KR102414640B1 - Apparatus and method for predicting concentration of algal - Google Patents

Abstract

The present disclosure relates to a method for an electronic device to predict the concentration of algae in a target underwater region. According to an embodiment of the present disclosure, a method for an electronic device to predict the concentration of algae in a target underwater region may include: acquiring algae information including factors related to algae generation from a measuring station of the target underwater region connected to the electronic device; training a neural network model for outputting a concentration of chlorophyll that generates algae in the target underwater region when the algae information is input, based on the acquired algae information; acquiring other tidal current information from a measuring station of the target underwater area; identifying the concentration of chlorophyll in the target underwater region by inputting other algae information obtained from the measuring station of the target underwater region to the learned neural network model; and predicting an algae occurrence probability in the target aquatic region based on the identified concentration of chlorophyll; may include

Description

A method for predicting algal concentrations and an electronic device for doing so

The present disclosure relates to a method for predicting an algal concentration and an electronic device for performing the same. More particularly, it relates to a method of determining the concentration of algae in a target underwater region using an artificial intelligence model, and an electronic device for performing the same.

As the era of the 4th industrial revolution arrives, the problem of water pollution is emerging. The characteristics of the outflow of pollutants discharged into public waters are diverse, and in order to manage the water quality of rivers and lakes, it is required to accurately understand the current water quality and to develop technologies to accurately predict the change pattern.

As a representative example of water pollution, green algae (Green Tide) is a case in which water appears blue or green due to excessive reproduction of algae called plankton. If there is an excess of blue-green algae, the risk of green algae can be very high.

When a problem such as algae occurs, there is a problem that the aquatic ecosystem is destroyed and the main drinking water source for humans is polluted. Therefore, various technologies have been developed to solve such algal blooms, but most technologies only allow removal of algal blooms after they occur, but have limitations in not fundamentally blocking the occurrence of algal blooms through advance prediction. Therefore, there is a demand for technology development for measuring the concentration of algae generating green algae and predicting a change in the concentration of algae based on the measured data.

Korean Patent No. 1384971

According to an embodiment, a method for predicting the concentration of algae in a target underwater region and an electronic device for performing the same may be provided.

According to an embodiment, a method for predicting the concentration of algae in a target underwater region using a neural network model and an electronic device for performing the same may be provided.

As a technical means for achieving the above-described technical problem, according to an embodiment, in a method for an electronic device to predict the concentration of algae in a target underwater area, the generation of algae from a measuring station of the target underwater area connected to the electronic device and obtaining bird information including related factors; training a neural network model for outputting a concentration of chlorophyll that generates algae in the target underwater region when the algae information is input, based on the acquired algae information; obtaining other tidal current information from a measuring station of the target underwater area; identifying the concentration of chlorophyll in the target underwater region by inputting other algae information obtained from the measuring station of the target underwater region to the learned neural network model; and predicting a probability of occurrence of algae in the target aquatic region based on the identified concentration of chlorophyll; may include

As a technical means for achieving the above-described technical problem, an electronic device for predicting the concentration of algae in a target underwater region includes: a network interface; display; a memory storing one or more instructions; and at least one processor executing the one or more instructions. including, wherein the at least one processor obtains tidal information including factors related to algae generation from a measuring station of the target underwater region connected to the electronic device by executing the one or more instructions, and adds to the acquired tidal information Based on the input of the algae information, a neural network model that outputs the concentration of chlorophyll that generates algae in the target underwater area is trained, other algae information is obtained from the measuring station of the target underwater area, and the learned neural network model , By inputting other algae information obtained from the measuring station of the target underwater region, the concentration of chlorophyll in the target underwater region is identified, and the probability of algae occurrence in the target underwater region can be predicted based on the identified concentration of chlorophyll.

According to another embodiment, in a method for an electronic device to predict the concentration of algae in a target underwater area, obtaining algae information including factors related to algae generation from a measuring station of the target underwater area connected to the electronic device ; training a neural network model for outputting a concentration of chlorophyll that generates algae in the target underwater region when the algae information is input, based on the acquired algae information; obtaining other tidal current information from a measuring station of the target underwater area; identifying the concentration of chlorophyll in the target underwater region by inputting other algae information obtained from the measuring station of the target underwater region to the learned neural network model; and predicting a probability of occurrence of algae in the target aquatic region based on the identified concentration of chlorophyll; A computer-readable recording medium in which a program for performing the method is stored, including the may be provided.

According to an embodiment, it is possible to effectively predict the concentration of algae.

According to an embodiment, it is possible to effectively predict a change in algae concentration using an artificial intelligence model.

1 is a diagram schematically illustrating a method for an electronic device to predict a concentration of algae in a target underwater area, according to an exemplary embodiment.
2 is a flowchart of a method of predicting, by an electronic device, a concentration of algae in a target underwater area, according to an exemplary embodiment.
3 is a diagram for explaining and comparing an output value of an artificial intelligence model used by an electronic device and actual data, according to an exemplary embodiment.
4 is a diagram for explaining and comparing an output value of an artificial intelligence model used by an electronic device and actual data according to another exemplary embodiment.
5 is a diagram for explaining and comparing an output value of an artificial intelligence model used by an electronic device and actual data according to another exemplary embodiment.
6 is a diagram for describing a process in which an electronic device trains an artificial intelligence model, according to an exemplary embodiment.
7 is a diagram for describing factors related to algae generation used by an electronic device according to an exemplary embodiment.
8 is a diagram for explaining a correlation between algae-related factors used by an electronic device and a concentration of chlorophyll, according to an embodiment.
FIG. 9 is a diagram for explaining a correlation between algae generation-related factors used by an electronic device and a concentration of chlorophyll, according to another embodiment.
10 is a diagram for explaining a scatter plot between factors related to algae generation used by an electronic device, according to an exemplary embodiment.
11 is a block diagram of an electronic device according to an embodiment.
12 is a block diagram of a server according to an embodiment.

Terms used in this specification will be briefly described, and the present disclosure will be described in detail.

The terms used in the present disclosure have been selected as currently widely used general terms as possible while considering the functions in the present disclosure, but these may vary depending on the intention or precedent of a person skilled in the art, the emergence of new technology, and the like. In addition, in a specific case, there is a term arbitrarily selected by the applicant, and in this case, the meaning will be described in detail in the description of the corresponding invention. Therefore, the terms used in the present disclosure should be defined based on the meaning of the term and the contents of the present disclosure, rather than the simple name of the term.

In the entire specification, when a part "includes" a certain element, this means that other elements may be further included, rather than excluding other elements, unless otherwise stated. In addition, terms such as "...unit" and "module" described in the specification mean a unit that processes at least one function or operation, which may be implemented as hardware or software, or a combination of hardware and software. .

Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art to which the present disclosure pertains can easily implement them. However, the present disclosure may be implemented in several different forms and is not limited to the embodiments described herein. And in order to clearly explain the present disclosure in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

1 is a diagram schematically illustrating a method for an electronic device to predict a concentration of algae in a target underwater area, according to an exemplary embodiment.

According to an embodiment, the electronic device 1000 may obtain the tide information 104 of the target underwater area from the observatory 102 installed in the predetermined target underwater area, and monitor the acquired tide information. The electronic device 1000 may predict in advance the concentration 106 of the algae in the target underwater region based on the tidal current information. The electronic device 1000 may determine the algae occurrence probability based on the predicted algae concentration. For example, the probability of occurrence of algae may mean a probability of occurrence of algae based on the concentration of algae. According to an embodiment, when the probability of occurrence of algae is greater than or equal to a predetermined threshold, the electronic device 1000 determines that there is a risk of algal bloom, alarm information related to algae occurrence, and measures to minimize damage caused by algae when algae occurs Contrast information and the like can be output.

According to an embodiment, the electronic device 1000 may include the artificial intelligence model 122 . According to an embodiment, the artificial intelligence model used by the electronic device 1000 may include a deep learning-based artificial intelligence model or an artificial neural network model. According to an embodiment, the artificial intelligence model may be an artificial intelligence model that can be learned according to an artificial intelligence learning algorithm.

According to an embodiment, the artificial intelligence model 122 may include a recurrent neural network (RNN). For example, artificial intelligence models include Convolutional Neural Network (CNN), Deep Neural Network (DNN), Recurrent Neural Network (RNN), Restricted Boltzmann Machine (RBM), Deep Belief Network (DBN), Bidirectional Recurrent Deep Neural Network (BRDNN). ) or deep Q-networks, but is also not limited thereto.

The artificial intelligence model may include at least one layer and a weight for connection strength of the layers. Artificial intelligence models can be trained by modifying and updating weights. According to an embodiment, the electronic device 1000 may acquire predetermined tide information from a plurality of observation stations at a preset period, and train an artificial intelligence model based on the acquired tide information. The artificial intelligence model 122 trained by the electronic device 1000 may output the algae concentration 106 when the algae information 104 is input. According to an embodiment, the artificial intelligence model 122 may output the algae concentration 106 as time-series data according to the passage of time.

According to an embodiment, the electronic device 1000 may use the algae concentration 106 output from the artificial intelligence model 122 as the feedback data 108 . The electronic device 1000 may retrain the artificial intelligence model by using the algae concentration output from the artificial intelligence model 122 as feedback data.

According to an embodiment, the electronic device 1000 may include a processor 110 , a memory 120 , and a network interface (not shown). According to an embodiment, the processor 110 may control the overall operation of components in the electronic device 1000 by executing one or more instructions in the memory 120 . According to an embodiment, the processor 110 may obtain tidal current information from an observatory by controlling components in the electronic device 1000 , and may predict the algae concentration after the observed time based on the acquired tidal current information.

According to an embodiment, when the predicted algae concentration is identified as greater than or equal to a predetermined threshold concentration, the electronic device 1000 determines that there is a high probability that algae will occur in the target underwater area, and provides alarm information or At least one of contrast information may be output.

According to an embodiment, the electronic device 1000 may be implemented in various forms. For example, the electronic device 1000 described herein includes a digital camera, a mobile terminal, a smart phone, a laptop computer, a tablet PC, an electronic book terminal, a digital broadcasting terminal, and a personal digital assistant (PDA). Digital Assistants), PMP (Portable Multimedia Player), navigation, MP3 player, etc. may be included, but is not limited thereto.

According to an embodiment, by interworking with the server 2000 , the electronic device 1000 may acquire tide information from a target underwater area and analyze the acquired tide information. For example, the electronic device 1000 may analyze algae information by interworking with the server 2000 and predict algae concentration based on the analyzed algae information. According to an embodiment, the server 2000 may be communicatively connected to the electronic device 1000 through a network, and may include other computing devices capable of transmitting and receiving data.

According to an embodiment, the network includes a local area network (LAN), a wide area network (WAN), a value added network (VAN), a mobile radio communication network, a satellite communication network, and It includes a combination of these and is a data communication network in a comprehensive sense that enables each network constituent entity shown in FIG. 1 to communicate smoothly with each other, and may include a wired Internet, a wireless Internet, and a mobile wireless communication network.

2 is a flowchart of a method of predicting, by an electronic device, a concentration of algae in a target underwater area, according to an exemplary embodiment.

In S210 , the electronic device 1000 may obtain tidal current information including factors related to algae generation from a measuring station of the target underwater region connected to the electronic device. According to an embodiment, algae information is chlorophyll concentration, water temperature, electrical conductivity, Ph concentration, dissolved oxygen amount, turbidity, air temperature, precipitation amount, wind speed, wind direction, humidity, vapor pressure, dew point, atmospheric pressure, sea level air pressure from the measuring station of the target underwater area , illuminance, sunlight, ground temperature, and may include at least one of an underwater temperature corresponding to a predetermined depth from the sea level.

According to an embodiment, the electronic device 1000 may acquire tidal current information including factors related to algae generation from a measuring station of a target underwater area according to a preset period. According to an embodiment, the electronic device 1000 may acquire tidal current information including factors related to algae generation from a measuring station of a target underwater area at a 15-minute cycle. Also, the tide information acquired by the electronic device 1000 or factors in the tide information may be time series data that changes with the passage of time.

According to another embodiment, the electronic device 1000 may acquire tidal current information on a plurality of target underwater regions in which a plurality of measuring stations are installed. According to an embodiment, the electronic device 1000 may acquire tide information from a plurality of measuring stations based on different periods. According to an embodiment, the period in which the electronic device 1000 acquires tidal current information from an observatory installed in a rural area without a factory is a period in which the electronic device 1000 acquires tidal current information from an observatory installed in an area with high risk of water pollution due to dense factories. can be longer The electronic device 1000 may reduce the cost of monitoring by monitoring algae information in a short period for an area with a high risk of water pollution, and may more precisely observe the area with high pollution sources in time.

In S220 , the electronic device 1000 may train a neural network model that outputs a concentration of chlorophyll that generates algae in a target underwater region when algae information is input based on the acquired algae information. According to an embodiment, the electronic device 1000 may train the neural network model by using only at least some of the factors included in the bird information. According to an embodiment, the concentration of chlorophyll may mean a concentration of chlorophyll a.

In S230, the electronic device 1000 may acquire other tidal current information from a measuring station of the target underwater area. For example, after learning of the neural network model used by the electronic device is completed, the electronic device 1000 may acquire current information from an observatory installed in the current target underwater area in real time. The data for training the neural network model and the data received from the observatory to predict the actual algae concentration may be prepared differently.

In S240 , the electronic device 1000 may identify the concentration of chlorophyll in the target underwater region by inputting other algae information obtained from a measuring station of the target underwater region to the learned neural network model. According to an embodiment, the neural network model used by the electronic device 1000 may predict a change in the concentration of chlorophyll for a preset time from a current time point. That is, according to an embodiment, the information on the chlorophyll concentration output from the neural network model may further include information on the chlorophyll change trend from the point in time when data is measured from the observation station to the point in time when a predetermined time elapses.

According to an embodiment, the electronic device 1000 may preset a time interval for identifying the concentration of chlorophyll. The electronic device 1000 may identify the concentration of chlorophyll in the target underwater region from the output value of the neural network model according to the determined time interval. According to another embodiment, the electronic device 1000 may output the chlorophyll concentration according to a predetermined time interval from the neural network model.

In S250 , the electronic device 1000 may predict the occurrence probability of algae in the target underwater region based on the identified concentration of chlorophyll. The algae occurrence probability may include a green algae occurrence probability. According to an embodiment, the electronic device 1000 may predict in advance the probability of algae occurrence not only in the present but also in the future when a predetermined time has elapsed from the current time point, based on the information on the chlorophyll concentration output from the neural network model. When the probability of occurrence of algae is greater than or equal to a predetermined threshold, the electronic device 1000 may determine that algae will occur, and may output alarm information to warn of danger or preparation information or countermeasure information to prepare for danger.

Although not shown in FIG. 2 , the electronic device 1000 compares the concentration of chlorophyll output from the trained neural network model with the actual concentration of chlorophyll in the target underwater region, and based on the comparison result, the concentration of chlorophyll output from the neural network model And, the neural network model may be retrained so that the difference in the concentration of actual chlorophyll in the target underwater region becomes smaller. In more detail, the electronic device 1000 may predict the algae concentration at a time point t2 when a predetermined time elapses from a time point t1 at which tide information is obtained from an observatory by using the neural network model. When the predetermined time elapses at the time point t2 (eg, when the time point t2 becomes the current time point), the electronic device 1000 re-acquires the actual tide information from the observatory, and includes the tide factors in the obtained tide information; Self-predicted algal concentrations can be compared. The electronic device 1000, together with the process of predicting the algae concentration in the target underwater region at the time t3, based on the algae concentration predicted at the time t2 and the algae information actually acquired at the time t2 By retraining the neural network model so that the difference in the determined algal concentration becomes small, the algae concentration in the target underwater region can be accurately predicted in real time.

Also, according to an embodiment, the electronic device 1000 may simultaneously train a plurality of neural network models based on the selected tidal current information, and may select a predetermined neural network model with high accuracy from among the plurality of learned neural network models. The electronic device 1000 selects one neural network model from among the plurality of neural network models based on the accuracy of each of the plurality of neural network models, and inputs the algae information obtained from the observatory to the selected neural network model, thereby increasing the amount of chlorophyll in the target underwater region. concentrations, or algal concentrations, can be predicted.

Also, according to an embodiment, when the predicted tide occurrence probability value is greater than or equal to a predetermined threshold, the electronic device 1000 generates alarm information for notifying the tide generation, and outputs the generated alarm information on the screen of the electronic device. can do. According to another embodiment, the electronic device 1000 may transmit the generated alarm information to a server connected to the electronic device or another electronic device.

3 is a diagram for explaining and comparing an output value of an artificial intelligence model used by an electronic device and actual data, according to an exemplary embodiment.

Referring to FIG. 3 , the comparison chart 302 between the algae concentration 304 predicted by the electronic device 1000 using a regression model and the algae data 306 in the actual target underwater area and the change of the explanatory variable A chart 314 is shown for explaining the degree of change of the dependent variable. When the electronic device 1000 uses the regression analysis model, if the amount of data change is not large, it can be seen that the current data in the actual target underwater area is tracked well, but the amount of change 306 of the data in the actual target underwater area is large. In the case (eg, when the amount of change of the actual data 306 increases and indicates a peak), it can be seen that the predicted algae concentration output from the regression analysis model does not match the actual data amount well. This is due to a limitation in which the change of the dependent variable y (response variable) is too large than the change of the explanatory variable x (explanatory variable), as shown in the chart 314 .

4 is a diagram for explaining and comparing an output value of an artificial intelligence model used by an electronic device and actual data according to another exemplary embodiment.

Referring to FIG. 4 , a comparison chart 402 of the algae concentration 406 predicted by the electronic device 1000 using the RNN model and the tidal current data 404 in the actual target underwater area and a schematic structure of the RNN model are shown. Figure 412 is shown. When the electronic device 1000 uses the RNN model, since time series data can be reflected, prediction accuracy is improved compared to the regression analysis model.

Referring to Figure 412 , in the case of the RNN model, since the information of the past includes a neural network structure that affects the current state by using tide information, which is time series data, the electronic device 1000 may use the RNN model. Algae concentration can be predicted with higher accuracy than when using the case regression model, but due to the problem of ignoring algae data from too long ago, if the change in actual algal data is large, there is still a limit to accurately predicting algae concentration have.

5 is a diagram for explaining and comparing an output value of an artificial intelligence model used by an electronic device and actual data according to another exemplary embodiment.

Referring to FIG. 5 , a comparison chart 502 between the algae concentration 504 predicted by the electronic device 1000 using the LSTM model and the algae data 506 in the actual target underwater area and a schematic structure of the LSTM model are shown. Figure 512 is shown. When the electronic device 1000 uses the LSTM model, it can be seen that the change trend of the algae concentration is tracked more accurately than when the model illustrated in FIGS. 3 to 4 is used.

According to an embodiment, the electronic device 1000 may display an improved phenomenon in response to a change in an existing variable by maintaining the explanatory power of the immediately preceding explanatory variable by using the LSTM (Long Short Term Memory) model. Prediction of algae concentrations within a target underwater region with accuracy.

6 is a diagram for describing a process in which an electronic device trains an artificial intelligence model, according to an exemplary embodiment.

In S610, the electronic device 1000 determines the correlation with the factors other than the factor related to the chlorophyll concentration among the factors in the algae information (eg, chlorophyll a) and the factors related to the chlorophyll concentration among the factors in the algae information. Based on it, at least one factor among the remaining factors may be extracted.

For example, the electronic device 1000 may obtain a chlorophyll concentration, water temperature, electrical conductivity, Ph concentration, dissolved oxygen amount, turbidity, air temperature, precipitation amount, wind speed, wind direction, humidity, vapor pressure, dew point, atmospheric pressure, and sea level air pressure from a predetermined observing station. , illuminance, sunlight, ground temperature, and factors for underwater temperature corresponding to a predetermined depth from sea level may be acquired as tide information. The electronic device 1000 may define a correlation between the chlorophyll concentration factor and the chlorophyll concentration factor, and determine the correlation between the chlorophyll concentration factor and other factors.

According to an embodiment, a correlation between factors used by the electronic device 1000 may be expressed as a Pearson correlation coefficient. According to an embodiment, the correlation between factors in the tide information determined by the electronic device 1000 is expressed as a peason correlation coefficient and may be expressed as a correlation coefficient of -1 to 1. According to an embodiment, the closer the correlation coefficient of the factors in the bird information to 1, the greater the degree of correlation of the factors.

According to an embodiment, the electronic device 1000 may determine the degree of correlation between the chlorophyll concentration factor and the chlorophyll concentration factor as a correlation score. According to an embodiment, the electronic device 1000 may determine the correlation score of the two factors as the correlation coefficient between the two factors in the tide information increases. The electronic device 1000 may extract at least one factor among a plurality of factors in the bird information based on the correlation score.

In more detail, the electronic device 1000 determines a correlation score for a chlorophyll concentration factor for each of the algal factors obtained by the electronic device from the observing station, and determines the priority of factors in the algae information based on the correlation score can The electronic device 1000 may extract at least one factor among the tidal factors acquired from the observatory based on the determined priority. According to an embodiment, the electronic device 1000 sets a chlorophyll concentration factor, a water temperature factor, a hydrogen ion concentration factor (pH), and a total phosphorus factor (eg, a concentration of phosphorus (p)) among factors acquired from an observing station based on the correlation score. , dissolved oxygen (DO) factor, total organic carbon factor, total nitrogen factor, turbidity factor and electrical conductivity factor can be extracted.

The electronic device 1000 extracts at least one factor among a plurality of factors in the algae information based on the correlation between the factors with respect to the chlorophyll concentration factor, and generates the extracted at least one factor as selective algae information. can The electronic device 1000 may train the neural network model based on predetermined selected selected bird information.

7 is a diagram for describing factors related to algae generation used by an electronic device according to an exemplary embodiment.

Referring to FIG. 7 , a plurality of tide factors obtained by the electronic device 1000 from an observatory are illustrated. According to an embodiment, the algae information obtained by the electronic device 1000 from the observatory includes chl-a (chlorophyll a concentration), wt (water temperature), tp (total phosphorus), do (dissolved oxygen amount), and total organic carbon (toc). , tn (total nitrogen), turbidity (ntu, turbidity), and ec (electric conductivity) may include at least one factor.

According to an embodiment, the electronic device 1000 selects some factors among the tide factors obtained by the electronic device from the observatory, and the data number, average, standard error, standard deviation, variance, minimum and maximum of the selected factors. value can be determined.

8 is a diagram for explaining a correlation between algae-related factors used by an electronic device and a concentration of chlorophyll, according to an embodiment.

According to an embodiment, the electronic device 1000 may acquire a greater number of factors 810 than the number illustrated in FIG. 7 from the observing station. According to an exemplary embodiment, the electronic device 1000 receives chlorophyll concentration (chl-a), water temperature, electrical conductivity, Ph concentration, dissolved oxygen amount, turbidity, air temperature (air-tmp), and precipitation from a measuring station of the target underwater area. ), wind-speed, wind direction, humidity, vapor pressure, dew-point tmp, spot atmospheric pressure, sea-level pressure, At least one of illuminance, sunlight, ground temperature, and underwater temperature corresponding to a predetermined depth from the sea level may be acquired as a tidal factor.

The electronic device 1000 may select predetermined factors from among the factors 810 acquired from the observatory, and predict the algae concentration in the target underwater region using only the selected factors.

9 is a diagram for explaining a correlation between algae generation-related factors used by an electronic device and a concentration of chlorophyll, according to another embodiment.

According to an embodiment, the electronic device 1000 may determine a degree of correlation between factors obtained from an observing station as a Pearson correlation coefficient. For example, in the electronic device 1000, water temperature (wt), hydrogen ion concentration (Ph), electrical conductivity (ec), dissolved oxygen amount (do), turbidity (ntu), total organic carbon (toc), and total nitrogen (tn) ) and the total phosphorus (tp) may determine a correlation between the factors as a correlation coefficient.

According to an embodiment, the electronic device 1000 may determine the correlation between the algal factor 906 and the algal factor 902 as well as the correlation between the algal factor 906 and the chlorophyll concentration 904 . may decide According to an embodiment, the electronic device 1000 may generate a correlation matrix by using the correlation between the algal factors 906 and the correlation between the algal factors 906 and the chlorophyll concentration.

The electronic device 1000 may set weights for each algal factor based on a correlation coefficient of the algal factors with respect to the chlorophyll concentration 904 in the correlation matrix, and apply the set weights to the algal factors. According to an embodiment, the electronic device 1000 sets a large weight to tidal factors having a high correlation coefficient and sets a low weight to tidal factors having a high correlation coefficient, thereby preferentially using factors having a high correlation coefficient. Thus, it is possible to predict the concentration of algae in the target underwater area.

10 is a diagram for explaining a scatter plot between factors related to algae generation used by an electronic device, according to an exemplary embodiment.

According to an embodiment, the electronic device 1000 may generate a scatterplot chart based on a correlation between tide factors obtained from an observatory. For example, referring to FIG. 10 , algal factors for chlorophyll concentration, total phosphorus, total nitrogen, electrical conductivity, hydrogen ion concentration, and water temperature are arranged on the horizontal axis 922 , and algal factors excluding the chlorophyll concentration on the vertical axis 924 . It can be seen that the algal factors for , total phosphorus, total nitrogen, electrical conductivity, hydrogen ion concentration, and water temperature are arranged.

In analyzing the correlation between the algal factors, the electronic device 1000 may represent the correlation between the two algal factors according to a change in the algal factors as a scatter plot. According to an embodiment, the scatterplot may be generated for each combination of two algal factors. The electronic device 1000 may extract two algal factors from among the plurality of algal factors, and may generate a scatter plot for the two extracted algal factors. The electronic device 1000 may generate a scatterplot matrix by generating scatterplots for a combination of two factors among tidal factors and combining the generated scatterplots.

The electronic device 1000 may three-dimensionally analyze the correlation within the tide factor based on the scatterplot matrix. The electronic device 1000 may determine an area ratio of a scatter plot to a background ratio for each scatter plot chart in the scatter plot matrix, and identify scatter plots in which the determined area ratio is equal to or greater than a predetermined threshold. The electronic device 1000 may extract tidal factors constituting scatter plots having an area ratio greater than or equal to a predetermined threshold among tidal factors obtained from the observatory, and train an artificial intelligence model using only the extracted tidal factors. Accordingly, the electronic device 1000 may more accurately predict a change in the concentration of algae in the target underwater region.

11 is a block diagram of an electronic device according to an embodiment.

According to an embodiment, the electronic device 1000 may include a processor 1300 , a network interface 1500 , and a memory 1700 . However, it is not limited to the above-described configuration, and the electronic device 1000 may include more components for predicting the algae concentration. According to an embodiment, the electronic device 1000 may further include a user input interface and an output unit in addition to the processor 1300 , the network interface 1500 , and the memory 1700 .

According to an embodiment, a user input interface (not shown) means a means for a user to input data for controlling the electronic device 1000 . For example, the user input interface includes a keypad, a dome switch, and a touch pad (contact capacitive method, pressure resistive film method, infrared sensing method, surface ultrasonic conduction method, integral tension measurement method). method, piezo effect method, etc.), a jog wheel, a jog switch, etc., but is not limited thereto.

The output unit (not shown) may output an audio signal, a video signal, or a vibration signal. For example, the output unit (not shown) may include a display unit, a sound output unit, and a vibration motor. For example, the display unit includes a screen for displaying and outputting information processed by the electronic device 1000 . Also, the screen may display an image. For example, the display unit (not shown) may display an algae occurrence probability, a green algae occurrence probability, a notification content for notifying when an algae is identified, and contents for contrast information, and the like.

The sound output unit (not shown) outputs audio data received from the network interface 1500 or stored in the memory 1700 . Also, the sound output unit (not shown) outputs a sound signal related to a function (eg, a call signal reception sound, a message reception sound, and a notification sound) performed by the electronic device 1000 .

The processor 1300 typically controls the overall operation of the electronic device 1000 . For example, the processor 1300 may control overall operations of the user input interface, the output unit, and the network interface 1500 by executing programs stored in the memory 1700 . Also, the processor 1300 may execute the programs stored in the memory 1700 to perform the functions of the electronic device 1000 described in FIGS. 1 to 10 .

According to an embodiment, the processor 1300 obtains tidal information including factors related to algae generation from the measuring station of the target underwater region connected to the electronic device by executing the one or more instructions, and the at least one processor obtains tidal information and , based on the obtained algae information, when the algae information is input, a neural network model that outputs the concentration of chlorophyll that generates algae in the target underwater area is trained, and other algae information is obtained from the measuring station of the target underwater area, and , to the learned neural network model, by inputting other algae information obtained from a measuring station of the target underwater area, the concentration of chlorophyll in the target underwater area is identified, and based on the identified concentration of chlorophyll, within the target underwater area The probability of algae occurrence can be predicted.

According to an embodiment, the at least one processor is predetermined from chlorophyll, air temperature, precipitation, wind speed, wind direction, humidity, vapor pressure, dew point, atmospheric pressure, sea level air pressure, illuminance, sunlight, ground temperature, and sea level from the measuring station of the target underwater area It is possible to obtain algae information including at least one of the underwater temperature corresponding to the depth of .

According to an embodiment, the at least one processor may acquire current information including factors related to the generation of the current from a measuring station of the target underwater area according to a preset period.

According to an embodiment, the at least one processor 1300 correlates with the factors other than the factors related to the chlorophyll concentration among the factors in the algae information and the factors related to the chlorophyll concentration among the factors in the algae information Based on , extract at least one factor among the remaining factors, and train the neural network model based on the selected tidal current information including the extracted at least one factor.

According to an embodiment, the at least one processor 1300 determines a correlation score for factors in the tide information based on the correlation, and prioritizes factors in the tide information based on the determined correlation score may be determined, and at least one factor among the remaining factors may be extracted based on the determined priority.

According to an embodiment, the at least one processor 1300 determines a time interval for identifying the concentration of chlorophyll, and determines the concentration of chlorophyll in the target underwater region from the output value of the neural network model according to the determined time interval. can be identified.

According to an embodiment, the at least one processor 1300 compares the concentration of chlorophyll output from the learned neural network model with the actual concentration of chlorophyll in the target underwater region, and based on the comparison result, from the neural network model. The neural network model may be retrained so that the difference between the output chlorophyll concentration and the actual concentration of chlorophyll in the target underwater region is reduced.

According to an embodiment, the at least one processor 1300 generates alarm information for notifying the occurrence of a tidal current when the predicted algae occurrence probability value is greater than or equal to a predetermined threshold, and connects the generated alarm information with the electronic device. It can be transferred to another device.

The network interface 1500 may be connected to a device in the observatory to acquire tide information at a predetermined period. Also, according to an embodiment, the network interface 1500 may transmit, to the server 2000 , algae occurrence notification information, warning information, and preparation information for algae removal in the event of algae occurrence, determined by the electronic device 1000 . According to an embodiment, the network interface 1500 may transmit information on the algae concentration predicted by the electronic device 1000 to the server 2000 .

The memory 1700 may store a program for processing and control of the processor 1300 , and may also store data input to or output from the electronic device 1000 . Also, the memory 1700 may further store an artificial intelligence model, a scatterplot matrix, and the like for the electronic device to predict the algae concentration. According to an embodiment, the artificial intelligence model used by the electronic device 1000 may include a deep neural network (DNN). For example, artificial intelligence models include Convolutional Neural Network (CNN), Deep Neural Network (DNN), Recurrent Neural Network (RNN), Restricted Boltzmann Machine (RBM), Deep Belief Network (DBN), Bidirectional Recurrent Deep Neural Network (BRDNN). ) or deep Q-networks (Deep Q-Networks), but is also not limited thereto.

Also, the memory 1700 may further store information on parameters of at least one neural network model used by the electronic device 1000 . For example, the memory 1700 may store layers in at least one neural network model, nodes, and weight values related to connection strengths of the layers. Also, the electronic device 1000 may further store training data generated by the electronic device 1000 to learn the neural network model.

The memory 1700 may include a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg, SD or XD memory), and a RAM. (RAM, Random Access Memory) SRAM (Static Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, magnetic disk , may include at least one type of storage medium among optical disks.

12 is a block diagram of a server according to an embodiment.

According to an embodiment, the server 2000 may include a processor 2300 , a network interface 2500 , and a database 2700 . According to an embodiment, the processor 2300 may predict the concentration of algae in the target underwater region in conjunction with the electronic device by executing one or more instructions stored in the database 2700 . According to an embodiment, the processor 2300 may control the overall operation of devices in the server 2000 .

The network interface 2500 may transmit tidal current information of an observatory to the electronic device by interworking with the electronic device 1000 , or may receive information on the algae concentration predicted by the electronic device 1000 from the electronic device. In addition, the network interface 2500 may receive, from the electronic device, alarm information about the occurrence of birds, information about countermeasures for mitigating the occurrence of birds, preparation information, information on action tips, and the like.

The database 2700 may correspond to the configuration of the memory 1700 of the electronic device 1000 . The database 2700 may further store information about an artificial intelligence model, a neural network model, a scatterplot matrix, a regression analysis model, etc. used by the electronic device 1000 . The server 2000 may perform at least a part of the method of the electronic device estimating the concentration of algae in the target underwater area, illustrated in FIGS. 1 to 11 , by interworking with the electronic device 1000 .

The method for the electronic device according to an embodiment of predicting the concentration of algae in a target underwater region may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software. Also, there may be provided a computer program device including a recording medium in which a program for performing a method for performing a method for performing a method for an electronic device to predict the concentration of algae in a target underwater region is stored.

Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.

Some embodiments may also be implemented in the form of a recording medium containing instructions executable by a computer, such as program modules to be executed by a computer. Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. In addition, computer-readable media may include both computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Communication media typically includes computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, or other transport mechanism, and includes any information delivery media. Also, some embodiments may be implemented as a computer program or computer program product comprising instructions executable by a computer, such as a computer program executed by a computer.

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements by those skilled in the art using the basic concept of the present invention as defined in the following claims are also provided. belong to the scope of the right.

Claims (20)

In the method of the electronic device predicting the concentration of algae in the target underwater area,
acquiring tidal current information including factors related to algae generation from a measuring station of the target underwater region connected to the electronic device;
generating a scatter plot matrix including scatter plots relating to correlation between two predetermined algae factors among the factors related to algae generation included in the acquired algae information;
determining an area ratio of a scatter plot chart to a background ratio for each of the scatter plots included in the scatter plot matrix;
identifying tide factors constituting scatter plots in which the determined area ratio is greater than or equal to a predetermined threshold;
training a neural network model that outputs a concentration of chlorophyll that generates algae in the target underwater region when the algae information is input based on the identified algal factors;
acquiring different tide information at a first time point from a measuring station of the target underwater area;
identifying the concentration of chlorophyll in the target underwater area at a second time point by inputting other algae information obtained from a measuring station of the target underwater area at the first time point into the learned neural network model; and
predicting an algae occurrence probability in the target aquatic region based on the identified concentration of chlorophyll; including,
The step of training the neural network model is
training a plurality of neural network models to output a concentration of chlorophyll that generates algae in the target underwater region when the algae information is input, based on the identified algal factors; and
selecting a predetermined neural network model with high accuracy from among the plurality of learned neural network models; including,
The step of identifying the concentration of chlorophyll in the target aquatic region comprises:
identifying the concentration of chlorophyll in the target underwater region at the second time point by inputting other algae information obtained from a measuring station of the target underwater region at the first time point into the selected predetermined neural network model; includes,
the method
acquiring actual current information obtained from a measuring station of the target underwater area at the second time point; and
re-learning the predetermined neural network model so that the difference between the concentration of chlorophyll in the target underwater region obtained from the real algae information and the neural network model is reduced; further comprising,
The step of obtaining the bird information
acquiring algae information at different intervals from a plurality of measuring stations of a target underwater area connected to the electronic device based on the risk of water pollution; including,
The step of acquiring algae information at different intervals from the plurality of measuring stations includes: acquiring algae information at a longer cycle from measuring stations of a target underwater area with a low risk of water pollution, than measuring stations of a target underwater area with a high risk of water pollution ; A method comprising According to claim 1, wherein the step of obtaining the bird information
Chlorophyll concentration, water temperature, electrical conductivity, Ph concentration, dissolved oxygen amount, turbidity, air temperature, precipitation, wind speed, wind direction, humidity, vapor pressure, dew point, atmospheric pressure, sea level air pressure, illuminance, sunlight, ground temperature, obtaining tidal current information including at least one of an underwater temperature corresponding to a predetermined depth from the sea level; A method comprising According to claim 1, wherein the step of obtaining the bird information
acquiring current information including factors related to the generation of the current from a measuring station of the target underwater area according to a preset period; including,
The factors in the tide information, characterized in that the time series data that changes with the passage of time, method. According to claim 1, wherein the step of training the neural network model
Among the factors in the algae information, based on the correlation with the factors other than the factors related to the chlorophyll concentration and the factors related to the chlorophyll concentration among the factors in the algae information, at least one factor among the remaining factors extracting; and
training the neural network model based on the selected tidal current information including the extracted at least one factor; A method comprising 5. The method of claim 4, wherein the extracting the at least one factor comprises:
determining, based on the correlation, correlation scores for factors in the bird information;
determining the priority of factors in the bird information based on the determined correlation score; and
extracting at least one factor among the remaining factors based on the determined priority; A method comprising 5. The method of claim 4, wherein the neural network model is
A Long Short Term Memory (LSTM) model, wherein the correlation is a Pearson correlation coefficient. 5. The method of claim 4, wherein the step of identifying the concentration of chlorophyll in the target aquatic region comprises:
determining a time interval for identifying the concentration of chlorophyll;
identifying the concentration of chlorophyll in the target aquatic region from the output value of the neural network model according to the determined time interval; A method comprising The method of claim 1, wherein the method
comparing the concentration of chlorophyll output from the learned neural network model with the concentration of actual chlorophyll in the target underwater region; and
re-training the neural network model so that a difference between the concentration of chlorophyll output from the neural network model and the concentration of actual chlorophyll in the target underwater region becomes smaller based on the comparison result; A method further comprising: The method of claim 1, wherein the method
generating alarm information for notifying the occurrence of algae when the predicted algae occurrence probability value is greater than or equal to a predetermined threshold; and
transmitting the generated alarm information to another device connected to the electronic device; A method comprising 5. The method of claim 4, wherein training the neural network model comprises:
training a plurality of neural network models based on the selected tidal current information including the extracted at least one factor; and
selecting one neural network model from among the plurality of neural network models based on the accuracy of the plurality of neural network models; including,
The step of identifying the concentration of chlorophyll in the target aquatic region comprises:
identifying the concentration of chlorophyll in the target aquatic region by inputting the other algae information into the selected one neural network model; A method comprising In the electronic device for predicting the concentration of algae in the target underwater area,
network interface;
display;
a memory storing one or more instructions; and
at least one processor executing the one or more instructions; including,
The at least one processor by executing the one or more instructions,
Obtaining tidal current information including factors related to algae generation from a measurement station of the target underwater area connected to the electronic device,
generating a scatter plot matrix including scatter plots relating to correlations between two predetermined algae factors among the factors related to algae generation included in the acquired algae information;
For each of the scatter plots included in the scatter plot matrix, determine an area ratio of the scatter plot to a background ratio;
Identifying tidal factors constituting the scatter plots in which the determined area ratio is greater than or equal to a predetermined threshold,
Based on the identified algal factors, when the algae information is input, a plurality of neural network models that output the concentration of chlorophyll that generate algae in the target underwater region are trained,
Selecting a predetermined neural network model with high accuracy from among the plurality of learned neural network models,
Obtaining different tide information at a first time point from the measuring station of the target underwater area,
By inputting other tidal current information obtained from a measuring station of the target underwater region at the first time point to the selected predetermined neural network model, the chlorophyll in the target underwater region at a second time point when a predetermined time has elapsed from the first time point identify the concentration,
Predicting the occurrence probability of algae in the target aquatic region based on the identified concentration of chlorophyll,
the at least one processor
Acquire the actual tide information obtained from the measuring station of the target underwater area at the second time point,
retraining the predetermined neural network model so that the difference in the concentration of chlorophyll in the target underwater region obtained from the real algae information and the neural network model becomes small,
the at least one processor
Based on the risk of water pollution, obtaining algae information at different intervals from a plurality of measuring stations in the target underwater area connected to the electronic device,
the at least one processor
An electronic device for acquiring tidal current information from measuring stations of a target underwater area having a low risk of water pollution, with a longer period than measurement stations of a target underwater area having a high risk of water pollution. 12. The method of claim 11, wherein the at least one processor comprises:
At least one of chlorophyll concentration, air temperature, precipitation, wind speed, wind direction, humidity, vapor pressure, dew point, atmospheric pressure, sea level pressure, illuminance, sunlight, surface temperature, and water temperature corresponding to a predetermined depth from the sea level from the measuring station of the target underwater area Acquiring bird information comprising a, electronic device. 12. The method of claim 11, wherein the at least one processor comprises:
According to a preset period, acquiring tidal information including factors related to the algae generation from a measuring station of the target underwater area,
The factors in the tide information are time series data that changes over time, the electronic device. 12. The method of claim 11, wherein the at least one processor comprises:
Among the factors in the algae information, based on the correlation with the factors other than the factors related to the chlorophyll concentration and the factors related to the chlorophyll concentration among the factors in the algae information, at least one factor among the remaining factors extract,
An electronic device for learning the neural network model based on the selected tidal current information including the extracted at least one factor. 15. The method of claim 14, wherein the at least one processor comprises:
Based on the correlation, determining a correlation score for the factors in the bird information,
Determine the priority of factors in the bird information based on the determined correlation score,
and extracting at least one factor among the remaining factors based on the determined priority. 15. The method of claim 14, wherein the neural network model is
A Long Short Term Memory (LSTM) model, wherein the correlation is a Pearson correlation coefficient. 15. The method of claim 14, wherein the at least one processor comprises:
determining a time interval for identifying the concentration of chlorophyll;
The electronic device identifies the concentration of chlorophyll in the target underwater region from the output value of the neural network model according to the determined time interval. 12. The method of claim 11, wherein the at least one processor comprises:
comparing the concentration of chlorophyll output from the learned neural network model with the concentration of actual chlorophyll in the target underwater region;
and retraining the neural network model so that a difference between the concentration of chlorophyll output from the neural network model and the concentration of actual chlorophyll in the target underwater region becomes smaller based on the comparison result. 12. The method of claim 11, wherein the at least one processor comprises:
When the predicted algae occurrence probability value is greater than or equal to a predetermined threshold, alarm information is generated to notify the algae occurrence,
and transmitting the generated alarm information to another device connected to the electronic device. A method for an electronic device to predict the concentration of algae in a target underwater region, the method comprising:
acquiring tidal current information including factors related to algae generation from a measuring station of the target underwater region connected to the electronic device;
generating a scatter plot matrix including scatter plots relating to correlation between two predetermined algae factors among the factors related to algae generation included in the acquired algae information;
determining an area ratio of a scatter plot chart to a background ratio for each of the scatter plots included in the scatter plot matrix;
identifying tidal factors constituting scatter plots in which the determined area ratio is greater than or equal to a predetermined threshold;
training a neural network model that outputs a concentration of chlorophyll that generates algae in the target underwater region when the algae information is input based on the identified algal factors;
acquiring different tide information at a first time point from a measuring station of the target underwater area;
identifying the concentration of chlorophyll in the target underwater area at a second time point by inputting other algae information obtained from a measuring station of the target underwater area at the first time point into the learned neural network model; and
predicting an algae occurrence probability in the target aquatic region based on the identified concentration of chlorophyll; including,
The step of training the neural network model is
training a plurality of neural network models to output a concentration of chlorophyll that generates algae in the target underwater region when the algae information is input, based on the identified algal factors; and
selecting a predetermined neural network model with high accuracy from among the plurality of learned neural network models; including,
The step of identifying the concentration of chlorophyll in the target aquatic region comprises:
identifying the concentration of chlorophyll in the target underwater region at the second time point by inputting other algae information obtained from a measuring station of the target underwater region at the first time point into the selected predetermined neural network model; includes,
the method
acquiring actual current information obtained from a measuring station of the target underwater area at the second time point; and
re-learning the predetermined neural network model so that the difference between the concentration of chlorophyll in the target underwater region obtained from the real algae information and the neural network model is reduced; further comprising,
The step of obtaining the bird information
acquiring algae information at different intervals from a plurality of measuring stations of a target underwater area connected to the electronic device based on the risk of water pollution; including,
The step of acquiring algae information at different intervals from the plurality of measuring stations includes: acquiring algae information at a longer cycle from measuring stations of a target underwater area with a low risk of water pollution, than measuring stations of a target underwater area with a high risk of water pollution ; A computer-readable recording medium in which a program for performing the method is stored, comprising a.

Images