Image analysis system, attention point detection device and image analysis method

Abstract

To provide an image analysis system which captures a wide-area image from the sky, detects an attention point, captures the attention point from the ground and analyzes the detail.SOLUTION: An image analysis system includes: an attention point detection device which analyzes an image capturing a wide area from the sky, or a set of numerical data mappable onto a plane observed from the sky, detects an attention point from the analyzed image or a set of numerical data, and instructs a moving imaging device to move from the ground to the attention point and capture an image of the attention point. The image analysis system analyzes the image captured by the moving imaging device. The attention point detection device, for example, converts the attention point into a designation of a ground location and instructs the moving imaging device to move.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an image analysis system, an interest point detection device, and an image analysis method. [Background technology]

[0002] Patent document 1 describes how smartphone location information is constantly collected on a specific server and shared with emergency agencies, limited to smartphone users within the service area of ​​the disaster response emergency dispatch HAPS, making it possible to identify the location of people in need of rescue. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-90383 Summary of the Invention [Problem to be solved by the invention]

[0004] However, there is no disclosure of capturing images from an aircraft, analyzing the images, and detecting points of interest. Therefore, one of the objectives of the present disclosure is to provide an image analysis system that captures or observes a wide area from the sky, detects points of interest, and captures the points of interest from the ground and analyzes the details. [Means for solving the problem]

[0005] The image analysis system of the present disclosure includes: a point of interest detection device that analyzes an image of a wide area taken from the sky or a set of numerical data that can be mapped on a plane of a wide area observed from the sky, detects a point of interest from the analyzed image or set of numerical data, and instructs a mobile imaging device that moves from the ground to the point of interest and takes an image of the point of interest to move to the point of interest; An image analysis system analyzes the images captured by the mobile imaging device.

[0006] The attention point detection device of the present disclosure comprises: Analyzing an image of a wide area taken from the sky or a set of numerical data that can be mapped onto a plane of a wide area observed from the sky, and detecting points of interest from the analyzed image or set of numerical data; The point-of-interest detection device instructs a mobile imaging device, which moves from the ground to the point of interest and takes an image of the point of interest, to move to the point of interest.

[0007] The image analysis method of the present disclosure includes: The server Analyze images of a wide area taken from the sky or a collection of numerical data that can be mapped onto a plane observed from the sky, Detecting points of interest from the analyzed image or set of numerical data; instructing a mobile imaging device that moves from the ground to the point of interest and captures an image of the point of interest to move to the point of interest; An image analysis method is provided for analyzing images captured by the mobile imaging device. [Effects of the Invention]

[0008] The present disclosure provides an image analysis system, an interest point detection device, and an image analysis method that photograph or observe a wide area from the sky, detect an interest point, and photograph the interest point from the ground and analyze the details. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a block diagram showing a configuration of an image analysis system according to the present disclosure. [Figure 2] 1 is a schematic diagram of an image analysis system according to the present disclosure. [Figure 3] 1 is a flowchart of an image analysis method according to the present disclosure. [Figure 4] FIG. 1 is a first diagram showing a specific example of an image analysis system according to the present disclosure. [Figure 5] FIG. 2 is a second diagram showing a specific example of an image analysis system according to the present disclosure. [Figure 6]FIG. 3 is a third diagram showing a specific example of an image analysis system according to the present disclosure. [Figure 7] FIG. 1 is a block diagram illustrating a configuration of an information processing device according to the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0010] (Description of Image Analysis System According to an Embodiment) Fig. 1 is a block diagram showing the configuration of an image analysis system according to the present disclosure. Fig. 2 is a schematic diagram of an image analysis system according to the present disclosure. The image analysis system 100 will be described with reference to Figs. 1 and 2. The image analysis system 100 photographs a wide area from the sky, detects points of interest, and photographs the points of interest from the ground to analyze the details.

[0011] As shown in FIG. 1, the image analysis system 100 includes an interest point detection device 101 and a moving image capture device 102.

[0012] The interest point detection device 101 analyzes an image of a wide area photographed from the sky or a set of numerical data that can be mapped on a plane obtained by observing a wide area from the sky, and detects an interest point from the analyzed image or set of numerical data. The interest point detection device 101 is configured, for example, as a server. The server is an information processing device. As shown in FIG. 7, the information processing device 700 includes a processor 701 that executes a program and performs processing, and a memory 702 that stores the program. The information processing device 700 may be configured as a single device or as multiple devices. The information processing device 700 may be a cloud server that executes some or all of its functions in a distributed manner.

[0013] As shown in Figure 2, interest point detection device 101 of image analysis system 100 wirelessly acquires a collection of images or numerical data captured over a wide area from the sky by artificial satellite 201 or flying drone 202. For example, the sky includes space where artificial satellites are located, high altitudes where airplanes fly, and low altitudes where flying drones fly. A wide area means an area wider than the range captured by a mobile imaging device in one shot.

[0014] These images taken from the sky include not only ordinary RGB cameras, but also special images obtained using infrared cameras, spectroscopic cameras, etc., stereo images and three-directional images obtained using stereo cameras or aircraft. The images may also include depth images obtained using laser depth cameras, stereo images obtained using radar, and images in each radio wave wavelength band.

[0015] Furthermore, a set of numerical data that can be mapped onto a plane by observing a wide area from the sky is, for example, a set of numerical data that can be obtained by measuring the earth's surface from an artificial satellite using electromagnetic waves, radio waves, radar, etc., and mapping the strength or characteristics of the electromagnetic waves onto a plane.

[0016] The artificial satellite 201 and the flying drone 202 transmit the position of the point of interest to the point of interest detection device 101. The point of interest is detected, for example, by recognizing a location that has changed over time or by detecting a difference between an object that should be in advance information and an actual object. The point of interest detection device 101 transmits the position of the point of interest to a mobile imaging device 102, such as a drive recorder mounted on a car. The mobile imaging device 102 moves as a car, a person, or a flying drone. The mobile imaging device 102 then captures an image of an object 203 at the point of interest.

[0017] Here, the object 203 is a building, nature, the environment, a car, a bicycle, farmland, a ship, etc. This object 203 is too large for the artificial satellite 201 to analyze sufficiently.

[0018] The point-of-interest detection device 101 then instructs the mobile imaging device 102 to move to the point of interest. The movement instruction may be given by converting the point of interest into a name for a location on the ground. For example, XX prefecture, △△ city, XX town, ???. The movement instruction may also be given by instructing the mobile imaging device on a route to move. For example, 4 km north on National Route XX. There may also be multiple points of interest. In this case, the movement instruction may be given by instructing a circular route to multiple points of interest. The point-of-interest detection device 101 may also transmit public data related to the point of interest to the mobile imaging device 102. The public data may include, for example, the name of a facility such as a dam, water purification plant, waste disposal plant, port, or bicycle parking lot, and building characteristics such as the number of floors. In this way, the mobile imaging device 102 can more easily reach the point of interest.

[0019] The mobile imaging device 102 moves from the ground to a point of interest and captures an image of the point of interest. The mobile imaging device 102 captures images for detailed inspection of the point of interest from the ground. The mobile imaging device 102 may be a drive recorder mounted on a vehicle such as a police vehicle, a construction vehicle, a passenger car, or a commercial vehicle. The mobile imaging device 102 may also be an imaging device mounted on a flying drone, a smartphone carried by a person, a body cam, or an action camera.

[0020] At the point of interest, the mobile image capture device 102 captures an image of the surroundings from a position higher than the surrounding objects, and by comparing this with an image captured over a wide area, it is possible to link the point of interest with the object captured by the mobile image capture device 102. By capturing an image from a position higher than the surrounding objects, it is possible to capture an image over a wide area. This makes it easier to compare this with an image captured from above, and a more detailed image can be obtained.

[0021] Furthermore, an image captured by the mobile imaging device 102 at the point of interest may be converted into an image seen from above, and the converted image may be compared with an image captured over a wide area, thereby linking the point of interest to an object captured by the mobile imaging device 102. The mobile imaging device 102 may capture multiple images. Having multiple images makes it easier to convert the images into an image viewed from above. Therefore, the converted image can be compared with an image captured over a wide area from above.

[0022] The image analysis system 100 analyzes images captured by the mobile image capture device 102. This analysis may be performed by a point of interest detection device. Alternatively, the image analysis may be performed by a separate information processing device 700 that receives data from the point of interest detection device. Image analysis can be performed by a variety of information processing devices 700.

[0023] The above configuration provides an image analysis system that photographs or observes a wide area from the sky, detects points of interest, and photographs the points of interest from the ground to analyze the details.

[0024] (Description of Image Analysis Method According to Embodiment) 3 is a flowchart of the image analysis method according to the present disclosure, which will be described with reference to FIG.

[0025] As shown in Fig. 3, first, a point-of-interest detection device 101, such as a server, analyzes an image of a wide area captured from the sky or a set of numerical data obtained by observing a wide area from the sky (step S301). Next, the point-of-interest detection device 101 detects a point of interest from the analyzed image or set of numerical data (step S302). Next, the point-of-interest detection device 101 instructs the mobile image capture device 102 to move to the point of interest (step S303). Finally, the image analysis system 100 analyzes the image captured by the mobile image capture device 102 (step S304).

[0026] The above configuration provides an image analysis method for photographing or observing a wide area from the sky, detecting a point of interest, and photographing the point of interest from the ground to analyze the details.

[0027] (Explanation of specific examples of image analysis system according to embodiments) Fig. 4 is a first diagram showing a specific example of an image analysis system according to the present disclosure. Fig. 5 is a second diagram showing a specific example of an image analysis system according to the present disclosure. Fig. 6 is a third diagram showing a specific example of an image analysis system according to the present disclosure. Specific examples of the image analysis system according to the present disclosure will be described with reference to Figs. 4 to 6.

[0028] As shown at the top of Figure 4, the Agricultural Land and Forestry Division has a need to determine whether or not there are any violations in permit applications for agricultural land conversion and whether or not the land uses are in line with the intended purpose. Violations of the intended purpose of land use include whether agricultural land has been turned into an illegal garbage dump, whether surplus soil has been illegally dumped, and whether or not an election office has been established on the land.

[0029] To make such a judgment, first, suspicious areas are selected from images taken by a satellite. For example, the interest point detection device 101 detects interest points based on the degree of deviation from hyperspectral data showing farmland, including fallow and uncultivated land. Next, an efficient investigation is carried out by route setting by the server using a drive recorder. For example, the image analysis system 100 analyzes the degree of deviation from farmland using Scene Understanding of the image data.

[0030] As shown in the second figure from the top of Figure 4, the Agricultural Land and Forestry Division, River Division, and Environmental Policy Division need to determine whether refrigerators and other items that easily reflect radio waves have been illegally dumped in forests and rivers. To make such a determination, suspicious areas are first identified using images from a satellite. A point-of-interest detection device 101 detects points of interest based on the electromagnetic wave reflectivity or radiation level of objects in the natural terrain. Next, roadside surveys are conducted using a dashcam and river surveys are conducted using an aerial drone. For example, an image analysis system analyzes the degree of color deviation from natural colors or the degree of artificial shapes from the image data.

[0031] As shown in the third figure from the top in Figure 4, the Urban Planning Guidance Division needs to detect new or additional illegal structures since the time of the building permit. For example, a satellite imagery is stored within 14 days of the building permit and monitored for changes. A point of interest detection device 101 detects points of interest based on the degree of pixel change at the address of the building permit. Next, a dashcam is used to confirm the image, and an inspector equipped with a body camera on the rooftop balcony conducts an on-site inspection. For example, an image analysis system 100 analyzes the degree of difference between the photographed building and the information contained in the application documents and architectural blueprints required for the building confirmation application. The application documents and blueprints required for the building confirmation application include the confirmation application, structural calculation summary, building plan summary, construction notification, layout plan, area calculation diagram, floor plan, elevation, cross section, and composition.

[0032] As shown in the fourth row from the top of Figure 4, the River Division needs to determine whether fields have been cultivated again on the riverbed since enforcement. To make such a determination, a satellite first identifies suspicious areas based on changes such as the absence of weeds. The interest point detection device 101 detects interest points that indicate the degree of artificial management using satellite images. Artificial management means a lack of weeds, the presence of a tool shed, etc. Next, an aerial drone flies and surveys a group of points on the riverbed that require inspection. For example, an image analysis system can analyze the degree of artificial management using images.

[0033] As shown in the fifth figure from the top of Figure 4, the River Department needs to determine whether large objects that require removal have washed up on the riverbed after a typhoon. To make such a determination, suspicious areas are first selected based on whether there are any additional objects in wide-area satellite images. An interest point detection device 101 detects interest points based on the degree of pixel change compared to past images. Next, an aerial drone flies and investigates a group of points on the riverbed that require checking. For example, an image analysis system analyzes the degree of object detection.

[0034] As shown in the sixth row from the top of Figure 4, the River Division needs to detect long-term sediment accumulation in sabo dams and sudden changes due to typhoons, heavy rain, etc., and determine whether action is required based on micro-level images. To make such a determination, satellite images are first used to observe sediment accumulation over time and identify suspicious areas. The interest point detection device 101 detects interest points from satellite images based on the overall increase or decrease in the height or width of natural terrain or the height or width of natural objects. Next, after typhoons and heavy rains, satellites, dashcams, and flying drones capture images on demand. For example, an image analysis system analyzes the degree of change in the topographical data on demand.

[0035] As shown in the seventh row from the top of Figure 4, the Road Management Department needs to use satellites and drones to create a list of areas with poor drainage after heavy rain. This list is created by first selecting suspected areas using satellite images. The interest point detection device 101 detects interest points based on the degree of reflection that differs from normal water. Next, the suspected cause areas are classified by a score from images taken by the flying drone and the dashcam. The classification is based on whether or not an analysis needs to be carried out quickly and efficiently before the water subsides. For example, an image analysis system can analyze the degree of change by comparing past image data, such as dashcam images, with on-demand image data.

[0036] As shown in the top row of Figure 5, the Urban Planning and Policy Division needs to detect groups of suspected abandoned bicycles in public spaces. The Division can improve the accuracy of identifying abandoned bicycles from micro-level images, identify priority areas for enforcement, tag them, and implement a recovery process. To achieve this, they first identify areas for priority enforcement of abandoned bicycles using wide-area images. The interest point detection device 101 detects interest points based on the degree of score that indicates a group of suspected abandoned bicycles using satellite or aerial drone images. Next, a dashcam or action camera virtually tags or identifies bicycles of repeat offenders in the area, records their registration numbers, and scores abandoned bicycles based on the degree of rust. For example, an image analysis system can analyze the volume of illegally parked bicycles.

[0037] As shown in the second figure from the top of Figure 5, the Fisheries and Coastal Affairs Division and the River Division have a need to detect illegal vessel storage in rivers and ports. To achieve this, they first create a list of suspicious moorings from satellite images. The interest point detection device 101 detects interest points based on the degree of deviation from the distribution of vessels with mooring permission, or the degree of deviation from the actual situation and mooring facility use permits, port entry / departure notifications, etc. Next, the ship's registration number is recorded using an airborne drone and a drive recorder. The airborne drone and drive recorder also record information such as the type of vessel and fishing gear. For example, an image analysis system analyzes the degree of the judgment score for the vessel.

[0038] As shown in the third figure from the top of Figure 5, police parking control departments and highway management organizations have a need to detect illegal and nuisance parking. For example, they need to identify trucks parking for long periods of time near service areas. To detect such illegal and nuisance parking, they first create a list of common parking locations over a wide area from satellite images. The interest point detection device 101 detects interest points by measuring the degree to which car objects are identified from satellite images and the number of vehicles within the area. Next, a drive recorder is used to conduct a focused investigation of the common parking locations. For example, an image analysis system can determine whether a vehicle is parked rather than in a traffic jam, and measure and analyze the parking time. For example, a parked vehicle can be identified by the presence or absence of a driver, and the parking time can be measured by patrols.

[0039] As shown in the top row of Figure 6, the Green Spaces and Parks Division has a need to detect suspected herbicide use and pest damage to planted areas in public areas. First, satellite images are used to detect wilting and other issues in specific locations of the same type of plant. An interest point detection device 101 detects interest points from the satellite images based on the degree to which they deviate from other planted areas. Next, a drive recorder acquires detailed images. For example, an image analysis system analyzes whether the observed points are the result of human intervention, such as replanting.

[0040] As shown in the second image from the top of Figure 6, the Road Management Department needs to analyze the amount of rust on infrastructure facilities such as footbridges and bridge steel frames using satellites and perform microscopic image analysis using action cameras. For such analysis, the degree of deterioration is first detected from satellite images. An interest point detection device 101 detects interest points by measuring the degree of increase in the amount of rust using hyperspectral data. Next, a detailed investigation is carried out using a drive recorder and an aerial drone. For example, an image analysis system analyzes the degree of rust at a more detailed pixel level.

[0041] As shown in the third row from the top of Figure 6, the Housing and Maintenance Division and the Urban Planning Guidance Division have a need to detect abandoned buildings, homeless camps, and illegal dumping in abandoned buildings. Abandoned buildings are areas with holes in the roof or gardens overgrown with weeds. To detect such areas, a list of abandoned buildings and suspected abandoned buildings is first compiled from satellite images. The interest point detection device 101 detects interest points based on the degree of similarity with training data on ruins, abandoned buildings, and homeless camps. Next, a detailed investigation is conducted using a drive recorder. For example, an image analysis system uses Scene Understanding to analyze whether the building is an abandoned building, a homeless camp, or a ruin.

[0042] Some or all of the processes in the attention point detection device 101 and the information processing device 700 described above can be realized as a computer program. Such a program can be stored in various types of non-transitory computer-readable media and provided to a computer. Non-transitory computer-readable media include various types of tangible recording media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The program may also be provided to a computer by various types of temporary computer-readable media. Examples of temporary computer-readable media include electrical signals, optical signals, and electromagnetic waves. The temporary computer-readable media can provide the program to a computer via a wired communication path such as an electric wire or optical fiber, or via a wireless communication path.

[0043] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0044] Each drawing is merely an example for describing one or more embodiments. Each drawing may relate not only to one particular embodiment, but also to one or more other embodiments. As will be understood by those skilled in the art, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings to create, for example, an embodiment not explicitly shown or described. Not all features or steps shown in any one drawing are necessary to describe an exemplary embodiment, and some features or steps may be omitted. The order of steps described in any drawing may be changed as appropriate.

[0045] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes. (Appendix 1) a point of interest detection device that analyzes an image of a wide area taken from the sky or a set of numerical data that can be mapped on a plane of a wide area observed from the sky, detects a point of interest from the analyzed image or set of numerical data, and instructs a mobile imaging device that moves from the ground to the point of interest and takes an image of the point of interest to move to the point of interest; an image analysis system that analyzes the images captured by the mobile imaging device; (Appendix 2) The image analysis system according to claim 1, wherein the interest point detection device converts the interest point into a name of a position on the ground and instructs the mobile imaging device to move. (Appendix 3) The image analysis system according to claim 1, wherein the interest point detection device instructs the mobile imaging device on a movement path. (Appendix 4) The image analysis system of claim 1, wherein the interest point detection device detects a plurality of interest points and instructs the mobile imaging device on a route to the plurality of interest points. (Appendix 5) 2. The image analysis system of claim 1, wherein the interest point detection device transmits public data related to the interest point to the mobile imaging device. (Appendix 6) The image analysis system described in Appendix 1, wherein at the point of interest, the mobile imaging device captures an image of the surroundings from a position higher than the surrounding objects, and compares the image with the image of a wide area, thereby linking the point of interest with the object captured by the mobile imaging device. (Appendix 7) The image analysis system described in Appendix 1, wherein the image captured by the mobile imaging device at the point of interest is converted into an image viewed from above and compared with the image captured over a wide area, thereby linking the point of interest to the object captured by the mobile imaging device. (Appendix 8) Using the image captured over a wide area, the point of interest is detected based on the degree of deviation from hyperspectral data representing farmland; The image analysis system described in Appendix 1 uses images captured by the mobile imaging device to analyze the degree of deviation from agricultural land using Scene Understanding. (Appendix 9) Using the image captured over a wide area, the point of interest is detected based on the reflectivity or radiation level of electromagnetic waves of an object within the natural terrain; An image analysis system according to claim 1, which uses images captured by the mobile imaging device to analyze the degree of deviation from natural colors or the degree of artificial shapes. (Appendix 10) Using the image captured over a wide area, the attention point is detected based on the degree of change in pixels of the address of the building permit; The image analysis system described in Appendix 1 uses images captured by the mobile imaging device to analyze the degree of difference between the photographed building and an architectural blueprint. (Appendix 11) Using the image captured over a wide area, the attention point is detected based on the degree of artificial management; 2. The image analysis system of claim 1, which uses images captured by the mobile imaging device to analyze the degree of artificial management. (Appendix 12) Using the image captured over a wide area, the attention point is detected based on the degree of pixel change compared to the past; 2. The image analysis system according to claim 1, wherein the degree of object determination is analyzed using images captured by the mobile imaging device. (Appendix 13) Using the image captured over a wide area, the attention point is detected based on the overall degree of increase or decrease in the height or width of a terrain or the height or width of a natural object; An image analysis system according to claim 1, which uses images captured by the mobile imaging device to analyze the degree of change in topographical data on demand. (Appendix 14) Using the image captured over a wide area, the point of interest is detected based on an unusual degree of reflection caused by water; The image analysis system of claim 1, which uses images captured by the mobile imaging device to analyze the degree of change by comparing past images with on-demand images. (Appendix 15) Using the image captured over a wide area, the attention points are detected based on the degree of score that indicates that the group of vehicles is suspected of being abandoned bicycles; The image analysis system described in Appendix 1 analyzes the amount of illegally parked bicycles using images captured by the mobile imaging device. (Appendix 16) Using the image of a wide area, the attention points are detected based on the degree of deviation from the distribution of ships permitted to moor, or the degree of deviation from the actual situation and the mooring facility use permit, port entry / departure notification, etc. An image analysis system as described in Appendix 1, which uses images captured by the mobile imaging device to analyze the degree of judgment score with a ship. (Appendix 17) Using the image captured over a wide area, the attention point is detected by measuring the degree of a judgment score of a vehicle object or the number of vehicles within the range; The image analysis system described in Appendix 1 uses images captured by the mobile imaging device to determine whether the vehicle is parked rather than in a traffic jam, and measures and analyzes the parking time. (Appendix 18) Using the image captured over a wide area, the attention point is detected based on the degree of deviation from other planting locations; An image analysis system as described in Appendix 1, which uses images captured by the mobile imaging device to analyze whether the work is human-made, such as replanting. (Appendix 19) Using the image captured over a wide area, the points of interest are detected based on the degree of increase in the amount of rust on infrastructure equipment using hyperspectral data; An image analysis system as described in Appendix 1, which uses images taken by the mobile imaging device to analyze the degree of rust on infrastructure equipment at a detailed pixel level. (Appendix 20) Using the image captured over a wide area, the points of interest are detected based on their similarity to ruins, abandoned houses, and homeless camps; An image analysis system as described in Appendix 1, which uses images taken by the mobile imaging device to analyze whether the images are ruins, abandoned houses, or homeless camps using Scene Understanding. (Appendix 21) Analyzing an image of a wide area taken from the sky or a set of numerical data that can be mapped onto a plane of a wide area observed from the sky, and detecting points of interest from the analyzed image or set of numerical data; A point of interest detection device instructs a mobile imaging device, which moves from the ground to the point of interest and takes an image of the point of interest, to move to the point of interest. (Appendix 22) The server Analyze images of a wide area taken from the sky or a collection of numerical data that can be mapped onto a plane observed from the sky, Detecting points of interest from the analyzed image or set of numerical data; instructing a mobile imaging device that moves from the ground to the point of interest and captures an image of the point of interest to move to the point of interest; An image analysis method for analyzing images captured by the mobile imaging device.

[0046] Some or all of the elements (e.g., configurations and functions) described in Supplementary Note 2 to Supplementary Note 20 that are dependent on Supplementary Note 1 {e.g., system} may also be dependent on Supplementary Note 21 {e.g., device} and Supplementary Note 22 {e.g., method} in the same dependency relationship as Supplementary Note 2 to Supplementary Note 20. Some or all of the elements described in any Supplementary Note may be applied to various hardware, software, recording means for recording software, systems, and methods. [Explanation of symbols]

[0047] 100 Image analysis system, 101 Interest point detection device, 102 Moving imaging device, 201 Artificial satellite, 202 Flying drone, 203 Object

Claims

1. a point of interest detection device that analyzes an image of a wide area taken from the sky or a set of numerical data that can be mapped on a plane of a wide area observed from the sky, detects a point of interest from the analyzed image or set of numerical data, and instructs a mobile imaging device that moves from the ground to the point of interest and takes an image of the point of interest to move to the point of interest; an image analysis system that analyzes the images captured by the mobile imaging device;

2. 2. The image analysis system according to claim 1, wherein the interest point detection device converts the interest point into a name of a position on the ground and instructs the mobile imaging device to move.

3. The image analysis system according to claim 1 , wherein the interest point detection device instructs the mobile imaging device on a movement path.

4. The image analysis system according to claim 1 , wherein the interest point detection device detects a plurality of interest points and instructs the mobile imaging device on a route to the plurality of interest points.

5. The image analysis system of claim 1 , wherein the point of interest detection device transmits public data related to the point of interest to the mobile imaging device.

6. 2. The image analysis system according to claim 1, wherein the mobile imaging device captures an image of the surroundings at the point of interest from a position higher than the surrounding objects, and compares the image with the image of a wide area, thereby linking the point of interest to the object captured by the mobile imaging device.

7. The image analysis system according to claim 1, wherein the image captured by the mobile imaging device at the point of interest is converted into an image viewed from above and compared with the image captured over a wide area, thereby linking the point of interest to the object captured by the mobile imaging device.

8. Using the image captured over a wide area, the point of interest is detected based on the degree of deviation from hyperspectral data representing farmland; The image analysis system according to claim 1 , wherein the image captured by the mobile imaging device is used to analyze the degree of deviation from farmland using Scene Understanding.

9. Analyzing an image of a wide area taken from the sky or a set of numerical data that can be mapped onto a plane of a wide area observed from the sky, and detecting points of interest from the analyzed image or set of numerical data; A point of interest detection device instructs a mobile imaging device, which moves from the ground to the point of interest and takes an image of the point of interest, to move to the point of interest.

10. The server Analyze images of a wide area taken from the sky or a collection of numerical data that can be mapped onto a plane observed from the sky, Detecting points of interest from the analyzed image or set of numerical data; instructing a mobile imaging device that moves from the ground to the point of interest and captures an image of the point of interest to move to the point of interest; An image analysis method for analyzing images captured by the mobile imaging device.

Images