Information processing apparatus

Abstract

To detect or predict, with a simple configuration, guerilla type heavy rain.SOLUTION: An information processing device is configured to: acquire Image data in which observation object sky is photographed; and detect or predict occurrence of guerilla type heavy rain in the observation object sky on the basis of the image data.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to an information processing apparatus.

Background Art

[0002] In recent years, the occurrence of locally and temporarily heavy rain (guerrilla heavy rain) has been increasing. Guerrilla heavy rain may be related to human life, and methods for detecting the occurrence of guerrilla heavy rain have been studied (for example, Non-Patent Document 1).

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the conventional detection of the occurrence of guerrilla heavy rain has required large-scale systems such as weather radars and observation satellites.

[0005] Therefore, an object of the present invention is to detect or predict the occurrence of guerrilla heavy rain with a simple configuration.

Means for Solving the Problems

[0006] To solve the above problems, an information processing apparatus according to the present invention includes an image data acquisition processing unit that acquires image data in which the sky of an observation target is imaged, and a detection unit that detects or predicts the occurrence of guerrilla heavy rain in the sky of the observation target based on the image data.

[0007] The information processing apparatus may further include an output processing unit that outputs a warning when the occurrence of a guerrilla heavy rain in the sky of the observation target is detected or predicted.

[0008] The detection unit may detect or predict the occurrence of a guerrilla heavy rain in the sky of the observation target by using a learned model that has learned the sky where the guerrilla heavy rain has occurred to check whether a guerrilla heavy rain is occurring in the sky shown in the image data or whether the sky shown in the image is the sky before the occurrence of the guerrilla heavy rain.

[0009] The detection unit may use a learned model that has learned the sky where the guerrilla heavy rain has occurred to check whether a guerrilla heavy rain is occurring in the sky shown in the image data when the image data satisfies a predetermined condition.

[0010] The detection unit may determine that a guerrilla heavy rain is occurring in the sky of the observation target when the image data satisfies a predetermined condition.

[0011] The predetermined condition may preferably include that the change in the brightness of the sky shown in the image data is rapid.

[0012] The predetermined condition may include that the light and dark of the sky shown in the image data are distinct.

[0013] The predetermined condition may include that the position of the cloud base of the cloud shown in the image data is below a predetermined height.

[0014] In addition, the information processing method according to the present invention is an information processing method executed by a computer, and includes an image data acquisition processing step of acquiring image data showing the sky of the observation target, and a detection step of detecting or predicting the occurrence of a guerrilla heavy rain in the sky of the observation target based on the image data.

[0015] In addition, the information processing program in the present invention causes a computer to execute the above information processing method.

[0016] In addition, the computer-readable storage medium in the present invention stores the above information processing program.

Advantages of the Invention

[0017] According to the present invention, it becomes possible to detect or predict the occurrence of a guerrilla heavy rain with a simple configuration.

Brief Description of the Drawings

[0018]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Embodiments for Carrying Out the Invention

[0019] <Information Processing Apparatus 100> FIG. 1 is a diagram showing an information processing apparatus 100 according to an embodiment of the present invention. The information processing apparatus 100 includes a control unit 110, an imaging unit 120, a storage unit 130, an output unit 140, and a communication unit 150. The information processing apparatus 100 acquires image data in which the sky to be observed is captured, and detects or predicts the occurrence of a guerrilla heavy rain in the sky to be observed based on the acquired image data.

[0020] As shown in FIG. 2(A), the information processing apparatus 100 may be a device located at a position where the air of the observation target can be imaged. Alternatively, as shown in FIG. 2(B), it may be a server apparatus that is located at a position away from the air of the observation target and communicates with a device O located at a position where the air of the observation target can be imaged. When the information processing apparatus 100 is a server apparatus that communicates with a device O located at a position where the air of the observation target can be imaged, the information processing apparatus 100 does not necessarily have the imaging unit 120 and the output unit 140.

[0021] The control unit 110 is configured by a computer, for example. The imaging unit 120 is an imaging device such as a camera, and is a device for imaging an object and generating image data in which the object is captured. The storage unit 130 is a storage device for storing information such as a hard disk and a memory. The output unit 140 is an output device for outputting information such as a display and a speaker. The communication unit 150 is a communication device for transmitting and receiving information to and from other devices.

[0022] FIG. 3 is a diagram showing the control unit 110. The control unit 110 includes an image data acquisition processing unit 111, a detection unit 112, and an output processing unit 113.

[0023] The image data acquisition processing unit 111 acquires image data in which the air of the observation target is captured at a predetermined time interval. For example, when the information processing apparatus 100 is a device located at a position where the air of the observation target can be imaged, the image data acquisition processing unit 111 uses the imaging unit 120 to acquire image data in which the air of the observation target is captured. On the other hand, for example, when the information processing apparatus 100 is a server apparatus that communicates with a device located at a position where the air of the observation target can be imaged, the image data acquisition processing unit 111 communicates using the communication unit 150, and uses an imaging device (for example, a camera) provided in the device located at a position where the air of the observation target can be imaged to acquire image data in which the air of the observation target is captured.

[0024] In this embodiment, the image data acquired by the image data acquisition processing unit 111 is image data obtained by photographing the sky of the observation target from the ground. Therefore, in this embodiment, the device (information processing device 100 or a device that communicates with the information processing device 100) located at a position where the sky of the observation target can be photographed is a device installed on the ground or a device installed on a moving body that moves on the ground.

[0025] When the image data image is acquired by the image data acquisition processing unit 111, the detection unit 112 detects or predicts the occurrence of a guerrilla heavy rain in the sky of the observation target based on the acquired image data.

[0026] As described above, in this embodiment, instead of using the observation results of a weather radar or an observation satellite, the occurrence of a guerrilla heavy rain is detected or predicted using image data photographed from the ground. Therefore, in this embodiment, it is possible to detect or predict the occurrence of a guerrilla heavy rain with a simple configuration without using a large-scale system such as a weather radar or an observation image.

[0027] When the occurrence of a guerrilla heavy rain in the sky of the observation target is detected or predicted by the detection unit 112, that is, when it is determined by the detection unit 112 that a guerrilla heavy rain is occurring or will occur in the sky of the observation target, the output processing unit 113 outputs a warning (for example, a warning that a guerrilla heavy rain is occurring or will occur). For example, when the information processing device 100 is a device located at a position where the sky of the observation target can be photographed, the output processing unit 113 outputs a warning by the output unit 140. On the other hand, for example, when the information processing device 100 is a server device that communicates with a device located at a position where the sky of the observation target can be photographed, the output processing unit 113 outputs a warning by a communication device (for example, a display or a speaker) provided in the device located at a position where the sky of the observation target can be photographed through communication using the communication unit 150.

[0028] Therefore, in this embodiment, it is possible to warn of the occurrence of a guerrilla heavy rain around the position where the sky where the guerrilla heavy rain has occurred can be photographed.

[0029] <Processing by the detection unit 112> The detection unit 112 checks, for example, whether the image data acquired by the image data acquisition processing unit 111 meets a predetermined requirement. If the image data acquired by the image data acquisition processing unit 111 meets the predetermined requirement, it is preferable to determine that a guerrilla heavy rain has occurred in the sky of the observation target.

[0030] The cumulonimbus cloud that causes guerrilla heavy rain is a very tall cloud, and sunlight cannot pass through the cloud. Therefore, the cloud base of the cumulonimbus cloud that causes guerrilla heavy rain is very dark and has a pitch-black color. Therefore, when the sky shown in the image data suddenly becomes dark, there is a possibility that cumulonimbus clouds have started to appear in the image data.

[0031] Therefore, the predetermined condition preferably includes the condition that "the change in the brightness of the sky shown in the image data is rapid". At this time, the detection unit 112 may use the pixel value of each pixel of the image data. The darker the pixel, the smaller the pixel value, and the brighter the pixel, the larger the pixel value. The detection unit 112 compares, for example, the average of the pixel values of the image data when the image data is acquired by the image data acquisition processing unit 111 with the average of the pixel values of the image data acquired before the image data, and if the change in the average of the pixel values is equal to or greater than a predetermined change, it is preferable to determine that the change in the brightness of the sky shown in the image data is rapid. The detection unit 111 may determine that the change in the average of the pixel values is equal to or greater than a predetermined change if, for example, the difference between the averages of the pixel values of the two image data is equal to or greater than a predetermined value. Further, the detection unit 111 may determine that the change in the average of the pixel values is equal to or greater than a predetermined change if, for example, the ratio of the averages of the pixel values of the two image data is equal to or greater than a predetermined value. At this time, it is preferable to use the image data in which only the sky of the observation target is shown as the image data.

[0032] The average of the pixel values of the image data may be the average of the pixel values for all the pixels of the image data, or may be the average of the pixel values for the pixels in which clouds (or the cloud base) in the image data are captured. Further, when comparing the averages of the pixel values of the image data, the detection unit 111 may convert the image data into grayscale image data when the image data is not grayscale.

[0033] Also, when cumulonimbus clouds start to form in the sky of the observation target, both the bright blue sky and the dark cloud base of the cumulonimbus clouds are captured in the image data. At this time, the pixel values of the pixels where the cloud base of the cumulonimbus clouds is captured are completely different from the pixel values of the pixels where the cloud base of the cumulonimbus clouds is not captured. The histogram of the image data in which both the bright blue sky and the dark cloud base of the cumulonimbus clouds are captured is as shown in FIG. 4. In FIG. 4, the horizontal axis represents the pixel value, and the vertical axis represents the number of pixels. In the histogram of the image data in which both the bright blue sky and the dark cloud base of the cumulonimbus clouds are captured, as shown in FIG. 4, peaks are formed at two locations, a location with a high pixel value and a location with a low pixel value. This is because all the pixel values of the pixels where the bright blue sky is captured are high, and all the pixel values of the pixels where the dark cumulonimbus clouds are captured are low.

[0034] Therefore, the predetermined condition may include the condition that "the brightness and darkness of the sky captured in the image data are clear". At this time, for example, when the image data is acquired by the image data acquisition processing unit 111, the detection unit 112 detects the peak of the histogram of the image data, calculates the difference in pixel values between adjacent peaks among the detected peaks, and if there is a difference in pixel values of a predetermined value or more among the calculated differences in pixel values, it is good to determine that the brightness and darkness of the sky captured in the image data are clear. At this time, it is good to use, as the image data, image data in which only the sky of the observation target is captured.

[0035] In addition, the cloud base of the cumulonimbus clouds that cause guerrilla heavy rain is at a relatively low altitude. Therefore, the predetermined conditions may include the condition that "the position of the cloud base of the clouds shown in the image data is at or below a predetermined height". At this time, for example, when the image data acquisition processing unit 111 acquires the image data, the detection unit 112 calculates the height from the ground surface of the cloud base of the clouds shown in the image data using, for example, a conventional image recognition technique. If the calculated height from the ground surface of the cloud base is at or below the predetermined height, it is preferable to determine that the position of the cloud base of the clouds shown in the image data is at or below the predetermined height. At this time, it is preferable to use image data in which both the sky and the ground surface of the observation target are shown as the image data.

[0036] In addition, for example, the detection unit 112 stores a learned model obtained by learning the sky where guerrilla heavy rain has occurred in the storage unit 130, and uses this learned model to determine whether the image data acquired by the image data acquisition processing unit 111 is image data showing the sky where guerrilla heavy rain has occurred, or whether the image data acquired by the image data acquisition processing unit 111 is the sky before guerrilla heavy rain occurs. If the image data acquired by the image data acquisition processing unit 111 is image data showing the sky where guerrilla heavy rain has occurred or the sky before guerrilla heavy rain occurs, it may be determined that guerrilla heavy rain is occurring or will occur in the sky of the observation target. The learned model can be obtained, for example, by learning teacher data including image data showing the sky where guerrilla heavy rain has occurred, image data showing the sky before guerrilla heavy rain occurs, and video data composed of these image data. By doing so, it becomes possible to learn the process of the occurrence of guerrilla heavy rain (changes in clouds (color and size)), and it becomes possible to detect and predict guerrilla heavy rain.

[0037] <Processing operations in the information processing apparatus 100> FIG. 5 is a diagram showing an example of the processing operations executed by the control unit 110 when the image data acquisition processing unit 111 acquires the image data.

[0038] The detection unit 112 of the control unit 110 checks whether the image data satisfies a predetermined condition (step S501). If the image data satisfies the predetermined condition (step S501, YES), it is determined that a guerrilla heavy rain has occurred in the sky of the observation target (step S502), and the output processing unit 113 outputs a warning (step S503). If the image data does not satisfy the predetermined condition (step S501, NO), the process ends.

[0039] By doing so, especially when the predetermined condition is the condition that "the change in the brightness of the sky shown in the image data is rapid" and / or the condition that "the light and dark of the sky shown in the image data is clear", it becomes possible to detect the occurrence of guerrilla heavy rain in the observation target only by using the pixel values of the image data. Therefore, it becomes possible to detect the occurrence of guerrilla heavy rain in the observation target and issue a warning with a light calculation.

[0040] FIG. 6 is a diagram showing another example of the processing operation executed in the control unit 110 when image data is acquired by the image data acquisition processing unit 111.

[0041] The detection unit 112 of the control unit 110 checks whether the image data satisfies a predetermined condition (step S601).

[0042] If the image data satisfies the predetermined condition (step S601, YES), it is determined whether the image data is image data showing the sky where guerrilla heavy rain has occurred by using a learned model learned with the sky where guerrilla heavy rain has occurred (step S602). If the image data does not satisfy the predetermined condition (step S601, NO), the process ends.

[0043] If the image data is the image data in which the sky with heavy rain is captured (step S602, YES), it is determined that heavy rain is occurring in the sky to be observed (step S603), and the output processing unit 113 outputs a warning (step S604). If the image data is not the image data in which the sky with heavy rain is captured (step S602, NO), the process ends.

[0044] By doing so, it becomes possible to more accurately detect the occurrence of heavy rain and issue a warning in the observation target. Also, by doing so, the detection of the occurrence of heavy rain using the learned model is performed only when a predetermined condition is satisfied. For this reason, it becomes possible to reduce the number of times of detecting the occurrence of heavy rain using the learned model that requires heavy calculation, and it becomes possible to reduce the amount of calculation. In particular, when the predetermined condition is the condition that "the change in the brightness of the sky captured in the image data is rapid" and / or the condition that "the light and dark of the sky captured in the image data are distinct", it is possible to confirm whether or not the predetermined condition is satisfied with light calculation, and it is possible to further reduce the amount of calculation.

[0045] The present invention has been described above according to the preferred embodiments of the present invention. Here, the present invention has been described by showing specific specific examples, but various modifications and changes can be made to these specific examples without departing from the spirit and scope of the present invention described in the claims.

Explanation of Reference Numerals

[0046] 100 Information processing apparatus 110 Control unit 111 Image data acquisition processing unit 112 Detection unit 113 Output processing unit 120 Imaging unit 130 Storage unit 140 Output unit 150 Communication unit

Claims

1. An image data acquisition processing unit that acquires image data in which the sky to be observed is reflected, A detection unit that detects or predicts the occurrence of guerrilla heavy rain in the sky to be observed based on the image data, and has, The detection unit is, Using a learned model that has learned the sky where guerrilla heavy rain has occurred, it is confirmed whether guerrilla heavy rain is occurring in the sky reflected in the image data, or whether the sky reflected in the image data is the sky before guerrilla heavy rain occurs. Thus, the occurrence of guerrilla heavy rain in the sky to be observed is detected or predicted, When the image data satisfies a predetermined condition, it is confirmed whether guerrilla heavy rain is occurring in the sky reflected in the image data using a learned model that has learned the sky where guerrilla heavy rain has occurred, The predetermined condition includes that the brightness and darkness of the sky reflected in the image data are clear, an information processing apparatus.

2. An image data acquisition processing unit that acquires image data in which the sky to be observed is reflected, A detection unit that detects or predicts the occurrence of guerrilla heavy rain in the sky to be observed based on the image data, and has, When the image data satisfies a predetermined condition, the detection unit determines that guerrilla heavy rain is occurring in the sky to be observed, The predetermined condition includes that the brightness and darkness of the sky reflected in the image data are clear, an information processing apparatus.

3. When the occurrence of guerrilla heavy rain in the sky to be observed is detected or predicted, it further has an output processing unit that outputs a warning, the information processing apparatus according to claim 1 or 2.

4. The predetermined condition includes that the change in the brightness of the sky reflected in the image data is rapid, the information processing apparatus according to any one of claims 1 to 3.

5. The predetermined condition includes that the position of the cloud base of the cloud reflected in the image data is at or below a predetermined height, the information processing apparatus according to any one of claims 1 to 4.

6. An information processing method executed by a computer, An image data acquisition processing step of acquiring image data in which the sky to be observed is reflected, A detection step of detecting or predicting the occurrence of guerrilla heavy rain in the sky to be observed based on the image data, and has, The detection step is, Using a trained model that has learned the sky during a flash flood, it is possible to detect or predict the occurrence of a flash flood in the sky of the observation target by checking whether a flash flood is occurring in the sky shown in the image data or whether the sky shown in the image data is the sky before a flash flood occurs. When the image data meets a predetermined condition, use a trained model that has learned the sky during a flash flood to check whether a flash flood is occurring in the sky shown in the image data. The predetermined condition includes that the brightness and darkness of the sky shown in the image data are clear. An information processing method. **Claim 7**: An information processing method executed by a computer, An image data acquisition processing step of acquiring image data showing the sky of the observation target, Based on the image data, a detection step of detecting or predicting the occurrence of a flash flood in the sky of the observation target, and having, In the detection step, when the image data meets a predetermined condition, it is determined that a flash flood is occurring in the sky of the observation target. The predetermined condition includes that the brightness and darkness of the sky shown in the image data are clear. An information processing method. **Claim 8** An information processing program that causes a computer to execute the information processing method according to claim 6 or 7. **Claim 9** A computer-readable storage medium storing the information processing program according to claim 8.

Images