Predictive image generation device, predictive image generation method, and program

Abstract

This invention provides a predictive image generation device, a predictive image generation method, and a program that generate predictive images that clearly show the predicted state of the ocean. [Solution] A predictive image generation device for generating a predictive image showing the predicted state of the coastal area of ​​the sea, comprising: a past information creation unit that creates past information by associating wave breaking information showing the past state of wave breaking in the coastal area, weather information for that coastal area, and an image of that coastal area; a weather information acquisition unit that acquires weather information for the coastal area at the time of prediction that is the target of generating the predictive image; and a predictive image generation unit that generates a predictive image showing the predicted state of the coastal area of ​​the sea based on the wave breaking information and image in the past information corresponding to the weather information at the time of prediction acquired by the weather information acquisition unit.

Description

Technical Field

[0001] The present invention relates to a predicted image generation device, a predicted image generation method, and a program.

Background Art

[0002] In Patent Document 1, a wave prediction device is disclosed that substitutes wave GPV data into a wave prediction model formula to predict wave information at the coast of a target location. Thereby, wave information regarding waves near the coast can be predicted.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in Patent Document 1, although the predicted wave information can be output by expressing it as a numerical value or the like, for beginners of marine activities such as surfing staying particularly in coastal areas, and users of marine leisure including swimming, it is difficult to intuitively understand the state of the coastal area of the sea just by reading the numerical information.

[0005] The present invention has been made in view of such problems, and an object thereof is to provide a predicted image generation device, a predicted image generation method, and a program that generate a predicted image in which the state of the predicted coastal area of the sea is easy to understand.

Means for Solving the Problems

[0006] One aspect of the present invention is a predictive image generation device that generates a predictive image showing the predicted state of the coastal area of ​​the sea, comprising: a past information creation unit that creates past information by associating wave breaking information showing the past state of wave breaking in the coastal area, weather information for that coastal area, and an image of that coastal area; a weather information acquisition unit that acquires weather information for the coastal area at the time of prediction that is the target of generating the predictive image; and a predictive image generation unit that generates a predictive image showing the predicted state of the coastal area of ​​the sea based on the wave breaking information and image in the past information corresponding to the weather information at the time of prediction acquired by the weather information acquisition unit.

[0007] One aspect of the present invention is a predictive image generation method for generating a predictive image showing the predicted state of the coastal area of ​​the sea, comprising: a past information creation step of creating past information that associates wave breaking information showing the past state of wave breaking in the coastal area, weather information for that coastal area, and an image of that coastal area; a weather information acquisition step of acquiring weather information for the coastal area at the time of prediction that is the target of generating the predictive image; and a predictive image generation step of generating a predictive image showing the predicted state of the coastal area of ​​the sea based on the wave breaking information and image in the past information that corresponds to the weather information at the time of prediction acquired in the weather information acquisition step.

[0008] A program according to one aspect of the present invention is a program that generates a predictive image showing the predicted state of the coastal area of ​​the sea, which can be executed by a computer capable of exchanging information with an input / output device, and includes: a past information creation step of creating past information that associates wave breaking information showing the past state of wave breaking in the coastal area, weather information for that coastal area, and an image of that coastal area; a weather information acquisition step of acquiring weather information for the coastal area at the time of prediction that is the target of generating the predictive image; and a predictive image generation step of generating a predictive image showing the predicted state of the coastal area of ​​the sea based on the wave breaking information and image in the past information that corresponds to the weather information at the time of prediction acquired in the weather information acquisition step. [Effects of the Invention]

[0009] According to the present invention, it is possible to generate predictive images that clearly show the predicted conditions of coastal areas of the sea. [Brief explanation of the drawing]

[0010] [Figure 1] Figure 1 shows the configuration of the predictive image generation system. [Figure 2] Figure 2 is a block diagram showing an example of the functional configuration of the predictive image generation device 13. [Figure 3] Figure 3 shows an example of a generated predicted image. [Figure 4] Figure 4 shows an example of a generated predicted image. [Figure 5] Figure 5 shows an example of a generated predicted image. [Figure 6] Figure 6 is a flowchart showing the flow of the predictive image generation process. [Figure 7] Figure 7 shows an example of a generated predicted image. [Figure 8] Figure 8 shows the configuration of the predictive image generation device 112 in another embodiment. [Modes for carrying out the invention]

[0011] [One Embodiment] (Configuration of the predictive image generation system) Figure 1 shows the configuration of the predictive image generation system in the first embodiment. The predictive image generation system comprises a camera 11, a weather information content device 12, and a predictive image generation device 13.

[0012] Camera 11 is mounted on land and photographs the coastal scenery.

[0013] The weather information content device 12 is, for example, a server and supplies weather information to the forecast image generation device 13. The weather information content device 12 stores, for example, AMeDAS information from the Japan Meteorological Agency and information from private weather companies, and transmits it along with location information, date and time information, etc. Weather information includes, for example, wind direction, wind speed, sunshine duration, precipitation, cloud cover, and tide level. The weather information may be weather information measured at a predetermined date and time, or forecast information that predicts the weather at a predetermined date and time.

[0014] The prediction image generation device 13 is, for example, a server, and generates a prediction image showing the state of the coastal area of the sea to be predicted based on the weather information at the time of prediction acquired from the weather information content device 12 and past information. Details will be described later.

[0015] (Configuration of the prediction image generation device 13) FIG. 2 is a block diagram showing a functional configuration example of the prediction image generation device ۱۳. The prediction image generation device 13 includes a communication unit 21, a storage unit 22, and a control unit 23.

[0016] The communication unit 21 communicates with the weather information content device 12 and the like. The communication unit 21 may realize various information by, for example, short-range wireless communication such as the communication unit 21, Bluetooth (registered trademark), or Wi-Fi (registered trademark). Further, information may be exchanged via the Internet or the like by wide-area wireless communication.

[0017] The storage unit 22 has, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), and a non-volatile memory. The storage unit 22 stores a control application program of the prediction image generation device 13, various information necessary for its execution, and information generated by processing.

[0018] The control unit 23 is composed of a CPU (Central Processing Unit), a storage part (ROM (Read Only Memory), RAM (Random Access Memory), non-volatile memory, etc.), hardware, and other elements. By executing a control application program (not shown) stored in the storage unit 22, the control unit 23 controls the entire prediction image generation device 13 and functions as a past information creation unit 31, a weather information acquisition unit 32, and a prediction image generation unit 33.

[0019] The past information creation unit 31 creates past information in which the breaker information indicating the state of the breaker of the past coastal area, the weather information of the coastal area, and the image of the coastal area are linked to each other and stores it in the storage unit 22.

[0020] Wave breaking information includes, for example, information such as the number of waves, wave height, and wave breaking patterns in past coastal areas. Wave breaking information may be information automatically generated by using known image analysis techniques on images taken by camera 11, information manually generated by the user visually observing images taken by camera 11 or the sea directly, or information generated by other methods.

[0021] When wave-breaking information is created using image analysis technology, an image analysis device (not shown) may be used to create the wave-breaking information. For example, the image analysis device extracts the outline of the "soup," which is the area where water breaks from the wave crest and becomes a white foam, from the image captured by the camera 11, creates wave-breaking information by quantifying the shape of the soup outline from the soup outline, and transmits it to the predictive image generation device 13. Alternatively, the image analysis device may create wave-breaking information by analyzing the behavior of people engaging in marine activities, such as surfers, to understand the state of wave breaking. For example, the state of wave breaking can be understood based on the number of times a surfer was able to ride a wave and the duration of time they rode the wave.

[0022] The coastal weather information includes, for example, wind direction, wind speed, sunshine duration, precipitation, cloud cover, tide level, and wave information, and is obtained from the weather information content device 12. This coastal weather information may be actual weather information or weather forecast information. Location information and date / time information are associated with the weather information.

[0023] The coastal area image is the image of the coastal area targeted by the wave breaking information, and is, for example, an image taken by camera 11. For example, if the above-mentioned image analysis device creates the wave breaking information, it may be the image used to create the wave breaking information.

[0024] The linking of wave breaking information, weather information, and images may be done manually by the user, or it may be done automatically based on information such as date and time.

[0025] Past information may be stored in a location other than the predictive image generation device 13, for example, in an external device.

[0026] The weather information acquisition unit 32 acquires weather forecast information for the coastal area to be used to generate the forecast image from the weather information content device 12, for example, via the communication unit 21. The weather forecast information includes, for example, wind direction, wind speed, sunshine duration, precipitation, cloud cover, and tide level.

[0027] The prediction image generation unit 33 generates a prediction image showing the predicted coastal conditions of the sea, based on past wave breaking information and images of the coast in the past information created by the past information creation unit 31, which corresponds to the prediction weather information acquired by the weather information acquisition unit 32.

[0028] For example, the predictive image generation unit 33 outputs a predictive image by inputting weather information acquired by the weather information acquisition unit 32 into a trained model that has been trained using machine learning to generate predictive images showing the condition of coastal areas. The trained model is generated using a well-known machine learning algorithm such as a neural network. The training information to be input into the machine learning algorithm can be created by linking wave breaking information, weather information, and images of coastal areas.

[0029] As a specific example, the predictive image generation unit 33 uses machine learning to analyze the correlation between wave breaking information, weather information, and coastal area images, and calculates patterns and trends. For example, when the weather is rough, the area of ​​the "soup" in the wave breaking information tends to be larger, and the wave height tends to be higher. Then, using the calculated patterns and trends, the predictive image generation unit 33 generates a predictive image showing the coastal area during that weather condition, based on the weather information that is the target of predictive image generation.

[0030] The predictive image generation unit 33 may create a predictive image by selecting an image with high similarity to weather information from images included in past information, or it may create a predictive image by combining multiple images with high similarity to weather information.

[0031] Figures 3 and 4 show examples of generated prediction images. Figure 3 is a prediction image generated, for example, when the input weather information indicates good weather, and Figure 4 is a prediction image generated, for example, when the weather information indicates bad weather. The prediction image may include not only the sea region W showing the sea surface, but also the air region S and the land region L. For example, the air region S is created based on weather information, and the land region L is created based on location information linked to the weather information. The prediction image is then created by combining the sea region W, the air region S, and the land region L. Alternatively, as shown in Figure 5, an image consisting only of the sea region W is also possible.

[0032] The predictive image generation unit 33 uploads the generated predictive images to the internet via the communication unit 21 and provides them to surfers and other marine activity participants, as well as users of marine leisure activities.

[0033] (Predictive image generation process) Next, the predictive image generation process by the predictive image generation device 13 will be described. Figure 6 is a flowchart showing the flow of the predictive image generation process. The past information creation unit 31 creates past information in advance, linking wave breaking information showing the past wave breaking conditions of the coastal area, weather information for that coastal area, and images of that coastal area, and stores it in the storage unit 22. The process starts, for example, when a start button (not shown) is pressed.

[0034] In step S11, the weather information acquisition unit 32 acquires weather information for the coastal area to be used to generate the prediction image. For example, the user operates an operation unit (not shown) to select the target coastal area, and the weather information acquisition unit 32 acquires the weather information for that coastal area from the weather information content device 12.

[0035] In step S12, the prediction image generation unit 33 generates a prediction image showing the predicted coastal conditions of the sea, based on past wave breaking information and images of the coast in the past information created by the past information creation unit 31, which corresponds to the prediction weather information acquired by the weather information acquisition unit 32.

[0036] In step S13, the prediction image generation unit 33 outputs the generated prediction image via the communication unit 21.

[0037] The process will then be completed.

[0038] [Other examples] (Overlapping congestion) As described above, the predictive image generation unit 33 generated a predictive image that included the sea area W, the air area S, and the land area L. However, it may also generate a predictive image that includes information on the congestion status of the sea.

[0039] In this case, the past information creation unit 31 further stores congestion information based on people staying in the coastal area in the storage unit 22, linking it to the past information. Congestion information may be automatically created, for example, using known image analysis techniques, based on images taken by the camera 11. Specifically, edge information can be extracted from the image, and people can be detected based on the extracted edge information. The number of people in the image and the percentage of the area occupied by people in the image can then be obtained as congestion information. The congestion information may also be obtained by manually counting the number of people staying in the coastal area, or by evaluating the congestion level with a score based on the count, etc. Furthermore, the congestion information may also be information regarding the congestion level of surfers.

[0040] The predictive image generation unit 33 further generates a predictive image showing the coastal area including the predicted congestion, based on past congestion information of the coastal area in the past information created by the past information creation unit 31, which corresponds to the predicted weather information acquired by the weather information acquisition unit 32.

[0041] For example, the predictive image generation unit 33 outputs a predictive image by inputting weather information acquired by the weather information acquisition unit 32 into a trained model that has been trained using machine learning to generate a predictive image showing the condition of the coastal area, including congestion. The trained model is generated using a known machine learning algorithm such as a neural network. The training information to be input into the machine learning algorithm can be created by linking congestion information with weather information, etc.

[0042] As a specific example, the predictive image generation unit 33 uses machine learning to analyze the correlation between congestion information and weather information, and calculates patterns and trends. For example, when the weather is rough, there is a tendency for congestion to be low. Using the calculated patterns and trends, the predictive image generation unit 33 predicts congestion based on the weather information that is the target of predictive image generation. Then, as shown in Figure 7, for example, it superimposes images of people that represent the congestion onto the sea area W and land area L of the predictive image.

[0043] (Text overlay) In the above description, the predictive image generation unit 33 generated a predictive image including the sea area W, the air area S, and the land area L, but it may also generate a predictive image that includes further predicted numerical information. Figure 8 shows the configuration of the predictive image generation device 112 in another embodiment. Components having the same function as the predictive image generation device 112 shown in Figure 1 are given the same reference numerals and their explanations are omitted as appropriate.

[0044] The control unit 123 of the predictive image generation device 112 further includes a text creation unit 131.

[0045] The text creation unit 131 creates text describing the predicted coastal conditions of the sea based on past wave breaking information in past data created by the past data creation unit 31, which corresponds to the predicted weather information acquired by the weather information acquisition unit 32. The text describing the coastal conditions includes, for example, text describing the wave breaking conditions and numerical information for each element of the wave breaking (wave height, wave number, etc.).

[0046] For example, the text generation unit 131 outputs text describing the predicted coastal conditions by inputting weather information acquired by the weather information acquisition unit 32 into a trained model that has undergone machine learning to generate text. The trained model is generated using a known machine learning algorithm such as a neural network. The training information to be input into the machine learning algorithm can be created by linking wave breaking information and weather information.

[0047] As a specific example, the predictive numerical information generation unit uses machine learning to analyze the correlation between text-based wave breaking information and weather information, and calculates patterns and trends. For example, when the weather is rough, the area of ​​the "soup" in the wave breaking information tends to increase, and the wave height tends to increase. Based on this trend, the text generation unit 131 predicts the wave height.

[0048] The predictive image generation unit 33 overlays the text created by the text creation unit 131 onto the predictive image, for example, the lower right portion.

[0049] Furthermore, the text generation unit 131 may factorize the predicted image generated by the predicted image generation unit 33, converting each element into a numerical value, and superimposing these numerical values ​​onto the predicted image. Alternatively, an easy-to-understand explanatory text may be created based on the numerically converted elements.

[0050] Furthermore, the text generation unit 131 may create an explanatory text describing the predicted congestion level, or it may calculate the predicted number of visitors. The prediction image generation unit 33 then superimposes the explanatory text describing the congestion level and the number of visitors onto the prediction image, for example, in the lower right corner.

[0051] (movie) The aforementioned predictive image generation devices 13 and 112 may generate not only still images but also videos as predictive images. For example, the movement of breaking waves in the ocean region W and the movement of clouds in the sky region S of the predictive image may be represented as videos.

[0052] [Supplementary explanation of the embodiment]

[0053] The embodiments described above are all preferred examples of the present invention. The numerical values, components, arrangement positions and connection configurations of the components, and processing order in the flowcharts shown in the following embodiments are examples only and are not intended to limit the present invention. Furthermore, the figures are not necessarily strictly illustrative.

[0054] The series of processes described above can be executed by hardware or by software. When the series of processes are executed by software, the programs that make up the software are installed from a program storage medium onto a computer that is built into dedicated hardware, or onto a general-purpose personal computer, for example, which can perform various functions by installing various programs.

[0055] The programs executed by the computer may be programs that are processed chronologically in the order described herein, or they may be programs that are processed in parallel or at necessary times, such as when a call is made. [Note] The contents described in some of the embodiments above can be understood, for example, as follows:

[0056] (1) Generation of predictive images showing the state of the sea A predictive image generation device 13 (Figure 2) generates a predictive image showing the predicted state of the coastal area of ​​the sea, A historical information creation unit 31 creates historical information by associating wave breaking information showing past wave breaking conditions along the coast, weather information for that coast, and images of that coast with each other. A weather information acquisition unit 32 acquires weather information for coastal areas that are the target of prediction image generation, A prediction image generation unit 33 generates prediction images (Figures 3-5) showing the predicted coastal conditions of the sea, based on wave breaking information and images from past data corresponding to the predicted weather information acquired by the weather information acquisition unit 32. It is equipped with.

[0057] Because of this configuration, predictive images allow people, for example, visitors to the sea, to intuitively understand the condition of the ocean. For example, it can provide predictive information that is easy to understand for beginners in marine activities such as surfing, as well as for users of marine leisure activities including swimming. For example, surfers can understand the condition of the ocean at a surf spot the next day and know in advance whether the waves are suitable for surfing or not. Furthermore, the predictive image generation device 13 creates predictive images based on both past coastal wave breaking information and images, enabling it to generate more accurate predictive images. For example, the wave breaking information includes not only information obtained from images captured by the camera 11, but also information obtained from the user's visual observation of the sea, allowing it to generate predictive images that better match human perception.

[0058] (2) How people gather, number of people The historical information creation unit 31 further associates the congestion information of the coastal area with the historical information. The predictive image generation unit 33 can further generate a predictive image (Figure 7) showing the condition of the coastal area, including the predicted congestion, based on past information on coastal congestion that corresponds to the predicted weather information.

[0059] With this configuration, if someone visiting the sea, for example, views the predictive images, they can understand the condition of the coastal area, including the level of congestion.

[0060] (3) Presentation of text A text generation unit 131 generates text that describes the predicted conditions of the coastal area of ​​the sea based on past wave breaking information for coastal areas, corresponding to the predicted weather information. Furthermore, The predictive image generation unit 33 can superimpose the text generated by the text creation unit onto an image showing the predicted state of the sea.

[0061] With this configuration, if someone visiting the sea, for example, views the predictive images, they can understand the condition of the coastal area not only through images but also through text. [Explanation of Symbols]

[0062] 11 Cameras 12 Weather Information Content Device 13 Predictive Image Generation Device 21 Communications Department 22 Memory section 23 Control Unit 31. Historical Information Creation Department 32 Weather Information Acquisition Department 33 Predictive Image Generation Unit 112 Predictive Image Generation Device 123 Control Unit 131 Text Creation Section

Claims

1. A predictive image generation device that generates predictive images showing the predicted state of coastal areas of the sea, A historical information creation unit creates historical information by associating wave breaking information showing the past wave breaking conditions of the coastal area, weather information for the coastal area, and images of the coastal area with each other. A weather information acquisition unit that acquires the weather information for the coastal area at the time of prediction, which is the target of generating the prediction image, A prediction image generation unit generates a prediction image showing the predicted state of the coastal area of ​​the sea, based on the wave breaking information and the image in the past information corresponding to the weather information at the time of prediction acquired by the weather information acquisition unit, A predictive image generation device characterized by comprising the following features.

2. A predictive image generation apparatus according to claim 1, The historical information creation unit further associates the aforementioned congestion information for the coastal area with the aforementioned historical information. The predictive image generation unit further generates a predictive image showing the coastal area including the predicted congestion, based on the past congestion information of the coastal area corresponding to the weather information acquired by the weather information acquisition unit at the time of prediction. A predictive image generation device characterized by the following features.

3. A predictive image generation apparatus according to claim 1 or 2, A text creation unit creates text representing the predicted conditions of the coastal area of ​​the sea, based on the wave breaking information of the coastal area in the past information, which corresponds to the weather information at the time of prediction acquired by the weather information acquisition unit. Furthermore, The predictive image generation unit superimposes the text generated by the text creation unit onto the predictive image showing the predicted state of the sea. A predictive image generation device characterized by the following features.

4. A method for generating predictive images that show the predicted state of coastal areas of the sea, A past information creation step involves creating past information by associating wave breaking information showing the past wave breaking conditions of the coastal area, weather information for the coastal area, and images of the coastal area with each other. A weather information acquisition step involves acquiring the weather information for the coastal area at the time of prediction, which is the target of generating the prediction image. A prediction image generation step generates a prediction image showing the predicted state of the coastal area of ​​the sea, based on the wave breaking information and the image in the past information that corresponds to the weather information at the time of prediction acquired in the weather information acquisition step, A method for generating predictive images, characterized by including the following:

5. In a program that generates predictive images showing the predicted state of coastal areas of the sea, which can be executed by an input / output device and a computer capable of exchanging information, A past information creation step involves creating past information by associating wave breaking information showing the past wave breaking conditions of the coastal area, weather information for the coastal area, and images of the coastal area with each other. A weather information acquisition step involves acquiring the weather information for the coastal area at the time of prediction, which is the target of generating the prediction image. A prediction image generation step generates a prediction image showing the predicted state of the coastal area of ​​the sea, based on the wave breaking information and the image in the past information that corresponds to the weather information at the time of prediction acquired in the weather information acquisition step, A program that causes a computer to perform a process, characterized by including [a certain element].

Images