WO2018216623A1 - Earth science data analyzing device, earth science data analyzing method, and computer-readable recording medium - Google Patents

Earth science data analyzing device, earth science data analyzing method, and computer-readable recording medium Download PDF

Info

Publication number
WO2018216623A1
WO2018216623A1 PCT/JP2018/019343 JP2018019343W WO2018216623A1 WO 2018216623 A1 WO2018216623 A1 WO 2018216623A1 JP 2018019343 W JP2018019343 W JP 2018019343W WO 2018216623 A1 WO2018216623 A1 WO 2018216623A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
earth science
specific
geoscience
specific area
Prior art date
Application number
PCT/JP2018/019343
Other languages
French (fr)
Japanese (ja)
Inventor
晨暉 黄
田能村　昌宏
武仁 芳木
Original Assignee
日本電気株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 日本電気株式会社 filed Critical 日本電気株式会社
Publication of WO2018216623A1 publication Critical patent/WO2018216623A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V8/00—Prospecting or detecting by optical means
    • G01V8/10—Detecting, e.g. by using light barriers
    • G01V8/12—Detecting, e.g. by using light barriers using one transmitter and one receiver

Definitions

  • the present invention relates to an earth science data analysis apparatus and an earth science data analysis method for analyzing earth science data indicating characteristics of a specific region, for example, the content of a substance existing on the ground surface, and further, The present invention relates to a computer-readable recording medium on which a program for realizing is recorded.
  • Patent Literature 1 discloses a ground estimation method for estimating a geological distribution and geological properties of a place where no boring is performed using boring data acquired at a plurality of places.
  • the contour map of the geological characteristic value for each stratum of the area to be estimated based on the geological characteristic value regarding each layer included in each boring data. is generated.
  • the area to be estimated is an area including a place where the bowling is performed. In other words, the bowling is performed at a plurality of locations in the area to be estimated.
  • the position of the estimated ground location is collated in the contour map of each region, and the geological characteristic value there is estimated. After that, the estimated geological characteristic values of each region are displayed.
  • An example of an object of the present invention is to solve the above-described problem and use the earth science data acquired in a certain area to estimate the earth science data in another area, the earth science data analysis apparatus and the earth science data analysis method And providing a computer-readable recording medium.
  • an earth science data analysis apparatus provides: A data acquisition unit for acquiring geoscience data indicating the characteristics of the specific area and satellite data indicating the characteristics of the specific area; Using the acquired earth science data and the satellite data, learning the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data, and generating a learning model A learning model generation unit; It is characterized by having.
  • an earth science data analysis method includes: (A) obtaining geoscience data indicating characteristics of a specific area, and satellite data indicating characteristics of the specific area; (B) Correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data using the earth science data and the satellite data acquired in the step (a). Learning relationships and generating learning models, steps, It is characterized by having.
  • a computer-readable recording medium On the computer, (A) obtaining geoscience data indicating characteristics of a specific area, and satellite data indicating characteristics of the specific area; (B) Correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data using the earth science data and the satellite data acquired in the step (a). Learning relationships and generating learning models, steps, A program including an instruction for executing is recorded.
  • FIG. 1 is a block diagram schematically showing the configuration of the earth science data analysis apparatus according to Embodiment 1 of the present invention.
  • FIG. 2 is a block diagram specifically showing the configuration of the earth science data analysis apparatus according to Embodiment 1 of the present invention.
  • FIG. 3 is a flowchart showing the operation of the earth science data analysis apparatus according to Embodiment 1 of the present invention.
  • FIG. 4 is a flowchart showing the operation of the earth science data analysis apparatus according to the second embodiment of the present invention.
  • FIG. 5 is a block diagram showing an example of a computer that implements the earth science data analysis apparatus according to the first and second embodiments of the present invention.
  • FIG. 6 is a diagram showing an example of sample data at a specific point used in the embodiment of the present invention.
  • FIG. 1 is a block diagram schematically showing the configuration of the earth science data analysis apparatus according to Embodiment 1 of the present invention.
  • FIG. 2 is a block diagram specifically showing the configuration of the earth science data analysis apparatus according to
  • FIG. 7 is a diagram showing an example of satellite data.
  • FIG. 7 (a) shows the reflectance distribution of light in the infrared region
  • FIG. 7 (b) shows elevation data
  • FIG. Geomagnetic measurement data is shown.
  • FIG. 8 is a diagram showing an example of a set of sample data used in the embodiment of the present invention.
  • FIG. 9 is a diagram illustrating an example of a specific area in which geoscience data is acquired and other areas.
  • FIG. 10 is a diagram illustrating the earth science data in which the deficit is corrected by the learning model generation unit.
  • FIG. 11 is a diagram illustrating a specific region after the earth science data is estimated by the data estimation unit and other regions.
  • Embodiment 1 an earth science data analysis apparatus, an earth science data analysis method, and a program according to Embodiment 1 of the present invention will be described with reference to FIGS.
  • FIG. 1 is a block diagram schematically showing the configuration of the earth science data analysis apparatus according to Embodiment 1 of the present invention.
  • the earth science data analysis device 10 is a device for analyzing earth science data.
  • the earth science data analysis device 10 according to the first embodiment shown in FIG.
  • the earth science data analysis apparatus 10 includes a data acquisition unit 11 and a learning model generation unit 12.
  • the data acquisition unit 11 acquires the earth science data indicating the characteristics of the specific area and the satellite data indicating the characteristics of the specific area.
  • the learning model generation unit 12 uses the acquired earth science data and satellite data to learn the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data, and obtains the learning model. Generate.
  • the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data is learned. For this reason, if satellite data in a region other than the specific region is applied to the learning model obtained by learning, it is possible to estimate geoscience data in a region other than the specific region. That is, according to the first embodiment, it is possible to estimate the earth science data using the earth science data acquired in another area in the area where the earth science data is not acquired.
  • FIG. 2 is a block diagram specifically showing the configuration of the earth science data analysis apparatus according to Embodiment 1 of the present invention.
  • the earth science data that can be used in the first embodiment, data indicating the existence of resources as characteristics of a specific region, for example, substances existing on the earth's surface, types of elements, component ratios, contents, etc. Data. Specifically, assuming that the prediction of the presence of copper in a certain area is required, the earth science data includes data indicating the copper content (ppm) per unit area, which is a characteristic of a specific area. .
  • Other geoscience data include gravity value, carbon dioxide concentration profile, temperature, humidity, wind direction, wind speed, atmospheric pressure, global solar radiation, spectral radiation, photosynthetic effective radiation, earth temperature, soil moisture, geothermal heat, direct delivery Data including radiation spectrum, ground stability, geological age, fault information, groundwater vein information, plant type distribution, evapotranspiration information, mineral production, etc.
  • the earth science data includes data indicating the abundance ratio of the element to be grasped. Can be mentioned.
  • Satellite data is data obtained from the sky above the earth, and is data indicating the characteristics of a specific area.
  • the satellite data includes data acquired by a satellite and data acquired by an aircraft such as an aircraft.
  • the satellite data that can be used in the first embodiment includes data indicating the intensity of electromagnetic waves reflected or emitted from the area to be acquired, data indicating the reflectance distribution of light of a specific wavelength, and geomagnetism.
  • the data indicating the reflectance distribution of light of a specific wavelength includes data measured by an aster (ASTER: “Advanced” Spaceborne “Thermal” Emission “and Reflection” Radiometer).
  • An aster is an optical sensor for observation mounted on the Terra satellite of NASA in the United States, and can observe 14 bands from visible to thermal infrared.
  • the 14 bands are wavelengths suitable for capturing a characteristic spectrum related to minerals.
  • the satellite data is not limited to the above, and includes data obtained by remote sensing.
  • the earth science data analysis apparatus 10 includes a data estimation unit 13, a display unit 14, and a data acquisition unit 11 and a learning model generation unit 12.
  • the storage unit 15 is provided.
  • a display device 20 is connected to the earth science data analysis device 10.
  • the data acquisition unit 11 acquires the earth science data and satellite data from the database 30 and passes the acquired earth science data and satellite data to the learning model generation unit 12.
  • the database 30 is connected to the earth science data analysis apparatus 10 via a network.
  • the database 30 stores earth science data and satellite data in a specific area.
  • geoscience data is data indicating the copper content (ppm) per unit area at each point
  • satellite data is data indicating the reflectance distribution of light at a specific wavelength, altitude data, and altitude slope
  • the database 30 stores data indicating the copper content (ppm) per unit area as geoscience data for each point (latitude and longitude), and reflects the reflectance of light of a specific wavelength as satellite data.
  • Store elevation values, and slope values In this case, an area obtained by superimposing a set range centering on a point where the earth science data and satellite data are acquired can be set as the specific area.
  • the value of the earth science data and the value of the satellite data for each point are associated with each other as one set. Furthermore, the value of the earth science data and the value of the satellite data constituting one set are handled as one sample data. In addition, since satellite data can be acquired over a wider range than earth science data, it may cover areas other than the specific area where earth science data is acquired.
  • the learning model generation unit 12 first receives a plurality of sample data from the data acquisition unit 11, and executes machine learning using each received sample data as teacher data.
  • the machine learning method used in the first embodiment includes a decision tree, support vector machine, neural network, logistic regression, nearest neighbor classification (K-NN), ensemble classification learning method, discrimination Analysis and the like.
  • the learning model generation unit 12 gives each sample data to the support vector machine, and the relationship between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data, for example, copper The relationship between the content (ppm) and the reflectance, altitude value, and slope value of light of a specific wavelength is learned. And the learning model production
  • the learning model generation unit 12 performs deep learning using each sample data, thereby determining the copper content according to the reflectance, altitude value, and slope value of light of a specific wavelength. It is also possible to create a classifier and use the created classifier as the learning model 16.
  • the data estimation unit 13 uses the learning model 16 generated by the learning model generation unit 12 to estimate geoscience data in a region other than the specific region (hereinafter referred to as “estimation region”).
  • the data estimation unit 13 estimates the earth science data in the estimation region by acquiring the satellite data in the estimation region and applying the acquired satellite data in the estimation region to the learning model 16.
  • the data estimation unit 13 first selects a plurality of points (latitude and longitude) on the estimation area when the estimation area is designated from the outside. Next, the data estimation unit 13 specifies the reflectance, altitude value, and slope value of light of a specific wavelength corresponding to the selected point from the satellite data stored in the database 30, and the copper content is blank. Create sample data. Then, the data estimation unit 13 applies the created sample data to the learning model 16 to calculate the blank copper content.
  • the display unit 14 displays the earth science data in the specific area and the earth science data in the estimation area on the map data on the screen of the display device 20 so as to overlap each other. For example, if the earth science data is the copper content (ppm) per unit area for each point, the copper content (predicted value) is also displayed on the screen for points where the copper content is not specified. ) Is displayed. For this reason, the user can formulate an efficient mining plan.
  • FIG. 3 is a flowchart showing the operation of the earth science data analysis apparatus according to Embodiment 1 of the present invention.
  • FIGS. 1 and 2 are referred to as appropriate.
  • the earth science data analysis method is implemented by operating the earth science data analysis apparatus 10. Therefore, the description of the earth science data analysis method in the first embodiment is replaced with the following description of the operation of the earth science data analysis apparatus 10.
  • the data acquisition unit 11 acquires geoscience data and satellite data in a specific area from the database 30 (step A1).
  • step A1 the data acquisition unit 11 acquires sample data for each point included in the specific area from the database 30, and passes the acquired sample data for each point to the learning model generation unit 12.
  • the learning model generation unit 12 learns the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data by using the earth science data and the satellite data acquired in step A1. Then, the learning model 16 is generated (step A2).
  • the learning model generation unit 12 when the learning model generation unit 12 receives the sample data for each point acquired in step A1, the learning model generation unit 12 performs machine learning using the received sample data as teacher data, thereby generating the learning model 16 To do. Further, the learning model generation unit 12 stores the generated learning model 16 in the storage unit 15.
  • the data estimation unit 13 estimates the earth science data in a region (estimated region) other than the specific region using the learning model 16 generated by the learning model generation unit 12 (step A3).
  • the data estimation unit 13 selects a plurality of points (latitude and longitude) on the estimation area. Next, the data estimation unit 13 specifies satellite data corresponding to the selected point from the satellite data stored in the database 30, and creates sample data in which the earth science data is blank. Then, the data estimation unit 13 applies the created sample data to the learning model 16 to calculate blank earth science data.
  • the display unit 14 displays the earth science data in the specific area and the earth science data in the estimation area on the map data on the screen of the display device 20 (step A4).
  • a learning model is generated that indicates the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data. It is possible to estimate geoscience data even in areas that have not been acquired.
  • the program in the first embodiment may be a program that causes a computer to execute steps A1 to A4 shown in FIG.
  • a CPU Central Processing Unit
  • the earth science data analysis apparatus 10 and the earth science data analysis method according to the first embodiment can be realized.
  • a CPU Central Processing Unit
  • the computer functions as the data acquisition unit 11, the learning model generation unit 12, the data estimation unit 13, and the display unit 14 to perform processing.
  • each computer may function as any one of the data acquisition unit 11, the learning model generation unit 12, the data estimation unit 13, and the display unit 14, respectively.
  • Embodiment 2 Next, an earth science data analysis device, an earth science data analysis method, and a program according to Embodiment 2 of the present invention will be described with reference to FIG.
  • FIG. 1 and FIG. Refer to In the following description, differences from the first embodiment will be mainly described.
  • the learning model generation unit 12 corrects the deficiency of the earth science data in the specific region using the mathematical model before the generation of the learning model 16. Then, the learning model generation unit 12 uses the corrected earth science data to learn the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data, and generates a learning model. To do.
  • the learning model generation unit 12 supplements the earth science data at a point where the satellite data is acquired but the earth science data is not acquired. In other words, the learning model generation unit 12 supplements the value of the earth science data in the sample data of the specific region where the value of the earth science data is missing.
  • each sample data is, for example, the point where it is associated. Classify into one or more groups based on latitude and longitude.
  • the learning model generation unit 12 selects, for example, three sample data, and specifies three points using the latitude and longitude values associated with each of the selected three sample data. Further, the learning model generation unit 12 calculates the area of the region determined by the three specified points, and determines whether the calculated area is equal to or less than a predetermined threshold.
  • the learning model generation unit 12 classifies the three sample data into one group. Conversely, if the obtained area is larger than the threshold, It is determined that the three sample data do not belong to one group.
  • the learning model generation unit 12 classifies the plurality of sample data into one or a plurality of groups by repeating these series of processes.
  • the learning model generation unit 12 determines, for each group, whether there is any sample data in which the value of the earth science data is missing. As a result of the determination, if there is sample data with missing earth science data value, out of the sample data included in the group, the value of sample data with missing earth science data value is Use to fill in missing sample data values.
  • methods for supplementing values include neighborhood data interpolation, linear interpolation, polynomial interpolation, non-parametric regression interpolation, and the like.
  • the proximity data interpolation is a method of obtaining the distance between the point C to be interpolated and the nearby points A and B, and interpolating the point C with the data of the points that are close to the point C.
  • This is a method of interpolating data.
  • the interpolation method by nonparametric regression is a method of creating a nonparametric regression model using all data and estimating gap data of all data using the created nonparametric regression model.
  • FIG. 4 is a flowchart showing the operation of the earth science data analysis apparatus according to the second embodiment of the present invention.
  • the earth science data analysis method is implemented by operating the earth science data analysis apparatus. Therefore, the explanation of the earth science data analysis method in the second embodiment is replaced with the following explanation of the operation of the earth science data analysis apparatus.
  • Step B1 is the same as step A1 shown in FIG.
  • the learning model generation unit 12 corrects the deficiency of the earth science data acquired in Step A1 using a preset mathematical model (Step B2). Specifically, the learning model generation unit 12 supplements the earth science data at a point where the satellite data is acquired but the earth science data is not acquired.
  • the learning model generation unit 12 uses the geoscience data corrected in step B2 and the satellite data acquired in step A1, and the characteristics of the specific area indicated by the geoscience data and the specific area indicated by the satellite data.
  • the learning model 16 is generated by learning the correlation with the characteristics (step B3). Step B3 is the same as step A2 shown in FIG.
  • Step B4 is the same as step A3 shown in FIG.
  • Step B5 is the same as step A4 shown in FIG.
  • the learning model is generated after the missing of the earth science data is supplemented, the estimation of the earth science data in the region where the earth science data is not acquired is more accurate. It will be expensive.
  • the program in the second embodiment may be a program that causes a computer to execute steps B1 to B5 shown in FIG.
  • the CPU Central Processing Unit
  • the CPU functions as the data acquisition unit 11, the learning model generation unit 12, the data estimation unit 13, and the display unit 14 to perform processing.
  • each computer may function as any one of the data acquisition unit 11, the learning model generation unit 12, the data estimation unit 13, and the display unit 14, respectively.
  • FIG. 5 is a block diagram showing an example of a computer that implements the earth science data analysis apparatus according to the first and second embodiments of the present invention.
  • the computer 110 includes a CPU 111, a main memory 112, a storage device 113, an input interface 114, a display controller 115, a data reader / writer 116, and a communication interface 117. These units are connected to each other via a bus 121 so that data communication is possible.
  • the computer 110 may include a GPU (GraphicsGraphProcessing Unit) or an FPGA (Field-Programmable Gate Array) in addition to or instead of the CPU 111.
  • the CPU 111 performs various operations by developing the program (code) in the present embodiment stored in the storage device 113 in the main memory 112 and executing them in a predetermined order.
  • the main memory 112 is typically a volatile storage device such as a DRAM (Dynamic Random Access Memory).
  • the program in the present embodiment is provided in a state of being stored in a computer-readable recording medium 120. Note that the program in the present embodiment may be distributed on the Internet connected via the communication interface 117.
  • the storage device 113 includes a hard disk drive and a semiconductor storage device such as a flash memory.
  • the input interface 114 mediates data transmission between the CPU 111 and an input device 118 such as a keyboard and a mouse.
  • the display controller 115 is connected to the display device 119 and controls display on the display device 119.
  • the data reader / writer 116 mediates data transmission between the CPU 111 and the recording medium 120, and reads a program from the recording medium 120 and writes a processing result in the computer 110 to the recording medium 120.
  • the communication interface 117 mediates data transmission between the CPU 111 and another computer.
  • the recording medium 120 include general-purpose semiconductor storage devices such as CF (Compact Flash (registered trademark)) and SD (Secure Digital), magnetic recording media such as a flexible disk, or CD- Optical recording media such as ROM (Compact Disk Read Only Memory) are listed.
  • CF Compact Flash
  • SD Secure Digital
  • magnetic recording media such as a flexible disk
  • CD- Optical recording media such as ROM (Compact Disk Read Only Memory) are listed.
  • the earth science data analysis apparatus can be realized by using hardware corresponding to each unit, not a computer in which a program is installed. Furthermore, part of the earth science data analysis apparatus may be realized by a program, and the remaining part may be realized by hardware.
  • FIG. 6 is a diagram showing an example of sample data at a specific point used in the embodiment of the present invention.
  • FIG. 7 is a diagram showing an example of satellite data.
  • FIG. 7 (a) shows the reflectance distribution of light in the infrared region
  • FIG. 7 (b) shows elevation data
  • FIG. Geomagnetic measurement data is shown.
  • FIG. 8 is a diagram showing an example of a set of sample data used in the embodiment of the present invention.
  • FIG. 9 is a diagram illustrating an example of a specific area in which geoscience data is acquired and other areas.
  • FIG. 10 is a diagram illustrating the earth science data in which the deficit is corrected by the learning model generation unit.
  • FIG. 11 is a diagram illustrating a specific region after the earth science data is estimated by the data estimation unit and other regions.
  • white broken lines indicate the boundaries of administrative divisions.
  • the database 30 registers a plurality of sample data shown in FIG.
  • the sample data includes a point (latitude and longitude), and corresponding earth science data and satellite data.
  • the geoscience data includes the copper content (ppm) per unit area
  • the satellite data includes the reflectance of light of a specific wavelength (Aster band data Band 1, Aster band data Band 14, Aster Including band reciprocal data (Band (1 ⁇ -1)), elevation value, and slope value.
  • satellite data is acquired in a wide range.
  • some of the sample data registered in the database 30 may lack earth science data (copper content). That is, as shown in FIG. 9, there is a point where the earth science data is not acquired even on the specific region. In other words, the copper content is acquired as geoscience data at the point of white spot, but the copper content is not acquired at the point without point.
  • the learning model generation unit 12 supplements the value of the earth science data in the sample data of the specific area where the value of the earth science data is missing.
  • the specific area where the earth science data is acquired is as shown in FIG.
  • the learning model generation unit 12 performs machine learning using a plurality of corrected sample data (see FIGS. 8 and 9) as teacher data.
  • a learning model 16 that specifies the correlation between the copper content (ppm) per unit area and the reflectance, altitude value, and slope value of light of a specific wavelength is generated. .
  • the data estimation unit 13 estimates the earth science data in the estimation region using the learning model 16 generated by the learning model generation unit 12.
  • an area where no white point exists in FIG. 10 is set as an estimated area.
  • the display unit 14 displays the earth science data in the specific area and the earth science data in the estimation area on the screen of the display device 20 so as to overlap the map data.
  • the display unit 14 since the copper content of the area
  • a learning model generation unit for generating An earth science data analysis apparatus characterized by comprising:
  • Appendix 3 Using the learning model, further comprising a data estimation unit for estimating geoscience data in a region other than the specific region, The earth science data analyzer according to appendix 1 or 2.
  • Appendix 4 A display unit for displaying the geoscience data in the specific region and the geoscience data in a region other than the estimated specific region on the screen by superimposing on the map data;
  • the earth science data analysis apparatus according to appendix 3.
  • the geoscience data is data indicating the presence of a specific substance in the specific region as a characteristic of the specific region;
  • the satellite data is data indicating a reflectance distribution of light of a specific wavelength in the specific region as a characteristic of the specific region.
  • the earth science data analysis device according to any one of appendices 1 to 4.
  • the method further includes the step of displaying the geoscience data in the specific area and the geoscience data in the area other than the estimated specific area on the screen so as to overlap the map data.
  • the geoscience data is data indicating the presence of a substance in the specific region as a characteristic of the specific region
  • the satellite data is data indicating a reflectance distribution of light of a specific wavelength in the specific region as a characteristic of the specific region.
  • the geoscience data is data indicating the presence of a substance in the specific region as a characteristic of the specific region
  • the satellite data is data indicating a reflectance distribution of light of a specific wavelength in the specific region as a characteristic of the specific region.
  • a computer-readable recording medium according to any one of appendices 11 to 14.
  • the present invention it is possible to estimate the earth science data of another area using the earth science data acquired in a certain area.
  • the present invention is useful for, for example, mining of mineral resources, ground survey, vegetation survey, and the like.

Landscapes

  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Geophysics (AREA)
  • Geophysics And Detection Of Objects (AREA)

Abstract

An earth science data analyzing device 10 is provided with: a data acquiring unit 11 which acquires earth science data indicating a characteristic of a specific region, and satellite data indicating a characteristic of the specific region; and a learning model generating unit 12 which uses the acquired earth science data and satellite data to generate a learning model by learning a correlation between the characteristic of the specific region indicated by the earth science data and the characteristic of the specific region indicated by the satellite data.

Description

地球科学データ解析装置、地球科学データ解析方法、及びコンピュータ読み取り可能な記録媒体Geoscience data analysis apparatus, geoscience data analysis method, and computer-readable recording medium
　本発明は、特定領域の特性、例えば、地表に存在する物質の含有量等を示す地球科学データを解析するための、地球科学データ解析装置、及び地球科学データ解析方法に関し、更には、これらを実現するためのプログラムを記録したコンピュータ読み取り可能な記録媒体に関する。 The present invention relates to an earth science data analysis apparatus and an earth science data analysis method for analyzing earth science data indicating characteristics of a specific region, for example, the content of a substance existing on the ground surface, and further, The present invention relates to a computer-readable recording medium on which a program for realizing is recorded.
　地球科学データは、それが取得された箇所における、地質、岩石成分、植生等の特性を表すデータである。従来から、特定のエリアで取得された地球科学データを用いて、地球科学データが取得されていない箇所の特性を推測することが行なわれている。例えば、特許文献１は、複数箇所で取得されたボーリングデータを用いて、ボーリングが行なわれていない箇所の地層分布及び地質性状を推定する地盤推定方法を開示している。 Earth science data is data representing the characteristics of the geology, rock composition, vegetation, etc. at the location where it was acquired. Conventionally, using the earth science data acquired in a specific area, it has been estimated that the characteristics of the location where the earth science data is not acquired. For example, Patent Literature 1 discloses a ground estimation method for estimating a geological distribution and geological properties of a place where no boring is performed using boring data acquired at a plurality of places.
　具体的には、特許文献１に開示された地盤推定方法では、まず、各ボーリングデータが含む各地層に関する地質特性値に基づいて、推定対象となるエリアの地層毎に、地質特性値の等高線図が生成される。このとき推定対象となるエリアは、ボーリングが行なわれた箇所を含むエリアである。言い換えると、ボーリングは、推定対象となるエリアの複数箇所において行なわれている。次に、地盤推定箇所の位置を、各地層の等高線図中に照合して、そこでの地質特性値が推定される。その後、推定された各地層の地質特性値が表示される。 Specifically, in the ground estimation method disclosed in Patent Document 1, first, the contour map of the geological characteristic value for each stratum of the area to be estimated based on the geological characteristic value regarding each layer included in each boring data. Is generated. At this time, the area to be estimated is an area including a place where the bowling is performed. In other words, the bowling is performed at a plurality of locations in the area to be estimated. Next, the position of the estimated ground location is collated in the contour map of each region, and the geological characteristic value there is estimated. After that, the estimated geological characteristic values of each region are displayed.
　このように、特許文献１に開示された地盤推定方法では、推定対象となるエリアにおいて、ボーリングが行なわれていない箇所の地質特性値を推定することができる。通常、ボーリングの実施には多額の費用がかかることから、特許文献１に開示された地盤推定方法によれば、地盤推定にかかる費用を削減することができる。 As described above, in the ground estimation method disclosed in Patent Document 1, it is possible to estimate the geological characteristic value of the portion where the drilling is not performed in the area to be estimated. Usually, since a large amount of cost is required to perform boring, according to the ground estimation method disclosed in Patent Document 1, the cost for ground estimation can be reduced.
特開２０１２－３７４２７号公報JP 2012-37427 A
　しかしながら、特許文献１に開示された地盤推定方法では、地質特性値の等高線図を作成する必要があるため、推定できる箇所は、ボーリングが行なわれた箇所の周辺に限られてしまう。 However, in the ground estimation method disclosed in Patent Document 1, it is necessary to create a contour map of geological characteristic values, and therefore the places that can be estimated are limited to the vicinity of the place where the boring has been performed.
　本発明の目的の一例は、上記問題を解消し、ある領域で取得された地球科学データを用いて、他の領域の地球科学データを推定し得る、地球科学データ解析装置、地球科学データ解析方法、及びコンピュータ読み取り可能な記録媒体を提供することにある。 An example of an object of the present invention is to solve the above-described problem and use the earth science data acquired in a certain area to estimate the earth science data in another area, the earth science data analysis apparatus and the earth science data analysis method And providing a computer-readable recording medium.
　上記目的を達成するため、本発明の一側面における地球科学データ解析装置は、
　特定領域の特性を示す地球科学データ、及び前記特定領域の特性を示す衛星データを取得する、データ取得部と、
　取得された前記地球科学データ及び前記衛星データを用いて、前記地球科学データが示す前記特定領域の特性と前記衛星データが示す前記特定領域の特性との相関関係を学習して、学習モデルを生成する、学習モデル生成部と、
を備えていることを特徴とする。 In order to achieve the above object, an earth science data analysis apparatus according to one aspect of the present invention provides:
A data acquisition unit for acquiring geoscience data indicating the characteristics of the specific area and satellite data indicating the characteristics of the specific area;
Using the acquired earth science data and the satellite data, learning the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data, and generating a learning model A learning model generation unit;
It is characterized by having.
　また、上記目的を達成するため、本発明の一側面における地球科学データ解析方法は、
（ａ）特定領域の特性を示す地球科学データ、及び前記特定領域の特性を示す衛星データを取得する、ステップと、
（ｂ）前記（ａ）のステップで取得された前記地球科学データ及び前記衛星データを用いて、前記地球科学データが示す前記特定領域の特性と前記衛星データが示す前記特定領域の特性との相関関係を学習して、学習モデルを生成する、ステップと、
を有する、ことを特徴とする。 In order to achieve the above object, an earth science data analysis method according to one aspect of the present invention includes:
(A) obtaining geoscience data indicating characteristics of a specific area, and satellite data indicating characteristics of the specific area;
(B) Correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data using the earth science data and the satellite data acquired in the step (a). Learning relationships and generating learning models, steps,
It is characterized by having.
　更に、上記目的を達成するため、本発明の一側面におけるコンピュータ読み取り可能な記録媒体は、
コンピュータに、
（ａ）特定領域の特性を示す地球科学データ、及び前記特定領域の特性を示す衛星データを取得する、ステップと、
（ｂ）前記（ａ）のステップで取得された前記地球科学データ及び前記衛星データを用いて、前記地球科学データが示す前記特定領域の特性と前記衛星データが示す前記特定領域の特性との相関関係を学習して、学習モデルを生成する、ステップと、
を実行させる命令を含む、プログラムを記録していることを特徴とする。 Furthermore, in order to achieve the above object, a computer-readable recording medium according to one aspect of the present invention is provided.
On the computer,
(A) obtaining geoscience data indicating characteristics of a specific area, and satellite data indicating characteristics of the specific area;
(B) Correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data using the earth science data and the satellite data acquired in the step (a). Learning relationships and generating learning models, steps,
A program including an instruction for executing is recorded.
　以上のように、本発明によれば、ある領域で取得された地球科学データを用いて、他の領域の地球科学データを推定することができる。 As described above, according to the present invention, it is possible to estimate the earth science data of another area using the earth science data acquired in a certain area.
図１は、本発明の実施の形態１における地球科学データ解析装置の構成を概略的に示すブロック図である。FIG. 1 is a block diagram schematically showing the configuration of the earth science data analysis apparatus according to Embodiment 1 of the present invention. 図２は、本発明の実施の形態１における地球科学データ解析装置の構成を具体的に示すブロック図である。FIG. 2 is a block diagram specifically showing the configuration of the earth science data analysis apparatus according to Embodiment 1 of the present invention. 図３は、本発明の実施の形態１における地球科学データ解析装置の動作を示すフロー図である。FIG. 3 is a flowchart showing the operation of the earth science data analysis apparatus according to Embodiment 1 of the present invention. 図４は、本発明の実施の形態２における地球科学データ解析装置の動作を示すフロー図である。FIG. 4 is a flowchart showing the operation of the earth science data analysis apparatus according to the second embodiment of the present invention. 図５は、本発明の実施の形態１及び２における地球科学データ解析装置を実現するコンピュータの一例を示すブロック図である。FIG. 5 is a block diagram showing an example of a computer that implements the earth science data analysis apparatus according to the first and second embodiments of the present invention. 図６は、本発明の実施例で用いられる特定の地点のサンプルデータの一例を示す図である。FIG. 6 is a diagram showing an example of sample data at a specific point used in the embodiment of the present invention. 図７は、衛星データの一例を示す図であり、図７（ａ）は赤外領域の光の反射率の分布を示し、図７（ｂ）は標高データを示し、図７（ｃ）は地磁気測定データを示している。FIG. 7 is a diagram showing an example of satellite data. FIG. 7 (a) shows the reflectance distribution of light in the infrared region, FIG. 7 (b) shows elevation data, and FIG. Geomagnetic measurement data is shown. 図８は、本発明の実施例で用いられるサンプルデータの集合の一例を示す図である。FIG. 8 is a diagram showing an example of a set of sample data used in the embodiment of the present invention. 図９は、地球科学データが取得されている特定領域とそれ以外の領域との一例を示す図である。FIG. 9 is a diagram illustrating an example of a specific area in which geoscience data is acquired and other areas. 図１０は、学習モデル生成部によって欠損が補正された地球科学データを示す図である。FIG. 10 is a diagram illustrating the earth science data in which the deficit is corrected by the learning model generation unit. 図１１は、データ推定部によって地球科学データが推定された後の特定領域とそれ以外の領域とを示す図である。FIG. 11 is a diagram illustrating a specific region after the earth science data is estimated by the data estimation unit and other regions.
（実施の形態１）
　以下、本発明の実施の形態１における、地球科学データ解析装置、地球科学データ解析方法、及びプログラムについて、図１～図３を参照しながら説明する。 (Embodiment 1)
Hereinafter, an earth science data analysis apparatus, an earth science data analysis method, and a program according to Embodiment 1 of the present invention will be described with reference to FIGS.
［装置構成］
　最初に、本実施の形態１における地球科学データ解析装置の構成について説明する。図１は、本発明の実施の形態１における地球科学データ解析装置の構成を概略的に示すブロック図である。 [Device configuration]
First, the configuration of the earth science data analysis apparatus according to the first embodiment will be described. FIG. 1 is a block diagram schematically showing the configuration of the earth science data analysis apparatus according to Embodiment 1 of the present invention.
　図１に示す本実施の形態１における地球科学データ解析装置１０は、地球科学データを解析するための装置である。図１に示すように、地球科学データ解析装置１０は、データ取得部１１と、学習モデル生成部１２とを備えている。 1 is a device for analyzing earth science data. The earth science data analysis device 10 according to the first embodiment shown in FIG. As shown in FIG. 1, the earth science data analysis apparatus 10 includes a data acquisition unit 11 and a learning model generation unit 12.
　データ取得部１１は、特定領域の特性を示す地球科学データ、及び該特定領域の特性を示す衛星データを取得する。学習モデル生成部１２は、取得された地球科学データ及び衛星データを用いて、地球科学データが示す特定領域の特性と衛星データが示す特定領域の特性との相関関係を学習して、学習モデルを生成する。 The data acquisition unit 11 acquires the earth science data indicating the characteristics of the specific area and the satellite data indicating the characteristics of the specific area. The learning model generation unit 12 uses the acquired earth science data and satellite data to learn the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data, and obtains the learning model. Generate.
　このように、本実施の形態１では、地球科学データが示す特定領域の特性と衛星データが示す特定領域の特性との相関関係が学習される。このため、学習によって得られた学習モデルに、特定領域以外の領域における衛星データを適用すれば、この特定領域以外の領域での地球科学データの推測が可能となる。つまり、本実施の形態１によれば、地球科学データが取得されていない領域において、別の領域で取得された地球科学データを用いて、地球科学データを推定することが可能となる。 Thus, in the first embodiment, the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data is learned. For this reason, if satellite data in a region other than the specific region is applied to the learning model obtained by learning, it is possible to estimate geoscience data in a region other than the specific region. That is, according to the first embodiment, it is possible to estimate the earth science data using the earth science data acquired in another area in the area where the earth science data is not acquired.
　続いて、図２を用いて、本実施の形態１における地球科学データ解析装置１０の構成をより具体的に説明する。図２は、本発明の実施の形態１における地球科学データ解析装置の構成を具体的に示すブロック図である。 Subsequently, the configuration of the earth science data analysis apparatus 10 according to the first embodiment will be described more specifically with reference to FIG. FIG. 2 is a block diagram specifically showing the configuration of the earth science data analysis apparatus according to Embodiment 1 of the present invention.
　まず、本実施の形態１において用いることができる地球科学データとしては、特定領域の特性として資源の存在を示すデータ、例えば、地表に存在する物質、元素の種類、成分比、含有量等を示すデータが挙げられる。具体的には、ある領域において銅の存在の予測が求められているとすると、地球科学データとしては、特定領域の特性である単位面積当たりの銅の含有量（ppm）を示すデータが挙げられる。 First, as the earth science data that can be used in the first embodiment, data indicating the existence of resources as characteristics of a specific region, for example, substances existing on the earth's surface, types of elements, component ratios, contents, etc. Data. Specifically, assuming that the prediction of the presence of copper in a certain area is required, the earth science data includes data indicating the copper content (ppm) per unit area, which is a characteristic of a specific area. .
　また、その他の地球科学データとしては、重力値、二酸化炭素の濃度プロファイル、気温、湿度、風向、風速、気圧、全天日射、分光放射、光合成有効放射、地温、土壌水分、地流熱量、直達放射スペクトル、地盤安定性、地層年代、断層情報、地下水脈情報、植物種類の分布、蒸発散情報、鉱物産量等を示すデータも挙げられる。また、特定の資源の探査又は存在の把握を目的とする場合は、資源の存在に関連のあるデータを用いることが好適である。例えば、地殻に存在する特定の元素の存在の把握を目的とする場合、鉱脈の存在確率の算出を目的とする場合は、地球科学データとしては、把握対象となる元素の存在比率を示すデータが挙げられる。 Other geoscience data include gravity value, carbon dioxide concentration profile, temperature, humidity, wind direction, wind speed, atmospheric pressure, global solar radiation, spectral radiation, photosynthetic effective radiation, earth temperature, soil moisture, geothermal heat, direct delivery Data including radiation spectrum, ground stability, geological age, fault information, groundwater vein information, plant type distribution, evapotranspiration information, mineral production, etc. For the purpose of exploring a specific resource or grasping its existence, it is preferable to use data related to the existence of the resource. For example, when the purpose is to understand the presence of a specific element present in the crust, or when the purpose is to calculate the existence probability of a mine, the earth science data includes data indicating the abundance ratio of the element to be grasped. Can be mentioned.
　衛星データは、地球の上空から得られたデータであり、特定領域の特性を示すデータである。衛星データは、衛星が取得したデータ、航空機等の飛行体が取得したデータを含む。また、本実施の形態１において用いることができる衛星データとしては、取得対象の領域から反射または放射される電磁波の強度を示すデータ、特定波長の光の反射率の分布を示すデータ、地磁気を示すデータ、標高を示すデータ、標高傾斜を示すデータ等が挙げられる。具体的には、特定波長の光の反射率の分布を示すデータとしては、アスター(ASTER: Advanced Spaceborne Thermal Emission and Reflection Radiometer)によって測定されたデータが挙げられる。アスターは、米国ＮＡＳＡのテラ(Terra)衛星に搭載された観測用の光学センサであり、可視から熱赤外にわたる１４バンドを観測することができる。また、この１４バンドは、鉱物に関する特徴的なスペクトルを捉えるのに適した波長である。なお、衛星データは上記のものに限定されずリモートセンシングによって得られたデータを含む。 Satellite data is data obtained from the sky above the earth, and is data indicating the characteristics of a specific area. The satellite data includes data acquired by a satellite and data acquired by an aircraft such as an aircraft. The satellite data that can be used in the first embodiment includes data indicating the intensity of electromagnetic waves reflected or emitted from the area to be acquired, data indicating the reflectance distribution of light of a specific wavelength, and geomagnetism. Data, data indicating altitude, data indicating altitude slope, and the like. Specifically, the data indicating the reflectance distribution of light of a specific wavelength includes data measured by an aster (ASTER: “Advanced” Spaceborne “Thermal” Emission “and Reflection” Radiometer). An aster is an optical sensor for observation mounted on the Terra satellite of NASA in the United States, and can observe 14 bands from visible to thermal infrared. The 14 bands are wavelengths suitable for capturing a characteristic spectrum related to minerals. The satellite data is not limited to the above, and includes data obtained by remote sensing.
　また、図２に示すように、本実施の形態１では、地球科学データ解析装置１０は、上述したデータ取得部１１及び学習モデル生成部１２に加えて、データ推定部１３と、表示部１４と、記憶部１５を備えている。また、地球科学データ解析装置１０には、表示装置２０が接続されている。 As shown in FIG. 2, in the first embodiment, the earth science data analysis apparatus 10 includes a data estimation unit 13, a display unit 14, and a data acquisition unit 11 and a learning model generation unit 12. The storage unit 15 is provided. A display device 20 is connected to the earth science data analysis device 10.
　データ取得部１１は、本実施の形態では、データベース３０から、地球科学データ及び衛星データを取得し、取得した地球科学データ及び衛星データを学習モデル生成部１２に渡す。データベース３０は、ネットワークを介して、地球科学データ解析装置１０に接続されている。 In this embodiment, the data acquisition unit 11 acquires the earth science data and satellite data from the database 30 and passes the acquired earth science data and satellite data to the learning model generation unit 12. The database 30 is connected to the earth science data analysis apparatus 10 via a network.
　データベース３０は、特定領域における地球科学データ及び衛星データを格納している。例えば、地球科学データが、地点毎の単位面積当たりの銅の含有量（ppm）を示すデータであり、衛星データが、特定波長の光の反射率の分布を示すデータ、標高データ、及び標高傾斜データであるとする。この場合、データベース３０は、地点（緯度及び経度）毎に、地球科学データとして、単位面積当たりの銅の含有量（ppm）を示すデータを格納し、衛星データとして、特定波長の光の反射率、標高値、及び傾斜値を格納する。また、この場合、地球科学データ及び衛星データが取得されている地点を中心とした設定範囲を重ね合わせて得られた領域を、特定領域とすることができる。 The database 30 stores earth science data and satellite data in a specific area. For example, geoscience data is data indicating the copper content (ppm) per unit area at each point, and satellite data is data indicating the reflectance distribution of light at a specific wavelength, altitude data, and altitude slope Suppose that it is data. In this case, the database 30 stores data indicating the copper content (ppm) per unit area as geoscience data for each point (latitude and longitude), and reflects the reflectance of light of a specific wavelength as satellite data. , Store elevation values, and slope values. In this case, an area obtained by superimposing a set range centering on a point where the earth science data and satellite data are acquired can be set as the specific area.
　また、データベース３０では、地点毎の地球科学データの値と衛星データの値とは、１つの組として互いに紐付けられる。更に、１つの組を構成する地球科学データの値と衛星データの値とは、１つのサンプルデータとして扱われる。なお、衛星データは、地球科学データに比べて広範囲にわたって取得できるため、地球科学データが取得されている特定領域以外の領域までもカバーしていても良い。 Further, in the database 30, the value of the earth science data and the value of the satellite data for each point are associated with each other as one set. Furthermore, the value of the earth science data and the value of the satellite data constituting one set are handled as one sample data. In addition, since satellite data can be acquired over a wider range than earth science data, it may cover areas other than the specific area where earth science data is acquired.
　学習モデル生成部１２は、本実施の形態では、まず、データ取得部１１から複数のサンプルデータを受け取り、受け取った各サンプルデータを教師データとして機械学習を実行する。本実施の形態１において用いられる機械学習の方式としては、決定木、サポートベクトルマシン、ニューラルネットワーク、ロジスティック回帰、最近傍分類法（K-NN: k-nearest neighbor algorithm）、アンサンブル分類学習法、判別分析等が挙げられる。 In the present embodiment, the learning model generation unit 12 first receives a plurality of sample data from the data acquisition unit 11, and executes machine learning using each received sample data as teacher data. The machine learning method used in the first embodiment includes a decision tree, support vector machine, neural network, logistic regression, nearest neighbor classification (K-NN), ensemble classification learning method, discrimination Analysis and the like.
　具体的には、学習モデル生成部１２は、サポートベクトルマシンに、各サンプルデータを与えて、地球科学データが示す特定領域の特性と衛星データが示す特定領域の特性との関係、例えば、銅の含有量（ppm）と、特定波長の光の反射率、標高値、及び傾斜値との関係を学習する。そして、学習モデル生成部１２は、特定波長の光の反射率、標高値、及び傾斜値が入力されると、入力値に応じて、銅の含有量を出力する学習モデル１６を生成し、これを記憶部１５に格納させる。 Specifically, the learning model generation unit 12 gives each sample data to the support vector machine, and the relationship between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data, for example, copper The relationship between the content (ppm) and the reflectance, altitude value, and slope value of light of a specific wavelength is learned. And the learning model production | generation part 12 will produce | generate the learning model 16 which outputs copper content according to an input value, if the reflectance, elevation value, and inclination value of the light of a specific wavelength are input, this Is stored in the storage unit 15.
　また、学習モデル生成部１２は、各サンプルデータを用いて、ディープラーニングを行ない、それによって、特定波長の光の反射率、標高値、及び傾斜値に応じて、銅の含有量を決定する、分類器を作成し、作成した分類器を学習モデル１６とすることもできる。 Further, the learning model generation unit 12 performs deep learning using each sample data, thereby determining the copper content according to the reflectance, altitude value, and slope value of light of a specific wavelength. It is also possible to create a classifier and use the created classifier as the learning model 16.
　データ推定部１３は、学習モデル生成部１２によって生成された学習モデル１６を用いて、特定領域以外の領域（以下「推定領域」と表記する。）における地球科学データを推定する。本実施の形態１では、データ推定部１３は、推定領域における衛星データを取得し、取得した推定領域の衛星データを学習モデル１６に適用することによって、推定領域における地球科学データを推定する。 The data estimation unit 13 uses the learning model 16 generated by the learning model generation unit 12 to estimate geoscience data in a region other than the specific region (hereinafter referred to as “estimation region”). In the first embodiment, the data estimation unit 13 estimates the earth science data in the estimation region by acquiring the satellite data in the estimation region and applying the acquired satellite data in the estimation region to the learning model 16.
　具体的には、データ推定部１３は、まず、外部から推定領域が指定されると、推定領域上の複数の地点（緯度及び経度）を選出する。次いで、データ推定部１３は、データベース３０に格納されている衛星データから、選出した地点に対応する、特定波長の光の反射率、標高値、及び傾斜値を特定し、銅の含有量がブランクとなったサンプルデータを作成する。そして、データ推定部１３は、作成したサンプルデータを、学習モデル１６に適用して、ブランクとなっている銅の含有量を算出する。 Specifically, the data estimation unit 13 first selects a plurality of points (latitude and longitude) on the estimation area when the estimation area is designated from the outside. Next, the data estimation unit 13 specifies the reflectance, altitude value, and slope value of light of a specific wavelength corresponding to the selected point from the satellite data stored in the database 30, and the copper content is blank. Create sample data. Then, the data estimation unit 13 applies the created sample data to the learning model 16 to calculate the blank copper content.
　表示部１４は、表示装置２０の画面上において、特定領域における地球科学データと、推定領域における地球科学データとを、地図データ上に重ねて表示する。例えば、地球科学データが地点毎の単位面積当たりの銅の含有量（ppm）であるとすると、画面上には、銅の含有量が特定されていない地点についても、銅の含有量（予測値）が表示される。このため、ユーザは、効率の良い採掘計画を策定することができる。 The display unit 14 displays the earth science data in the specific area and the earth science data in the estimation area on the map data on the screen of the display device 20 so as to overlap each other. For example, if the earth science data is the copper content (ppm) per unit area for each point, the copper content (predicted value) is also displayed on the screen for points where the copper content is not specified. ) Is displayed. For this reason, the user can formulate an efficient mining plan.
［装置動作］
　次に、本発明の実施の形態１における地球科学データ解析装置１０の動作について、図３を用いて説明する。図３は、本発明の実施の形態１における地球科学データ解析装置の動作を示すフロー図である。以下の説明においては、適宜図１及び図２を参酌する。また、本実施の形態１では、地球科学データ解析装置１０を動作させることによって、地球科学データ解析方法が実施される。よって、本実施の形態１における地球科学データ解析方法の説明は、以下の地球科学データ解析装置１０の動作説明に代える。 [Device operation]
Next, the operation of the earth science data analysis apparatus 10 according to the first embodiment of the present invention will be described with reference to FIG. FIG. 3 is a flowchart showing the operation of the earth science data analysis apparatus according to Embodiment 1 of the present invention. In the following description, FIGS. 1 and 2 are referred to as appropriate. In the first embodiment, the earth science data analysis method is implemented by operating the earth science data analysis apparatus 10. Therefore, the description of the earth science data analysis method in the first embodiment is replaced with the following description of the operation of the earth science data analysis apparatus 10.
　図３に示すように、最初に、データ取得部１１は、データベース３０から、特定領域における地球科学データ及び衛星データを取得する（ステップＡ１）。 As shown in FIG. 3, first, the data acquisition unit 11 acquires geoscience data and satellite data in a specific area from the database 30 (step A1).
　具体的には、ステップＡ１では、データ取得部１１は、データベース３０から、特定領域に含まれる地点毎のサンプルデータを取得し、取得した地点毎のサンプルデータを学習モデル生成部１２に渡す。 Specifically, in step A1, the data acquisition unit 11 acquires sample data for each point included in the specific area from the database 30, and passes the acquired sample data for each point to the learning model generation unit 12.
　次に、学習モデル生成部１２は、ステップＡ１で取得された地球科学データ及び衛星データを用いて、地球科学データが示す特定領域の特性と衛星データが示す特定領域の特性との相関関係を学習して、学習モデル１６を生成する（ステップＡ２）。 Next, the learning model generation unit 12 learns the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data by using the earth science data and the satellite data acquired in step A1. Then, the learning model 16 is generated (step A2).
　具体的には、学習モデル生成部１２は、ステップＡ１で取得された地点毎のサンプルデータを受け取ると、受け取った各サンプルデータを教師データとして機械学習を実行し、それによって、学習モデル１６を生成する。また、学習モデル生成部１２は、生成した学習モデル１６を記憶部１５に格納する。 Specifically, when the learning model generation unit 12 receives the sample data for each point acquired in step A1, the learning model generation unit 12 performs machine learning using the received sample data as teacher data, thereby generating the learning model 16 To do. Further, the learning model generation unit 12 stores the generated learning model 16 in the storage unit 15.
　次に、データ推定部１３は、学習モデル生成部１２によって生成された学習モデル１６を用いて、特定領域以外の領域（推定領域）における地球科学データを推定する（ステップＡ３）。 Next, the data estimation unit 13 estimates the earth science data in a region (estimated region) other than the specific region using the learning model 16 generated by the learning model generation unit 12 (step A3).
　具体的には、データ推定部１３は、外部から推定領域が指定されると、推定領域上の複数の地点（緯度及び経度）を選出する。次いで、データ推定部１３は、データベース３０に格納されている衛星データから、選出した地点に対応する衛星データを特定し、地球科学データがブランクとなったサンプルデータを作成する。そして、データ推定部１３は、作成したサンプルデータを、学習モデル１６に適用して、ブランクとなっている地球科学データを算出する。 Specifically, when the estimation area is designated from the outside, the data estimation unit 13 selects a plurality of points (latitude and longitude) on the estimation area. Next, the data estimation unit 13 specifies satellite data corresponding to the selected point from the satellite data stored in the database 30, and creates sample data in which the earth science data is blank. Then, the data estimation unit 13 applies the created sample data to the learning model 16 to calculate blank earth science data.
　次に、表示部１４は、表示装置２０の画面上において、特定領域における地球科学データと、推定領域における地球科学データとを、地図データ上に重ねて表示する（ステップＡ４）。 Next, the display unit 14 displays the earth science data in the specific area and the earth science data in the estimation area on the map data on the screen of the display device 20 (step A4).
　以上のように、本実施の形態１によれば、地球科学データが示す特定領域の特性と衛星データが示す特定領域の特性との相関関係を示す学習モデルが生成されるので、地球科学データが取得されていない領域においても、地球科学データの推定が可能となる。 As described above, according to the first embodiment, a learning model is generated that indicates the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data. It is possible to estimate geoscience data even in areas that have not been acquired.
［プログラム］
　本実施の形態１におけるプログラムは、コンピュータに、図３に示すステップＡ１～Ａ４を実行させるプログラムであれば良い。このプログラムをコンピュータにインストールし、実行することによって、本実施の形態１における地球科学データ解析装置１０と地球科学データ解析方法とを実現することができる。この場合、コンピュータのＣＰＵ（Central Processing Unit）は、データ取得部１１、学習モデル生成部１２、データ推定部１３、及び表示部１４として機能し、処理を行なう。 [program]
The program in the first embodiment may be a program that causes a computer to execute steps A1 to A4 shown in FIG. By installing and executing this program in a computer, the earth science data analysis apparatus 10 and the earth science data analysis method according to the first embodiment can be realized. In this case, a CPU (Central Processing Unit) of the computer functions as the data acquisition unit 11, the learning model generation unit 12, the data estimation unit 13, and the display unit 14 to perform processing.
　また、本実施の形態１におけるプログラムは、複数のコンピュータによって構築されたコンピュータシステムによって実行されても良い。この場合は、例えば、各コンピュータが、それぞれ、データ取得部１１、学習モデル生成部１２、データ推定部１３、及び表示部１４のいずれかとして機能しても良い。 Further, the program in the first embodiment may be executed by a computer system constructed by a plurality of computers. In this case, for example, each computer may function as any one of the data acquisition unit 11, the learning model generation unit 12, the data estimation unit 13, and the display unit 14, respectively.
（実施の形態２）
　次に、本発明の実施の形態２における、地球科学データ解析装置、地球科学データ解析方法、及びプログラムについて、図４を参照しながら説明する。 (Embodiment 2)
Next, an earth science data analysis device, an earth science data analysis method, and a program according to Embodiment 2 of the present invention will be described with reference to FIG.
［装置構成］
　最初に、本実施の形態２における地球科学データ解析装置の構成について説明する。但し、本実施の形態２における地球科学データ解析装置は、図１及び図２に示した地球科学データ解析装置１０と同様の構成を有しているため、以下の説明では、図１及び図２を参照する。また、以下においては、実施の形態１との相違点を中心に説明する。 [Device configuration]
First, the configuration of the earth science data analysis apparatus according to the second embodiment will be described. However, since the earth science data analysis apparatus in the second embodiment has the same configuration as the earth science data analysis apparatus 10 shown in FIGS. 1 and 2, in the following description, FIG. 1 and FIG. Refer to In the following description, differences from the first embodiment will be mainly described.
　本実施の形態２においては、学習モデル生成部１２は、学習モデル１６の生成の前に、数理モデルを用いて、特定領域における地球科学データの欠損を補正する。そして、学習モデル生成部１２は、補正後の地球科学データを用いて、地球科学データが示す特定領域の特性と衛星データが示す特定領域の特性との相関関係を学習して、学習モデルを生成する。 In the second embodiment, the learning model generation unit 12 corrects the deficiency of the earth science data in the specific region using the mathematical model before the generation of the learning model 16. Then, the learning model generation unit 12 uses the corrected earth science data to learn the correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data, and generates a learning model. To do.
　これは、地球科学データの取得は、衛星データの取得に比べて難しいことから、特定領域内の地点の中には、衛星データは取得されているが、地球科学データは取得されていない地点も存在するからである。本実施の形態２では、学習モデル生成部１２は、衛星データは取得されているが、地球科学データは取得されていない、地点において、地球科学データを補充する。言い換えると、学習モデル生成部１２は、地球科学データの値が欠損している特定領域のサンプルデータにおいて、地球科学データの値を補充する。 This is because acquisition of geoscience data is more difficult than acquisition of satellite data, so there are some points in a specific area where satellite data has been acquired but geoscience data has not been acquired. Because it exists. In the second embodiment, the learning model generation unit 12 supplements the earth science data at a point where the satellite data is acquired but the earth science data is not acquired. In other words, the learning model generation unit 12 supplements the value of the earth science data in the sample data of the specific region where the value of the earth science data is missing.
　具体的には、本実施の形態２では、学習モデル生成部１２は、まず、データ取得部１１から複数のサンプルデータを受け取ると、各サンプルデータを、例えば、それが紐付けられている地点の緯度及び経度に基づいて、１又は２以上のグループに分類する。 Specifically, in the second embodiment, when the learning model generation unit 12 first receives a plurality of sample data from the data acquisition unit 11, each sample data is, for example, the point where it is associated. Classify into one or more groups based on latitude and longitude.
　次いで、学習モデル生成部１２は、例えば、３つのサンプルデータを選択し、選択した３つのサンプルデータそれぞれに紐付けられている緯度及び経度の値を用いて、３つの地点を特定する。更に、学習モデル生成部１２は、特定した３つの地点で定まる領域の面積を求め、求めた面積が予め定めた閾値以下であるかどうかを判定する。 Next, the learning model generation unit 12 selects, for example, three sample data, and specifies three points using the latitude and longitude values associated with each of the selected three sample data. Further, the learning model generation unit 12 calculates the area of the region determined by the three specified points, and determines whether the calculated area is equal to or less than a predetermined threshold.
　そして、学習モデル生成部１２は、判定の結果、求めた面積が閾値以下である場合は、当該３つのサンプルデータを１つのグループに分類し、反対に、求めた面積が閾値より大きい場合は、当該３つのサンプルデータは１つのグループに属さないと判断する。学習モデル生成部１２は、これらの一連の処理を繰り返すことにより、複数のサンプルデータを１または複数のグループに分類する。 And as a result of determination, if the obtained area is less than or equal to the threshold, the learning model generation unit 12 classifies the three sample data into one group. Conversely, if the obtained area is larger than the threshold, It is determined that the three sample data do not belong to one group. The learning model generation unit 12 classifies the plurality of sample data into one or a plurality of groups by repeating these series of processes.
　次に、学習モデル生成部１２は、グループ毎に、地球科学データの値が欠損しているサンプルデータが存在しないかどうかを判定する。そして、判定の結果、地球科学データの値が欠損しているサンプルデータが存在している場合は、グループに含まれるサンプルデータのうち、地球科学データの値が欠損していないサンプルデータの値を用いて、欠損しているサンプルデータの値を補充する。 Next, the learning model generation unit 12 determines, for each group, whether there is any sample data in which the value of the earth science data is missing. As a result of the determination, if there is sample data with missing earth science data value, out of the sample data included in the group, the value of sample data with missing earth science data value is Use to fill in missing sample data values.
　また、値の補充方法としては、近傍データ補間法、線形補間法、多項式補間法、ノンパラメトリック回帰による補間法等が挙げられる。このうち、近傍データ補間は、補間対象となる点Ｃと、近傍の点Ａ及び点Ｂとの距離を求め、点Ｃとの距離が近い点のデータによって、点Ｃを補間する方法である。線形補間法は、点Ａと点Ｂとを満たす線形関係Y＝cX＋dを求め、求めた線形関係によって、補間対象となる点Ｃのデータを補間する方法である。また、多項式補間法は、点Ａと点Ｂとを満たす多項式関係Y=c1X^n+c2X^n-1+…+cnX+dを求め、求めた線形関係によって、補間対象となる点Ｃのデータを補間する方法である。ノンパラメトリック回帰による補間法は、全データを用いて、ノンパラメトリック回帰モデルを作成し、作成したノンパラメトリック回帰モデルを用いて、全データの隙間のデータを推定する方法である。 Also, methods for supplementing values include neighborhood data interpolation, linear interpolation, polynomial interpolation, non-parametric regression interpolation, and the like. Among these, the proximity data interpolation is a method of obtaining the distance between the point C to be interpolated and the nearby points A and B, and interpolating the point C with the data of the points that are close to the point C. The linear interpolation method is a method of obtaining a linear relationship Y = cX + d satisfying the points A and B and interpolating the data of the point C to be interpolated based on the obtained linear relationship. Further, the polynomial interpolation method obtains a polynomial relationship Y = c1X ^ n + c2X ^ n-1 +... + CnX + d satisfying points A and B, and the point C to be interpolated is determined by the obtained linear relationship. This is a method of interpolating data. The interpolation method by nonparametric regression is a method of creating a nonparametric regression model using all data and estimating gap data of all data using the created nonparametric regression model.
［装置動作］
　次に、本発明の実施の形態２における地球科学データ解析装置の動作について、図４を用いて説明する。図４は、本発明の実施の形態２における地球科学データ解析装置の動作を示すフロー図である。 [Device operation]
Next, the operation of the earth science data analysis apparatus according to Embodiment 2 of the present invention will be described with reference to FIG. FIG. 4 is a flowchart showing the operation of the earth science data analysis apparatus according to the second embodiment of the present invention.
　以下の説明においては、適宜図１及び図２を参酌する。また、本実施の形態２でも、地球科学データ解析装置を動作させることによって、地球科学データ解析方法が実施される。よって、本実施の形態２における地球科学データ解析方法の説明は、以下の地球科学データ解析装置の動作説明に代える。 In the following description, refer to FIGS. 1 and 2 as appropriate. Also in the second embodiment, the earth science data analysis method is implemented by operating the earth science data analysis apparatus. Therefore, the explanation of the earth science data analysis method in the second embodiment is replaced with the following explanation of the operation of the earth science data analysis apparatus.
　図４に示すように、最初に、データ取得部１１は、データベース３０から、特定領域における地球科学データ及び衛星データを取得する（ステップＢ１）。ステップＢ１は、図３に示したステップＡ１と同様のステップである。 As shown in FIG. 4, first, the data acquisition unit 11 acquires geoscience data and satellite data in a specific area from the database 30 (step B1). Step B1 is the same as step A1 shown in FIG.
　次に、学習モデル生成部１２は、予め設定されている数理モデルを用いて、ステップＡ１で取得された地球科学データの欠損を補正する（ステップＢ２）。具体的には、学習モデル生成部１２は、衛星データは取得されているが、地球科学データは取得されていない、地点において、地球科学データを補充する。 Next, the learning model generation unit 12 corrects the deficiency of the earth science data acquired in Step A1 using a preset mathematical model (Step B2). Specifically, the learning model generation unit 12 supplements the earth science data at a point where the satellite data is acquired but the earth science data is not acquired.
　次に、学習モデル生成部１２は、ステップＢ２で補正された地球科学データ、及びステップＡ１で取得された衛星データを用いて、地球科学データが示す特定領域の特性と衛星データが示す特定領域の特性との相関関係を学習して、学習モデル１６を生成する（ステップＢ３）。ステップＢ３は、図３に示したステップＡ２と同様のステップである。 Next, the learning model generation unit 12 uses the geoscience data corrected in step B2 and the satellite data acquired in step A1, and the characteristics of the specific area indicated by the geoscience data and the specific area indicated by the satellite data. The learning model 16 is generated by learning the correlation with the characteristics (step B3). Step B3 is the same as step A2 shown in FIG.
　次に、データ推定部１３は、学習モデル生成部１２によって生成された学習モデル１６を用いて、推定領域における地球科学データを推定する（ステップＢ４）。ステップＢ４は、図３に示したステップＡ３と同様のステップである。 Next, the data estimation unit 13 estimates the earth science data in the estimation area using the learning model 16 generated by the learning model generation unit 12 (step B4). Step B4 is the same as step A3 shown in FIG.
　次に、表示部１４は、表示装置２０の画面上において、特定領域における地球科学データと、推定領域における地球科学データとを、地図データ上に重ねて表示する（ステップＢ５）。ステップＢ５は、図３に示したステップＡ４と同様のステップである。 Next, the display unit 14 displays the earth science data in the specific area and the earth science data in the estimation area on the map data on the screen of the display device 20 (step B5). Step B5 is the same as step A4 shown in FIG.
　以上のように、本実施の形態２では、地球科学データの欠損が補充されてから、学習モデルが生成されるので、地球科学データが取得されていない領域における地球科学データの推定がより精度の高いものとなる。 As described above, in the second embodiment, since the learning model is generated after the missing of the earth science data is supplemented, the estimation of the earth science data in the region where the earth science data is not acquired is more accurate. It will be expensive.
　本実施の形態２におけるプログラムは、コンピュータに、図４に示すステップＢ１～Ｂ５を実行させるプログラムであれば良い。このプログラムをコンピュータにインストールし、実行することによって、本実施の形態２における地球科学データ解析装置と地球科学データ解析方法とを実現することができる。この場合、コンピュータのＣＰＵ（Central Processing Unit）は、データ取得部１１、学習モデル生成部１２、データ推定部１３、及び表示部１４として機能し、処理を行なう。 The program in the second embodiment may be a program that causes a computer to execute steps B1 to B5 shown in FIG. By installing and executing this program on a computer, the earth science data analysis apparatus and the earth science data analysis method according to the second embodiment can be realized. In this case, the CPU (Central Processing Unit) of the computer functions as the data acquisition unit 11, the learning model generation unit 12, the data estimation unit 13, and the display unit 14 to perform processing.
　また、本実施の形態２におけるプログラムは、複数のコンピュータによって構築されたコンピュータシステムによって実行されても良い。この場合は、例えば、各コンピュータが、それぞれ、データ取得部１１、学習モデル生成部１２、データ推定部１３、及び表示部１４のいずれかとして機能しても良い。 Further, the program according to the second embodiment may be executed by a computer system constructed by a plurality of computers. In this case, for example, each computer may function as any one of the data acquisition unit 11, the learning model generation unit 12, the data estimation unit 13, and the display unit 14, respectively.
（物理構成）
　ここで、実施の形態１及び２におけるプログラムを実行することによって、地球科学データ解析装置を実現するコンピュータについて図５を用いて説明する。図５は、本発明の実施の形態１及び２における地球科学データ解析装置を実現するコンピュータの一例を示すブロック図である。 (Physical configuration)
Here, a computer that realizes the earth science data analysis apparatus by executing the programs in the first and second embodiments will be described with reference to FIG. FIG. 5 is a block diagram showing an example of a computer that implements the earth science data analysis apparatus according to the first and second embodiments of the present invention.
　図５に示すように、コンピュータ１１０は、ＣＰＵ１１１と、メインメモリ１１２と、記憶装置１１３と、入力インターフェイス１１４と、表示コントローラ１１５と、データリーダ／ライタ１１６と、通信インターフェイス１１７とを備える。これらの各部は、バス１２１を介して、互いにデータ通信可能に接続される。なお、コンピュータ１１０は、ＣＰＵ１１１に加えて、又はＣＰＵ１１１に代えて、ＧＰＵ（Graphics Processing Unit）、又はＦＰＧＡ（Field-Programmable Gate Array）を備えていても良い。 As shown in FIG. 5, the computer 110 includes a CPU 111, a main memory 112, a storage device 113, an input interface 114, a display controller 115, a data reader / writer 116, and a communication interface 117. These units are connected to each other via a bus 121 so that data communication is possible. The computer 110 may include a GPU (GraphicsGraphProcessing Unit) or an FPGA (Field-Programmable Gate Array) in addition to or instead of the CPU 111.
　ＣＰＵ１１１は、記憶装置１１３に格納された、本実施の形態におけるプログラム（コード）をメインメモリ１１２に展開し、これらを所定順序で実行することにより、各種の演算を実施する。メインメモリ１１２は、典型的には、ＤＲＡＭ（Dynamic Random Access Memory）等の揮発性の記憶装置である。また、本実施の形態におけるプログラムは、コンピュータ読み取り可能な記録媒体１２０に格納された状態で提供される。なお、本実施の形態におけるプログラムは、通信インターフェイス１１７を介して接続されたインターネット上で流通するものであっても良い。 The CPU 111 performs various operations by developing the program (code) in the present embodiment stored in the storage device 113 in the main memory 112 and executing them in a predetermined order. The main memory 112 is typically a volatile storage device such as a DRAM (Dynamic Random Access Memory). Further, the program in the present embodiment is provided in a state of being stored in a computer-readable recording medium 120. Note that the program in the present embodiment may be distributed on the Internet connected via the communication interface 117.
　また、記憶装置１１３の具体例としては、ハードディスクドライブの他、フラッシュメモリ等の半導体記憶装置が挙げられる。入力インターフェイス１１４は、ＣＰＵ１１１と、キーボード及びマウスといった入力機器１１８との間のデータ伝送を仲介する。表示コントローラ１１５は、ディスプレイ装置１１９と接続され、ディスプレイ装置１１９での表示を制御する。 Further, specific examples of the storage device 113 include a hard disk drive and a semiconductor storage device such as a flash memory. The input interface 114 mediates data transmission between the CPU 111 and an input device 118 such as a keyboard and a mouse. The display controller 115 is connected to the display device 119 and controls display on the display device 119.
　データリーダ／ライタ１１６は、ＣＰＵ１１１と記録媒体１２０との間のデータ伝送を仲介し、記録媒体１２０からのプログラムの読み出し、及びコンピュータ１１０における処理結果の記録媒体１２０への書き込みを実行する。通信インターフェイス１１７は、ＣＰＵ１１１と、他のコンピュータとの間のデータ伝送を仲介する。 The data reader / writer 116 mediates data transmission between the CPU 111 and the recording medium 120, and reads a program from the recording medium 120 and writes a processing result in the computer 110 to the recording medium 120. The communication interface 117 mediates data transmission between the CPU 111 and another computer.
　また、記録媒体１２０の具体例としては、ＣＦ（Compact Flash（登録商標））及びＳＤ（Secure Digital）等の汎用的な半導体記憶デバイス、フレキシブルディスク（Flexible Disk）等の磁気記録媒体、又はＣＤ－ＲＯＭ（Compact Disk Read Only Memory）などの光学記録媒体が挙げられる。 Specific examples of the recording medium 120 include general-purpose semiconductor storage devices such as CF (Compact Flash (registered trademark)) and SD (Secure Digital), magnetic recording media such as a flexible disk, or CD- Optical recording media such as ROM (Compact Disk Read Only Memory) are listed.
　なお、本実施の形態における地球科学データ解析装置は、プログラムがインストールされたコンピュータではなく、各部に対応したハードウェアを用いることによっても実現可能である。更に、地球科学データ解析装置は、一部がプログラムで実現され、残りの部分がハードウェアで実現されていてもよい。 Note that the earth science data analysis apparatus according to the present embodiment can be realized by using hardware corresponding to each unit, not a computer in which a program is installed. Furthermore, part of the earth science data analysis apparatus may be realized by a program, and the remaining part may be realized by hardware.
　続いて、本発明の実施の形態２における地球科学データ解析装置の実施例について図６～図１１を用いて説明する。図６は、本発明の実施例で用いられる特定の地点のサンプルデータの一例を示す図である。図７は、衛星データの一例を示す図であり、図７（ａ）は赤外領域の光の反射率の分布を示し、図７（ｂ）は標高データを示し、図７（ｃ）は地磁気測定データを示している。 Subsequently, examples of the earth science data analysis apparatus according to the second embodiment of the present invention will be described with reference to FIGS. FIG. 6 is a diagram showing an example of sample data at a specific point used in the embodiment of the present invention. FIG. 7 is a diagram showing an example of satellite data. FIG. 7 (a) shows the reflectance distribution of light in the infrared region, FIG. 7 (b) shows elevation data, and FIG. Geomagnetic measurement data is shown.
　図８は、本発明の実施例で用いられるサンプルデータの集合の一例を示す図である。図９は、地球科学データが取得されている特定領域とそれ以外の領域との一例を示す図である。図１０は、学習モデル生成部によって欠損が補正された地球科学データを示す図である。図１１は、データ推定部によって地球科学データが推定された後の特定領域とそれ以外の領域とを示す図である。なお、図９～図１１において、白色の破線は行政区画の境界を示している。 FIG. 8 is a diagram showing an example of a set of sample data used in the embodiment of the present invention. FIG. 9 is a diagram illustrating an example of a specific area in which geoscience data is acquired and other areas. FIG. 10 is a diagram illustrating the earth science data in which the deficit is corrected by the learning model generation unit. FIG. 11 is a diagram illustrating a specific region after the earth science data is estimated by the data estimation unit and other regions. In FIGS. 9 to 11, white broken lines indicate the boundaries of administrative divisions.
　まず、図６に示すように、本実施例では、データベース３０は、図６に示すサンプルデータを複数登録している。図６に示すように、サンプルデータは、地点（緯度及び経度）と、それに対応する地球科学データと衛星データとを含む。図６の例では、地球科学データは、単位面積当たりの銅の含有量（ppm）を含み、衛星データは、特定波長の光の反射率（Asterバンドデータ Band 1、AsterバンドデータBand 14、Asterバンド逆数データ Band 1^-1）、標高値、及び傾斜値を含む。また、図７（ａ）～（ｃ）に示すように、衛星データは、広範な範囲において取得されている。 First, as shown in FIG. 6, in this embodiment, the database 30 registers a plurality of sample data shown in FIG. As shown in FIG. 6, the sample data includes a point (latitude and longitude), and corresponding earth science data and satellite data. In the example of FIG. 6, the geoscience data includes the copper content (ppm) per unit area, and the satellite data includes the reflectance of light of a specific wavelength (Aster band data Band 1, Aster band data Band 14, Aster Including band reciprocal data (Band (1 ^ -1)), elevation value, and slope value. In addition, as shown in FIGS. 7A to 7C, satellite data is acquired in a wide range.
　また、図８に示すように、データベース３０に登録されている複数のサンプルデータの中には、地球科学データ（銅の含有量）が欠損しているものがある。つまり、図９に示すように、特定領域上であっても、地球科学データが取得されていない地点が存在している。言い換えると、白点の地点では地球科学データとして銅の含有量が取得されているが、点が無い地点では銅の含有量は取得されていない。 Also, as shown in FIG. 8, some of the sample data registered in the database 30 may lack earth science data (copper content). That is, as shown in FIG. 9, there is a point where the earth science data is not acquired even on the specific region. In other words, the copper content is acquired as geoscience data at the point of white spot, but the copper content is not acquired at the point without point.
　このため、学習モデル生成部１２は、地球科学データの値が欠損している特定領域のサンプルデータにおいて、地球科学データの値を補充する。この結果、地球科学データが取得されている特定領域は、図１０に示す通りとなる。 For this reason, the learning model generation unit 12 supplements the value of the earth science data in the sample data of the specific area where the value of the earth science data is missing. As a result, the specific area where the earth science data is acquired is as shown in FIG.
　次に、学習モデル生成部１２は、補正後の複数のサンプルデータ（図８及び図９参照）を教師データとして、機械学習を実行する。これにより、本実施例では、単位面積当たりの銅の含有量（ppm）と、特定波長の光の反射率、標高値、及び傾斜値と、の相関関係を特定する学習モデル１６が生成される。 Next, the learning model generation unit 12 performs machine learning using a plurality of corrected sample data (see FIGS. 8 and 9) as teacher data. Thereby, in the present embodiment, a learning model 16 that specifies the correlation between the copper content (ppm) per unit area and the reflectance, altitude value, and slope value of light of a specific wavelength is generated. .
　続いて、データ推定部１３は、学習モデル生成部１２によって生成された学習モデル１６を用いて、推定領域における地球科学データを推定する。本実施の形態では、図１０における白点が全く存在していない領域が推定領域に設定されている。 Subsequently, the data estimation unit 13 estimates the earth science data in the estimation region using the learning model 16 generated by the learning model generation unit 12. In the present embodiment, an area where no white point exists in FIG. 10 is set as an estimated area.
　結果は、図１１に示す通りである。図１１に示すように、表示部１４は、表示装置２０の画面上において、特定領域における地球科学データと、推定領域における地球科学データとを、地図データ上に重ねて表示する。このように、本実施例によれば、銅の含有量が取得されていない領域の銅の含有量を推定できるので、効率良く採掘計画を策定することができる。 The result is as shown in FIG. As shown in FIG. 11, the display unit 14 displays the earth science data in the specific area and the earth science data in the estimation area on the screen of the display device 20 so as to overlap the map data. Thus, according to the present Example, since the copper content of the area | region where the copper content is not acquired can be estimated, a mining plan can be formulated efficiently.
　上述した実施の形態の及び実施例の一部又は全部は、以下に記載する（付記１）～（付記１５）によって表現することができるが、以下の記載に限定されるものではない。　　　 Some or all of the above-described embodiments and examples can be expressed by the following (Appendix 1) to (Appendix 15), but is not limited to the following description. *
（付記１）
　特定領域の特性を示す地球科学データ、及び前記特定領域の特性を示す衛星データを取得する、データ取得部と、
　取得された前記地球科学データ及び前記衛星データを用いて、前記地球科学データが示す前記特定領域の特性と、前記衛星データが示す前記特定領域の特性と、の相関関係を学習して、学習モデルを生成する、学習モデル生成部と、
を備えていることを特徴とする地球科学データ解析装置。 (Appendix 1)
A data acquisition unit for acquiring geoscience data indicating the characteristics of the specific area and satellite data indicating the characteristics of the specific area;
Using the acquired earth science data and the satellite data, learning a correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data, and a learning model A learning model generation unit for generating
An earth science data analysis apparatus characterized by comprising:
（付記２）
　前記学習モデル生成部が、数理モデルを用いて、前記特定領域における前記地球科学データの欠損を補正し、補正後の前記地球科学データを用いて、前記相関関係を学習する、付記１に記載の地球科学データ解析装置。 (Appendix 2)
The supplementary note 1, wherein the learning model generation unit corrects the deficiency of the earth science data in the specific region using a mathematical model, and learns the correlation using the corrected earth science data. Earth science data analyzer.
（付記３）
　前記学習モデルを用いて、前記特定領域以外の領域における地球科学データを推定する、データ推定部を、更に備えている、
付記１または２に記載の地球科学データ解析装置。 (Appendix 3)
Using the learning model, further comprising a data estimation unit for estimating geoscience data in a region other than the specific region,
The earth science data analyzer according to appendix 1 or 2.
（付記４）
　前記特定領域における地球科学データと、推定された前記特定領域以外の領域における地球科学データとを、地図データ上に重ねて画面上に表示する、表示部を、更に備えている、
付記３に記載の地球科学データ解析装置。 (Appendix 4)
A display unit for displaying the geoscience data in the specific region and the geoscience data in a region other than the estimated specific region on the screen by superimposing on the map data;
The earth science data analysis apparatus according to appendix 3.
（付記５）
　前記地球科学データが、前記特定領域の特性として、前記特定領域における、特定の物質の存在を示すデータであり、
　前記衛星データが、前記特定領域の特性として、前記特定領域における、特定波長の光の反射率の分布を示すデータである、
付記１～４のいずれかに記載の地球科学データ解析装置。 (Appendix 5)
The geoscience data is data indicating the presence of a specific substance in the specific region as a characteristic of the specific region;
The satellite data is data indicating a reflectance distribution of light of a specific wavelength in the specific region as a characteristic of the specific region.
The earth science data analysis device according to any one of appendices 1 to 4.
（付記６）
（ａ）特定領域の特性を示す地球科学データ、及び前記特定領域の特性を示す衛星データを取得する、ステップと、
（ｂ）前記（ａ）のステップで取得された前記地球科学データ及び前記衛星データを用いて、前記地球科学データが示す前記特定領域の特性と、前記衛星データが示す前記特定領域の特性と、の相関関係を学習して、学習モデルを生成する、ステップと、
を有する、ことを特徴とする地球科学データ解析方法。 (Appendix 6)
(A) obtaining geoscience data indicating characteristics of a specific area, and satellite data indicating characteristics of the specific area;
(B) Using the geoscience data and the satellite data acquired in the step (a), the characteristics of the specific area indicated by the geoscience data, the characteristics of the specific area indicated by the satellite data, Learning the correlation of and generating a learning model,
An earth science data analysis method characterized by comprising:
（付記７）
　前記（ｂ）のステップにおいて、数理モデルを用いて、前記特定領域における前記地球科学データの欠損を補正し、補正後の前記地球科学データを用いて、前記相関関係を学習する、
付記６に記載の地球科学データ解析方法。 (Appendix 7)
In the step (b), using the mathematical model, the deficiency of the earth science data in the specific region is corrected, and the correlation is learned using the corrected earth science data.
The earth science data analysis method according to appendix 6.
（付記８）
（ｃ）前記学習モデルを用いて、前記特定領域以外の領域における地球科学データを推定する、ステップを、更に有する、
付記６または７に記載の地球科学データ解析方法。 (Appendix 8)
(C) using the learning model, further including the step of estimating geoscience data in a region other than the specific region,
The earth science data analysis method according to appendix 6 or 7.
（付記９）
（ｄ）前記特定領域における地球科学データと、推定された前記特定領域以外の領域における地球科学データとを、地図データ上に重ねて画面上に表示する、ステップを、更に有する、
付記８に記載の地球科学データ解析方法。 (Appendix 9)
(D) The method further includes the step of displaying the geoscience data in the specific area and the geoscience data in the area other than the estimated specific area on the screen so as to overlap the map data.
The earth science data analysis method according to attachment 8.
（付記１０）
　前記地球科学データが、前記特定領域の特性として、前記特定領域における、物質の存在を示すデータであり、
　前記衛星データが、前記特定領域の特性として、前記特定領域における、特定波長の光の反射率の分布を示すデータである、
付記６～９のいずれかに記載の地球科学データ解析方法。 (Appendix 10)
The geoscience data is data indicating the presence of a substance in the specific region as a characteristic of the specific region,
The satellite data is data indicating a reflectance distribution of light of a specific wavelength in the specific region as a characteristic of the specific region.
The method for analyzing earth science data according to any one of appendices 6 to 9.
（付記１１）
　コンピュータに、
（ａ）特定領域の特性を示す地球科学データ、及び前記特定領域の特性を示す衛星データを取得する、ステップと、
（ｂ）前記（ａ）のステップで取得された前記地球科学データ及び前記衛星データを用いて、前記地球科学データが示す前記特定領域の特性と前記衛星データが示す前記特定領域の特性との相関関係を学習して、学習モデルを生成する、ステップと、
を実行させる命令を含む、プログラムを記録したコンピュータ読み取り可能な記録媒体。 (Appendix 11)
On the computer,
(A) obtaining geoscience data indicating characteristics of a specific area, and satellite data indicating characteristics of the specific area;
(B) Correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data using the earth science data and the satellite data acquired in the step (a). Learning relationships and generating learning models, steps,
The computer-readable recording medium which recorded the program containing the instruction | indication which performs this.
（付記１２）
　前記（ｂ）のステップにおいて、数理モデルを用いて、前記特定領域における前記地球科学データの欠損を補正し、補正後の前記地球科学データを用いて、前記相関関係を学習する、
付記１１に記載のコンピュータ読み取り可能な記録媒体。 (Appendix 12)
In the step (b), using the mathematical model, the deficiency of the earth science data in the specific region is corrected, and the correlation is learned using the corrected earth science data.
The computer-readable recording medium according to appendix 11.
（付記１３）
前記プログラムが、前記コンピュータに、
（ｃ）前記学習モデルを用いて、前記特定領域以外の領域における地球科学データを推定する、ステップを実行させる命令を、更に含む、
付記１１または１２に記載のコンピュータ読み取り可能な記録媒体。 (Appendix 13)
The program is stored in the computer.
(C) further including an instruction for executing a step of estimating geoscience data in a region other than the specific region using the learning model;
The computer-readable recording medium according to appendix 11 or 12.
（付記１４）
前記プログラムが、前記コンピュータに、
（ｄ）前記特定領域における地球科学データと、推定された前記特定領域以外の領域における地球科学データとを、地図データ上に重ねて画面上に表示する、ステップを実行させる命令を、更に含む、
付記１３に記載のコンピュータ読み取り可能な記録媒体。 (Appendix 14)
The program is stored in the computer.
(D) further including an instruction to execute a step of displaying the earth science data in the specific area and the earth science data in the area other than the estimated specific area on the screen by superimposing them on the map data.
The computer-readable recording medium according to attachment 13.
（付記１５）
　前記地球科学データが、前記特定領域の特性として、前記特定領域における、物質の存在を示すデータであり、
　前記衛星データが、前記特定領域の特性として、前記特定領域における、特定波長の光の反射率の分布を示すデータである、
付記１１～１４のいずれかに記載のコンピュータ読み取り可能な記録媒体。 (Appendix 15)
The geoscience data is data indicating the presence of a substance in the specific region as a characteristic of the specific region,
The satellite data is data indicating a reflectance distribution of light of a specific wavelength in the specific region as a characteristic of the specific region.
15. A computer-readable recording medium according to any one of appendices 11 to 14.
　以上、実施の形態及び実施例を参照して本願発明を説明したが、本願発明は上記実施の形態に限定されるものではない。本願発明の構成や詳細には、本願発明のスコープ内で当業者が理解し得る様々な変更をすることができる。 Although the present invention has been described with reference to the embodiments and examples, the present invention is not limited to the above embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.
　この出願は、２０１７年５月２５日に出願された日本出願特願２０１７－１０３８５３を基礎とする優先権を主張し、その開示の全てをここに取り込む。 This application claims priority based on Japanese Patent Application No. 2017-103853 filed on May 25, 2017, the entire disclosure of which is incorporated herein.
　以上のように、本発明によれば、ある領域で取得された地球科学データを用いて、他の領域の地球科学データを推定することができる。本発明は、例えば、鉱物資源の採掘、地盤調査、植生調査等に有用である。 As described above, according to the present invention, it is possible to estimate the earth science data of another area using the earth science data acquired in a certain area. The present invention is useful for, for example, mining of mineral resources, ground survey, vegetation survey, and the like.
　１０　地球科学データ解析装置
　１１　データ取得部
　１２　学習モデル生成部
　１３　データ推定部
　１４　表示部
　１５　記憶部
　１６　学習モデル
　２０　表示装置
　３０　データベース
　１１０　コンピュータ
　１１１　ＣＰＵ
　１１２　メインメモリ
　１１３　記憶装置
　１１４　入力インターフェイス
　１１５　表示コントローラ
　１１６　データリーダ／ライタ
　１１７　通信インターフェイス
　１１８　入力機器
　１１９　ディスプレイ装置
　１２０　記録媒体
　１２１　バス DESCRIPTION OF SYMBOLS 10 Geoscience data analyzer 11 Data acquisition part 12 Learning model production | generation part 13 Data estimation part 14 Display part 15 Memory | storage part 16 Learning model 20 Display apparatus 30 Database 110 Computer 111 CPU
112 Main Memory 113 Storage Device 114 Input Interface 115 Display Controller 116 Data Reader / Writer 117 Communication Interface 118 Input Device 119 Display Device 120 Recording Medium 121 Bus

Claims (15)

  1. 　特定領域の特性を示す地球科学データ、及び前記特定領域の特性を示す衛星データを取得する、データ取得部と、
    　取得された前記地球科学データ及び前記衛星データを用いて、前記地球科学データが示す前記特定領域の特性と、前記衛星データが示す前記特定領域の特性と、の相関関係を学習して、学習モデルを生成する、学習モデル生成部と、
    を備えていることを特徴とする地球科学データ解析装置。     A data acquisition unit for acquiring geoscience data indicating the characteristics of the specific area and satellite data indicating the characteristics of the specific area;
    Using the acquired earth science data and the satellite data, learning a correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data, and a learning model A learning model generation unit for generating
    An earth science data analysis apparatus characterized by comprising:
  2. 　前記学習モデル生成部が、数理モデルを用いて、前記特定領域における前記地球科学データの欠損を補正し、補正後の前記地球科学データを用いて、前記相関関係を学習する、請求項１に記載の地球科学データ解析装置。 The said learning model production | generation part correct | amends the defect | deletion of the said earth science data in the said specific area | region using a mathematical model, and learns the said correlation using the said earth science data after correction | amendment. Earth science data analysis device.
  3. 　前記学習モデルを用いて、前記特定領域以外の領域における地球科学データを推定する、データ推定部を、更に備えている、
    請求項１または２に記載の地球科学データ解析装置。     Using the learning model, further comprising a data estimation unit for estimating geoscience data in a region other than the specific region,
    The earth science data analysis apparatus according to claim 1 or 2.
  4. 　前記特定領域における地球科学データと、推定された前記特定領域以外の領域における地球科学データとを、地図データ上に重ねて画面上に表示する、表示部を、更に備えている、
    請求項３に記載の地球科学データ解析装置。     A display unit for displaying the geoscience data in the specific region and the geoscience data in a region other than the estimated specific region on the screen by superimposing on the map data;
    The earth science data analysis apparatus according to claim 3.
  5. 　前記地球科学データが、前記特定領域の特性として、前記特定領域における、特定の物質の存在を示すデータであり、
    　前記衛星データが、前記特定領域の特性として、前記特定領域における、特定波長の光の反射率の分布を示すデータである、
    請求項１～４のいずれかに記載の地球科学データ解析装置。     The geoscience data is data indicating the presence of a specific substance in the specific region as a characteristic of the specific region;
    The satellite data is data indicating a reflectance distribution of light of a specific wavelength in the specific region as a characteristic of the specific region.
    The earth science data analysis apparatus according to any one of claims 1 to 4.
  6. （ａ）特定領域の特性を示す地球科学データ、及び前記特定領域の特性を示す衛星データを取得する、ステップと、
    （ｂ）前記（ａ）のステップで取得された前記地球科学データ及び前記衛星データを用いて、前記地球科学データが示す前記特定領域の特性と、前記衛星データが示す前記特定領域の特性と、の相関関係を学習して、学習モデルを生成する、ステップと、
    を有する、ことを特徴とする地球科学データ解析方法。     (A) obtaining geoscience data indicating characteristics of a specific area, and satellite data indicating characteristics of the specific area;
    (B) Using the geoscience data and the satellite data acquired in the step (a), the characteristics of the specific area indicated by the geoscience data, the characteristics of the specific area indicated by the satellite data, Learning the correlation of and generating a learning model,
    An earth science data analysis method characterized by comprising:
  7. 　前記（ｂ）のステップにおいて、数理モデルを用いて、前記特定領域における前記地球科学データの欠損を補正し、補正後の前記地球科学データを用いて、前記相関関係を学習する、
    請求項６に記載の地球科学データ解析方法。     In the step (b), using the mathematical model, the deficiency of the earth science data in the specific region is corrected, and the correlation is learned using the corrected earth science data.
    The geoscience data analysis method according to claim 6.
  8. （ｃ）前記学習モデルを用いて、前記特定領域以外の領域における地球科学データを推定する、ステップを、更に有する、
    請求項６または７に記載の地球科学データ解析方法。     (C) using the learning model, further including the step of estimating geoscience data in a region other than the specific region,
    The earth science data analysis method according to claim 6 or 7.
  9. （ｄ）前記特定領域における地球科学データと、推定された前記特定領域以外の領域における地球科学データとを、地図データ上に重ねて画面上に表示する、ステップを、更に有する、
    請求項８に記載の地球科学データ解析方法。     (D) The method further includes the step of displaying the geoscience data in the specific area and the geoscience data in the area other than the estimated specific area on the screen so as to overlap the map data.
    The earth science data analysis method according to claim 8.
  10. 　前記地球科学データが、前記特定領域の特性として、前記特定領域における、物質の存在を示すデータであり、
    　前記衛星データが、前記特定領域の特性として、前記特定領域における、特定波長の光の反射率の分布を示すデータである、
    請求項６～９のいずれかに記載の地球科学データ解析方法。     The geoscience data is data indicating the presence of a substance in the specific region as a characteristic of the specific region,
    The satellite data is data indicating a reflectance distribution of light of a specific wavelength in the specific region as a characteristic of the specific region.
    The geoscience data analysis method according to any one of claims 6 to 9.
  11. コンピュータに、
    （ａ）特定領域の特性を示す地球科学データ、及び前記特定領域の特性を示す衛星データを取得する、ステップと、
    （ｂ）前記（ａ）のステップで取得された前記地球科学データ及び前記衛星データを用いて、前記地球科学データが示す前記特定領域の特性と前記衛星データが示す前記特定領域の特性との相関関係を学習して、学習モデルを生成する、ステップと、
    を実行させる命令を含む、プログラムを記録したコンピュータ読み取り可能な記録媒体。     On the computer,
    (A) obtaining geoscience data indicating characteristics of a specific area, and satellite data indicating characteristics of the specific area;
    (B) Correlation between the characteristics of the specific area indicated by the earth science data and the characteristics of the specific area indicated by the satellite data using the earth science data and the satellite data acquired in the step (a). Learning relationships and generating learning models, steps,
    The computer-readable recording medium which recorded the program containing the instruction | indication which performs this.
  12. 　前記（ｂ）のステップにおいて、数理モデルを用いて、前記特定領域における前記地球科学データの欠損を補正し、補正後の前記地球科学データを用いて、前記相関関係を学習する、
    請求項１１に記載のコンピュータ読み取り可能な記録媒体。     In the step (b), using the mathematical model, the deficiency of the earth science data in the specific region is corrected, and the correlation is learned using the corrected earth science data.
    The computer-readable recording medium according to claim 11.
  13. 前記プログラムが、前記コンピュータに、
    （ｃ）前記学習モデルを用いて、前記特定領域以外の領域における地球科学データを推定する、ステップを実行させる命令を、更に含む、
    請求項１１または１２に記載のコンピュータ読み取り可能な記録媒体。     The program is stored in the computer.
    (C) further including an instruction for executing a step of estimating geoscience data in a region other than the specific region using the learning model;
    The computer-readable recording medium according to claim 11 or 12.
  14. 前記プログラムが、前記コンピュータに、
    （ｄ）前記特定領域における地球科学データと、推定された前記特定領域以外の領域における地球科学データとを、地図データ上に重ねて画面上に表示する、ステップを実行させる命令を、更に含む、
    請求項１３に記載のコンピュータ読み取り可能な記録媒体。     The program is stored in the computer.
    (D) further including an instruction to execute a step of displaying the earth science data in the specific area and the earth science data in the area other than the estimated specific area on the screen by superimposing them on the map data.
    The computer-readable recording medium according to claim 13.
  15. 　前記地球科学データが、前記特定領域の特性として、前記特定領域における、物質の存在を示すデータであり、
    　前記衛星データが、前記特定領域の特性として、前記特定領域における、特定波長の光の反射率の分布を示すデータである、
    請求項１１～１４のいずれかに記載のコンピュータ読み取り可能な記録媒体。     The geoscience data is data indicating the presence of a substance in the specific region as a characteristic of the specific region,
    The satellite data is data indicating a reflectance distribution of light of a specific wavelength in the specific region as a characteristic of the specific region.
    The computer-readable recording medium according to any one of claims 11 to 14.
PCT/JP2018/019343 2017-05-25 2018-05-18 Earth science data analyzing device, earth science data analyzing method, and computer-readable recording medium WO2018216623A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2017-103853 2017-05-25
JP2017103853 2017-05-25

Publications (1)

Publication Number Publication Date
WO2018216623A1 true WO2018216623A1 (en) 2018-11-29

Family

ID=64395641

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2018/019343 WO2018216623A1 (en) 2017-05-25 2018-05-18 Earth science data analyzing device, earth science data analyzing method, and computer-readable recording medium

Country Status (1)

Country Link
WO (1) WO2018216623A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPWO2020255413A1 (en) * 2019-06-21 2020-12-24
WO2021246027A1 (en) * 2020-05-30 2021-12-09 ソニーグループ株式会社 Evaluation device, evaluation method, and navigation control device

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06138251A (en) * 1992-10-28 1994-05-20 N T T Data Tsushin Kk Cover material identification device
JP2004536367A (en) * 2001-01-23 2004-12-02 バイオウルフ テクノロジーズ エルエルスィー Computer-based image analysis
JP2005189100A (en) * 2003-12-25 2005-07-14 National Institute Of Information & Communication Technology Vegetative cover classification observation method by sar
US20140351183A1 (en) * 2012-06-11 2014-11-27 Landmark Graphics Corporation Methods and related systems of building models and predicting operational outcomes of a drilling operation

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06138251A (en) * 1992-10-28 1994-05-20 N T T Data Tsushin Kk Cover material identification device
JP2004536367A (en) * 2001-01-23 2004-12-02 バイオウルフ テクノロジーズ エルエルスィー Computer-based image analysis
JP2005189100A (en) * 2003-12-25 2005-07-14 National Institute Of Information & Communication Technology Vegetative cover classification observation method by sar
US20140351183A1 (en) * 2012-06-11 2014-11-27 Landmark Graphics Corporation Methods and related systems of building models and predicting operational outcomes of a drilling operation

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPWO2020255413A1 (en) * 2019-06-21 2020-12-24
WO2020255413A1 (en) * 2019-06-21 2020-12-24 日本電気株式会社 Data analysis device, data analysis method, and computer-readable recording medium
JP7235111B2 (en) 2019-06-21 2023-03-08 日本電気株式会社 Data analysis device, data analysis method, and program
WO2021246027A1 (en) * 2020-05-30 2021-12-09 ソニーグループ株式会社 Evaluation device, evaluation method, and navigation control device

Similar Documents

Publication Publication Date Title
Wang et al. High resolution mapping of soil organic carbon stocks using remote sensing variables in the semi-arid rangelands of eastern Australia
Zhang et al. Estimating daily air temperatures over the Tibetan Plateau by dynamically integrating MODIS LST data
Poggio et al. Bayesian spatial modelling of soil properties and their uncertainty: The example of soil organic matter in Scotland using R-INLA
Xu et al. In situ SST quality monitor (iQuam)
Hengl et al. SoilGrids250m: Global gridded soil information based on machine learning
Kantakumar et al. Multi-temporal land use classification using hybrid approach
Peres et al. Land surface temperature and emissivity estimation based on the two-temperature method: Sensitivity analysis using simulated MSG/SEVIRI data
Zhang et al. A system for operational aerosol optical depth data assimilation over global oceans
Hansen et al. Comparing annual MODIS and PRODES forest cover change data for advancing monitoring of Brazilian forest cover
Guevara et al. Downscaling satellite soil moisture using geomorphometry and machine learning
Hazaymeh et al. Fusion of MODIS and Landsat-8 surface temperature images: A new approach
Malone et al. A general method for downscaling earth resource information
JP7235111B2 (en) Data analysis device, data analysis method, and program
Mallick Land characterization analysis of surface temperature of semi-arid mountainous city Abha, Saudi Arabia using remote sensing and GIS
Kofler et al. An inventory-driven rock glacier status model (intact vs. relict) for South Tyrol, Eastern Italian Alps
Benitez et al. Evaluation of geostatistical techniques to estimate the spatial distribution of aboveground biomass in the Amazon rainforest using high-resolution remote sensing data
Chang et al. A modified method for inferring upper troposphere cloud top height using the GOES 12 imager 10.7 and 13.3 μm data
WO2018216623A1 (en) Earth science data analyzing device, earth science data analyzing method, and computer-readable recording medium
Kumar et al. A machine learning approach to estimating the error in satellite sea surface temperature retrievals
Lee et al. Machine learning for targeted assimilation of satellite data
Karimi et al. Land use analysis on land surface temperature in urban areas using a geographically weighted regression and Landsat 8 imagery, a case study: Tehran, Iran
Park et al. Delineation of high resolution climate regions over the Korean Peninsula using machine learning approaches
Mackie et al. Generalized Bayesian cloud detection for satellite imagery. Part 2: Technique and validation for daytime imagery
Norman et al. Fusion of multispectral imagery and LiDAR data for roofing materials and roofing surface conditions assessment
Shao et al. Assessment of Four Near‐Surface Soil Freeze/Thaw Detection Algorithms Based on Calibrated Passive Microwave Remote Sensing Data Over China

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18806492

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18806492

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP