An example of a small-sized protozoa is the bodo.

[0057] In the second method, small-sized protozoa are detected in sludge images by detecting moving objects from multiple frames of sludge images. Specifically, as shown in FIG. 5 , the information processing device 1 generates an average image by averaging all frames using a moving average of multiple frames in chronological order, i.e., a moving image over a predetermined unit time. Moving objects are then detected from the difference between the generated average image and each frame. The detection condition for moving objects may be to detect only objects that are less than a predetermined value (e.g., a maximum length in the object region that is less than 100 μm). Note that the method for detecting moving objects is not particularly limited, and may be performed by calculating the difference between adjacent frames.

[0058] The information processing device 1 counts the detected small-sized protozoa based on the detection result of the moving object. The information processing device 1 may determine the final number of small-sized protozoa by calculating the average value, mode, median, or the like for all frames based on the number of small-sized protozoa determined for each frame.

[0059] The information processing device 1 determines whether the number of small-sized protozoa is appropriate based on the determined number of small-sized protozoa. Whether the number of small-sized protozoa is appropriate can be determined based on the number of protozoa or the increase in the number, as in the case of large-sized protozoa. Note that the details of the determination may differ depending on the method, such as determining whether the number of protozoa is appropriate based on the trend in the increase in the number of protozoa in the first method, and determining whether the number of protozoa is appropriate based on the number of protozoa in the second method.

[0060] If the number of small-sized protozoa is appropriate, the health of the activated sludge can be determined to be good. If the number of small-sized protozoa is not appropriate, the health of the activated sludge can be determined to be poor.

[0061] The increase in protozoa such as bodo is likely due to an increase in sludge load. Therefore, if the number of small-sized protozoa is not appropriate and there is an increase in small-sized protozoa, the health condition may be determined to be poor (high sludge load) in detail.

[0062] Fig. 6 is a conceptual diagram showing a method for estimating the health state of activated sludge using the third method. The information processing device 1 analyzes the state of filamentous bacteria using the third method. Note that the lower image of the binary image shown in the upper right of Fig. 6 is an enlarged view of a partial region of the upper binary image.

[0063] In the third method, filamentous bacteria are detected in a sludge image by performing a binarization process on the sludge image. Specifically, the information processing device 1 performs a binarization process on the sludge image, converting the pixel values ​​of the sludge image to 0 (black) or 1 (white). The method of the binarization process is not particularly limited, but for example, the brightness value (pixel value) of a pixel is compared with a threshold calculated by a predetermined method, and if the pixel value is equal to or greater than the threshold, the pixel value is set to 1.

[0064] The binarized image obtained by the binarization process is analyzed to identify regions with pixel values ​​equal to or greater than a threshold as blob regions. Predetermined feature quantities related to the identified blob regions are calculated, and it is determined whether each calculated feature quantity satisfies a preset threshold condition. Examples of feature quantities include area, circularity, aspect ratio, maximum height, and maximum width.

[0065] The aspect ratio is the ratio of the height to the width of a blob region. In this embodiment, the aspect ratio is the ratio of the width to the height of a blob region (width / height). In the image at the bottom of the binarized image in Figure 6, the long side of the white rectangular region represents the height, and the short side of the white rectangular region represents the width.

[0066] The circularity is expressed by the following formula, where S is the area of ​​the blob region and L is the perimeter of the blob region: Circularity=4πS / L 2

[0067] The threshold conditions for each feature amount are, for example, an area of ​​50 μm 2 or more, the circularity is 0.7 or less, the aspect ratio is more than 2.5, the maximum height is 100 μm or less, and the maximum width is 100 μm or less. It is determined whether each calculated feature quantity satisfies the above threshold conditions, and a blob region that satisfies the threshold conditions for all feature quantities is finally extracted as a filamentous bacterium. Note that if a predetermined number or more feature quantities related to a blob region satisfy the threshold conditions, the blob region may be recognized as a filamentous bacterium. Furthermore, the feature quantities and the threshold conditions for the feature quantities are not limited to the above examples, and any condition may be used as long as filamentous bacteria can be recognized from a binarized image.

[0068] The information processing device 1 counts the extracted filamentous bacteria and determines whether the number of filamentous bacteria is appropriate. Whether the number of filamentous bacteria is appropriate can be determined based on the number of filamentous bacteria or the increase in the number, as in the case of protozoa. If the number of filamentous bacteria is appropriate, it can be determined that the health state of the activated sludge is good. If the number of filamentous bacteria is not appropriate and the number of filamentous bacteria is increasing, it can be determined that the health state of the activated sludge is poor.

[0069] The increase in filamentous bacteria is likely due to a decrease in the dissolved oxygen concentration in the wastewater or a shortage of nutrient sources (e.g., nitrogen sources, phosphorus sources, etc.). Therefore, if the number of filamentous bacteria is not appropriate, the health condition may be determined in detail as poor (a decrease in the dissolved oxygen concentration or a shortage of nutrient sources).

[0070] In the determination process in the first to third methods, instead of or in addition to the number of protozoa or filamentous bacteria in the detected sludge image, it may be determined whether the area of ​​the detected protozoa or filamentous bacteria is appropriate. If the total area of ​​the regions corresponding to the detected protozoa or filamentous bacteria or the increase in the total area is within an acceptable range, it can be determined that the area of ​​the protozoa or filamentous bacteria is appropriate and the health state of the activated sludge is good. If the area of ​​the protozoa or filamentous bacteria is not appropriate, it can be determined that the number of protozoa or filamentous bacteria has increased and the health state of the activated sludge is poor.

[0071] Furthermore, in the determination process of the first to third methods, since the number and area of ​​microorganisms depend on the amount of activated sludge itself, the number and area may be calculated relatively to the amount of activated sludge in the sludge image. In this case, the sludge image is analyzed to identify the activated sludge region in the sludge image, and the area of ​​the identified activated sludge region is calculated. The appropriateness of the number and area is determined based on the calculated number of protozoa or filamentous bacteria relative to the area of ​​the activated sludge region, and the area of ​​the protozoa or filamentous bacteria relative to the area of ​​the activated sludge region.

[0072] 7 is a conceptual diagram showing a method for estimating the health state of activated sludge by the fourth method. The information processing device 1 applies the fourth method to analyze the state of flocs.

[0073] In the fourth method, flocs in a sludge image are detected by performing a binarization process on the sludge image. Specifically, the information processing device 1 binarizes the sludge image using a threshold calculated by a predetermined method, and converts the pixel values ​​of the sludge image to 0 or 1. For example, if a pixel value is equal to or greater than the threshold, the pixel value is set to 1. The threshold for the binarization process in the fourth method may be a value different from the threshold for the binarization process in the third method.

[0074] In the fourth method, the sludge image may be filtered to extract predetermined spatial frequency components, and the extracted spatial frequency components may be binarized. The filtering may be performed by, for example, a fast Fourier transform (FFT). Note that the method of binarizing the sludge image is not limited to the above example.

[0075] By analyzing the binarized image obtained by the binarization process, regions with pixel values ​​equal to or greater than a threshold are identified as blob regions corresponding to flocs. The information processing device 1 calculates predetermined feature quantities for each identified blob region and identifies the formation state of flocs based on the calculated feature quantities for each blob region. Examples of feature quantities include area, circularity, and color.

[0076] The information processing device 1, for example, calculates statistical values ​​(e.g., average, sum, median, etc.) of the calculated feature quantities for each feature quantity, and determines whether the floc size is appropriate based on the obtained statistical values. For example, if the decrease in the average area value of flocs calculated based on a comparison between the previous average area value and the current average area value is within a predetermined allowable range of decrease, it can be determined that the floc size is appropriate. If the decrease in the average area value of flocs is not within the predetermined allowable range, it can be determined that the floc size is not appropriate and the flocs have dispersed or broken down. The determination of whether the floc size is appropriate may be made by determining whether the current statistical value (e.g., average area value) is within a predetermined allowable range of statistical values. Alternatively, the information processing device 1 may determine the color of the flocs using a known color analysis method for the extracted floc region, and determine whether the determined color of the flocs is appropriate.

[0077] If the floc size is appropriate, the health of the activated sludge can be determined to be good. If the floc size is inappropriate and the flocs are dispersed or disintegrated, the health of the activated sludge can be determined to be poor.

[0078] The dispersion or disintegration of flocs is likely due to an increase in sludge load or excessive aeration. Therefore, if the floc size is not appropriate, the health condition may be determined to be poor (increased sludge load or excessive aeration).

[0079] The image analysis techniques in the first to fourth techniques described above are merely examples, and the image analysis techniques and the types and sizes of the objects to be analyzed by these techniques are not limited to the above examples. The health condition can be estimated by combining appropriate analysis techniques as long as they are capable of detecting the microorganisms to be detected from sludge images and include at least two or more techniques that correspond to the types of microorganisms to be detected.

[0080] The information processing device 1 of this embodiment further generates, in accordance with the estimation result of the health state of the activated sludge, proposal information for optimizing the operation of the wastewater treatment system 3. The proposal information includes, for example, at least one of proposals regarding the operating conditions of the wastewater treatment system 3 and proposals regarding the conditions for adding nutrients.

[0081] The information processing device 1 stores in advance in the storage unit 12 a proposed information table that associates, for example, the health state of activated sludge with proposed information, and generates proposed information using the proposed information table. The information processing device 1 refers to the proposed information table to identify proposed operating conditions and addition conditions that are associated with the details of the estimated poor health state.

[0082] For example, in the case of a problem caused by an increase in colpoda (increased ammonia concentration), dilution of the wastewater is suggested. In the case of a problem caused by an increase in paramecium (decreased dissolved oxygen concentration), an increase in the aeration flow rate is suggested. In the case of a problem caused by an increase in small-sized protozoa (increased sludge load), reduction in the treatment flow rate, an increase in MLSS, etc. are suggested. In the case of a problem caused by an increase in filamentous bacteria (decreased dissolved oxygen concentration or lack of nutrient sources), an increase in the aeration flow rate, addition of nutrient sources, etc. are suggested. In the case of a problem caused by the dispersion or disintegration of flocs (increased sludge load or over-aeration), reduction in the treatment flow rate, an increase in MLSS, an increase in the aeration flow rate, etc. are suggested.

[0083] 8 is a flowchart showing an example of a processing procedure executed by the information processing device 1. The processing in the flowchart below is executed by the control unit 11 in accordance with a program 1P stored in the storage unit 12 of the information processing device 1.

[0084] The control unit 11 of the information processing device 1 receives a sludge image of the wastewater treatment system 3 from the control device 33 (step S11) and stores the received sludge image in the memory unit 12. The sludge image is associated with identification information that identifies the control device 33 or the wastewater treatment system 3.

[0085] The control unit 11 analyzes the state of large-sized protozoa using the first method based on the acquired sludge image (step S12). In step S12, the control unit 11 inputs the sludge image to the detection model 1M and acquires a mask image in which large-sized protozoa are detected and output from the detection model 1M. The mask image is, for example, information indicating large-sized protozoa by type. Based on the mask image, the control unit 11 determines the types and numbers of large-sized protozoa in the sludge image and, based on the determined numbers of each type of large-sized protozoa, determines whether the number of large-sized protozoa is appropriate for each type.

[0086] The control unit 11 analyzes the state of small-sized protozoa based on the acquired sludge image using the second method (step S13). In step S13, the control unit 11 detects small-sized protozoa in the sludge image by detecting moving objects from multiple frames of sludge image in time series, and calculates the number of small-sized protozoa. Based on the detected number of small-sized protozoa, the control unit 11 determines whether the number of small-sized protozoa is appropriate.

[0087] The control unit 11 analyzes the state of filamentous bacteria using the third method based on the acquired sludge image (step S14). In step S14, the control unit 11 performs a binarization process on the sludge image to recognize blob regions, and identifies blob regions whose feature amounts in the recognized blob regions satisfy a preset threshold condition, thereby detecting filamentous bacteria in the sludge image and calculating the number of filamentous bacteria. Based on the detected number of filamentous bacteria, the control unit 11 determines whether the number of filamentous bacteria is appropriate.

[0088] The control unit 11 analyzes the state of flocs using the fourth method based on the acquired sludge image (step S15). In step S15, the control unit 11 performs binarization processing on the sludge image to detect blob regions representing flocs. The control unit 11 determines whether the size of the flocs is appropriate based on statistics of predetermined feature values ​​related to the detected blob regions.

[0089] The control unit 11 estimates the health state of the activated sludge based on the analysis results of the states of large-sized protozoa, small-sized protozoa, filamentous bacteria, and flocs (step S16). The health state of the activated sludge may be estimated for each group of microorganisms or for each type in each group. Specifically, the control unit 11 determines whether the health state of the activated sludge is good or bad based on whether the number of large-sized protozoa is appropriate, whether the number of small-sized protozoa is appropriate, whether the number of filamentous bacteria is appropriate, and whether the size of the flocs is appropriate. In addition to determining whether the health state is good or bad, the control unit 11 may also determine details if the health state is poor. The control unit 11 stores the analysis results of the states of various microorganisms and the health state determination results in the memory unit 12.

[0090] The control unit 11 refers to the proposed information table of the estimated data 121 and derives proposed information according to the estimated health state of the activated sludge (step S17).

[0091] The control unit 11 generates screen information showing the estimated results of the activated sludge health state and the derived proposed information (step S18). The control unit 11 transmits the generated screen information showing the estimated results and proposed information to the terminal device 2 of the user of the wastewater treatment system 3 (step S19). The control unit 11 may identify the terminal device 2 to which the screen information should be transmitted based on the identification information of the control device 33 or the wastewater treatment system 3 associated with the sludge information. The control unit 11 then terminates the series of processes.

[0092] The information processing device 1 repeatedly executes the above-described process each time a sludge image is acquired. The information processing device 1 preferably acquires sludge images at predetermined intervals and executes the above-described process in real time, thereby enabling early notification of a deterioration in health condition.

[0093] Although the above describes an example in which the information processing device 1 executes a series of processes, the entity that executes each process is not limited. Some or all of the processes in the flowchart may be executed by, for example, the terminal device 2 or the control device 33.

[0094] In the above-described process, if the control unit 11 determines that the health state of the activated sludge is poor, it may send an alert notifying that the health state is poor to the terminal device 2 or the control device 33. The estimation result and the proposal information may be sent only when it is determined that the health state of the activated sludge is poor.

[0095] 9 is a schematic diagram showing an example of a display screen 240 showing the estimation result and the proposal information. The control unit 21 of the terminal device 2 displays the display screen 240 shown in FIG. 9 on the display unit 24 based on the screen information received from the information processing device 1.

[0096] The display screen 240 includes, for example, an estimation result field 241 that displays estimation results, and a proposal information field 242 that displays proposal information. The estimation result field 241 and the proposal information field 242 display estimation results and proposal information for each type and size of microorganism. The estimation result field 241 displays the results of determining the state of microorganisms and the results of determining the health state of activated sludge. The proposal information field 242 displays various proposals for conditions according to the health state of the activated sludge.

[0097] The information processing device 1 displays the state of the microorganisms derived by each method (for example, whether or not they have increased, dispersed, or broken down) and the good or bad health condition of the activated sludge based on the state of the microorganisms in an estimation result field 241. In addition, proposed information identified based on the health condition of the activated sludge is displayed in a proposed information field 242. Note that if the health condition is good, proposed information need not be generated, or proposed information indicating no changes may be presented.

[0098] The information processing device 1 may also display an image showing the state of a microorganism on the display screen 240 for a microorganism determined to be inappropriate. In the example shown in FIG. 9 , a hyperlink 243 for accessing the image showing the state of the microorganism is displayed on the display screen 240. When the information processing device 1 determines that the state of any microorganism is inappropriate, it generates an image showing the state of the microorganism based on the information stored in the storage unit 12, and displays the generated image or a hyperlink to the image on the display screen 240. The image showing the state of the microorganism is an image that presents the state of the microorganism in a visually recognizable manner, and is, for example, an image that has been processed by superimposing a mask, frame, or the like representing the detection result of the microorganism on a sludge image. The image showing the state of the microorganism may be a mask image generated by the detection model 1M, or may be the sludge image itself.

[0099] According to this embodiment, the state of microorganisms and the health state of activated sludge can be automatically estimated based on sludge images. The state of microorganisms and the health state of activated sludge are estimated using different algorithms depending on the group to which the microorganisms belong. By applying different algorithms to each group to which the microorganisms belong to recognize microorganisms in sludge images, the accuracy of microorganism recognition is improved. Since microorganisms in an image can be detected with high accuracy, the accuracy of recognizing the number of microorganisms is improved. Furthermore, according to this embodiment, microorganisms can be observed at a fixed magnification, which eliminates the need to change the magnification, shortens observation time, and enables efficient analysis processing at low magnification.

[0100] By determining the algorithm to be applied depending on the characteristics of the microbial group to be analyzed and the detection content, it is possible to reduce the overall processing load while performing accurate analysis. In the case of protozoa, by using the first method and the second method depending on the size, it is possible to detect relatively large protozoa by distinguishing between the types of protozoa, and to detect relatively small protozoa without distinguishing between the types of protozoa. In the case of filamentous bacteria, by setting extraction conditions using feature amounts, filamentous bacteria with characteristic shapes can be efficiently detected.

[0101] For example, by quantitatively evaluating the state of microorganisms, such as the number of microorganisms and the area, the state of microorganisms can be accurately grasped. Since the health state of activated sludge is estimated based on the results of quantitative evaluation of the state of microorganisms, the estimation accuracy is improved. Automatic estimation of the health state of activated sludge can be realized without relying on the ability of the operator.

[0102] According to this embodiment, it is possible to provide proposal information according to the health state of the activated sludge, and therefore it is possible to realize optimization of the operation management of the wastewater treatment system 3.

[0103] By providing the user with a screen showing the estimated health status and suggested information, the user can clearly understand the status of the wastewater treatment system 3 through the screen.

[0104] The state of wastewater in the wastewater treatment system 3 varies daily. Furthermore, since the environment of the wastewater treatment system 3 differs depending on the facility, the state of activated sludge differs for each wastewater treatment system 3. In this system, by using sludge images representing the actual state of wastewater collected from each wastewater treatment system 3, it is possible to estimate the health state of each wastewater treatment system 3 in accordance with the state of the wastewater treatment system 3.

[0105] The computer program, information processing method, and information processing device can be applied to applications other than wastewater treatment systems. For example, the information processing method and information processing device can be applied to the cultivation of fish feed. When cultivating rotifers as feed, it is desirable to co-cultivate them with other microorganisms. However, a large number of Vorticella worms can weaken the rotifers. Therefore, it is useful to apply this method to appropriately count each microorganism and manage the cultivation conditions.

[0106] The following supplementary notes are further disclosed with respect to the above embodiments. (Supplementary Note 1) A computer program causing a computer to execute a process of acquiring an image in which a plurality of microorganisms are captured, and recognizing the microorganisms in the image by applying a plurality of techniques to the acquired image according to the type or size of the microorganisms. (Supplementary Note 2) The computer program according to Supplementary Note 1, wherein recognizing the microorganisms in the image includes recognizing the number of microorganisms in the image or the type of microorganisms in the image. (Supplementary Note 3) The computer program according to Supplementary Note 1 or Supplementary Note 2, wherein the plurality of techniques include a first technique corresponding to microorganisms of a first size and a second technique corresponding to microorganisms of a second size smaller than the first size. (Supplementary Note 4) The computer program according to any one of Supplements 1 to 3, wherein the plurality of techniques include a first technique for detecting microorganisms of a first size using an object detection model, and a second technique for detecting microorganisms of a second size smaller than the first size by detecting moving objects from the time-series images. (Supplementary Note 5) The computer program according to Supplementary Note 4, wherein the image is composed of a plurality of frames in chronological order, and the second method detects the moving object based on a difference between an average image obtained by averaging a plurality of frames for a predetermined period of time and each frame. (Supplementary Note 6) The computer program according to any one of Supplements 1 to 5, wherein the image is a sludge image obtained by capturing sludge in a wastewater treatment system using an activated sludge method. (Supplementary Note 7) The computer program according to Supplementary Note 6, wherein the plurality of methods recognize protozoa as microorganisms in the sludge image, and estimate a state of the activated sludge based on the recognized state of the protozoa in the sludge image. (Supplementary Note 8) The computer program according to Supplementary Note 7, wherein suggested information regarding operating conditions of the wastewater treatment system is derived based on the estimated state of the activated sludge. (Supplementary Note 9) The computer program according to any one of Supplements 6 to 8, wherein the plurality of methods include a third method corresponding to filamentous bacteria and a fourth method corresponding to flocs. (Supplementary Note 10) The computer program according to any one of Supplementary Note 6 to Supplementary Note 9, wherein information indicating the estimated state of the activated sludge is transmitted to a user of the wastewater treatment system.(Supplementary Note 11) The computer program according to any one of Supplementary Note 6 to Supplementary Note 10, wherein screen information is generated that displays a state of the microorganisms in association with a state of the activated sludge for each type or size of the microorganisms.

[0107] The embodiments disclosed herein are illustrative in all respects and should not be considered limiting. The technical features described in each embodiment can be combined with each other, and the scope of the present invention is intended to include all modifications within the scope of the claims and equivalents thereto. The sequences shown in each embodiment are not limited, and within the scope of no contradiction, each processing step may be executed in a different order, or multiple processes may be executed in parallel. The entity that performs each process is not limited, and within the scope of no contradiction, the process of each device may be executed by another device.

[0108] The matters described in each embodiment can be combined with each other. Furthermore, the independent claims and dependent claims described in the claims can be combined with each other in any and all combinations, regardless of the reference format. Furthermore, the claims use a format in which a claim references two or more other claims (multiple claim format), but this is not limited to this. A multiple claim (multi-multi claim) that references at least one other multiple claim may also be used.

[0109] REFERENCE SIGNS LIST 100 Management system 1 Information processing device 11 Control unit 12 Storage unit 13 Communication unit 1P Program 1A Recording medium 2 Terminal device 21 Control unit 22 Storage unit 23 Communication unit 24 Display unit 25 Operation unit 3 Wastewater treatment system

Claims

1. A computer program for causing a computer to execute a process of acquiring an image of a plurality of microorganisms, and recognizing the microorganisms in the image by applying a plurality of techniques to the acquired image according to the type or size of the microorganisms.

2. The computer program product of claim 1, wherein recognizing the microorganisms in the image includes recognizing the number of microorganisms in the image or the type of microorganisms in the image.

3. The computer program according to claim 1 or claim 2, wherein the plurality of techniques include a first technique corresponding to microorganisms of a first size and a second technique corresponding to microorganisms of a second size smaller than the first size.

4. A computer program according to any one of claims 1 to 3, wherein the plurality of techniques include a first technique for detecting microorganisms of a first size using an object detection model, and a second technique for detecting microorganisms of a second size smaller than the first size by detecting moving objects from the time series of images.

5. The computer program according to claim 4, wherein the image is composed of multiple frames in chronological order, and the second method detects the moving object based on the difference between an average image obtained by averaging multiple frames over a predetermined period of time and each frame.

6. The computer program according to any one of claims 1 to 5, wherein the image is a sludge image captured of sludge in a wastewater treatment system using an activated sludge process.

7. The computer program according to claim 6, further comprising: recognizing protozoa as microorganisms in the sludge image using the plurality of techniques; and estimating the state of the activated sludge based on the state of the recognized protozoa in the sludge image.

8. The computer program according to claim 7, further comprising the step of: deriving suggested information regarding operating conditions of the wastewater treatment system based on the estimated state of the activated sludge.

9. An information processing method in which a computer executes a process of acquiring an image of a plurality of microorganisms, and recognizing the microorganisms in the image by applying a plurality of techniques to the acquired image according to the type or size of the microorganisms.

10. An information processing device comprising a control unit that executes a process of acquiring an image in which a plurality of microorganisms are captured, and recognizing the microorganisms in the image by applying a plurality of techniques to the acquired image according to the type or size of the microorganisms.

Images