WO2024195139A1 - Image analysis system, image analysis method, and recording medium - Google Patents

Abstract

This image analysis system is provided with a satellite image acquisition unit, an environmental data acquisition unit, a detection unit, an estimation unit, and an output unit. The satellite image acquisition unit acquires a satellite image of an area to be analyzed. The environmental data acquisition unit acquires environmental data of the area to be analyzed. The detection unit detects an object to be analyzed in the satellite image. The estimation unit estimates the accuracy of detection, on the basis of the environmental data. The output unit outputs a result of the detection and information indicating the accuracy of the detection.

Description

Image analysis system, image analysis method, and recording medium

This disclosure relates to image analysis systems, etc.

Satellite images taken by artificial satellites are widely used to analyze objects on the Earth's surface. For example, SAR (Synthetic Aperture Radar) can make observations regardless of the presence or absence of sunlight and is not affected by clouds, so satellite images taken by SAR are suitable for analyzing objects on the Earth's surface. For this reason, satellite images taken by SAR may be used to identify objects that exist in the area being analyzed. On the other hand, since images are taken by observing the reflection of electromagnetic waves on the Earth's surface, for example, the mapping images taken by SAR are images expressed in monochrome gradations. In images expressed in monochrome gradations, for example, the contrast ratio between the sea or land and the object being analyzed is insufficient, and it may be difficult to distinguish the object in the satellite image. For this reason, for example, analysis of objects in satellite images may be performed using data other than satellite images.

The data analysis device of Patent Document 1 acquires satellite images and observation data of the earth's surface environment. The data analysis device of Patent Document 1 generates analysis data by correcting the analysis data based on the satellite images, based on the observation data of the earth's surface environment.

International Publication No. 2022/107619

The technology described in Patent Document 1 can make it difficult to identify objects that exist in the area being analyzed.

The present disclosure aims to provide an image analysis system etc. that can easily identify objects present in the area being analyzed in order to solve the above problems.

In order to solve the above problems, the image analysis system disclosed herein includes an image acquisition means for acquiring satellite images of the area to be analyzed, an environmental data acquisition means for acquiring environmental data of the area to be analyzed, a detection means for detecting the object to be analyzed that appears in the satellite images, an estimation means for estimating the accuracy of detection based on the environmental data, and an output means for outputting the detection results and information indicating the accuracy of detection.

The image analysis method disclosed herein acquires satellite images of the area to be analyzed, acquires environmental data of the area to be analyzed, detects the object to be analyzed that appears in the satellite images, estimates the accuracy of detection based on the environmental data, and outputs the detection results and information indicating the accuracy of detection.

The recording medium of the present disclosure non-temporarily records an image analysis program that causes a computer to execute the following processes: acquiring satellite images of an area to be analyzed; acquiring environmental data of the area to be analyzed; detecting an object to be analyzed that appears in the satellite images; estimating the accuracy of detection based on the environmental data; and outputting the detection results and information indicating the accuracy of detection.

The present disclosure makes it easier to identify objects present in the area being analyzed.

FIG. 1 is a diagram illustrating an example of a configuration according to an embodiment of the present disclosure. FIG. 1 is a diagram illustrating a form in which the earth's surface is imaged by an artificial satellite. FIG. 2 is a diagram illustrating an example of a satellite image according to an embodiment of the present disclosure. FIG. 2 is a diagram illustrating an example of a satellite image according to an embodiment of the present disclosure. FIG. 2 is a diagram illustrating an example of a satellite image according to an embodiment of the present disclosure. FIG. 1 is a diagram illustrating an example of a configuration of an image analysis system according to an embodiment of the present disclosure. FIG. 13 is a diagram illustrating an example of a display screen of an analysis result according to an embodiment of the present disclosure. FIG. 2 is a diagram illustrating an example of an operation flow of the image analysis system according to an embodiment of the present disclosure. FIG. 2 is a diagram illustrating an example of a hardware configuration of an image analysis system according to an embodiment of the present disclosure.

Embodiments of the present disclosure will be described in detail with reference to the drawings. FIG. 1 is a diagram showing an example of the configuration of an analysis system. The analysis system includes, for example, an image analysis system 10, a terminal device 20, a satellite image management server 30, and an environmental data management server 40. The image analysis system 10 is connected to the terminal device 20, for example, via a network. The image analysis system 10 is connected to the satellite image management server 30, for example, via a network. The image analysis system 10 is connected to the environmental data management server 40, for example, via a network. There may be multiple terminal devices 20. The number of terminal devices 20 may be set as appropriate.

The image analysis system 10 is, for example, a system that analyzes satellite images. A satellite image is, for example, an image of the earth's surface captured by an imaging device mounted on an artificial satellite. A satellite image is, for example, captured by a synthetic aperture radar (SAR). A satellite image may be captured by an imaging device other than a SAR. A satellite image is, for example, used to detect an object captured in the satellite image. An object captured in the satellite image is detected, for example, by image recognition. However, for example, an image captured by a SAR is a monochrome image based on electromagnetic waves reflected by the earth's surface, so the accuracy of object detection by image recognition may not be sufficient. For example, when the subject of analysis is a ship, an object other than a ship may be detected as a ship by image recognition. For this reason, when an analysis is performed using an image captured by a SAR, for example, after an object is detected by image recognition, the person in charge of analysis may identify the object captured in the image. The analyst, for example, visually checks the images captured by the SAR to identify objects that appear in the images captured by the SAR.

FIG. 2 is a diagram showing a schematic example of an image of the earth's surface captured by a satellite. In the example of FIG. 2, a ship is sailing on the sea. In the example of FIG. 2, the area in which the ship is sailing is captured by an imaging device mounted on the satellite. The image of the earth's surface is captured by, for example, a SAR. The satellite image captured by the imaging device mounted on the satellite is output to a satellite image management server 30 via, for example, a ground station. In addition, in the example of FIG. 2, environmental data is observed by an observation sensor installed on the sea. The observation sensor installed on the sea is, for example, an observation buoy that observes water temperature and waves. The waves are, for example, wave height, wave direction, and wave period. The observation sensor is not limited to an observation buoy. In addition, the items of environmental data observed by the observation sensor are not limited to those mentioned above. The observed environmental data is output to, for example, an environmental data management server 40.

The image analysis system 10 acquires satellite images captured by an artificial satellite from, for example, the satellite image management server 30. The image analysis system 10 also acquires environmental data observed by an observation sensor from, for example, the environmental data management server 40. The image analysis system 10 detects an object to be analyzed that appears in the acquired satellite image, for example, using an image analysis model. The image analysis system 10 then estimates the accuracy of detection of the object to be analyzed in the detection result by the image recognition model, for example, based on the acquired environmental data of the area to be analyzed. The accuracy of detection of the object to be analyzed is, for example, an index indicating the accuracy of the detection result of the object. In other words, the accuracy of detection of the object to be analyzed is an index indicating the accuracy of whether the object detected by the image recognition model is the object to be analyzed. For example, when the object to be analyzed is a ship, when the image analysis system 10 detects that the ship is captured in the satellite image, the accuracy of detection of the object to be analyzed is an index indicating the possibility that the object is a ship. The image recognition model is a learning model that detects objects captured in satellite images. The image recognition model will be explained later.

Furthermore, the environmental data is, for example, data related to the environment that may affect the presence or absence of an object. The environmental data is, for example, observational data obtained by observing the environment that may affect the presence or absence of an object. The environmental data may be data related to topography. The environmental data may be data on water depth. The water depth data may be data obtained by observing changes in water depth due to tides. For example, when the subject of analysis is a ship, the navigation or anchoring of the ship may be affected by water depth, tides, waves, wind speed, and topography. For example, a large ship cannot navigate in shallow water. Also, a normal small ship avoids places with fast tides and high waves because it is dangerous to navigate in places with fast tides and high waves. Also, areas near steep cliffs are not suitable for anchoring a ship. In this way, the possibility of a ship being present at a target point may change depending on the situation indicated by the environmental data at the target point. Therefore, when the subject of analysis is a ship, the likelihood that the ship is present can be estimated using the environmental data. Therefore, for example, when the analysis subject is a ship, when the image recognition model detects a ship, the image analysis system 10 can estimate the accuracy of the detection of the ship in the detection results based on the environmental data of the area being analyzed.

Furthermore, when the subject of analysis is a ship, for example, it may be difficult to distinguish between objects other than the ship and the ship. Objects other than the ship are, for example, marine life. The presence or absence of marine life may be affected, for example, by the environment. The presence or absence of marine life may be affected, for example, by water depth, water temperature, air temperature, tides, waves, wind speed, and topography. Environmental items that may affect the presence of marine life are not limited to the above. For example, when the area to be analyzed is an ocean area where marine life may exist, when a ship is detected by image recognition, the image analysis system 10 estimates the accuracy of detection by image recognition based on environmental data regarding the presence of marine life. For example, when it is estimated from the environmental data that the possibility of it being marine life is low, the image analysis system 10 can estimate that the accuracy of the detection of the ship in the detection results by the image recognition model is high. Also, for example, when it is estimated from the environmental data that the possibility of it being marine life is high, the image analysis system 10 estimates that the accuracy of the detection of the ship in the detection results by the image recognition model is low.

The image analysis system 10 outputs, for example, to the terminal device 20, the detection result of the object to be analyzed that appears in the satellite image and the detection accuracy. The terminal device 20 outputs, for example, the detection result of the object to be analyzed that appears in the satellite image and the detection accuracy to a display device not shown. The analyst, for example, refers to the detection result of the object to be analyzed that appears in the satellite image displayed on the display device and the detection accuracy to identify the object. For example, when a ship to be analyzed is detected and the estimation result of the detection accuracy based on the environmental data indicates that it is highly likely that it is not a ship, the analyst refers to the estimation result of the detection accuracy to determine whether it is a ship. When the object is detected as a ship and the estimation result based on the environmental data indicates that it is highly likely that it is not a ship, the analyst, for example, refers to a satellite image when it was identified as an object other than a ship in a past analysis to determine whether the object detected as a ship is a ship. When the object is detected as a ship and the estimation result based on the environmental data indicates that it is highly likely that it is not a ship, the analyst, for example, identifies the object by taking into account the possibility of the presence of a suspicious ship. This is because if a ship is present in an area in the environmental data where the presence of ships is unlikely, it may be navigating or anchoring with an unusual intention. An unusual intention means, for example, a purpose that is different from the usual purpose in the sea area in which the ship is navigating or anchoring. In this way, by referring to the detection results of the object to be analyzed and the detection accuracy in the detection results to identify the object shown in the satellite image, the task of identifying the object shown in the satellite image can be made easier.

Figure 3 shows an example of a satellite image of the earth's surface captured by SAR. The example of the satellite image in Figure 3 is an example of a satellite image of the area around a bay. The example of the satellite image in Figure 3 is an example of a satellite image in which there are no ships or other objects in the captured area. In the example of the satellite image in Figure 3, the areas with low brightness are land areas. Also, in the example of the satellite image in Figure 3, the areas with high brightness are sea areas.

FIG. 4 is an example of a satellite image captured by SAR of the same area as the example satellite image of FIG. 3, where a ship and objects other than ships are present in the same area. In the example image of FIG. 4, a ship and other objects are captured in an elliptical shape in the sea area. In the example image of FIG. 4, objects are also captured in an elliptical and circular shape in the land area. The image analysis system 10 detects, for example, an object to be analyzed that is captured in the satellite image of the example of FIG. 4. If the object to be analyzed is a ship, the image analysis system 10 detects the ship captured in the satellite image of the example of FIG. 4, for example, using an image recognition model. Then, when a ship is detected from the satellite image of the example of FIG. 4, the image analysis system 10 estimates the likelihood that the detected object is a ship based on environmental data for the area captured in the satellite image of the example of FIG. 4.

5 is an example of a display screen showing the detection result of the object to be analyzed and the estimation result of the detection accuracy. The example of the display screen of FIG. 5 is an example of a display screen showing the detection result of the object to be analyzed and the estimation result of the detection accuracy when the object to be analyzed captured in the satellite image of the example of FIG. 4 is detected. In the example of the display screen of FIG. 5, the area where the object to be analyzed is detected is surrounded by a solid or dashed rectangle. In the example of the display screen of FIG. 5, the area surrounded by the solid rectangle is, for example, an area where the detection accuracy is equal to or higher than the standard. In the example of the display screen of FIG. 5, the area surrounded by the dashed rectangle is, for example, an area where the detection accuracy is lower than the standard. By outputting an image such as the example of the display screen of FIG. 5, the person in charge of analysis can grasp, for example, an area where the detection accuracy by the image recognition model is low and a more detailed analysis is required. For example, the person in charge of analysis can identify the object captured in the satellite image by performing a detailed analysis of the area surrounded by the dashed line, taking into account that the estimation result of the accuracy based on the environmental data shows that the object is unlikely to be a ship.

Here, we will explain a specific example of the configuration of the image analysis system 10. Figure 6 is a diagram showing an example of the configuration of the image analysis system 10.

The image analysis system 10 basically comprises a satellite image acquisition unit 12, an environmental data acquisition unit 13, a detection unit 14, an estimation unit 15, and an output unit 16. The image analysis system 10 also comprises, for example, an acquisition unit 11 and a storage unit 17.

The acquisition unit 11, for example, acquires information about the analysis target. The acquisition unit 11 acquires information about the analysis target, for example, from the terminal device 20. The information about the analysis target is, for example, information indicating at least one of the region of the analysis target and the object of the analysis target.

The area to be analyzed is, for example, information indicating the range of the area to be analyzed. The area to be analyzed may be set in advance. For example, if the area to be analyzed by the image analysis system 10 is only a specific area, the acquisition unit 11 does not need to acquire information indicating the area to be analyzed.

The information on the object to be analyzed is, for example, information indicating what is to be detected. The object to be analyzed is, for example, a ship. When the object to be analyzed is a ship, the ship to be analyzed is, for example, a ship that is sailing or anchored. The ship to be analyzed may be both a ship that is sailing and a ship that is anchored. The ship to be analyzed may also be a ship that is located on land. The object to be analyzed may also be an aircraft, vehicle, structure, or stored item on land. The object to be analyzed is not limited to the above. The object to be analyzed may also be set in advance. For example, when the image analysis system 10 is used only to analyze the presence or absence of a ship, the acquisition unit 11 does not need to newly acquire information indicating that the presence of a ship is to be detected.

The information indicating the object to be analyzed may be the type of the object to be analyzed. For example, if the object to be analyzed is a ship, the type of object to be analyzed is, for example, the type of ship. The type of ship may be, for example, a large ship, a medium-sized ship, or a small ship. The type of ship may be, for example, a tanker, a passenger ship, a cargo ship, a ferry, a work boat, or a fishing boat. The type of ship may be, for example, a ship capable of carrying an aircraft squadron, a conventional ship, or a submarine. The types of ships are not limited to the above. Furthermore, the types of objects are not limited to the above.

The information on the subject of analysis may include information indicating the timing to be analyzed. The information indicating the timing to be analyzed is, for example, information indicating the timing when the satellite image to be analyzed was captured and the timing when the environmental data was observed. The information indicating the timing to be analyzed is set by date, time, day, or period. The period to be analyzed is set, for example, using the first day of the period and the last day of the period. How the period to be analyzed is set is not limited to the above. The information on the subject of analysis may also be information indicating which area within the region to be analyzed is to be analyzed. The information on the subject of analysis is not limited to the above. The information on the subject of analysis is set, for example, by the person in charge of analysis.

The satellite image acquisition unit 12 acquires satellite images of the area to be analyzed. The satellite image acquisition unit 12 acquires satellite images to which information on the location where the image was taken and the date and time of image capture is added, for example. The information on the location where the image was taken is, for example, the latitude and longitude of the location at the center of the image. The information on the location where the image was taken is not limited to the above, and may be any information for identifying the location where the image was taken. The satellite image acquisition unit 12 acquires satellite images of the area to be analyzed, for example, via the satellite image management server 30. The satellite image acquisition unit 12 may also acquire satellite images of the area to be analyzed via a storage medium.

When the acquisition unit 11 has acquired information indicating the region to be analyzed, the satellite image acquisition unit 12 may acquire a satellite image of the region indicated by the information indicating the region to be analyzed. When the acquisition unit 11 has acquired information indicating the timing of the region to be analyzed, the satellite image acquisition unit 12 may acquire a satellite image captured at the timing indicated by the information indicating the timing of the region to be analyzed.

The satellite image is, for example, an image captured by a SAR mounted on an artificial satellite. The satellite image is not limited to an image captured by a SAR. Furthermore, the satellite image acquisition unit 12 may acquire satellite images captured by a plurality of imaging methods. The satellite image acquisition unit 12 acquires, for example, a satellite image captured by a SAR and an optical image in the visible light region. The satellite image acquisition unit 12 may acquire a satellite image in the infrared region. The satellite images captured by a plurality of imaging methods are not limited to satellite images captured over the same range, so long as each of the images includes the area to be analyzed.

The environmental data acquisition unit 13 acquires environmental data for the area to be analyzed. The environmental data acquisition unit 13 acquires environmental data to which, for example, the observation point and the observation date and time are added. The information on the observation point is, for example, the latitude and longitude of the point where the environmental data was observed. The information on the observation point is not limited to the above. The environmental data may also be time series data of observed values of the environment. The environmental data acquisition unit 13 acquires environmental data for the area to be analyzed, for example, from the environmental data management server 40. The environmental data acquisition unit 13 may acquire environmental data for the area to be analyzed from multiple environmental data management servers 40. The environmental data acquisition unit 13 may acquire environmental data as environmental data from an environmental data providing server operated by the Japan Meteorological Agency or other government agency.

The environmental data is, for example, data on at least one of the following items: water depth, ocean weather, and topography. Ocean weather data is data on at least one of the following items: water temperature, air temperature, waves, and wind speed. Ocean weather is not limited to the above. Furthermore, the environmental data is not limited to the above. The environmental data acquisition unit 13 may acquire environmental data for items that are set based on the object to be analyzed. For example, when the object to be analyzed is a ship, the environmental data acquisition unit 13 acquires data on water depth, waves, and topography, which are necessary for estimating the presence or absence of a ship, as environmental data. The items for estimating the presence or absence of a ship are set, for example, by the person in charge of analysis or the operator of the image analysis system 10.

The environmental data acquisition unit 13 may acquire environmental data for items that are set based on at least one of the region and timing of the analysis target. For example, if the analysis target is a region where marine life exists in winter, when analyzing satellite images captured in winter, the environmental data acquisition unit 13 acquires environmental data for items necessary to estimate the presence or absence of marine life, and which was observed in winter.

When the acquisition unit 11 has acquired information indicating the region to be analyzed, the environmental data acquisition unit 13 may acquire environmental data observed in the region indicated by the information indicating the region to be analyzed. When the acquisition unit 11 has acquired information indicating the timing of the analysis target, the environmental data acquisition unit 13 may acquire environmental data observed at the timing indicated by the information indicating the timing of the analysis target.

The environmental data acquisition unit 13 may acquire environmental data such that the environmental data includes environmental data corresponding to the time when the satellite image was captured and the period of observation. The time when the satellite image was captured and the period of observation of the environmental data correspond means that the time when the satellite image was captured are included within a period within the range of fluctuations that can occur under the assumed conditions of fluctuations in the environmental data. For example, if the tidal current is constant during the winter season, the time when the satellite image was captured and the data of the tidal current at any time during the winter season may correspond to the time when the satellite image was captured and the period of observation. In addition, if the average value affects the presence or absence of an object, the environmental data acquisition unit 13 acquires data of the average value of the environmental data for a predetermined period including the time when the satellite image was captured. The predetermined period is set, for example, by the person in charge of analysis. In addition, in the case of data that changes depending on the date and time, such as wave height, the environmental data acquisition unit 13 acquires data of wave height such that the date and time when the satellite image was captured and the date and time when the observation were made are the same. Data that are the same include, for example, data that differs within a range that can be considered to be the same in the processing by the estimation unit 15. Furthermore, if the environmental data is data that does not normally change, such as topography, the environmental data acquisition unit 13 may acquire environmental data at any time when no changes have occurred. The timing at which observed environmental data is acquired is not limited to the above.

The environmental data acquisition unit 13 may acquire information that identifies ships that are present within the area being analyzed. The environmental data acquisition unit 13 may acquire identification signals of ships by AIS (Automatic Identification System) in the area being analyzed. The environmental data acquisition unit 13 acquires identification signals of ships by AIS from a monitoring server that monitors the navigation of ships in the area being analyzed. The information that identifies ships that are present within the area being analyzed is not limited to identification signals of ships by AIS.

The detection unit 14 detects an object to be analyzed that appears in a satellite image. The detection unit 14 detects an object to be analyzed that appears in a satellite image, for example, using an image recognition model. The detection unit 14 detects an area in the satellite image in which the object to be analyzed appears, for example, using an image recognition model. The detection unit 14 may further detect the type of object to be analyzed in the satellite image, using an image recognition model.

The image recognition model is, for example, a learning model that uses a satellite image as input and estimates objects that appear in the satellite image. The detection unit 14 may use the image recognition model to detect objects other than the object being analyzed that appear in the satellite image. For example, when the object being analyzed is a ship, if marine life is present in the area being analyzed, the detection unit 14 may use the image recognition model to detect marine life that appears in the satellite image.

The image recognition model is generated, for example, by learning the relationship between a satellite image and an object shown in the satellite image. The image recognition model is generated, for example, by learning the relationship between a satellite image and an area in which an object to be analyzed appears in the satellite image. The image recognition model may be generated by learning the relationship between a satellite image and the name of an object shown in the satellite image. The image recognition model may be generated by learning the relationship between a satellite image and an area in which an object appears and the name of the object shown. The image recognition model is generated, for example, by deep learning using a neural network. The learning data and learning algorithms used to generate the image recognition model are not limited to those described above. In addition, the image recognition model is generated, for example, in a system external to the image analysis system 10. The image recognition model may be generated in a generation unit (not shown) included in the image analysis system 10.

The detection unit 14 may detect the object to be analyzed based on the brightness change of the satellite image. For example, when a satellite image shows only an area of ocean and there is no change in the topography, the detection unit 14 detects the contour of the object shown in the satellite image based on the brightness change between pixels of the satellite image. The detection unit 14 then identifies whether the object shown in the satellite image is the object to be analyzed based on at least one of the size of the area surrounded by the contour and the shape of the contour. If the detection unit 14 identifies the object shown in the image as the object to be analyzed, it detects, for example, that the object to be analyzed is present in the area surrounded by the contour. How the object to be analyzed is detected is not limited to the above.

The estimation unit 15 estimates the accuracy of detection based on the environmental data. For example, based on the environmental data, the estimation unit 15 estimates an index indicating the likelihood that the object detected by the detection unit 14 is the object to be analyzed as the accuracy of detection. For example, based on the environmental data, the estimation unit 15 estimates the accuracy of detection of the object to be analyzed in the detection result by the image recognition model. That is, based on the environmental data, the estimation unit 15 estimates the likelihood of the detection result by the image recognition model. For example, based on the environmental data, the estimation unit 15 estimates the presence or absence of the object to be analyzed. Then, the estimation unit 15 estimates the accuracy of detection based on the estimated presence or absence of the object to be analyzed.

The estimation unit 15 may estimate the accuracy of detection based on whether the object detected by the image recognition model is an object other than the analysis target. For example, the estimation unit 15 estimates a candidate for the object when the object detected by the image recognition model is an object other than the analysis target. Then, the estimation unit 15 estimates the accuracy of detection based on the possibility that the candidate object may exist. For example, when the environmental data is suitable for the existence of the candidate object, the possibility that the candidate object may exist increases. When the possibility of the existence of an object other than the analysis target is high, the likelihood that the object detected by the image recognition model is the object to be analyzed decreases. When the object to be analyzed is a ship, the estimation unit 15 estimates a candidate for the object other than the ship when the object detected by the image recognition model is other than the ship. When the object to be analyzed is a ship, when it is estimated from the environmental data of the area to be analyzed that marine life exists, the estimation unit 15 estimates, for example, marine life as a candidate for the object other than the ship. When the environmental data indicates that there is a high possibility that marine life is present, the estimation unit 15 estimates, for example, that the object detected as a ship by the image recognition model is unlikely to be a ship. In other words, the estimation unit 15 estimates that the accuracy of the detection result that the image recognition model detects as a ship is low.

The estimation unit 15 estimates the accuracy of detection, for example, based on the detection result of the image recognition model and the presence or absence of the object to be analyzed in the area to be analyzed estimated using the environmental data. The presence or absence of the object to be analyzed in the area to be analyzed estimated using the environmental data is, for example, the possibility that the object to be analyzed may exist in the area to be analyzed estimated based on the environmental data. The estimation unit 15 estimates the presence or absence of the object to be analyzed in the area to be analyzed, for example, based on a criterion that defines the relationship between the object to be analyzed and the environmental data. Then, when the image recognition model detects the object to be analyzed, the estimation unit 15 estimates the accuracy of the detection result based on the estimation result of the presence or absence of the object to be analyzed. The criterion that defines the relationship between the object to be analyzed and the environmental data is a criterion for estimating the presence or absence of the object to be analyzed based on the environmental data. The criterion that defines the relationship between the object to be analyzed and the environmental data is a criterion for determining in what kind of environment the object to be analyzed is likely to exist. The criterion that defines the relationship between the object to be analyzed and the environmental data may also be a criterion for determining in what kind of environment the object to be analyzed is likely not to exist.

When the object to be analyzed is a ship, the estimation unit 15 estimates the accuracy of detection based on the result of the image recognition model detecting the ship and the possibility that the ship may exist in the area to be analyzed estimated using the environmental data. For example, when the image recognition model detects a ship, which is the object to be analyzed, and the topographical data included in the environmental data indicates a topographical topography that is not suitable for anchoring, the estimation unit 15 estimates that the detected object is unlikely to be a ship because of the topography that is not suitable for anchoring a ship. Also, for example, when the image recognition model detects a large ship, which is the object to be analyzed, and the water depth data included in the environmental data indicates a depth that is not suitable for navigation and anchoring, the estimation unit 15 estimates that the detected object is unlikely to be a ship because of the water depth that is not suitable for navigation and anchoring a ship. Also, for example, when the image recognition model detects multiple ships approaching each other, which are the object to be analyzed, and the ocean current and wave data included in the environmental data indicate values that are not suitable for the approach of ships, the estimation unit 15 estimates that the detected objects are unlikely to be ships. The criteria that define the relationship between the object to be analyzed and the environmental data are not limited to the above. The criteria that define the relationship between the object to be analyzed and the environmental data are set, for example, by the person in charge of analysis or the operator of the image analysis system 10.

The estimation unit 15 estimates the accuracy of detection based on a score indicating the possibility of the presence of an object estimated using environmental data, for example. The estimation unit 15 estimates a score indicating the possibility of the presence of an object to be analyzed, for example, based on environmental data. Then, the estimation unit 15 estimates the accuracy of detection based on the estimated score. The score indicating the possibility of the presence of an object is, for example, an index indicating whether an environment is suitable for an object to exist. For example, a score indicating the possibility of a ship to exist is an index indicating whether an environment is suitable for a ship to exist. An environment in which a ship exists is, for example, an environment related to water depth, topography, and waves that is suitable for at least one of ship navigation and anchoring. The estimation unit 15 calculates a score for each item of environmental data. Then, the estimation unit 15 estimates a score indicating the possibility of the presence of an object by, for example, adding up the scores for each item of environmental data. The relationship between the value of environmental data and the score is set, for example, as a table for each item of environmental data. The relationship between the value of environmental data and the score may be set using a function with the value of environmental data as an explanatory variable and the score as a target variable. The relationship between the environmental data value and the score is set, for example, by the person in charge of analysis or the operator of the image analysis system 10.

The estimation unit 15 may estimate a score indicating the possibility that an object other than the object to be analyzed exists based on the environmental data. An object other than the object to be analyzed is, for example, an object that the image recognition model may mistakenly recognize as the object to be analyzed. When a score indicating the possibility that an object other than the object to be analyzed exists is calculated, the estimation unit 15 estimates the accuracy of detection based on, for example, the score indicating the possibility that an object other than the object to be analyzed exists. The estimation unit 15 estimates a score indicating the possibility that an object exists for an object other than the object to be analyzed that may exist in the area to be analyzed. The possibility that an object exists in the area to be analyzed means, for example, that an object has existed in the area to be analyzed in the past. The possibility that an object exists in the area to be analyzed may mean that an object exists frequently in an area having at least one of the same topography and environment as the area to be analyzed, or similar thereto. For example, when the object to be analyzed is a ship and there is a possibility that marine life exists, the estimation unit 15 estimates a score indicating the possibility that marine life exists. For example, when the score of marine life is higher than that of a ship, the estimation unit 15 estimates that the object detected as a ship is highly likely to be marine life. That is, if the score for marine life is higher than that for ships, the estimation unit 15 estimates that the object detected as a ship is unlikely to be a ship.

The estimation unit 15 may estimate the accuracy of detection based on the probability of being the analysis target in the detection results of the image recognition model and a score indicating the possibility of the existence of an object estimated from the environmental data. For example, if the probability of being the analysis target in the estimation results of the image recognition model is Pm, the score indicating the possibility of the existence of an object estimated from the environmental data is S, and the accuracy of detection is Pd, the estimation unit 15 estimates the accuracy of detection using the formula Pd = Pm x S. The formula for estimating the accuracy of detection is not limited to the above. Furthermore, how the accuracy of detection is estimated is not limited to the above.

The estimation unit 15 estimates the accuracy of detection, for example, based on environmental factors that affect the accuracy of the detection result of the image recognition model estimated using the environmental data. The estimation unit 15 may also estimate the accuracy of detection, based on environmental factors that affect the presence of an object that the image recognition model estimated using the environmental data may mistakenly recognize as the object to be analyzed. The environmental factors that affect the accuracy of the detection result of the image recognition model that detects the object to be analyzed are, for example, items of environmental data that have a large effect on the presence or absence of the object. The estimation unit 15 estimates, for example, that the item with the highest score is the environmental factor that affects the accuracy of the detection result. The estimation unit 15 may estimate, for example, that the item whose score meets a criterion is the environmental factor that affects the accuracy of the detection result. The estimation unit 15 may also estimate that the items with the highest scores up to a predetermined rank are the environmental factors that affect the accuracy of the detection result.

The estimation unit 15 may also estimate, based on the environmental data, environmental factors that affect the likelihood of the detection result of the image recognition model that detects the object to be analyzed as the estimated reason for the detection accuracy. For example, the estimation unit 15 estimates environmental factors that affect the possibility of the existence of the object to be analyzed estimated using the environmental data as the estimated reason for the detection accuracy. For example, the estimation unit 15 estimates environmental factors that affect the possibility of the existence of an object other than the object to be analyzed estimated using the environmental data as the estimated reason for the detection accuracy. When the object to be analyzed is a ship, the estimation unit 15 estimates the estimated reason for the detection accuracy based on environmental factors that affect the possibility of the existence of a ship or environmental factors that affect the possibility of the existence of an object other than a ship.

The estimation unit 15 may estimate the accuracy of detection based on objects present in the area to be analyzed that are estimated using an estimation model that estimates the objects present from environmental data. The estimation model is, for example, a learning model that learns the relationship between environmental data and existing objects. The estimation model is, for example, generated by deep learning using a neural network.

The estimation unit 15 may estimate the accuracy of detection based on an object present in the area to be analyzed, which is estimated using an estimation model capable of identifying the reason for estimation. The estimation model capable of outputting the reason for estimation may be generated using, for example, a learning algorithm based on factorized asymptotic Bayesian inference. When learning is performed using a learning algorithm based on factorized asymptotic Bayesian inference, the learning device performs case classification using a decision tree-type rule with the data of each item of environmental data as input data and the existing object as correct answer data. The learning device then generates a learning model that predicts the degree of realization using a linear model that combines different explanatory variables in each case. The learning device generates the learning model by sequentially performing the processes of optimizing the case classification conditions of the data, generating a prediction model by optimizing the combination of explanatory variables, and deleting unnecessary prediction models. The estimation model may also be a learning model that identifies the reason for estimation based on the change in the estimation result relative to the amount of variation of each item of environmental data. The learning algorithm that generates the estimation model is not limited to the above. The estimation model is also generated, for example, in a system external to the image analysis system 10. The estimation model may be generated in a generation unit (not shown) included in the image analysis system 10.

The estimation unit 15 may estimate the accuracy of detection based on areas where the object to be analyzed is not present, which is estimated using environmental data. When the object to be analyzed is a ship, the estimation unit 15 estimates areas where the ship is not present, for example, based on environmental data. When the object to be analyzed is a ship, the estimation unit 15 estimates that the accuracy of detection is low when a ship is detected in an area where it is estimated that no ship is present. The estimation unit 15 estimates a score indicating the possibility that the object to be analyzed is present for each area, based on environmental data. Then, the estimation unit 15 estimates that the accuracy of detection is low, for example, when a ship is detected in an area where the score is below a standard.

The estimation unit 15 may estimate the accuracy of detection based on an area where a specific object other than the object to be analyzed, which is estimated using the environmental data, does not exist. The specific object is, for example, an object that is expected to be erroneously recognized as the object to be analyzed by the image recognition model. The estimation unit 15 estimates a score indicating the possibility that a specific object exists for each area based on the environmental data. Then, the estimation unit 15 estimates that a specific object does not exist in an area where the score is below a standard, for example. The estimation unit 15 estimates that the accuracy of detection is high, for example, when the object to be analyzed is detected in an area where a specific object does not exist. The estimation unit 15 may estimate an area where a specific object other than the object to be analyzed is likely to exist based on the environmental data. For example, when the object to be analyzed is detected in an area where a specific object is likely to exist, the estimation unit 15 estimates that the accuracy of detection is low. The estimation unit 15 estimates a score indicating the possibility that a specific object exists for each area based on the environmental data. Then, the estimation unit 15 estimates that an area where a score is equal to or higher than a standard is an area where a specific object is likely to exist. The specified object is set, for example, by the person in charge of the analysis.

The estimation unit 15 may further use the ship's identification signal to estimate the accuracy of detection. The estimation unit 15 estimates the accuracy of detection based on, for example, the AIS identification signal. The estimation unit 15 estimates that the accuracy of detection is high, for example, when the type of ship indicated by the detection result matches the type of ship indicated by the AIS identification signal. The estimation unit 15 may estimate that the accuracy of detection is high, for example, the more items that match between the detection result and the information indicated by the AIS identification signal. For example, the estimation unit 15 estimates that the accuracy of detection is higher when the type of ship and course match in the detection result and the information indicated by the AIS identification signal than when only the type of ship matches.

The output unit 16 outputs the detection result and information indicating the accuracy of the detection. The output unit 16 outputs the detection result and information indicating the accuracy of the detection to, for example, the terminal device 20. The output unit 16 outputs, for example, a satellite image, the detection result of the object to be analyzed in the satellite image, and information indicating the accuracy of the detection. The output unit 16 outputs the detection result indicating the area in which the object to be analyzed is detected by surrounding the area in which the object to be analyzed is detected on the satellite image with a figure. The figure surrounding the area in which the object to be analyzed is detected is, for example, a rectangle. The figure surrounding the area in which the object to be analyzed is detected is not limited to, for example, a rectangle. The output unit 16 outputs, for example, a satellite image in which the line surrounding the area in which the object to be analyzed is detected on the satellite image is changed according to the level of the accuracy of the detection. The output unit 16 outputs, for example, a satellite image in which at least one of the shape, thickness, and color of the line surrounding the area in which the object to be analyzed is detected on the satellite image is changed according to the level of the accuracy of the detection. The output unit 16 may output a satellite image in which the shape of a figure surrounding an area in which the object to be analyzed is detected on the satellite image is changed according to the level of detection accuracy. The output unit 16 may output a numerical value indicating the level of detection accuracy by superimposing it on the satellite image. The output unit 16 may also output a numerical value, letter, or symbol indicating the level of detection accuracy by superimposing it on the satellite image. The form in which the detection accuracy is output is not limited to the above.

The output unit 16 may output candidates for objects other than the object to be analyzed that are estimated by the estimation unit 15 based on the environmental data. The output unit 16 outputs information on candidates for objects other than the object to be analyzed that are estimated by the estimation unit 15 based on the environmental data, in association with an area in which the image recognition model detects the object. The output unit 16 outputs names of candidates for objects other than the object to be analyzed that are estimated by the estimation unit 15 based on the environmental data, in association with an area in which the image recognition model detects the object.

The output unit 16 may further output the reason for estimating the accuracy of detection. For example, the output unit 16 outputs the reason for estimating the accuracy of detection that the estimation unit 15 estimates based on the environmental data. For example, the output unit 16 outputs the reason for estimating that the detected object is the object to be analyzed. For example, the output unit 16 may output the reason for estimating that the detected object is an object other than the object to be analyzed.

The output unit 16 may output, as a reference image, a satellite image in which an object of the same type as the detected object is identified. For example, if the detected object is a large ship, the output unit 16 may output, as a reference image, a satellite image in which a large ship has been identified in a past analysis as showing the large ship. The output unit 16 may output, as a reference image, a satellite image of an object other than the analysis target that may be present in the area being analyzed. For example, if the analysis target is a ship, the output unit 16 may output a satellite image of an object other than a ship.

The output unit 16 may output a satellite image captured using a method different from that of the satellite image being analyzed. For example, if the satellite image being analyzed is an image captured by a SAR, the output unit 16 may further output an optical image in the visible light region that includes the same range as the range in which the image captured by the SAR was captured.

The output unit 16 may, for example, output an image of an enlarged area selected by the person in charge of analysis. When the output unit 16 detects, for example, that an area on the display screen in which the object to be analyzed is detected has been selected by the operation of the person in charge of analysis, it outputs an image of an enlarged area of the selected area. The output unit 16 may also output an image of the area selected by the person in charge of analysis that has a higher resolution than other areas. The output unit 16 may also output, as a reference image, a satellite image that has been identified in a past analysis as containing the object to be analyzed in the area selected by the person in charge of analysis. The output unit 16 may also output, as a reference image, a satellite image that has been identified in a past analysis as containing a candidate object other than the object to be analyzed in the area selected by the person in charge of analysis.

FIG. 7 is a diagram showing an example of a display screen that displays the reason for the estimated detection accuracy. In the example of the display screen in FIG. 7, the area surrounded by a solid-line rectangle is, for example, an area where the detection accuracy is equal to or higher than a standard. In the example of the display screen in FIG. 7, the area surrounded by a dashed-line rectangle is, for example, an area where the detection accuracy is lower than a standard. In the example of the display screen in FIG. 7, the area surrounded by the dashed line displays the possibility that the detected object is a "seal." In the example of the display screen in FIG. 7, the reason for the estimation that the object is other than the object being analyzed is displayed as "temperature" and "tides." In the example of the display screen in FIG. 7, the observation data for temperature and tides, which are part of the environmental data, are suitable for the presence of a seal, so it is shown that the possibility that it is a ship is low and the possibility that it is a seal is high.

The storage unit 17 stores, for example, data used in the analysis of satellite images. The storage unit 17 stores, for example, satellite images acquired by the satellite image acquisition unit 12. The storage unit 17 stores, for example, environmental data acquired by the environmental data acquisition unit 13. The storage unit 17 stores, for example, a table showing the relationship between environmental data and a score indicating the possibility of the presence of an object. The storage unit 17 stores, for example, an image recognition model. The storage unit 17 stores, for example, an estimation model. The image recognition model and the estimation model may be stored in a storage means other than the storage unit 17. The storage unit 17 stores, for example, the detection results of the image recognition model. The storage unit 17 may store, for example, an estimation reason for the accuracy of detection. The storage unit 17 may store, as a reference image, a satellite image identified in a past analysis as containing an object to be analyzed. The storage unit 17 may store, as a reference image, a satellite image identified in a past analysis as containing an object other than the object to be analyzed.

The terminal device 20 is, for example, a terminal device used by an analyst to analyze satellite images. The terminal device 20 acquires satellite images, detection results of objects captured in the satellite images, and the accuracy of detection, for example, from the output unit 16 of the image analysis system 10. The terminal device 20 then outputs the satellite images, detection results of objects captured in the satellite images, and the accuracy of detection, for example, to a display device not shown. The terminal device 20 acquires an estimated reason for the accuracy of detection, for example, from the output unit 16 of the image analysis system 10. The terminal device 20 then outputs an estimated reason for the accuracy of detection, for example, to a display device not shown.

The terminal device 20 may, for example, acquire information about the analysis target input by a user's operation. The terminal device 20 outputs the information about the analysis target to the acquisition unit 11 of the image analysis system 10, for example.

The terminal device 20 may be, for example, a personal computer, a tablet computer, or a smartphone. The terminal device 20 is not limited to the above examples.

The satellite image management server 30 manages, for example, images of the earth's surface captured by an imaging device mounted on an artificial satellite. The satellite image management server 30 acquires satellite images of the earth's surface captured by an imaging device mounted on an artificial satellite, for example, from a ground station that communicates with the artificial satellite. The satellite images are associated with, for example, the image capture date and time and the image capture location. The satellite image management server 30 acquires satellite images of the earth's surface captured by an imaging device mounted on an artificial satellite, for example, from a ground station that communicates with the artificial satellite. The satellite image management server 30 then associates and stores the acquired satellite images with the image capture date and time and the image capture location. The satellite image management server 30 may identify the image capture location from the position of the satellite at the time of image capture and imaging-related parameters added to the satellite image. The imaging-related parameters are, for example, the direction in which electromagnetic waves in the band used for image capture are transmitted, the transmission angle of the electromagnetic waves relative to the earth's surface, and the reception accuracy of the electromagnetic waves reflected from the earth's surface. For example, when the satellite image management server 30 receives a request for a satellite image from the image analysis system 10, it outputs the requested satellite image and the image capture date and time and capture location of the satellite image to the satellite image acquisition unit 12 of the image analysis system 10. There may also be multiple satellite image management servers 30. For example, satellite images may be stored in different servers for each management entity of the artificial satellite that captured the image. The number of satellite image management servers 30 may be set as appropriate.

The environmental data management server 40, for example, manages environmental data. The environmental data management server 40, for example, acquires environmental data. The environmental data management server 40 then stores the acquired environmental data in association with information on the observation date and time and the observation point. The environmental data management server 40 stores, for example, observation data on water depth, water temperature, air temperature, tides, wave height, and wind speed. The environmental data management server 40 also stores, for example, topographical data. When the environmental data management server 40 receives a request for environmental data from the image analysis system 10, for example, it outputs the requested environmental data and the observation date and time and observation point of the environmental data to the environmental data acquisition unit 13 of the image analysis system 10. There may also be multiple environmental data management servers 40. For example, the environmental data may be stored in a different server for each observer of the environmental data. The number of environmental data management servers 40 may be set as appropriate.

An example of the operation of the image analysis system 10 will be described. Figure 8 shows an example of the operation flow when the image analysis system 10 detects an object captured in a satellite image and estimates the accuracy of the detection.

The satellite image acquisition unit 12 acquires satellite images of the area to be analyzed (step S11). The satellite image acquisition unit 12 acquires satellite images of the area to be analyzed, for example, from the satellite image management server 30.

The environmental data acquisition unit 13 also acquires environmental data for the area to be analyzed (step S12). The environmental data acquisition unit 13 acquires the environmental data for the area to be analyzed from, for example, the environmental data management server 40.

When the satellite image and environmental data of the area to be analyzed are acquired, the detection unit 14 detects the object to be analyzed that appears in the satellite image (step S13). The detection unit 14 detects the object to be analyzed that appears in the satellite image, for example, using an image recognition model.

If the object to be analyzed is detected in the satellite image (Yes in step S14), the estimation unit 15 estimates the accuracy of detection of the object to be analyzed based on the environmental data (step S15). The estimation unit 15 estimates, for example, based on the environmental data, the likelihood of the detection result of the image recognition model detecting the object to be analyzed as the detection accuracy.

Once the accuracy of detection of the object to be analyzed is estimated, the output unit 16 outputs the result of detection of the object to be analyzed and the accuracy of detection (step S16). The output unit 16 outputs the result of detection of the object to be analyzed and the accuracy of detection to, for example, the terminal device 20.

When the detection result of the object to be analyzed and the detection accuracy are output, if there is an image in which the process to detect the object to be analyzed has not been performed (No in step S17), the process returns to step S13, and the detection unit 14 detects the object to be analyzed that is captured in a satellite image in which the detection of the object to be analyzed has not been performed.

In step S16, when the detection result of the object to be analyzed and the detection accuracy are output, if the process of detecting the object to be analyzed has been performed for all satellite images (Yes in step S17), the image analysis system 10 ends the process of detecting the object in the satellite image and estimating the detection accuracy.

If the object to be analyzed is not detected in the satellite image in step S14 (No in step S14), or there is a satellite image for which processing to detect the object to be analyzed has not been performed (No in step S17), the process returns to step S13, and the detection unit 14 detects the object to be analyzed in other satellite images for which analysis has not been performed.

If the object to be analyzed is not detected in the satellite image in step S14 (No in step S14), and the process of detecting the object to be analyzed has been performed for all satellite images (Yes in step S17), the image analysis system 10 ends the process of detecting the object in the satellite image and estimating the accuracy of detection.

The image analysis system 10 detects an object to be analyzed from a satellite image of the area to be analyzed. The image analysis system 10 estimates the accuracy of detection of the object to be analyzed based on the environmental data of the area to be analyzed. The image analysis system 10 then outputs the result of detection of the object to be analyzed and information indicating the accuracy of detection. In this way, by outputting the result of detection of the object to be analyzed and information indicating the accuracy of detection, the person in charge of analyzing the satellite image can confirm whether the object shown in the satellite image is the object to be analyzed while referring to the information indicating the accuracy of detection output as the accuracy of detection. For example, if the environmental data indicates that there is a high possibility of marine life being present, the person in charge of analysis can confirm whether the object shown in the satellite image is the object to be analyzed while taking into account the possibility of marine life being present. Also, if there is a high possibility that there is no ship based on the topography and water currents, the person in charge of analysis can analyze the satellite image while taking into account that if it is a ship, it is an unusual ship. In this way, by analyzing the satellite image using the result of detection of the object to be analyzed and information indicating the accuracy of detection, it becomes easier to identify objects present in the area to be analyzed. Therefore, by using the image analysis system 10, it is possible to easily analyze objects present in the area being analyzed.

In addition, when outputting the estimated reason for the detection accuracy, the analyst can determine whether an object shown in the image is the object to be analyzed by considering, for example, the reasons why it may be the object to be analyzed or the reasons why it may be something other than the object to be analyzed. Therefore, when outputting the estimated reason for the detection accuracy, it is easier to analyze objects that exist in the area to be analyzed.

Each process in the image analysis system 10 may be distributed and executed in multiple information processing devices connected via a network. For example, the process in the detection unit 14 and the process in the estimation unit 15 may be executed in different information processing devices. It may be set as appropriate which of the multiple information processing devices each process in the image analysis system 10 is executed in.

Each process in the image analysis system 10 can be realized by executing a computer program on a computer. Figure 9 shows an example of the configuration of a computer 100 that executes a computer program that performs each process in the image analysis system 10. The computer 100 comprises a CPU (Central Processing Unit) 101, memory 102, a storage device 103, an input/output I/F (Interface) 104, and a communication I/F 105.

The CPU 101 reads out and executes computer programs for performing each process from the storage device 103. The CPU 101 may be configured by a combination of multiple CPUs. The CPU 101 may also be configured by a combination of a CPU and another type of processor. For example, the CPU 101 may be configured by a combination of a CPU and a GPU (Graphics Processing Unit). The memory 102 is configured by a DRAM (Dynamic Random Access Memory) or the like, and temporarily stores the computer programs executed by the CPU 101 and data being processed. The storage device 103 stores the computer programs executed by the CPU 101. The storage device 103 is configured by, for example, a non-volatile semiconductor storage device. Other storage devices such as a hard disk drive may be used for the storage device 103. The input/output I/F 104 is an interface that accepts input from an operator and outputs display data, etc. The communication I/F 105 is an interface that transmits and receives data between the terminal device 20, the satellite image management server 30, the environmental data management server 40, and other information processing devices. Furthermore, the terminal device 20, the satellite image management server 30, and the environmental data management server 40 may have the same configuration as the computer 100.

The computer programs used to execute each process can also be distributed by storing them on a computer-readable recording medium that non-temporarily records data. As the recording medium, for example, a magnetic tape for recording data or a magnetic disk such as a hard disk can be used. Alternatively, an optical disk such as a CD-ROM (Compact Disc Read Only Memory) can also be used as the recording medium. A non-volatile semiconductor memory device can also be used as the recording medium.

Some or all of the above embodiments can be described as follows, but are not limited to the following:

[Appendix 1]
An image acquisition means for acquiring a satellite image of an area to be analyzed;
An environmental data acquisition means for acquiring environmental data of an area to be analyzed;
A detection means for detecting an object to be analyzed that is captured in a satellite image;
an estimation means for estimating a degree of accuracy of the detection based on the environmental data;
and an output means for outputting a result of the detection and information indicating a degree of accuracy of the detection.

[Appendix 2]
The detection accuracy is an index indicating the likelihood that the object detected by the detection means is the object to be analyzed.
2. The image analysis system of claim 1.

[Appendix 3]
The estimation means estimates the accuracy of the detection based on environmental factors that affect the accuracy of a detection result of an image recognition model that detects the object to be analyzed, the detection result being estimated using environmental data.
3. The image analysis system of claim 2.

[Appendix 4]
The estimation means estimates the accuracy of the detection based on environmental factors that affect the presence of an object that the image recognition model estimated using environmental data may mistakenly recognize as the object to be analyzed.
4. The image analysis system of claim 3.

[Appendix 5]
The estimation means estimates a degree of accuracy of the detection based on a detection result of the image recognition model and the presence or absence of an object to be analyzed estimated using the environmental data.
4. The image analysis system of claim 3.

[Appendix 6]
The object to be analyzed is a ship,
The estimation means estimates the accuracy of the detection based on an environmental factor indicating that the object detected by the image recognition model is other than the ship.
6. The image analysis system of claim 3.

[Appendix 7]
The estimation means estimates the accuracy of the detection based on an area where the ship is not present, which is estimated using the environmental data.
7. The image analysis system of claim 6.

[Appendix 8]
The output means outputs a satellite image relating to the object other than the ship estimated by the estimation means.
8. The image analysis system of claim 6 or 7.

[Appendix 9]
The environmental data is data on at least one of air temperature, water temperature, water flow, water depth, and topography.
9. An image analysis system according to any one of claims 6 to 8.

[Appendix 10]
The estimation means further uses an identification signal of the vessel to estimate the accuracy of the detection.
10. The image analysis system of any one of claims 6 to 9.

[Appendix 11]
The output means further outputs a reason for the estimation of the accuracy of the detection.
11. The image analysis system of claim 1.

[Appendix 12]
the output means outputs a satellite image in which a line surrounding an area in which the object to be analyzed is detected on the satellite image is changed according to a level of accuracy of the detection.
12. An image analysis system according to any one of claims 1 to 11.

[Appendix 13]
Obtain satellite images of the area to be analyzed,
Obtain environmental data for the area to be analyzed;
Detects objects to be analyzed in satellite images,
Estimating the accuracy of the detection based on the environmental data;
outputting a result of the detection and information indicating a degree of accuracy of the detection;
Image analysis methods.

[Appendix 14]
Acquiring satellite images of the area to be analyzed;
Obtaining environmental data for the area to be analyzed;
A process for detecting an object to be analyzed that is captured in satellite imagery;
A process of estimating the accuracy of the detection based on the environmental data;
and outputting a result of the detection and information indicating the accuracy of the detection.

The present disclosure has been described above using the above-mentioned embodiment as an example. However, the present disclosure is not limited to the above-mentioned embodiment. In other words, the present disclosure can be applied in various aspects that can be understood by a person skilled in the art within the scope of the present disclosure.

REFERENCE SIGNS LIST 10 Image analysis system 11 Acquisition unit 12 Satellite image acquisition unit 13 Environmental data acquisition unit 14 Detection unit 15 Estimation unit 16 Output unit 17 Storage unit 20 Terminal device 30 Satellite image management server 40 Environmental data management server 100 Computer 101 CPU
102 Memory 103 Storage device 104 Input/output I/F
105 Communication I/F

Claims (14)

An image acquisition means for acquiring a satellite image of an area to be analyzed;
An environmental data acquisition means for acquiring environmental data of an area to be analyzed;
A detection means for detecting an object to be analyzed that is captured in a satellite image;
an estimation means for estimating a degree of accuracy of the detection based on the environmental data;
and an output means for outputting a result of the detection and information indicating a degree of accuracy of the detection. The detection accuracy is an index indicating the likelihood that the object detected by the detection means is the object to be analyzed.
The image analysis system of claim 1 . The estimation means estimates the accuracy of the detection based on environmental factors that affect the accuracy of a detection result of an image recognition model that detects the object to be analyzed, the detection result being estimated using environmental data.
The image analysis system of claim 2. The estimation means estimates the accuracy of the detection based on environmental factors that affect the presence of an object that the image recognition model estimated using environmental data may mistakenly recognize as the object to be analyzed.
The image analysis system of claim 3. The estimation means estimates a degree of accuracy of the detection based on a detection result of the image recognition model and the presence or absence of an object to be analyzed estimated using the environmental data.
The image analysis system of claim 3. The object to be analyzed is a ship,
The estimation means estimates the accuracy of the detection based on an environmental factor indicating that the object detected by the image recognition model is other than the ship.
6. An image analysis system according to any one of claims 3 to 5. The estimation means estimates the accuracy of the detection based on an area where the ship is not present, which is estimated using the environmental data.
The image analysis system of claim 6. The output means outputs a satellite image relating to the object other than the ship estimated by the estimation means.
8. An image analysis system according to claim 6 or 7. The environmental data is data on at least one of air temperature, water temperature, water flow, water depth, and topography.
An image analysis system according to any one of claims 6 to 8. The estimation means further uses an identification signal of the vessel to estimate the accuracy of the detection.
10. An image analysis system according to any one of claims 6 to 9. The output means further outputs a reason for the estimation of the accuracy of the detection.
11. An image analysis system according to any one of claims 1 to 10. the output means outputs a satellite image in which a line surrounding an area in which the object to be analyzed is detected on the satellite image is changed according to a level of accuracy of the detection.
12. An image analysis system according to any preceding claim. Obtain satellite images of the area to be analyzed,
Obtain environmental data for the area to be analyzed;
Detects objects to be analyzed in satellite images,
Estimating the accuracy of the detection based on the environmental data;
outputting a result of the detection and information indicating the accuracy of the detection;
Image analysis methods. Acquiring satellite images of the area to be analyzed;
Obtaining environmental data for the area to be analyzed;
A process for detecting an object to be analyzed that is captured in satellite imagery;
A process of estimating an accuracy of the detection based on the environmental data;
and outputting a result of the detection and information indicating the accuracy of the detection.

Images